Evidence · FSL

Head-motion correction

Head-motion correction estimates, for every volume of a BOLD run, the rigid transform that aligns it to a reference volume, in mcflirt’s coarse-to-fine schedule, and optionally resamples the run; fMRIPrep runs FSL’s mcflirt with its BOLD reference and keeps the matrices. The replica of mcflirt (larmorx-nc) gives FSL 6.0.7.23’s files, messages and exit status byte for byte in all 205 of its compared cases, out of 237, fMRIPrep’s calls among them, and FSL 6.0.7.17’s in all 173 of its full comparison’s, out of 194. The clean room (larmorx, lx.mri.hmc) is within its record’s thresholds in 145 of its 149 compared cases, out of 170; the other 4 are expected divergences, each with its reason. On fMRIPrep’s own run its motion estimates stay within mcflirt’s own variability, and it runs 1.2 to 2.3 times as fast as mcflirt on one thread, 4.9 to 9.4 times on 12.

Ports
  • replica larmorx-nc FSL Licence: non-commercial use only: translated from FSL 6.0.7.17's source, and following FSL 6.0.7.23's since the version update of 2026-10-29; it runs as a separate program
  • clean room larmorx Apache-2.0: crate larmorx-mri, written from specs/mcflirt.md and runs of mcflirt, without FSL's source
Status
replica validated (marked before the statuses' definition); clean room experimental
Records
the replica's: standard comparison (FSL 6.0.7.23) and full comparison (FSL 6.0.7.17) · the clean room's: standard comparison (FSL 6.0.7.23) · benchmark mri-hmc
Reference
mcflirt · API page mri-hmc

What is compared

Head-motion correction estimates, for every volume of a BOLD run, the rigid transform that aligns it to a reference volume, in mcflirt’s coarse-to-fine schedule, and optionally resamples the run. fMRIPrep runs mcflirt with its BOLD reference and keeps the matrices.

The two ports run beside FSL’s mcflirt on the same files with the same arguments: FSL 6.0.7.23’s since 2026-10-29, FSL 6.0.7.17’s before. fMRIPrep’s own matrices are the files it wrote in its run.

bash
larmorx mri hmc -in bold.nii.gz -out bold_mcf.nii.gz -reffile hmc_boldref.nii.gz -mats -plots
larmorx fsl mcflirt -in bold.nii.gz -out bold_mcf.nii.gz -reffile hmc_boldref.nii.gz -mats -plots

Both commands run lx.mri.hmc; the second is mcflirt’s own name for it.

python
import larmorx as lx

hmc = lx.mri.hmc("sub-01_task-rest_bold.nii.gz", reference="hmc_boldref.nii.gz",
                 resample=False, n_threads=8)
# hmc.matrices: the (N, 4, 4) FSL matrices, as mcflirt's MAT_* files
# hmc.implementation: which port ran

The calls above leave implementation at "auto". The API page describes every option. How to read what follows: how to read the evidence.

Its two ports

FSL comes under the FSL Licence, which allows non-commercial use only. So LARMORX has two ports of its FSL programs. The replica is translated from mcflirt’s source and lives in its own package, larmorx-nc. The clean room is in larmorx itself, written from mcflirt’s behavioural specification (specs/mcflirt.md) and runs of mcflirt as a black box.

implementation="auto" | "replica" | "clean-room" chooses between them ("original" in 0.1.0rc1). On the command line, larmorx --implementation clean-room fsl mcflirt ... does the same from the next release; in 0.1.0rc1 it is the environment variable LARMORX_IMPLEMENTATION. auto takes the replica when larmorx-nc is installed, else the clean room. larmorx-nc is installed by name, pip install larmorx-nc, or with pip install "larmorx[all]", so that its non-commercial licence is accepted knowingly (the packages).

The C library’s last digits (since 2026-11-04). With -cost mutualinfo and -cost normmi, mcflirt’s cost takes logarithms (logf) from glibc, the C library of Linux, whose results are now and then one unit in the last place from the exact value rounded once. LARMORX computes them correctly rounded, the same on every platform.

  • Since 2026-11-04 the clean room’s reproduce computes them exactly as glibc 2.35 does, with LARMORX’s versions of glibc’s functions (larmorx_core::math), and its fix correctly rounded.
  • GLIBC-07 in the register is named nowhere there, as decided on 2026-11-04: whether a fit moved with such last bits cannot be told without running it twice.
  • fMRIPrep’s calls use the normalised correlation, which calls neither, so the example below is unchanged.
  • Three cases of the record moved by thousandths of a millimetre, their verdicts unchanged: -cost mutualinfo toward mcflirt (0.0048 → 0.0047 mm, the 95th percentile of the matrices’ RMS deviation), and -cost normmi the other way (0.0108 → 0.0126 mm of its 0.1 mm), the size of the fit’s own sensitivity to those last bits (its two modes part by 0.0109 mm). So did random testing’s normmi case.

The weighted correlation’s count (since 2026-11-07). fMRIPrep’s cost, the normalised correlation with mcflirt’s edge weights, divides by the count of the points it weighs. The specification’s answer of 2026-11-06 (specs/mcflirt.md §6.4) says mcflirt adds each point’s weight to a running total as the point is visited. The clean room had added each row’s total at the row’s end, which rounds otherwise where the weights are fractional (near the volume’s bounds).

  • With the count per point, the clean room’s evaluated matrices are mcflirt’s for longer on every recorded run traced (findings, “The count per point”), and the example below moves within mcflirt’s band.
  • Its record, made again with it (the record), holds 145 of its compared cases within the thresholds.
  • Two synthetic runs end just beyond the thresholds made for real BOLD runs, each within mcflirt’s own spread on that run, measured by moving its inputs in their last bits (expected divergences, each with its numbers in the case explorer): syn/orientation/ras-unflipped/reffile, and the ramp syn/content/ramp/reffile, which only one translation can align.
  • The leastsquares cost is FLIRT’s since the same day (three single-precision levels, the specification’s appendix A).

The verdicts at a glance

The replica is a translation of mcflirt’s own source, so the question for it is whether it gives mcflirt’s bytes: every output file, the exit status and both message streams, compared with mcflirt as built (with the C library’s maths it calls on Linux).

Since the version update of 2026-10-29 the replica follows FSL 6.0.7.23, and its standard comparison runs against it (the chart below; the record): 137 runs recorded again with 6.0.7.23’s mcflirt, fMRIPrep’s calls and the edge cases, with new ones for 6.0.7.23’s zstd and bzip2 formats, for random testing’s finds and for the 33 failure cases of specs/mcflirt.md §14.1. 205 of its 205 compared cases are bit-identical, and in 32 both fail alike.

Agreement, replica against the original: FSL mcflirt 6.0.7.23 (standard comparison, 237 cases of the record), by category and verdict
real runs30
options, on ds00000525
synthetic: the command line11
synthetic: image content6
synthetic: data types10
synthetic: geometry10
synthetic: motion36
synthetic: orientation6
failure cases (specs/mcflirt.md §14.1)33
the replica's own: edge cases43
the replica's own: fMRIPrep's calls21
random testing6

Select a segment to list its cases below.

Each case also says what changed between FSL’s two versions, and why:

  • 127 of the 137 recorded runs give the same bytes.
  • -sinc_final and -nn_final now resample trilinearly in mcflirt itself (FSL-59, which the replica reproduces and corrects by default).
  • In five cases the last bits of a matrix change, because 6.0.7.23 inverts its 4 × 4 matrices through LAPACK (Armadillo 14.2.1 on OpenBLAS 0.3.33). The replica follows that arithmetic since the version update’s third part (2026-10-29), checked bit for bit against FSL’s own library on 30.2 million seeded matrices: those five, expected divergences of at most 10⁻¹² until then, are bit-identical now.
  • OpenBLAS chooses its kernels by the CPU, and on CPUs with AVX-512 they fuse multiplications and additions, so FSL’s own last bits depend on the machine (FSL-64). The 137 runs were recorded again on the kernels of CPUs without AVX-512, the order the replica follows on every platform. One of them changed: the relative rms of identical volumes, 8.07 × 10⁻¹⁵ against 7.98 × 10⁻¹⁵ mm.

The replica’s full comparison, against FSL 6.0.7.17 (the chart below; the record), runs the 137 recorded mcflirt runs, fMRIPrep’s own mcflirt calls on the twelve fMRIPrep runs that have BOLD runs, and edge cases made for it. fMRIPrep’s 22 calls include the data panel’s: a child’s extreme motion, 1.5 T, multiband-8 with 600 volumes, multi-echo, a stroke, a glioma; fMRIPrep’s 4,517 matrices are identical too. The edge cases are the failure cases, every output type, -gdt, -verbose 20, histogram costs with few bins and moving phantoms. 173 of its 173 compared cases are bit-identical, and in 21 both fail alike. The four cases only the full comparison runs (fMRIPrep’s 280-volume multiband run of ds005454 and three -sinc_final and -stages 4 runs) wait for the final validation phase against 6.0.7.23.

Agreement, replica against the original: FSL mcflirt 6.0.7.17 (full comparison, 194 cases of the record), by category and verdict
real runs30
options, on ds00000528
synthetic: the command line11
synthetic: image content6
synthetic: data types10
synthetic: geometry10
synthetic: motion36
synthetic: orientation6
the replica's own: edge cases35
the replica's own: fMRIPrep's calls22

Select a segment to list its cases below.

Two of the data panel’s calls, and 58 of the first 59 differences random testing found, came from one constant of the C library’s atanf in the replica’s translation. aT[0] had been taken from fdlibm’s source comment (0x3eaaaaaa) instead of its decimal value (0x3eaaaaab, the value glibc holds), which moved one rotation angle by one unit in the last place (on ds002422’s rest run, in 1 of mcflirt’s 600 rotation decompositions). With it corrected, every one of them is bit-identical; the last difference was a missing FSL-08 notice, now given (findings).

The clean room is written without mcflirt’s source, so it is not expected to give mcflirt’s bits. The question for it is whether its estimates differ from mcflirt’s by no more than mcflirt’s own differ when its input changes imperceptibly. mcflirt is deterministic, so that variability is measured by perturbing its input: noise of half a quantisation step (three realisations) and the volumes in reverse order (the record).

  • A case is within the original’s variability when every compared estimate is inside that band: 33 of them.
  • It is open when it exceeds the band, or when the band is not measured for that case (the options and synthetic series), even though it meets the record’s thresholds: 112 (below).
  • 4 are expected divergences: the two synthetic runs above, and two of random testing’s phantoms with large motion, whose estimates end in other minima than mcflirt’s, aligning the series as well or better.
  • In 21 both fail alike.
Agreement, clean room against the original: FSL mcflirt 6.0.7.23 (standard comparison, 170 cases of the record), by category and verdict
real runs30
options, on ds00000525
synthetic: the command line11
synthetic: image content6
synthetic: data types10
synthetic: geometry10
synthetic: motion36
synthetic: orientation6
failure cases (specs/mcflirt.md §14.1)33
random testing3

Select a segment to list its cases below.

The clean room follows FSL 6.0.7.23 too since 2026-10-29, and its record runs against 6.0.7.23’s recorded runs:

  • -sinc_final and -nn_final as asked by default, and trilinear in reproduce (FSL-59);
  • every output type, the zstd and bzip2 ones included, files read by their content and names completed as FSL completes them, and FSL-60’s renamed names in reproduce;
  • -cost mutualinfo and normmi computed exactly as the specification’s appendix B describes them. With -fudge -stages 1 they give mcflirt’s evaluated matrices, evaluation by evaluation, on 22 of 30 random phantoms. On the 169 phantoms random testing had found for these costs, the corrected series are as well aligned as mcflirt’s (median ratio of the temporal standard deviations 1.000; normmi was 1.051 before).

A worked example: fMRIPrep’s mcflirt

fMRIPrep’s own call on OpenNeuro ds000005’s first run: its command, its BOLD and its HMC reference, given to FSL 6.0.7.17’s mcflirt and to LARMORX’s two ports, and to FSL 6.0.7.23’s mcflirt with both ports’ runs compared with it. FSL mcflirt 6.0.7.23 − 6.0.7.17 is mcflirt’s own step: the same bytes on this command. The chart under the viewer has the six parameters and the framewise displacement of every volume, and each port’s deviation from mcflirt’s matrix next to mcflirt’s own band: the replica’s is zero at every volume.

Worked example

fMRIPrep's head-motion estimation of a real BOLD run

What this shows: On fMRIPrep’s own call, the replica gives mcflirt’s bytes, the realigned series and all 240 matrices, against FSL 6.0.7.17 and 6.0.7.23 alike (6.0.7.23 gives 6.0.7.17’s values on this call), at 1 and 3 threads. The clean room’s matrices lie within mcflirt’s own variability: their 95th-percentile RMS deviation from mcflirt’s is 0.0193 mm, where mcflirt’s own under the perturbations reaches 0.0241 mm; it gives the same bytes at 1 thread and on all CPUs. In fix, their default, both correct FSL-01 and FSL-09, which moves the matrices by up to 0.0217 mm in the replica and 0.023 mm in the clean room.

The mcflirt node fMRIPrep 25.2.6 ran on OpenNeuro ds000005 sub-01’s first BOLD run (240 volumes, its own HMC reference): its command given to FSL 6.0.7.17’s mcflirt, to the replica (larmorx-nc) and to the clean room, and to FSL 6.0.7.23’s mcflirt with both ports’ runs compared with it (the replica’s reference since the version update of 2026-10-29, and the clean room’s). fMRIPrep keeps only the matrices; the realigned series are shown at 6 of the 240 volumes (where the clean room deviates most, the largest motion, the ends).

Data
OpenNeuro ds000005 (Tom et al., 2007), snapshot 00001, sub-01, task mixedgamblestask run 1 (PDDL-1.0); shown inside fMRIPrep’s BOLD brain mask, dilated by one voxel
Node
fMRIPrep 25.2.6, bold_fit_wf/bold_hmc_wf/mcflirt
Command
mcflirt -in /data/sub-01/func/sub-01_task-mixedgamblestask_run-01_bold.nii.gz -out …/sub-01_task-mixedgamblestask_run-01_bold_mcf.nii.gz -reffile /out/sub-01/func/sub-01_task-mixedgamblestask_run-01_desc-hmc_boldref.nii.gz -mats
Programs
FSL 6.0.7.17 (mcflirt), run as a black box; the clean room (lx.mri.hmc, experimental) at commit 80cc208fa4a3; the replica (larmorx-nc 0.0.1, lx.mri.hmc(implementation="replica")) at commit d6ba67e5ccc3
Volumes
shown: file volumes 0, 19, 160, 166, 182, 239
FSL versions
FSL 6.0.7.17’s panels as first computed; FSL 6.0.7.23 (the replica’s reference since 2026-10-29) on the same arguments, with the replica’s run compared with it: on fMRIPrep’s command it gives 6.0.7.17’s values (the headers’ build string aside). FSL 6.0.7.23 (mcflirt) and larmorx-nc 0.0.1 at commit caefd642193a. The case explorer below shows the replica’s full comparison’s record (against FSL 6.0.7.17) and its standard comparison’s record (against FSL 6.0.7.23) side by side; the clean room’s record runs against 6.0.7.23’s recorded runs since its alignment (2026-10-29).
Inputds000005 sub-01 BOLD, run 1int16, 240 volumes; 6 shown
OriginalFSL mcflirt 6.0.7.17fMRIPrep's command
Replicareplicalarmorx-nc, FSL Licence · reproduce, 3 threads · against FSL mcflirt 6.0.7.17
Clean roomclean roomlarmorx, Apache-2.0 · reproduce, all CPUs · against FSL mcflirt 6.0.7.23, which it follows
Clean roomclean roomlarmorx, Apache-2.0 · fix (the default)
the original against itself: the volumes in reverse orderreversed input − unperturbedscale ±10

Bit-identical835,584 of 835,584 voxels equal

the replica against the originalreplica − FSL mcflirt 6.0.7.17
the clean room against the originalclean room − FSL mcflirt 6.0.7.17scale ±10
the clean room against itself: fix and reproduceclean room: fix − reproducescale ±10

Axial slice 16 of 34, volume 5 of 6; file volume 182, where the clean room's matrix deviates most from mcflirt's; the differences on one symmetric scale. The interactive viewer needs JavaScript; the metrics below are complete without it.

The six parameters and the framewise displacement over the 240 volumes, both ports against the original: FSL mcflirt 6.0.7.17
  • FSL mcflirt 6.0.7.17
  • replica, reproduce (mcflirt's bytes)
  • clean room, reproduce
  • the original's own variability (the worst of its four perturbations)
Rotation x (°)0.0425−0.0903Rotation y (°)0.0116−0.0924Rotation z (°)0.0119−0.05Translation x (mm)0.0184−0.0605Translation y (mm)0.113−0.0358Translation z (mm)0.239−0.162Framewise displacement (Power, mm)0.1860Deviation from mcflirt's matrix (RMS over an 80 mm sphere, mm)0.03430fix − reproduce, both ports (RMS deviation, mm)0.0230

Parameters as the record decomposes the matrices (about the centre of the reference’s field of view); FD as Power et al. (50 mm). Move the pointer over a chart to read every value at one volume.

The matrices compared, volume by volume, both ports against the original: FSL mcflirt 6.0.7.17 and 6.0.7.23 (RMS deviation over an 80 mm sphere; parameter differences about the reference’s centre; FD’s correlation)

Comparisonmedian (mm)95th pct (mm)largest (mm)median Δ translation (mm)median Δ rotation (°)FD r
FSL mcflirt 6.0.7.17: noise − unperturbed0.01140.0190.02780.001660.003210.8954
FSL mcflirt 6.0.7.17: noise2 − unperturbed0.01070.01940.03110.001910.002490.8863
FSL mcflirt 6.0.7.17: noise3 − unperturbed0.01080.02090.02730.001570.003210.8901
FSL mcflirt 6.0.7.17: reversed − unperturbed0.01480.02410.03430.00360.005670.8499
FSL mcflirt 6.0.7.17: second run − firstidentical as text8.31e−152.32e−142.86e−14001
FSL mcflirt 6.0.7.23 − 6.0.7.17the version step; 240 of 240 matrices identical as text8.31e−152.32e−142.86e−14001
fMRIPrep’s output − FSL mcflirt 6.0.7.17identical as text8.31e−152.32e−142.86e−14001
replica, reproduce − FSL mcflirt 6.0.7.17240 of 240 matrices identical as text8.31e−152.32e−142.86e−14001
replica, fix − FSL mcflirt 6.0.7.178.31e−152.32e−140.0217001
replica, fix − reproduceFSL-01, FSL-09 corrected8.31e−152.32e−140.0217001
replica, 1 thread − 3 threads8.31e−152.32e−142.86e−14001
replica, reproduce − FSL mcflirt 6.0.7.23240 of 240 matrices identical as text8.31e−152.32e−142.86e−14001
clean room, reproduce − FSL mcflirt 6.0.7.170 of 240 matrices identical as text0.01010.01930.02830.001820.00120.8919
clean room, fix − FSL mcflirt 6.0.7.170.01020.01980.02830.001830.00120.891
clean room, fix − reproduceFSL-01, FSL-09 corrected8.3e−152.33e−140.023000.9994
clean room, reproduce − FSL mcflirt 6.0.7.23its reference since 2026-10-29; 0 of 240 matrices identical as text0.01010.01930.02830.001820.00120.8919
clean room − replica, both fix0 of 240 matrices identical as text0.01020.01930.02830.001820.00120.8908

The clean room’s record’s thresholds (mri-hmc): 95th percentile ≤ 0.1 mm, median parameter differences ≤ 0.05 mm and 0.05°, FD r ≥ 0.99 where FD reaches 0.5 mm, each widened to mcflirt’s own variability where that is larger. The replica’s records’ threshold (fsl-mcflirt-replica-full and fsl-mcflirt-replica): every output file bit-identical, with mcflirt’s exit status and both message streams. The shaded rows are mcflirt’s own variability: the worst of them is the band of the chart above.

Every comparison, computed in float64 over every voxel of the outputs

Comparison (A − B), verdict and reasonmax |Δ|Values that differRMS of Δmax |Δ| ÷ range1 − NCC
The original against itself: the yardstick
second run − first runreferencemcflirt is deterministic: the same command gives the same bytes.identical: 0 of 835,584 values differ
reversed input − unperturbedreferencemcflirt’s own variability: the same volumes registered in another order end at other estimates (no change of the data).1066,890 of 835,584 (8.01 %)0.3690.008831.27e−6
FSL mcflirt 6.0.7.23 − 6.0.7.17referenceFSL 6.0.7.23 gives FSL 6.0.7.17’s values on fMRIPrep’s command: the version step changes nothing computed here. 240 of 240 matrices identical as text.identical: 0 of 835,584 values differ
The replica against the original
replica − FSL mcflirt 6.0.7.17Bit-identicalThe replica gives mcflirt’s bytes: the realigned series and 240 of 240 matrices identical as text (larmorx-nc 0.0.1 at the time, a translation of FSL 6.0.7.17’s source with glibc 2.35’s float functions). replica recordidentical: 0 of 835,584 values differ
replica − FSL mcflirt 6.0.7.23Bit-identicalThe same values as FSL 6.0.7.23, every one. The realigned series and 240 of 240 matrices identical as text (larmorx-nc 0.0.1, following FSL 6.0.7.23’s source). replica recordidentical: 0 of 835,584 values differ
The replica against itself: thread counts, fix and reproduce
replica, 1 thread − 3 threadsBit-identicalThe same bytes on any number of threads.identical: 0 of 835,584 values differ
replica: fix − reproduceDeliberate correctionThe same corrections of mcflirt’s defects as the clean room’s, by default with a warning: FSL-01, FSL-09; they move the matrices by up to 0.0217 mm (RMS deviation). reproduce gives mcflirt’s bytes. FSL-01 FSL-097152,591 of 835,584 (18.3 %)0.4480.006181.28e−6
The clean room against the original and the replica
clean room − FSL mcflirt 6.0.7.17Within the original's own variabilityThe matrices: 95th-percentile RMS deviation 0.0193 mm, against mcflirt’s own 0.0241 mm under the perturbations (the chart below); the images then differ by interpolating at slightly other positions. Written without mcflirt’s source, the clean room is not bit-identical by design.885,708 of 835,584 (10.3 %)0.390.007061.41e−6
clean room − FSL mcflirt 6.0.7.23Within the original's own variabilityFSL 6.0.7.23’s mcflirt is the clean room’s reference since 2026-10-29 (its record is against it). On fMRIPrep’s command the clean room’s matrices against it: 95th-percentile RMS deviation 0.0193 mm, against mcflirt’s own 0.0241 mm under the perturbations; 0 of 240 matrices identical as text. The clean room’s run is the one above: fMRIPrep’s command meets none of the changes of the alignment (FSL-59’s final interpolations, the zstd and bzip2 files, FSL-60’s names, the histogram costs). record885,708 of 835,584 (10.3 %)0.390.007061.41e−6
clean room − replica, both fixWithin the original's own variabilityBoth ports correct the same defects; their matrices differ as the clean room’s differ from mcflirt’s: 95th-percentile RMS deviation 0.0193 mm against mcflirt’s own 0.0241 mm (0 of 240 matrices identical as text).883,992 of 835,584 (10.1 %)0.3860.007051.38e−6
The clean room against itself: fix and reproduce, 1 thread and all CPUs
clean room, 1 thread − all CPUsBit-identicalThe same bytes on any number of threads.identical: 0 of 835,584 values differ
clean room: fix − reproduceDeliberate correctionCorrections of mcflirt’s deterministic defects, by default with a warning: FSL-01, FSL-09; they move the matrices by up to 0.023 mm (RMS deviation). reproduce gives mcflirt’s behaviour. FSL-01 FSL-093149,652 of 835,584 (17.9 %)0.4280.002651.1e−6

Δ is A − B over the 6 volumes shown (835,584 values), not only the brain; the matrices are compared over every volume in the chart and its table.

Provenance of this example
  • Data: OpenNeuro ds000005 (Tom et al., 2007), snapshot 00001, sub-01, PDDL-1.0
  • Inputs: BOLD (sha256 609ae0a21822318f…); HMC reference (sha256 5030b453bcc3bbd4…)
  • Runs: figures cache entry mri-hmc/fmriprep-ds000005-run-01, computed 2026-10-21T15:08:32+00:00; 3 of its 9 runs made again since, last on 2026-11-07T02:03:53+00:00, with larmorx 80cc208fa4a3; FSL 6.0.7.17 (mcflirt); every command, its environment, threads and exit status are in its RUN_INFO.json
  • Inputs: BOLD (sha256 609ae0a21822318f…); HMC reference (sha256 5030b453bcc3bbd4…)
  • Runs: figures cache entry mri-hmc/fmriprep-ds000005-run-01-replica, computed 2026-10-24T19:03:54+00:00; its runs made again since, last on 2026-10-25T15:20:38+00:00, with larmorx d6ba67e5ccc3; larmorx-nc 0.0.1; every command, its environment, threads and exit status are in its RUN_INFO.json
  • Inputs: BOLD (sha256 609ae0a21822318f…); HMC reference (sha256 5030b453bcc3bbd4…)
  • Runs: figures cache entry mri-hmc/fmriprep-ds000005-run-01-fsl-6.0.7.23, computed 2026-10-29T10:14:47+00:00 with larmorx caefd642193a; FSL 6.0.7.23 (mcflirt); larmorx-nc 0.0.1; every command, its environment, threads and exit status are in its RUN_INFO.json

Every number above is in mri-hmc.json, exported by python -m larmorx_validation evidence mri-hmc.

The clean room’s three runs of this call were made again on 2026-11-05 with the clean-room catch-up’s build, mcflirt’s runs and its band kept. Since then the clean room takes the centres of mass as mcflirt’s blocked sums (specs/mcflirt.md §6.1), prints mcflirt’s warning where a volume’s total is 0, and takes atan2 from LARMORX’s own maths (findings). They were made again on 2026-11-07 with the correlation’s count per point (above).

Every case

Each case of the three records, the clean room’s against FSL 6.0.7.23 and the replica’s against FSL 6.0.7.17 and 6.0.7.23 (a column each), with what it tests, its command line, each port’s verdict and the reason. The replica’s edge cases and fMRIPrep’s calls are cases of their own. Search, filter by category or verdict (or select a segment of a chart above), sort by a column; #case=<name> links to one case.

244 cases

CaseCategoryWhat it testsclean roomreplica, FSL 6.0.7.17replica, FSL 6.0.7.23Checks
real/ds000005/defaultreal runsmcflirt's defaults (middle volume as reference)
command linemcflirt -in sub-01_task-mixedgamblestask_run-01_bold.nii.gz -out out_mcf.nii.gz -mats -plots -rmsrel -rmsabs
Within the original's own variability

95th-percentile RMS deviation 0.0192 mm, median Δ 0.0012 mm / 0.0009°; 35/240 matrices identical as text; mcflirt’s own: 0.0219 mm, 0.00224 mm / 0.0035°

Bit-identical

Every file mcflirt writes (246/246), its exit status and both streams identical to mcflirt’s (FSL 6.0.7.17 as built), with the replica in reproduce.

Bit-identical

Every file mcflirt writes (246/246), its exit status and both streams identical to mcflirt’s (FSL 6.0.7.23 as built), with the replica in reproduce.

16/16 · 6/6 · 7/7
real/ds000005/fmriprepreal runsfMRIPrep's command (nipype MCFLIRT(save_mats=True), ref_file)
command linemcflirt -in sub-01_task-mixedgamblestask_run-01_bold.nii.gz -out out_mcf.nii.gz -reffile sub-01_task-mixedgamblestask_run-1_boldref.nii.gz -mats
Within the original's own variability

95th-percentile RMS deviation 0.0257 mm, median Δ 0.0019 mm / 0.0023°; 2/240 matrices identical as text; mcflirt’s own: 0.101 mm, 0.00507 mm / 0.00679°

Bit-identical

Every file mcflirt writes (241/241), its exit status and both streams identical to mcflirt’s (FSL 6.0.7.17 as built), with the replica in reproduce.

Bit-identical

Every file mcflirt writes (241/241), its exit status and both streams identical to mcflirt’s (FSL 6.0.7.23 as built), with the replica in reproduce.

11/11 · 6/6 · 7/7
real/ds000005/fmriprep-repeat2real runsrepeat of real/ds000005/fmriprep (run-to-run determinism)
command linemcflirt -in sub-01_task-mixedgamblestask_run-01_bold.nii.gz -out out_mcf.nii.gz -reffile sub-01_task-mixedgamblestask_run-1_boldref.nii.gz -mats
Within the original's own variability

95th-percentile RMS deviation 0.0257 mm, median Δ 0.0019 mm / 0.0023°; 2/240 matrices identical as text; mcflirt’s own: 0.101 mm, 0.00507 mm / 0.00679°

Bit-identical

Every file mcflirt writes (241/241), its exit status and both streams identical to mcflirt’s (FSL 6.0.7.17 as built), with the replica in reproduce.

Bit-identical

Every file mcflirt writes (241/241), its exit status and both streams identical to mcflirt’s (FSL 6.0.7.23 as built), with the replica in reproduce.

11/11 · 6/6 · 7/7
real/ds000005/fmriprep-repeat3real runsrepeat of real/ds000005/fmriprep (run-to-run determinism)
command linemcflirt -in sub-01_task-mixedgamblestask_run-01_bold.nii.gz -out out_mcf.nii.gz -reffile sub-01_task-mixedgamblestask_run-1_boldref.nii.gz -mats
Within the original's own variability

95th-percentile RMS deviation 0.0257 mm, median Δ 0.0019 mm / 0.0023°; 2/240 matrices identical as text; mcflirt’s own: 0.101 mm, 0.00507 mm / 0.00679°

Bit-identical

Every file mcflirt writes (241/241), its exit status and both streams identical to mcflirt’s (FSL 6.0.7.17 as built), with the replica in reproduce.

Bit-identical

Every file mcflirt writes (241/241), its exit status and both streams identical to mcflirt’s (FSL 6.0.7.23 as built), with the replica in reproduce.

11/11 · 6/6 · 7/7
real/ds000005/fmriprep-reportsreal runsfMRIPrep's command plus -plots -rmsrel -rmsabs
command linemcflirt -in sub-01_task-mixedgamblestask_run-01_bold.nii.gz -out out_mcf.nii.gz -reffile sub-01_task-mixedgamblestask_run-1_boldref.nii.gz -mats -plots -rmsrel -rmsabs
Within the original's own variability

95th-percentile RMS deviation 0.0257 mm, median Δ 0.0019 mm / 0.0023°; 2/240 matrices identical as text; mcflirt’s own: 0.101 mm, 0.00507 mm / 0.00679°

Bit-identical

Every file mcflirt writes (246/246), its exit status and both streams identical to mcflirt’s (FSL 6.0.7.17 as built), with the replica in reproduce.

Bit-identical

Every file mcflirt writes (246/246), its exit status and both streams identical to mcflirt’s (FSL 6.0.7.23 as built), with the replica in reproduce.

16/16 · 6/6 · 7/7
real/ds000005/opt-2doptions, on ds000005option sweep: -2d
command linemcflirt -in sub-01_task-mixedgamblestask_run-01_bold.nii.gz -out out_mcf.nii.gz -reffile sub-01_task-mixedgamblestask_run-1_boldref.nii.gz -mats -plots -rmsrel -rmsabs -2d
Open difference

95th-percentile RMS deviation 0.0125 mm, median Δ 0.0002 mm / 0°; 0/240 matrices identical as text: within the record’s thresholds (0.1 mm; 0.05 mm / 0.05°); mcflirt’s own variability is not measured for this run, so the difference is not yet set against it.

Bit-identical

Every file mcflirt writes (246/246), its exit status and both streams identical to mcflirt’s (FSL 6.0.7.17 as built), with the replica in reproduce.

Bit-identical

Every file mcflirt writes (246/246), its exit status and both streams identical to mcflirt’s (FSL 6.0.7.23 as built), with the replica in reproduce.

16/16 · 6/6 · 7/7
real/ds000005/opt-bins64options, on ds000005option sweep: -bins 64
command linemcflirt -in sub-01_task-mixedgamblestask_run-01_bold.nii.gz -out out_mcf.nii.gz -reffile sub-01_task-mixedgamblestask_run-1_boldref.nii.gz -mats -plots -rmsrel -rmsabs -bins 64
Open difference

95th-percentile RMS deviation 0.0257 mm, median Δ 0.0019 mm / 0.0023°; 2/240 matrices identical as text: within the record’s thresholds (0.1 mm; 0.05 mm / 0.05°); mcflirt’s own variability is not measured for this run, so the difference is not yet set against it.

Bit-identical

Every file mcflirt writes (246/246), its exit status and both streams identical to mcflirt’s (FSL 6.0.7.17 as built), with the replica in reproduce.

Bit-identical

Every file mcflirt writes (246/246), its exit status and both streams identical to mcflirt’s (FSL 6.0.7.23 as built), with the replica in reproduce.

16/16 · 6/6 · 7/7
real/ds000005/opt-cost-corratiooptions, on ds000005option sweep: -cost corratio
command linemcflirt -in sub-01_task-mixedgamblestask_run-01_bold.nii.gz -out out_mcf.nii.gz -reffile sub-01_task-mixedgamblestask_run-1_boldref.nii.gz -mats -plots -rmsrel -rmsabs -cost corratio
Open difference

95th-percentile RMS deviation 0.0284 mm, median Δ 0.002 mm / 0.0029°; 1/240 matrices identical as text: within the record’s thresholds (0.1 mm; 0.05 mm / 0.05°); mcflirt’s own variability is not measured for this run, so the difference is not yet set against it.

Bit-identical

Every file mcflirt writes (246/246), its exit status and both streams identical to mcflirt’s (FSL 6.0.7.17 as built), with the replica in reproduce.

Bit-identical

Every file mcflirt writes (246/246), its exit status and both streams identical to mcflirt’s (FSL 6.0.7.23 as built), with the replica in reproduce.

16/16 · 6/6 · 7/7
real/ds000005/opt-cost-leastsquaresoptions, on ds000005option sweep: -cost leastsquares
command linemcflirt -in sub-01_task-mixedgamblestask_run-01_bold.nii.gz -out out_mcf.nii.gz -reffile sub-01_task-mixedgamblestask_run-1_boldref.nii.gz -mats -plots -rmsrel -rmsabs -cost leastsquares
Open difference

95th-percentile RMS deviation 0.0227 mm, median Δ 0.0003 mm / 0.0005°; 0/240 matrices identical as text: within the record’s thresholds (0.1 mm; 0.05 mm / 0.05°); mcflirt’s own variability is not measured for this run, so the difference is not yet set against it.

Bit-identical

Every file mcflirt writes (246/246), its exit status and both streams identical to mcflirt’s (FSL 6.0.7.17 as built), with the replica in reproduce.

Bit-identical

Every file mcflirt writes (246/246), its exit status and both streams identical to mcflirt’s (FSL 6.0.7.23 as built), with the replica in reproduce.

16/16 · 6/6 · 7/7
real/ds000005/opt-cost-mutualinfooptions, on ds000005option sweep: -cost mutualinfo
command linemcflirt -in sub-01_task-mixedgamblestask_run-01_bold.nii.gz -out out_mcf.nii.gz -reffile sub-01_task-mixedgamblestask_run-1_boldref.nii.gz -mats -plots -rmsrel -rmsabs -cost mutualinfo
Open difference

95th-percentile RMS deviation 0.0047 mm, median Δ 0.0001 mm / 0.0001°; 1/240 matrices identical as text: within the record’s thresholds (0.1 mm; 0.05 mm / 0.05°); mcflirt’s own variability is not measured for this run, so the difference is not yet set against it.

Bit-identical

Every file mcflirt writes (246/246), its exit status and both streams identical to mcflirt’s (FSL 6.0.7.17 as built), with the replica in reproduce.

Bit-identical

Every file mcflirt writes (246/246), its exit status and both streams identical to mcflirt’s (FSL 6.0.7.23 as built), with the replica in reproduce.

16/16 · 6/6 · 7/7
real/ds000005/opt-cost-normcorroptions, on ds000005option sweep: -cost normcorr
command linemcflirt -in sub-01_task-mixedgamblestask_run-01_bold.nii.gz -out out_mcf.nii.gz -reffile sub-01_task-mixedgamblestask_run-1_boldref.nii.gz -mats -plots -rmsrel -rmsabs -cost normcorr
Open difference

95th-percentile RMS deviation 0.0257 mm, median Δ 0.0019 mm / 0.0023°; 2/240 matrices identical as text: within the record’s thresholds (0.1 mm; 0.05 mm / 0.05°); mcflirt’s own variability is not measured for this run, so the difference is not yet set against it.

Bit-identical

Every file mcflirt writes (246/246), its exit status and both streams identical to mcflirt’s (FSL 6.0.7.17 as built), with the replica in reproduce.

Bit-identical

Every file mcflirt writes (246/246), its exit status and both streams identical to mcflirt’s (FSL 6.0.7.23 as built), with the replica in reproduce.

16/16 · 6/6 · 7/7
real/ds000005/opt-cost-normmioptions, on ds000005option sweep: -cost normmi
command linemcflirt -in sub-01_task-mixedgamblestask_run-01_bold.nii.gz -out out_mcf.nii.gz -reffile sub-01_task-mixedgamblestask_run-1_boldref.nii.gz -mats -plots -rmsrel -rmsabs -cost normmi
Open difference

95th-percentile RMS deviation 0.0126 mm, median Δ 0.0005 mm / 0.0007°; 0/240 matrices identical as text: within the record’s thresholds (0.1 mm; 0.05 mm / 0.05°); mcflirt’s own variability is not measured for this run, so the difference is not yet set against it.

Bit-identical

Every file mcflirt writes (246/246), its exit status and both streams identical to mcflirt’s (FSL 6.0.7.17 as built), with the replica in reproduce.

Bit-identical

Every file mcflirt writes (246/246), its exit status and both streams identical to mcflirt’s (FSL 6.0.7.23 as built), with the replica in reproduce.

16/16 · 6/6 · 7/7
real/ds000005/opt-cost-woodsoptions, on ds000005option sweep: -cost woods
command linemcflirt -in sub-01_task-mixedgamblestask_run-01_bold.nii.gz -out out_mcf.nii.gz -reffile sub-01_task-mixedgamblestask_run-1_boldref.nii.gz -mats -plots -rmsrel -rmsabs -cost woods
Open difference

Within the record’s thresholds (files and messages compared); no matrices to set against mcflirt’s own variability.

Bit-identical

Every file mcflirt writes (246/246), its exit status and both streams identical to mcflirt’s (FSL 6.0.7.17 as built), with the replica in reproduce.

Bit-identical

Every file mcflirt writes (246/246), its exit status and both streams identical to mcflirt’s (FSL 6.0.7.23 as built), with the replica in reproduce.

5/5 · 6/6 · 7/7
real/ds000005/opt-dof12options, on ds000005option sweep: -dof 12
command linemcflirt -in sub-01_task-mixedgamblestask_run-01_bold.nii.gz -out out_mcf.nii.gz -reffile sub-01_task-mixedgamblestask_run-1_boldref.nii.gz -mats -plots -rmsrel -rmsabs -dof 12
Open difference

95th-percentile RMS deviation 0.0239 mm, median Δ 0.0024 mm / 0.0053°; 0/240 matrices identical as text: within the record’s thresholds (0.1 mm; 0.05 mm / 0.05°); mcflirt’s own variability is not measured for this run, so the difference is not yet set against it.

Bit-identical

Every file mcflirt writes (246/246), its exit status and both streams identical to mcflirt’s (FSL 6.0.7.17 as built), with the replica in reproduce.

Bit-identical

Every file mcflirt writes (246/246), its exit status and both streams identical to mcflirt’s (FSL 6.0.7.23 as built), with the replica in reproduce.

16/16 · 6/6 · 7/7
real/ds000005/opt-dof7options, on ds000005option sweep: -dof 7
command linemcflirt -in sub-01_task-mixedgamblestask_run-01_bold.nii.gz -out out_mcf.nii.gz -reffile sub-01_task-mixedgamblestask_run-1_boldref.nii.gz -mats -plots -rmsrel -rmsabs -dof 7
Open difference

95th-percentile RMS deviation 0.0189 mm, median Δ 0.0021 mm / 0.0042°; 0/240 matrices identical as text: within the record’s thresholds (0.1 mm; 0.05 mm / 0.05°); mcflirt’s own variability is not measured for this run, so the difference is not yet set against it.

Bit-identical

Every file mcflirt writes (246/246), its exit status and both streams identical to mcflirt’s (FSL 6.0.7.17 as built), with the replica in reproduce.

Bit-identical

Every file mcflirt writes (246/246), its exit status and both streams identical to mcflirt’s (FSL 6.0.7.23 as built), with the replica in reproduce.

16/16 · 6/6 · 7/7
real/ds000005/opt-dof9options, on ds000005option sweep: -dof 9
command linemcflirt -in sub-01_task-mixedgamblestask_run-01_bold.nii.gz -out out_mcf.nii.gz -reffile sub-01_task-mixedgamblestask_run-1_boldref.nii.gz -mats -plots -rmsrel -rmsabs -dof 9
Open difference

95th-percentile RMS deviation 0.0199 mm, median Δ 0.0023 mm / 0.003°; 0/240 matrices identical as text: within the record’s thresholds (0.1 mm; 0.05 mm / 0.05°); mcflirt’s own variability is not measured for this run, so the difference is not yet set against it.

Bit-identical

Every file mcflirt writes (246/246), its exit status and both streams identical to mcflirt’s (FSL 6.0.7.17 as built), with the replica in reproduce.

Bit-identical

Every file mcflirt writes (246/246), its exit status and both streams identical to mcflirt’s (FSL 6.0.7.23 as built), with the replica in reproduce.

16/16 · 6/6 · 7/7
real/ds000005/opt-fudgeoptions, on ds000005option sweep: -fudge
command linemcflirt -in sub-01_task-mixedgamblestask_run-01_bold.nii.gz -out out_mcf.nii.gz -reffile sub-01_task-mixedgamblestask_run-1_boldref.nii.gz -mats -plots -rmsrel -rmsabs -fudge
Open difference

95th-percentile RMS deviation 0.0121 mm, median Δ 0.0002 mm / 0°; 57/240 matrices identical as text: within the record’s thresholds (0.1 mm; 0.05 mm / 0.05°); mcflirt’s own variability is not measured for this run, so the difference is not yet set against it.

Bit-identical

Every file mcflirt writes (246/246), its exit status and both streams identical to mcflirt’s (FSL 6.0.7.17 as built), with the replica in reproduce.

Bit-identical

Every file mcflirt writes (246/246), its exit status and both streams identical to mcflirt’s (FSL 6.0.7.23 as built), with the replica in reproduce.

16/16 · 6/6 · 7/7
real/ds000005/opt-gdtoptions, on ds000005option sweep: -gdt (writes grefvol_<out>, impossible with an absolute -out)
command linemcflirt -in sub-01_task-mixedgamblestask_run-01_bold.nii.gz -out out_mcf.nii.gz -reffile sub-01_task-mixedgamblestask_run-1_boldref.nii.gz -mats -plots -rmsrel -rmsabs -gdt
Rejected by both

mcflirt exit [-6]: terminate called after throwing an instance of ‘NiftiIO::NiftiException’ what(): Error: cant open file grefvol_/oracl; the clean room exit [1]: Error: -gdt (registration of gradient images) is not supported

Rejected by both

mcflirt and the replica fail alike, with the same messages and status: mcflirt exits [-6]; the same from the replica.

Rejected by both

mcflirt and the replica fail alike, with the same messages and status: mcflirt exits [-6]; the same from the replica.

0/0 · 5/5 · 6/6
real/ds000005/opt-meanvoloptions, on ds000005option sweep: -meanvol
command linemcflirt -in sub-01_task-mixedgamblestask_run-01_bold.nii.gz -out out_mcf.nii.gz -reffile sub-01_task-mixedgamblestask_run-1_boldref.nii.gz -mats -plots -rmsrel -rmsabs -meanvol
Within the original's own variability

95th-percentile RMS deviation 0.023 mm, median Δ 0.0025 mm / 0.0034°; 0/240 matrices identical as text; mcflirt’s own: 0.0537 mm, 0.00511 mm / 0.00667°

Bit-identical

Every file mcflirt writes (247/247), its exit status and both streams identical to mcflirt’s (FSL 6.0.7.17 as built), with the replica in reproduce.

Bit-identical

Every file mcflirt writes (247/247), its exit status and both streams identical to mcflirt’s (FSL 6.0.7.23 as built), with the replica in reproduce.

18/18 · 6/6 · 7/7
real/ds000005/opt-meanvol-norefoptions, on ds000005-meanvol without -reffile
command linemcflirt -in sub-01_task-mixedgamblestask_run-01_bold.nii.gz -out out_mcf.nii.gz -mats -plots -rmsrel -rmsabs -meanvol
Open difference

95th-percentile RMS deviation 0.0227 mm, median Δ 0.0019 mm / 0.004°; 0/240 matrices identical as text: within the record’s thresholds (0.1 mm; 0.05 mm / 0.05°); mcflirt’s own variability is not measured for this run, so the difference is not yet set against it.

Bit-identical

Every file mcflirt writes (247/247), its exit status and both streams identical to mcflirt’s (FSL 6.0.7.17 as built), with the replica in reproduce.

Bit-identical

Every file mcflirt writes (247/247), its exit status and both streams identical to mcflirt’s (FSL 6.0.7.23 as built), with the replica in reproduce.

18/18 · 6/6 · 7/7
real/ds000005/opt-nn_finaloptions, on ds000005option sweep: -nn_final
command linemcflirt -in sub-01_task-mixedgamblestask_run-01_bold.nii.gz -out out_mcf.nii.gz -reffile sub-01_task-mixedgamblestask_run-1_boldref.nii.gz -mats -plots -rmsrel -rmsabs -nn_final
Open difference

95th-percentile RMS deviation 0.0257 mm, median Δ 0.0019 mm / 0.0023°; 2/240 matrices identical as text: within the record’s thresholds (0.1 mm; 0.05 mm / 0.05°); mcflirt’s own variability is not measured for this run, so the difference is not yet set against it.

Bit-identical

Every file mcflirt writes (246/246), its exit status and both streams identical to mcflirt’s (FSL 6.0.7.17 as built), with the replica in reproduce.

Bit-identical

Every file mcflirt writes (246/246), its exit status and both streams identical to mcflirt’s (FSL 6.0.7.23 as built), with the replica in reproduce.

16/16 · 6/6 · 7/7
real/ds000005/opt-refvol0options, on ds000005default with -refvol 0
command linemcflirt -in sub-01_task-mixedgamblestask_run-01_bold.nii.gz -out out_mcf.nii.gz -mats -plots -rmsrel -rmsabs -refvol 0
Open difference

95th-percentile RMS deviation 0.0188 mm, median Δ 0.0013 mm / 0.0029°; 18/240 matrices identical as text: within the record’s thresholds (0.1 mm; 0.05 mm / 0.05°); mcflirt’s own variability is not measured for this run, so the difference is not yet set against it.

Bit-identical

Every file mcflirt writes (246/246), its exit status and both streams identical to mcflirt’s (FSL 6.0.7.17 as built), with the replica in reproduce.

Bit-identical

Every file mcflirt writes (246/246), its exit status and both streams identical to mcflirt’s (FSL 6.0.7.23 as built), with the replica in reproduce.

16/16 · 6/6 · 7/7
real/ds000005/opt-rotation2options, on ds000005option sweep: -rotation 2
command linemcflirt -in sub-01_task-mixedgamblestask_run-01_bold.nii.gz -out out_mcf.nii.gz -reffile sub-01_task-mixedgamblestask_run-1_boldref.nii.gz -mats -plots -rmsrel -rmsabs -rotation 2
Open difference

95th-percentile RMS deviation 0.0213 mm, median Δ 0.0015 mm / 0.0015°; 4/240 matrices identical as text: within the record’s thresholds (0.1 mm; 0.05 mm / 0.05°); mcflirt’s own variability is not measured for this run, so the difference is not yet set against it.

Bit-identical

Every file mcflirt writes (246/246), its exit status and both streams identical to mcflirt’s (FSL 6.0.7.17 as built), with the replica in reproduce.

Bit-identical

Every file mcflirt writes (246/246), its exit status and both streams identical to mcflirt’s (FSL 6.0.7.23 as built), with the replica in reproduce.

16/16 · 6/6 · 7/7
real/ds000005/opt-scaling3options, on ds000005option sweep: -scaling 3
command linemcflirt -in sub-01_task-mixedgamblestask_run-01_bold.nii.gz -out out_mcf.nii.gz -reffile sub-01_task-mixedgamblestask_run-1_boldref.nii.gz -mats -plots -rmsrel -rmsabs -scaling 3
Open difference

95th-percentile RMS deviation 0.0257 mm, median Δ 0.0019 mm / 0.0023°; 2/240 matrices identical as text: within the record’s thresholds (0.1 mm; 0.05 mm / 0.05°); mcflirt’s own variability is not measured for this run, so the difference is not yet set against it.

Bit-identical

Every file mcflirt writes (246/246), its exit status and both streams identical to mcflirt’s (FSL 6.0.7.17 as built), with the replica in reproduce.

Bit-identical

Every file mcflirt writes (246/246), its exit status and both streams identical to mcflirt’s (FSL 6.0.7.23 as built), with the replica in reproduce.

16/16 · 6/6 · 7/7
real/ds000005/opt-smooth0options, on ds000005option sweep: -smooth 0
command linemcflirt -in sub-01_task-mixedgamblestask_run-01_bold.nii.gz -out out_mcf.nii.gz -reffile sub-01_task-mixedgamblestask_run-1_boldref.nii.gz -mats -plots -rmsrel -rmsabs -smooth 0
Open difference

95th-percentile RMS deviation 0.0247 mm, median Δ 0.0013 mm / 0.0001°; 3/240 matrices identical as text: within the record’s thresholds (0.1 mm; 0.05 mm / 0.05°); mcflirt’s own variability is not measured for this run, so the difference is not yet set against it.

Bit-identical

Every file mcflirt writes (246/246), its exit status and both streams identical to mcflirt’s (FSL 6.0.7.17 as built), with the replica in reproduce.

Bit-identical

Every file mcflirt writes (246/246), its exit status and both streams identical to mcflirt’s (FSL 6.0.7.23 as built), with the replica in reproduce.

16/16 · 6/6 · 7/7
real/ds000005/opt-smooth2options, on ds000005option sweep: -smooth 2
command linemcflirt -in sub-01_task-mixedgamblestask_run-01_bold.nii.gz -out out_mcf.nii.gz -reffile sub-01_task-mixedgamblestask_run-1_boldref.nii.gz -mats -plots -rmsrel -rmsabs -smooth 2
Open difference

95th-percentile RMS deviation 0.0151 mm, median Δ 0.0009 mm / 0.0002°; 3/240 matrices identical as text: within the record’s thresholds (0.1 mm; 0.05 mm / 0.05°); mcflirt’s own variability is not measured for this run, so the difference is not yet set against it.

Bit-identical

Every file mcflirt writes (246/246), its exit status and both streams identical to mcflirt’s (FSL 6.0.7.17 as built), with the replica in reproduce.

Bit-identical

Every file mcflirt writes (246/246), its exit status and both streams identical to mcflirt’s (FSL 6.0.7.23 as built), with the replica in reproduce.

16/16 · 6/6 · 7/7
real/ds000005/opt-spline_finaloptions, on ds000005option sweep: -spline_final
command linemcflirt -in sub-01_task-mixedgamblestask_run-01_bold.nii.gz -out out_mcf.nii.gz -reffile sub-01_task-mixedgamblestask_run-1_boldref.nii.gz -mats -plots -rmsrel -rmsabs -spline_final
Open difference

95th-percentile RMS deviation 0.0257 mm, median Δ 0.0019 mm / 0.0023°; 2/240 matrices identical as text: within the record’s thresholds (0.1 mm; 0.05 mm / 0.05°); mcflirt’s own variability is not measured for this run, so the difference is not yet set against it.

Bit-identical

Every file mcflirt writes (246/246), its exit status and both streams identical to mcflirt’s (FSL 6.0.7.17 as built), with the replica in reproduce.

Bit-identical

Every file mcflirt writes (246/246), its exit status and both streams identical to mcflirt’s (FSL 6.0.7.23 as built), with the replica in reproduce.

16/16 · 6/6 · 7/7
real/ds000005/opt-stages1options, on ds000005option sweep: -stages 1
command linemcflirt -in sub-01_task-mixedgamblestask_run-01_bold.nii.gz -out out_mcf.nii.gz -reffile sub-01_task-mixedgamblestask_run-1_boldref.nii.gz -mats -plots -rmsrel -rmsabs -stages 1
Open difference

95th-percentile RMS deviation 0.0251 mm, median Δ 0.0003 mm / 0.0003°; 4/240 matrices identical as text: within the record’s thresholds (0.1 mm; 0.05 mm / 0.05°); mcflirt’s own variability is not measured for this run, so the difference is not yet set against it.

Bit-identical

Every file mcflirt writes (246/246), its exit status and both streams identical to mcflirt’s (FSL 6.0.7.17 as built), with the replica in reproduce.

Bit-identical

Every file mcflirt writes (246/246), its exit status and both streams identical to mcflirt’s (FSL 6.0.7.23 as built), with the replica in reproduce.

16/16 · 6/6 · 7/7
real/ds000005/opt-stages2options, on ds000005option sweep: -stages 2
command linemcflirt -in sub-01_task-mixedgamblestask_run-01_bold.nii.gz -out out_mcf.nii.gz -reffile sub-01_task-mixedgamblestask_run-1_boldref.nii.gz -mats -plots -rmsrel -rmsabs -stages 2
Open difference

95th-percentile RMS deviation 0.0317 mm, median Δ 0.0007 mm / 0.0003°; 3/240 matrices identical as text: within the record’s thresholds (0.1 mm; 0.05 mm / 0.05°); mcflirt’s own variability is not measured for this run, so the difference is not yet set against it.

Bit-identical

Every file mcflirt writes (246/246), its exit status and both streams identical to mcflirt’s (FSL 6.0.7.17 as built), with the replica in reproduce.

Bit-identical

Every file mcflirt writes (246/246), its exit status and both streams identical to mcflirt’s (FSL 6.0.7.23 as built), with the replica in reproduce.

16/16 · 6/6 · 7/7
real/ds000005/opt-statsoptions, on ds000005option sweep: -stats
command linemcflirt -in sub-01_task-mixedgamblestask_run-01_bold.nii.gz -out out_mcf.nii.gz -reffile sub-01_task-mixedgamblestask_run-1_boldref.nii.gz -mats -plots -rmsrel -rmsabs -stats
Open difference

95th-percentile RMS deviation 0.0257 mm, median Δ 0.0019 mm / 0.0023°; 2/240 matrices identical as text: within the record’s thresholds (0.1 mm; 0.05 mm / 0.05°); mcflirt’s own variability is not measured for this run, so the difference is not yet set against it.

Bit-identical

Every file mcflirt writes (249/249), its exit status and both streams identical to mcflirt’s (FSL 6.0.7.17 as built), with the replica in reproduce.

Bit-identical

Every file mcflirt writes (249/249), its exit status and both streams identical to mcflirt’s (FSL 6.0.7.23 as built), with the replica in reproduce.

22/22 · 6/6 · 7/7
real/ds000117/defaultreal runsmcflirt's defaults (middle volume as reference)
command linemcflirt -in sub-01_ses-mri_task-facerecognition_run-01_bold.nii.gz -out out_mcf.nii.gz -mats -plots -rmsrel -rmsabs
Within the original's own variability

95th-percentile RMS deviation 0.0214 mm, median Δ 0.0016 mm / 0.0026°; 10/208 matrices identical as text; mcflirt’s own: 0.0231 mm, 0.00256 mm / 0.00384°

Bit-identical

Every file mcflirt writes (214/214), its exit status and both streams identical to mcflirt’s (FSL 6.0.7.17 as built), with the replica in reproduce.

Bit-identical

Every file mcflirt writes (214/214), its exit status and both streams identical to mcflirt’s (FSL 6.0.7.23 as built), with the replica in reproduce.

16/16 · 6/6 · 7/7
real/ds000117/fmriprepreal runsfMRIPrep's command (nipype MCFLIRT(save_mats=True), ref_file)
command linemcflirt -in sub-01_ses-mri_task-facerecognition_run-01_bold.nii.gz -out out_mcf.nii.gz -reffile ds000117_median10.nii.gz -mats
Within the original's own variability

95th-percentile RMS deviation 0.0203 mm, median Δ 0.0023 mm / 0.0031°; 0/208 matrices identical as text; mcflirt’s own: 0.0203 mm, 0.00356 mm / 0.00407°

Bit-identical

Every file mcflirt writes (209/209), its exit status and both streams identical to mcflirt’s (FSL 6.0.7.17 as built), with the replica in reproduce.

Bit-identical

Every file mcflirt writes (209/209), its exit status and both streams identical to mcflirt’s (FSL 6.0.7.23 as built), with the replica in reproduce.

11/11 · 6/6 · 7/7
real/ds000117/fmriprep-reportsreal runsfMRIPrep's command plus -plots -rmsrel -rmsabs
command linemcflirt -in sub-01_ses-mri_task-facerecognition_run-01_bold.nii.gz -out out_mcf.nii.gz -reffile ds000117_median10.nii.gz -mats -plots -rmsrel -rmsabs
Within the original's own variability

95th-percentile RMS deviation 0.0203 mm, median Δ 0.0023 mm / 0.0031°; 0/208 matrices identical as text; mcflirt’s own: 0.0203 mm, 0.00356 mm / 0.00407°

Bit-identical

Every file mcflirt writes (214/214), its exit status and both streams identical to mcflirt’s (FSL 6.0.7.17 as built), with the replica in reproduce.

Bit-identical

Every file mcflirt writes (214/214), its exit status and both streams identical to mcflirt’s (FSL 6.0.7.23 as built), with the replica in reproduce.

16/16 · 6/6 · 7/7
real/ds000122/defaultreal runsmcflirt's defaults (middle volume as reference)
command linemcflirt -in sub-17_task-visualattentiontask_run-01_bold.nii.gz -out out_mcf.nii.gz -mats -plots -rmsrel -rmsabs
Open difference

95th-percentile RMS deviation 0.0291 mm, median Δ 0.0025 mm / 0.0053°; 3/258 matrices identical as text, FD r 0.9987: above mcflirt’s own variability (mcflirt’s own: 0.0289 mm, 0.00257 mm / 0.00501°, FD r 0.9987), within the record’s thresholds

Bit-identical

Every file mcflirt writes (264/264), its exit status and both streams identical to mcflirt’s (FSL 6.0.7.17 as built), with the replica in reproduce.

Bit-identical

Every file mcflirt writes (264/264), its exit status and both streams identical to mcflirt’s (FSL 6.0.7.23 as built), with the replica in reproduce.

16/16 · 6/6 · 7/7
real/ds000122/fmriprepreal runsfMRIPrep's command (nipype MCFLIRT(save_mats=True), ref_file)
command linemcflirt -in sub-17_task-visualattentiontask_run-01_bold.nii.gz -out out_mcf.nii.gz -reffile ds000122_median10.nii.gz -mats
Within the original's own variability

95th-percentile RMS deviation 0.0269 mm, median Δ 0.0025 mm / 0.0054°; 0/258 matrices identical as text, FD r 0.9985; mcflirt’s own: 0.0312 mm, 0.0042 mm / 0.00733°, FD r 0.9977

Bit-identical

Every file mcflirt writes (259/259), its exit status and both streams identical to mcflirt’s (FSL 6.0.7.17 as built), with the replica in reproduce.

Bit-identical

Every file mcflirt writes (259/259), its exit status and both streams identical to mcflirt’s (FSL 6.0.7.23 as built), with the replica in reproduce.

11/11 · 6/6 · 7/7
real/ds000122/fmriprep-reportsreal runsfMRIPrep's command plus -plots -rmsrel -rmsabs
command linemcflirt -in sub-17_task-visualattentiontask_run-01_bold.nii.gz -out out_mcf.nii.gz -reffile ds000122_median10.nii.gz -mats -plots -rmsrel -rmsabs
Within the original's own variability

95th-percentile RMS deviation 0.0269 mm, median Δ 0.0025 mm / 0.0054°; 0/258 matrices identical as text, FD r 0.9985; mcflirt’s own: 0.0312 mm, 0.0042 mm / 0.00733°, FD r 0.9977

Bit-identical

Every file mcflirt writes (264/264), its exit status and both streams identical to mcflirt’s (FSL 6.0.7.17 as built), with the replica in reproduce.

Bit-identical

Every file mcflirt writes (264/264), its exit status and both streams identical to mcflirt’s (FSL 6.0.7.23 as built), with the replica in reproduce.

16/16 · 6/6 · 7/7
real/ds000210-rest-echo2/defaultreal runsmcflirt's defaults (middle volume as reference)
command linemcflirt -in sub-02_task-rest_run-01_echo-2_bold.nii.gz -out out_mcf.nii.gz -mats -plots -rmsrel -rmsabs
Within the original's own variability

95th-percentile RMS deviation 0.0215 mm, median Δ 0.0021 mm / 0.0031°; 9/204 matrices identical as text, FD r 0.9925; mcflirt’s own: 0.0221 mm, 0.00229 mm / 0.00338°, FD r 0.992

Bit-identical

Every file mcflirt writes (210/210), its exit status and both streams identical to mcflirt’s (FSL 6.0.7.17 as built), with the replica in reproduce.

Bit-identical

Every file mcflirt writes (210/210), its exit status and both streams identical to mcflirt’s (FSL 6.0.7.23 as built), with the replica in reproduce.

16/16 · 6/6 · 7/7
real/ds000210-rest-echo2/fmriprepreal runsfMRIPrep's command (nipype MCFLIRT(save_mats=True), ref_file)
command linemcflirt -in sub-02_task-rest_run-01_echo-2_bold.nii.gz -out out_mcf.nii.gz -reffile ds000210-rest-echo2_median10.nii.gz -mats
Within the original's own variability

95th-percentile RMS deviation 0.0221 mm, median Δ 0.0022 mm / 0.0032°; 0/204 matrices identical as text, FD r 0.9867; mcflirt’s own: 0.0229 mm, 0.00361 mm / 0.0051°, FD r 0.9799

Bit-identical

Every file mcflirt writes (205/205), its exit status and both streams identical to mcflirt’s (FSL 6.0.7.17 as built), with the replica in reproduce.

Bit-identical

Every file mcflirt writes (205/205), its exit status and both streams identical to mcflirt’s (FSL 6.0.7.23 as built), with the replica in reproduce.

11/11 · 6/6 · 7/7
real/ds000210-rest-echo2/fmriprep-reportsreal runsfMRIPrep's command plus -plots -rmsrel -rmsabs
command linemcflirt -in sub-02_task-rest_run-01_echo-2_bold.nii.gz -out out_mcf.nii.gz -reffile ds000210-rest-echo2_median10.nii.gz -mats -plots -rmsrel -rmsabs
Within the original's own variability

95th-percentile RMS deviation 0.0221 mm, median Δ 0.0022 mm / 0.0032°; 0/204 matrices identical as text, FD r 0.9867; mcflirt’s own: 0.0229 mm, 0.00361 mm / 0.0051°, FD r 0.9799

Bit-identical

Every file mcflirt writes (210/210), its exit status and both streams identical to mcflirt’s (FSL 6.0.7.17 as built), with the replica in reproduce.

Bit-identical

Every file mcflirt writes (210/210), its exit status and both streams identical to mcflirt’s (FSL 6.0.7.23 as built), with the replica in reproduce.

16/16 · 6/6 · 7/7
real/ds000258-bigendian/defaultreal runsmcflirt's defaults (middle volume as reference)
command linemcflirt -in sub-21262_task-rest_echo-4_bold.nii.gz -out out_mcf.nii.gz -mats -plots -rmsrel -rmsabs
Within the original's own variability

95th-percentile RMS deviation 0.0166 mm, median Δ 0.0011 mm / 0.0026°; 1/239 matrices identical as text, FD r 0.996; mcflirt’s own: 0.0186 mm, 0.00157 mm / 0.00269°, FD r 0.9948

Bit-identical

Every file mcflirt writes (245/245), its exit status and both streams identical to mcflirt’s (FSL 6.0.7.17 as built), with the replica in reproduce.

Bit-identical

Every file mcflirt writes (245/245), its exit status and both streams identical to mcflirt’s (FSL 6.0.7.23 as built), with the replica in reproduce.

16/16 · 6/6 · 7/7
real/ds000258-bigendian/fmriprepreal runsfMRIPrep's command (nipype MCFLIRT(save_mats=True), ref_file)
command linemcflirt -in sub-21262_task-rest_echo-4_bold.nii.gz -out out_mcf.nii.gz -reffile ds000258-bigendian_median10.nii.gz -mats
Within the original's own variability

95th-percentile RMS deviation 0.0186 mm, median Δ 0.0014 mm / 0.0029°; 0/239 matrices identical as text, FD r 0.9953; mcflirt’s own: 0.0193 mm, 0.00245 mm / 0.00395°, FD r 0.9949

Bit-identical

Every file mcflirt writes (240/240), its exit status and both streams identical to mcflirt’s (FSL 6.0.7.17 as built), with the replica in reproduce.

Bit-identical

Every file mcflirt writes (240/240), its exit status and both streams identical to mcflirt’s (FSL 6.0.7.23 as built), with the replica in reproduce.

11/11 · 6/6 · 7/7
real/ds000258-bigendian/fmriprep-reportsreal runsfMRIPrep's command plus -plots -rmsrel -rmsabs
command linemcflirt -in sub-21262_task-rest_echo-4_bold.nii.gz -out out_mcf.nii.gz -reffile ds000258-bigendian_median10.nii.gz -mats -plots -rmsrel -rmsabs
Within the original's own variability

95th-percentile RMS deviation 0.0186 mm, median Δ 0.0014 mm / 0.0029°; 0/239 matrices identical as text, FD r 0.9953; mcflirt’s own: 0.0193 mm, 0.00245 mm / 0.00395°, FD r 0.9949

Bit-identical

Every file mcflirt writes (245/245), its exit status and both streams identical to mcflirt’s (FSL 6.0.7.17 as built), with the replica in reproduce.

Bit-identical

Every file mcflirt writes (245/245), its exit status and both streams identical to mcflirt’s (FSL 6.0.7.23 as built), with the replica in reproduce.

16/16 · 6/6 · 7/7
real/ds003345/defaultreal runsmcflirt's defaults (middle volume as reference)
command linemcflirt -in sub-22973_task-PenaltyKik_run-02_bold.nii.gz -out out_mcf.nii.gz -mats -plots -rmsrel -rmsabs
Within the original's own variability

95th-percentile RMS deviation 0.0235 mm, median Δ 0.0022 mm / 0.0037°; 12/216 matrices identical as text, FD r 0.9907; mcflirt’s own: 0.0261 mm, 0.00256 mm / 0.00424°, FD r 0.9893

Bit-identical

Every file mcflirt writes (222/222), its exit status and both streams identical to mcflirt’s (FSL 6.0.7.17 as built), with the replica in reproduce.

Bit-identical

Every file mcflirt writes (222/222), its exit status and both streams identical to mcflirt’s (FSL 6.0.7.23 as built), with the replica in reproduce.

16/16 · 6/6 · 7/7
real/ds003345/fmriprepreal runsfMRIPrep's command (nipype MCFLIRT(save_mats=True), ref_file)
command linemcflirt -in sub-22973_task-PenaltyKik_run-02_bold.nii.gz -out out_mcf.nii.gz -reffile ds003345_median10.nii.gz -mats
Within the original's own variability

95th-percentile RMS deviation 0.0274 mm, median Δ 0.0028 mm / 0.0042°; 0/216 matrices identical as text, FD r 0.9854; mcflirt’s own: 0.0299 mm, 0.00387 mm / 0.00584°, FD r 0.984

Bit-identical

Every file mcflirt writes (217/217), its exit status and both streams identical to mcflirt’s (FSL 6.0.7.17 as built), with the replica in reproduce.

Bit-identical

Every file mcflirt writes (217/217), its exit status and both streams identical to mcflirt’s (FSL 6.0.7.23 as built), with the replica in reproduce.

11/11 · 6/6 · 7/7
real/ds003345/fmriprep-reportsreal runsfMRIPrep's command plus -plots -rmsrel -rmsabs
command linemcflirt -in sub-22973_task-PenaltyKik_run-02_bold.nii.gz -out out_mcf.nii.gz -reffile ds003345_median10.nii.gz -mats -plots -rmsrel -rmsabs
Within the original's own variability

95th-percentile RMS deviation 0.0274 mm, median Δ 0.0028 mm / 0.0042°; 0/216 matrices identical as text, FD r 0.9854; mcflirt’s own: 0.0299 mm, 0.00387 mm / 0.00584°, FD r 0.984

Bit-identical

Every file mcflirt writes (222/222), its exit status and both streams identical to mcflirt’s (FSL 6.0.7.17 as built), with the replica in reproduce.

Bit-identical

Every file mcflirt writes (222/222), its exit status and both streams identical to mcflirt’s (FSL 6.0.7.23 as built), with the replica in reproduce.

16/16 · 6/6 · 7/7
real/ds003763-truncated/defaultreal runsa .nii.gz truncated as published (65,536 bytes)
command linemcflirt -in sub-16111_task-heart_bold.nii.gz -out out_mcf.nii.gz -mats -plots -rmsrel -rmsabs
Rejected by both

mcflirt exit [-6]: Image Exception : #22 :: Failed to read volume /openneuro/ds003763/sub-16111/func/sub-16111_task-heart_bold.nii.gz Erro; the clean room exit [1]: Error: /openneuro/ds003763/sub-16111/func/sub-16111_task-heart_bold.n…

Rejected by both

mcflirt and the replica fail alike, with the same messages and status: mcflirt exits [-6]; the same from the replica.

Rejected by both

mcflirt and the replica fail alike, with the same messages and status: mcflirt exits [-6]; the same from the replica.

0/0 · 5/5 · 6/6
real/ds005040/defaultreal runsmcflirt's defaults (middle volume as reference)
command linemcflirt -in sub-002_task-clip_run-1_bold.nii.gz -out out_mcf.nii.gz -mats -plots -rmsrel -rmsabs
Open difference

95th-percentile RMS deviation 0.0268 mm, median Δ 0.0022 mm / 0.0046°; 3/63 matrices identical as text: above mcflirt’s own variability (mcflirt’s own: 0.0263 mm, 0.00317 mm / 0.00521°), within the record’s thresholds

Bit-identical

Every file mcflirt writes (69/69), its exit status and both streams identical to mcflirt’s (FSL 6.0.7.17 as built), with the replica in reproduce.

Bit-identical

Every file mcflirt writes (69/69), its exit status and both streams identical to mcflirt’s (FSL 6.0.7.23 as built), with the replica in reproduce.

16/16 · 6/6 · 7/7
real/ds005040/fmriprepreal runsfMRIPrep's command (nipype MCFLIRT(save_mats=True), ref_file)
command linemcflirt -in sub-002_task-clip_run-1_bold.nii.gz -out out_mcf.nii.gz -reffile ds005040_median10.nii.gz -mats
Within the original's own variability

95th-percentile RMS deviation 0.0225 mm, median Δ 0.0024 mm / 0.0038°; 1/63 matrices identical as text; mcflirt’s own: 0.0257 mm, 0.00382 mm / 0.0047°

Bit-identical

Every file mcflirt writes (64/64), its exit status and both streams identical to mcflirt’s (FSL 6.0.7.17 as built), with the replica in reproduce.

Bit-identical

Every file mcflirt writes (64/64), its exit status and both streams identical to mcflirt’s (FSL 6.0.7.23 as built), with the replica in reproduce.

11/11 · 6/6 · 7/7
real/ds005040/fmriprep-reportsreal runsfMRIPrep's command plus -plots -rmsrel -rmsabs
command linemcflirt -in sub-002_task-clip_run-1_bold.nii.gz -out out_mcf.nii.gz -reffile ds005040_median10.nii.gz -mats -plots -rmsrel -rmsabs
Within the original's own variability

95th-percentile RMS deviation 0.0225 mm, median Δ 0.0024 mm / 0.0038°; 1/63 matrices identical as text; mcflirt’s own: 0.0257 mm, 0.00382 mm / 0.0047°

Bit-identical

Every file mcflirt writes (69/69), its exit status and both streams identical to mcflirt’s (FSL 6.0.7.17 as built), with the replica in reproduce.

Bit-identical

Every file mcflirt writes (69/69), its exit status and both streams identical to mcflirt’s (FSL 6.0.7.23 as built), with the replica in reproduce.

16/16 · 6/6 · 7/7
real/ds006010/defaultreal runsmcflirt's defaults (middle volume as reference)
command linemcflirt -in sub-206_task-category_run-01_bold.nii.gz -out out_mcf.nii.gz -mats -plots -rmsrel -rmsabs
Within the original's own variability

95th-percentile RMS deviation 0.0134 mm, median Δ 0.0007 mm / 0.0001°; 2/113 matrices identical as text; mcflirt’s own: 0.0165 mm, 0.000776 mm / 0.000172°

Bit-identical

Every file mcflirt writes (119/119), its exit status and both streams identical to mcflirt’s (FSL 6.0.7.17 as built), with the replica in reproduce.

Bit-identical

Every file mcflirt writes (119/119), its exit status and both streams identical to mcflirt’s (FSL 6.0.7.23 as built), with the replica in reproduce.

16/16 · 6/6 · 7/7
real/ds006010/fmriprepreal runsfMRIPrep's command (nipype MCFLIRT(save_mats=True), ref_file)
command linemcflirt -in sub-206_task-category_run-01_bold.nii.gz -out out_mcf.nii.gz -reffile ds006010_median10.nii.gz -mats
Within the original's own variability

95th-percentile RMS deviation 0.0151 mm, median Δ 0.0016 mm / 0.0023°; 0/113 matrices identical as text; mcflirt’s own: 0.0164 mm, 0.00207 mm / 0.00298°

Bit-identical

Every file mcflirt writes (114/114), its exit status and both streams identical to mcflirt’s (FSL 6.0.7.17 as built), with the replica in reproduce.

Bit-identical

Every file mcflirt writes (114/114), its exit status and both streams identical to mcflirt’s (FSL 6.0.7.23 as built), with the replica in reproduce.

11/11 · 6/6 · 7/7
real/ds006010/fmriprep-reportsreal runsfMRIPrep's command plus -plots -rmsrel -rmsabs
command linemcflirt -in sub-206_task-category_run-01_bold.nii.gz -out out_mcf.nii.gz -reffile ds006010_median10.nii.gz -mats -plots -rmsrel -rmsabs
Within the original's own variability

95th-percentile RMS deviation 0.0151 mm, median Δ 0.0016 mm / 0.0023°; 0/113 matrices identical as text; mcflirt’s own: 0.0164 mm, 0.00207 mm / 0.00298°

Bit-identical

Every file mcflirt writes (119/119), its exit status and both streams identical to mcflirt’s (FSL 6.0.7.17 as built), with the replica in reproduce.

Bit-identical

Every file mcflirt writes (119/119), its exit status and both streams identical to mcflirt’s (FSL 6.0.7.23 as built), with the replica in reproduce.

16/16 · 6/6 · 7/7
real/ds006736/defaultreal runsmcflirt's defaults (middle volume as reference)
command linemcflirt -in sub-004_task-freeRecall_bold.nii.gz -out out_mcf.nii.gz -mats -plots -rmsrel -rmsabs
Open difference

95th-percentile RMS deviation 0.0211 mm, median Δ 0.0021 mm / 0.0031°; 1/69 matrices identical as text, FD r 0.9986: above mcflirt’s own variability (mcflirt’s own: 0.0233 mm, 0.00237 mm / 0.00258°, FD r 0.9989), within the record’s thresholds

Bit-identical

Every file mcflirt writes (75/75), its exit status and both streams identical to mcflirt’s (FSL 6.0.7.17 as built), with the replica in reproduce.

Bit-identical

Every file mcflirt writes (75/75), its exit status and both streams identical to mcflirt’s (FSL 6.0.7.23 as built), with the replica in reproduce.

16/16 · 6/6 · 7/7
real/ds006736/fmriprepreal runsfMRIPrep's command (nipype MCFLIRT(save_mats=True), ref_file)
command linemcflirt -in sub-004_task-freeRecall_bold.nii.gz -out out_mcf.nii.gz -reffile ds006736_median10.nii.gz -mats
Within the original's own variability

95th-percentile RMS deviation 0.0232 mm, median Δ 0.0022 mm / 0.0026°; 0/69 matrices identical as text, FD r 0.9989; mcflirt’s own: 0.0259 mm, 0.00351 mm / 0.0059°, FD r 0.9974

Bit-identical

Every file mcflirt writes (70/70), its exit status and both streams identical to mcflirt’s (FSL 6.0.7.17 as built), with the replica in reproduce.

Bit-identical

Every file mcflirt writes (70/70), its exit status and both streams identical to mcflirt’s (FSL 6.0.7.23 as built), with the replica in reproduce.

11/11 · 6/6 · 7/7
real/ds006736/fmriprep-reportsreal runsfMRIPrep's command plus -plots -rmsrel -rmsabs
command linemcflirt -in sub-004_task-freeRecall_bold.nii.gz -out out_mcf.nii.gz -reffile ds006736_median10.nii.gz -mats -plots -rmsrel -rmsabs
Within the original's own variability

95th-percentile RMS deviation 0.0232 mm, median Δ 0.0022 mm / 0.0026°; 0/69 matrices identical as text, FD r 0.9989; mcflirt’s own: 0.0259 mm, 0.00351 mm / 0.0059°, FD r 0.9974

Bit-identical

Every file mcflirt writes (75/75), its exit status and both streams identical to mcflirt’s (FSL 6.0.7.17 as built), with the replica in reproduce.

Bit-identical

Every file mcflirt writes (75/75), its exit status and both streams identical to mcflirt’s (FSL 6.0.7.23 as built), with the replica in reproduce.

16/16 · 6/6 · 7/7
syn/cli/bare-inputsynthetic: the command lineinput given without -in
command linemcflirt small.nii.gz -out out_mcf.nii.gz -mats
Open difference

95th-percentile RMS deviation 0.0601 mm, median Δ 0.0019 mm / 0.0097°; 1/12 matrices identical as text: within the record’s thresholds (0.1 mm; 0.05 mm / 0.05°); mcflirt’s own variability is not measured for this run, so the difference is not yet set against it.

Bit-identical

Every file mcflirt writes (13/13), its exit status and both streams identical to mcflirt’s (FSL 6.0.7.17 as built), with the replica in reproduce.

Bit-identical

Every file mcflirt writes (13/13), its exit status and both streams identical to mcflirt’s (FSL 6.0.7.23 as built), with the replica in reproduce.

11/11 · 6/6 · 7/7
syn/cli/dof5synthetic: the command line-dof 5 (below 6)
command linemcflirt -in small.nii.gz -out out_mcf.nii.gz -reffile ref.nii.gz -mats -plots -rmsrel -rmsabs -dof 5
Open difference

95th-percentile RMS deviation 0.0667 mm, median Δ 0.0037 mm / 0.016°; 0/12 matrices identical as text: within the record’s thresholds (0.1 mm; 0.05 mm / 0.05°); mcflirt’s own variability is not measured for this run, so the difference is not yet set against it.

Bit-identical

Every file mcflirt writes (18/18), its exit status and both streams identical to mcflirt’s (FSL 6.0.7.17 as built), with the replica in reproduce.

Bit-identical

Every file mcflirt writes (18/18), its exit status and both streams identical to mcflirt’s (FSL 6.0.7.23 as built), with the replica in reproduce.

17/17 · 6/6 · 7/7
syn/cli/helpsynthetic: the command line-help alone
command linemcflirt -help
Rejected by both

mcflirt exit [1]: ; the clean room exit [1]: larmorx: upstream defect FSL-07 (mcflirt) reproduced: -help alone exits 1, as in mcflirt [LARMORX_UPSTREAM_DEFECTS=fix corrects it]

Rejected by both

mcflirt and the replica fail alike, with the same messages and status: mcflirt exits [1]; the same from the replica.

Rejected by both

mcflirt and the replica fail alike, with the same messages and status: mcflirt exits [1]; the same from the replica.

0/0 · 5/5 · 6/6
syn/cli/help-with-inputsynthetic: the command line-help after -in
command linemcflirt -in small.nii.gz -help
Open difference

Within the record’s thresholds (files and messages compared); no matrices to set against mcflirt’s own variability.

Bit-identical

Every file mcflirt writes (0/0), its exit status and both streams identical to mcflirt’s (FSL 6.0.7.17 as built), with the replica in reproduce.

Bit-identical

Every file mcflirt writes (0/0), its exit status and both streams identical to mcflirt’s (FSL 6.0.7.23 as built), with the replica in reproduce.

3/3 · 6/6 · 7/7
syn/cli/no-argumentssynthetic: the command lineno arguments
command linemcflirt
Rejected by both

mcflirt exit [1]: ; the clean room exit [1]:

Rejected by both

mcflirt and the replica fail alike, with the same messages and status: mcflirt exits [1]; the same from the replica.

Rejected by both

mcflirt and the replica fail alike, with the same messages and status: mcflirt exits [1]; the same from the replica.

0/0 · 5/5 · 6/6
syn/cli/plots-onlysynthetic: the command line-plots without -mats: is a .mat directory created
command linemcflirt -in small.nii.gz -out out_mcf.nii.gz -plots
Open difference

Within the record’s thresholds (median .par difference 0.0004 / 0.0098 (mm / °)); no matrices to set against mcflirt’s own variability.

Bit-identical

Every file mcflirt writes (2/2), its exit status and both streams identical to mcflirt’s (FSL 6.0.7.17 as built), with the replica in reproduce.

Bit-identical

Every file mcflirt writes (2/2), its exit status and both streams identical to mcflirt’s (FSL 6.0.7.23 as built), with the replica in reproduce.

6/6 · 7/7
syn/cli/refvol-minus1synthetic: the command line-refvol -1
command linemcflirt -in small.nii.gz -out out_mcf.nii.gz -mats -plots -rmsrel -rmsabs -refvol -1
Open difference

95th-percentile RMS deviation 0.0601 mm, median Δ 0.0019 mm / 0.0097°; 1/12 matrices identical as text: within the record’s thresholds (0.1 mm; 0.05 mm / 0.05°); mcflirt’s own variability is not measured for this run, so the difference is not yet set against it.

Bit-identical

Every file mcflirt writes (18/18), its exit status and both streams identical to mcflirt’s (FSL 6.0.7.17 as built), with the replica in reproduce.

Bit-identical

Every file mcflirt writes (18/18), its exit status and both streams identical to mcflirt’s (FSL 6.0.7.23 as built), with the replica in reproduce.

16/16 · 6/6 · 7/7
syn/cli/refvol-minus2synthetic: the command line-refvol -2
command linemcflirt -in small.nii.gz -out out_mcf.nii.gz -mats -plots -rmsrel -rmsabs -refvol -2
Rejected by both

mcflirt exit [-6]: Image Exception : #61 :: Invalid t index in [] operator terminate called after throwing an instance of ‘std::runtime_error’ what(): Invalid t index in [] operat; the clean room exit [1]: Error: -refvol -2: invalid volume index for a series of 12 v…

Rejected by both

mcflirt and the replica fail alike, with the same messages and status: mcflirt exits [-6]; the same from the replica.

Rejected by both

mcflirt and the replica fail alike, with the same messages and status: mcflirt exits [-6]; the same from the replica.

0/0 · 5/5 · 6/6
syn/cli/rmsrel-onlysynthetic: the command line-rmsrel without -mats: which files appear in the .mat directory
command linemcflirt -in small.nii.gz -out out_mcf.nii.gz -rmsrel
Open difference

95th-percentile RMS deviation 0 mm, median Δ 0 mm / 0°; 1/1 matrices identical as text: within the record’s thresholds (0.1 mm; 0.05 mm / 0.05°); mcflirt’s own variability is not measured for this run, so the difference is not yet set against it.

Bit-identical

Every file mcflirt writes (4/4), its exit status and both streams identical to mcflirt’s (FSL 6.0.7.17 as built), with the replica in reproduce.

Bit-identical

Every file mcflirt writes (4/4), its exit status and both streams identical to mcflirt’s (FSL 6.0.7.23 as built), with the replica in reproduce.

12/12 · 6/6 · 7/7
syn/cli/stages0synthetic: the command line-stages 0 with -reffile: the reference is never read
command linemcflirt -in small.nii.gz -out out_mcf.nii.gz -reffile ref.nii.gz -mats -plots -rmsrel -rmsabs -stages 0
Rejected by both

mcflirt exit [-6]: Image Exception : #8 :: Attempted to use affine transform with no voxels in vout terminate called after throwing an instance of ‘std::runtime_error’ what(): Att; the clean room exit [1]: Error: -stages 0 with a reference file: nothing would be reg…

Rejected by both

mcflirt and the replica fail alike, with the same messages and status: mcflirt exits [-6]; the same from the replica.

Rejected by both

mcflirt and the replica fail alike, with the same messages and status: mcflirt exits [-6]; the same from the replica.

0/0 · 5/5 · 6/6
syn/cli/unknown-option-middlesynthetic: the command linean unknown option followed by others
command linemcflirt -in small.nii.gz -edge -out out_mcf.nii.gz
Rejected by both

mcflirt exit [255]: Unrecognised option -edge; the clean room exit [255]: Unrecognised option -edge

Rejected by both

mcflirt and the replica fail alike, with the same messages and status: mcflirt exits [255]; the same from the replica.

Rejected by both

mcflirt and the replica fail alike, with the same messages and status: mcflirt exits [255]; the same from the replica.

0/0 · 5/5 · 6/6
syn/content/background/defaultsynthetic: image contentdefaults
command linemcflirt -in background.nii.gz -out out_mcf.nii.gz -mats -plots -rmsrel -rmsabs
Within the original's own variability

95th-percentile RMS deviation 0.0482 mm, median Δ 0.0079 mm / 0.0106°; 1/6 matrices identical as text, FD r 1; mcflirt’s own: 3.96 mm, 0.0122 mm / 0.0297°, FD r 1

Bit-identical

Every file mcflirt writes (12/12), its exit status and both streams identical to mcflirt’s (FSL 6.0.7.17 as built), with the replica in reproduce.

Bit-identical

Every file mcflirt writes (12/12), its exit status and both streams identical to mcflirt’s (FSL 6.0.7.23 as built), with the replica in reproduce.

16/16 · 6/6 · 7/7
syn/content/background/reffilesynthetic: image content-reffile phantom/ref
command linemcflirt -in background.nii.gz -out out_mcf.nii.gz -reffile ref.nii.gz -mats -plots -rmsrel -rmsabs
Open difference

95th-percentile RMS deviation 0.0694 mm, median Δ 0.0018 mm / 0.0045°; 1/6 matrices identical as text: within the record’s thresholds (0.1 mm; 0.05 mm / 0.05°); mcflirt’s own variability is not measured for this run, so the difference is not yet set against it.

Bit-identical

Every file mcflirt writes (12/12), its exit status and both streams identical to mcflirt’s (FSL 6.0.7.17 as built), with the replica in reproduce.

Bit-identical

Every file mcflirt writes (12/12), its exit status and both streams identical to mcflirt’s (FSL 6.0.7.23 as built), with the replica in reproduce.

17/17 · 6/6 · 7/7
syn/content/nan/defaultsynthetic: image contentdefaults
command linemcflirt -in nan.nii.gz -out out_mcf.nii.gz -mats -plots -rmsrel -rmsabs
Open difference

95th-percentile RMS deviation 0.0502 mm, median Δ 0.0016 mm / 0.0092°; 2/12 matrices identical as text: within the record’s thresholds (0.1 mm; 0.05 mm / 0.05°); mcflirt’s own variability is not measured for this run, so the difference is not yet set against it.

Bit-identical

Every file mcflirt writes (18/18), its exit status and both streams identical to mcflirt’s (FSL 6.0.7.17 as built), with the replica in reproduce.

Bit-identical

Every file mcflirt writes (18/18), its exit status and both streams identical to mcflirt’s (FSL 6.0.7.23 as built), with the replica in reproduce.

16/16 · 6/6 · 7/7
syn/content/nan/reffilesynthetic: image content-reffile phantom/ref
command linemcflirt -in nan.nii.gz -out out_mcf.nii.gz -reffile ref.nii.gz -mats -plots -rmsrel -rmsabs
Open difference

95th-percentile RMS deviation 0.0829 mm, median Δ 0.0042 mm / 0.0165°; 1/12 matrices identical as text: within the record’s thresholds (0.1 mm; 0.05 mm / 0.05°); mcflirt’s own variability is not measured for this run, so the difference is not yet set against it.

Bit-identical

Every file mcflirt writes (18/18), its exit status and both streams identical to mcflirt’s (FSL 6.0.7.17 as built), with the replica in reproduce.

Bit-identical

Every file mcflirt writes (18/18), its exit status and both streams identical to mcflirt’s (FSL 6.0.7.23 as built), with the replica in reproduce.

16/16 · 6/6 · 7/7
syn/content/ramp/reffilesynthetic: image contentramp moved along x: the default cost finds the shifts
command linemcflirt -in ramp.nii.gz -out out_mcf.nii.gz -reffile ramp-ref.nii.gz -mats -plots -rmsrel -rmsabs
Expected divergence

A ramp along x, which only the x translation can align (the other parameters are not constrained by the data: mcflirt’s own matrices move 11 to 17 mm at the 95th percentile when its inputs move in their last bits): the x translation ends 0.342 mm from mcflirt…

Bit-identical

Every file mcflirt writes (11/11), its exit status and both streams identical to mcflirt’s (FSL 6.0.7.17 as built), with the replica in reproduce.

Bit-identical

Every file mcflirt writes (11/11), its exit status and both streams identical to mcflirt’s (FSL 6.0.7.23 as built), with the replica in reproduce.

4/5 · 6/6 · 7/7
syn/content/ramp/reffile-smooth0synthetic: image contentramp with -smooth 0 (the unweighted cost)
command linemcflirt -in ramp.nii.gz -out out_mcf.nii.gz -reffile ramp-ref.nii.gz -mats -plots -rmsrel -rmsabs -smooth 0
Open difference

Within the record’s thresholds (files and messages compared); no matrices to set against mcflirt’s own variability.

Bit-identical

Every file mcflirt writes (11/11), its exit status and both streams identical to mcflirt’s (FSL 6.0.7.17 as built), with the replica in reproduce.

Bit-identical

Every file mcflirt writes (11/11), its exit status and both streams identical to mcflirt’s (FSL 6.0.7.23 as built), with the replica in reproduce.

5/5 · 6/6 · 7/7
syn/dtypes/float32/defaultsynthetic: data typesdefaults
command linemcflirt -in float32.nii.gz -out out_mcf.nii.gz -mats -plots -rmsrel -rmsabs
Open difference

95th-percentile RMS deviation 0.0601 mm, median Δ 0.0019 mm / 0.0097°; 1/12 matrices identical as text: within the record’s thresholds (0.1 mm; 0.05 mm / 0.05°); mcflirt’s own variability is not measured for this run, so the difference is not yet set against it.

Bit-identical

Every file mcflirt writes (18/18), its exit status and both streams identical to mcflirt’s (FSL 6.0.7.17 as built), with the replica in reproduce.

Bit-identical

Every file mcflirt writes (18/18), its exit status and both streams identical to mcflirt’s (FSL 6.0.7.23 as built), with the replica in reproduce.

16/16 · 6/6 · 7/7
syn/dtypes/float32/reffilesynthetic: data types-reffile phantom/ref
command linemcflirt -in float32.nii.gz -out out_mcf.nii.gz -reffile ref.nii.gz -mats -plots -rmsrel -rmsabs
Open difference

95th-percentile RMS deviation 0.0667 mm, median Δ 0.0037 mm / 0.016°; 0/12 matrices identical as text: within the record’s thresholds (0.1 mm; 0.05 mm / 0.05°); mcflirt’s own variability is not measured for this run, so the difference is not yet set against it.

Bit-identical

Every file mcflirt writes (18/18), its exit status and both streams identical to mcflirt’s (FSL 6.0.7.17 as built), with the replica in reproduce.

Bit-identical

Every file mcflirt writes (18/18), its exit status and both streams identical to mcflirt’s (FSL 6.0.7.23 as built), with the replica in reproduce.

16/16 · 6/6 · 7/7
syn/dtypes/float64/defaultsynthetic: data typesdefaults
command linemcflirt -in float64.nii.gz -out out_mcf.nii.gz -mats -plots -rmsrel -rmsabs
Open difference

95th-percentile RMS deviation 0.0601 mm, median Δ 0.0019 mm / 0.0097°; 1/12 matrices identical as text: within the record’s thresholds (0.1 mm; 0.05 mm / 0.05°); mcflirt’s own variability is not measured for this run, so the difference is not yet set against it.

Bit-identical

Every file mcflirt writes (18/18), its exit status and both streams identical to mcflirt’s (FSL 6.0.7.17 as built), with the replica in reproduce.

Bit-identical

Every file mcflirt writes (18/18), its exit status and both streams identical to mcflirt’s (FSL 6.0.7.23 as built), with the replica in reproduce.

16/16 · 6/6 · 7/7
syn/dtypes/float64/reffilesynthetic: data types-reffile phantom/ref
command linemcflirt -in float64.nii.gz -out out_mcf.nii.gz -reffile ref.nii.gz -mats -plots -rmsrel -rmsabs
Open difference

95th-percentile RMS deviation 0.0667 mm, median Δ 0.0037 mm / 0.016°; 0/12 matrices identical as text: within the record’s thresholds (0.1 mm; 0.05 mm / 0.05°); mcflirt’s own variability is not measured for this run, so the difference is not yet set against it.

Bit-identical

Every file mcflirt writes (18/18), its exit status and both streams identical to mcflirt’s (FSL 6.0.7.17 as built), with the replica in reproduce.

Bit-identical

Every file mcflirt writes (18/18), its exit status and both streams identical to mcflirt’s (FSL 6.0.7.23 as built), with the replica in reproduce.

16/16 · 6/6 · 7/7
syn/dtypes/int16-scaled/defaultsynthetic: data typesdefaults
command linemcflirt -in int16-scaled.nii.gz -out out_mcf.nii.gz -mats -plots -rmsrel -rmsabs
Open difference

95th-percentile RMS deviation 0.0601 mm, median Δ 0.0019 mm / 0.0097°; 1/12 matrices identical as text: within the record’s thresholds (0.1 mm; 0.05 mm / 0.05°); mcflirt’s own variability is not measured for this run, so the difference is not yet set against it.

Bit-identical

Every file mcflirt writes (18/18), its exit status and both streams identical to mcflirt’s (FSL 6.0.7.17 as built), with the replica in reproduce.

Bit-identical

Every file mcflirt writes (18/18), its exit status and both streams identical to mcflirt’s (FSL 6.0.7.23 as built), with the replica in reproduce.

16/16 · 6/6 · 7/7
syn/dtypes/int16-scaled/reffilesynthetic: data types-reffile phantom/ref
command linemcflirt -in int16-scaled.nii.gz -out out_mcf.nii.gz -reffile ref.nii.gz -mats -plots -rmsrel -rmsabs
Open difference

95th-percentile RMS deviation 0.0667 mm, median Δ 0.0037 mm / 0.016°; 0/12 matrices identical as text: within the record’s thresholds (0.1 mm; 0.05 mm / 0.05°); mcflirt’s own variability is not measured for this run, so the difference is not yet set against it.

Bit-identical

Every file mcflirt writes (18/18), its exit status and both streams identical to mcflirt’s (FSL 6.0.7.17 as built), with the replica in reproduce.

Bit-identical

Every file mcflirt writes (18/18), its exit status and both streams identical to mcflirt’s (FSL 6.0.7.23 as built), with the replica in reproduce.

16/16 · 6/6 · 7/7
syn/dtypes/uint16/defaultsynthetic: data typesdefaults
command linemcflirt -in uint16.nii.gz -out out_mcf.nii.gz -mats -plots -rmsrel -rmsabs
Open difference

95th-percentile RMS deviation 0.0601 mm, median Δ 0.0019 mm / 0.0097°; 1/12 matrices identical as text: within the record’s thresholds (0.1 mm; 0.05 mm / 0.05°); mcflirt’s own variability is not measured for this run, so the difference is not yet set against it.

Bit-identical

Every file mcflirt writes (18/18), its exit status and both streams identical to mcflirt’s (FSL 6.0.7.17 as built), with the replica in reproduce.

Bit-identical

Every file mcflirt writes (18/18), its exit status and both streams identical to mcflirt’s (FSL 6.0.7.23 as built), with the replica in reproduce.

16/16 · 6/6 · 7/7
syn/dtypes/uint16/reffilesynthetic: data types-reffile phantom/ref
command linemcflirt -in uint16.nii.gz -out out_mcf.nii.gz -reffile ref.nii.gz -mats -plots -rmsrel -rmsabs
Open difference

95th-percentile RMS deviation 0.0667 mm, median Δ 0.0037 mm / 0.016°; 0/12 matrices identical as text: within the record’s thresholds (0.1 mm; 0.05 mm / 0.05°); mcflirt’s own variability is not measured for this run, so the difference is not yet set against it.

Bit-identical

Every file mcflirt writes (18/18), its exit status and both streams identical to mcflirt’s (FSL 6.0.7.17 as built), with the replica in reproduce.

Bit-identical

Every file mcflirt writes (18/18), its exit status and both streams identical to mcflirt’s (FSL 6.0.7.23 as built), with the replica in reproduce.

16/16 · 6/6 · 7/7
syn/dtypes/uint8/defaultsynthetic: data typesdefaults
command linemcflirt -in uint8.nii.gz -out out_mcf.nii.gz -mats -plots -rmsrel -rmsabs
Open difference

95th-percentile RMS deviation 0.0374 mm, median Δ 0.0023 mm / 0.0119°; 1/12 matrices identical as text: within the record’s thresholds (0.1 mm; 0.05 mm / 0.05°); mcflirt’s own variability is not measured for this run, so the difference is not yet set against it.

Bit-identical

Every file mcflirt writes (18/18), its exit status and both streams identical to mcflirt’s (FSL 6.0.7.17 as built), with the replica in reproduce.

Bit-identical

Every file mcflirt writes (18/18), its exit status and both streams identical to mcflirt’s (FSL 6.0.7.23 as built), with the replica in reproduce.

16/16 · 6/6 · 7/7
syn/dtypes/uint8/reffilesynthetic: data types-reffile phantom/ref
command linemcflirt -in uint8.nii.gz -out out_mcf.nii.gz -reffile ref.nii.gz -mats -plots -rmsrel -rmsabs
Open difference

95th-percentile RMS deviation 0.0619 mm, median Δ 0.0037 mm / 0.0126°; 0/12 matrices identical as text: within the record’s thresholds (0.1 mm; 0.05 mm / 0.05°); mcflirt’s own variability is not measured for this run, so the difference is not yet set against it.

Bit-identical

Every file mcflirt writes (18/18), its exit status and both streams identical to mcflirt’s (FSL 6.0.7.17 as built), with the replica in reproduce.

Bit-identical

Every file mcflirt writes (18/18), its exit status and both streams identical to mcflirt’s (FSL 6.0.7.23 as built), with the replica in reproduce.

16/16 · 6/6 · 7/7
syn/geometry/single-volume-3d/defaultsynthetic: geometrydefaults
command linemcflirt -in single-volume-3d.nii.gz -out out_mcf.nii.gz -mats -plots -rmsrel -rmsabs
Open difference

95th-percentile RMS deviation 0 mm, median Δ 0 mm / 0°; 1/1 matrices identical as text: within the record’s thresholds (0.1 mm; 0.05 mm / 0.05°); mcflirt’s own variability is not measured for this run, so the difference is not yet set against it.

Bit-identical

Every file mcflirt writes (7/7), its exit status and both streams identical to mcflirt’s (FSL 6.0.7.17 as built), with the replica in reproduce.

Bit-identical

Every file mcflirt writes (7/7), its exit status and both streams identical to mcflirt’s (FSL 6.0.7.23 as built), with the replica in reproduce.

15/15 · 6/6 · 7/7
syn/geometry/single-volume-3d/reffilesynthetic: geometry-reffile phantom/ref
command linemcflirt -in single-volume-3d.nii.gz -out out_mcf.nii.gz -reffile ref.nii.gz -mats -plots -rmsrel -rmsabs
Open difference

95th-percentile RMS deviation 0.0435 mm, median Δ 0.0014 mm / 0.0155°; 0/1 matrices identical as text: within the record’s thresholds (0.1 mm; 0.05 mm / 0.05°); mcflirt’s own variability is not measured for this run, so the difference is not yet set against it.

Bit-identical

Every file mcflirt writes (7/7), its exit status and both streams identical to mcflirt’s (FSL 6.0.7.17 as built), with the replica in reproduce.

Bit-identical

Every file mcflirt writes (7/7), its exit status and both streams identical to mcflirt’s (FSL 6.0.7.23 as built), with the replica in reproduce.

15/15 · 6/6 · 7/7
syn/geometry/single-volume/defaultsynthetic: geometrydefaults
command linemcflirt -in single-volume.nii.gz -out out_mcf.nii.gz -mats -plots -rmsrel -rmsabs
Open difference

95th-percentile RMS deviation 0 mm, median Δ 0 mm / 0°; 1/1 matrices identical as text: within the record’s thresholds (0.1 mm; 0.05 mm / 0.05°); mcflirt’s own variability is not measured for this run, so the difference is not yet set against it.

Bit-identical

Every file mcflirt writes (7/7), its exit status and both streams identical to mcflirt’s (FSL 6.0.7.17 as built), with the replica in reproduce.

Bit-identical

Every file mcflirt writes (7/7), its exit status and both streams identical to mcflirt’s (FSL 6.0.7.23 as built), with the replica in reproduce.

15/15 · 6/6 · 7/7
syn/geometry/single-volume/reffilesynthetic: geometry-reffile phantom/ref
command linemcflirt -in single-volume.nii.gz -out out_mcf.nii.gz -reffile ref.nii.gz -mats -plots -rmsrel -rmsabs
Open difference

95th-percentile RMS deviation 0.0435 mm, median Δ 0.0014 mm / 0.0155°; 0/1 matrices identical as text: within the record’s thresholds (0.1 mm; 0.05 mm / 0.05°); mcflirt’s own variability is not measured for this run, so the difference is not yet set against it.

Bit-identical

Every file mcflirt writes (7/7), its exit status and both streams identical to mcflirt’s (FSL 6.0.7.17 as built), with the replica in reproduce.

Bit-identical

Every file mcflirt writes (7/7), its exit status and both streams identical to mcflirt’s (FSL 6.0.7.23 as built), with the replica in reproduce.

16/16 · 6/6 · 7/7
syn/geometry/thin-slab/defaultsynthetic: geometry16 mm slab: in-plane registration
command linemcflirt -in thin-slab.nii.gz -out out_mcf.nii.gz -mats -plots -rmsrel -rmsabs -report
Open difference

95th-percentile RMS deviation 0.0344 mm, median Δ 0.0014 mm / 0°; 1/8 matrices identical as text: within the record’s thresholds (0.1 mm; 0.05 mm / 0.05°); mcflirt’s own variability is not measured for this run, so the difference is not yet set against it.

Bit-identical

Every file mcflirt writes (14/14), its exit status and both streams identical to mcflirt’s (FSL 6.0.7.17 as built), with the replica in reproduce.

Bit-identical

Every file mcflirt writes (14/14), its exit status and both streams identical to mcflirt’s (FSL 6.0.7.23 as built), with the replica in reproduce.

17/17 · 6/6 · 7/7
syn/geometry/thin-slab/fov10synthetic: geometry-fov 10: the 8 mm reference grid still has fewer than 3 slices
command linemcflirt -in thin-slab.nii.gz -out out_mcf.nii.gz -mats -plots -rmsrel -rmsabs -fov 10
Open difference

95th-percentile RMS deviation 0.0344 mm, median Δ 0.0014 mm / 0°; 1/8 matrices identical as text: within the record’s thresholds (0.1 mm; 0.05 mm / 0.05°); mcflirt’s own variability is not measured for this run, so the difference is not yet set against it.

Bit-identical

Every file mcflirt writes (14/14), its exit status and both streams identical to mcflirt’s (FSL 6.0.7.17 as built), with the replica in reproduce.

Bit-identical

Every file mcflirt writes (14/14), its exit status and both streams identical to mcflirt’s (FSL 6.0.7.23 as built), with the replica in reproduce.

16/16 · 6/6 · 7/7
syn/geometry/thin-slab/rotation2synthetic: geometryin-plane mode prints the tolerances
command linemcflirt -in thin-slab.nii.gz -out out_mcf.nii.gz -mats -plots -rmsrel -rmsabs -rotation 2
Open difference

95th-percentile RMS deviation 0.0155 mm, median Δ 0 mm / 0°; 2/8 matrices identical as text: within the record’s thresholds (0.1 mm; 0.05 mm / 0.05°); mcflirt’s own variability is not measured for this run, so the difference is not yet set against it.

Bit-identical

Every file mcflirt writes (14/14), its exit status and both streams identical to mcflirt’s (FSL 6.0.7.17 as built), with the replica in reproduce.

Bit-identical

Every file mcflirt writes (14/14), its exit status and both streams identical to mcflirt’s (FSL 6.0.7.23 as built), with the replica in reproduce.

16/16 · 6/6 · 7/7
syn/geometry/tiny/defaultsynthetic: geometrysyn/geometry/tiny/default
command linemcflirt -in tiny.nii.gz -out out_mcf.nii.gz -mats -plots -rmsrel -rmsabs
Within the original's own variability

95th-percentile RMS deviation 0.119 mm, median Δ 0.003 mm / 0.0345°; 1/6 matrices identical as text, FD r 0.9994; mcflirt’s own: 0.36 mm, 0.00572 mm / 0.0992°, FD r 0.9963

Bit-identical

Every file mcflirt writes (12/12), its exit status and both streams identical to mcflirt’s (FSL 6.0.7.17 as built), with the replica in reproduce.

Bit-identical

Every file mcflirt writes (12/12), its exit status and both streams identical to mcflirt’s (FSL 6.0.7.23 as built), with the replica in reproduce.

16/16 · 6/6 · 7/7
syn/geometry/two-volumes/defaultsynthetic: geometrydefaults
command linemcflirt -in two-volumes.nii.gz -out out_mcf.nii.gz -mats -plots -rmsrel -rmsabs
Open difference

95th-percentile RMS deviation 0.0416 mm, median Δ 0.0002 mm / 0.0081°; 1/2 matrices identical as text: within the record’s thresholds (0.1 mm; 0.05 mm / 0.05°); mcflirt’s own variability is not measured for this run, so the difference is not yet set against it.

Bit-identical

Every file mcflirt writes (8/8), its exit status and both streams identical to mcflirt’s (FSL 6.0.7.17 as built), with the replica in reproduce.

Bit-identical

Every file mcflirt writes (8/8), its exit status and both streams identical to mcflirt’s (FSL 6.0.7.23 as built), with the replica in reproduce.

15/15 · 6/6 · 7/7
syn/geometry/two-volumes/reffilesynthetic: geometry-reffile phantom/ref
command linemcflirt -in two-volumes.nii.gz -out out_mcf.nii.gz -reffile ref.nii.gz -mats -plots -rmsrel -rmsabs
Open difference

95th-percentile RMS deviation 0.0425 mm, median Δ 0.0022 mm / 0.0068°; 0/2 matrices identical as text: within the record’s thresholds (0.1 mm; 0.05 mm / 0.05°); mcflirt’s own variability is not measured for this run, so the difference is not yet set against it.

Bit-identical

Every file mcflirt writes (8/8), its exit status and both streams identical to mcflirt’s (FSL 6.0.7.17 as built), with the replica in reproduce.

Bit-identical

Every file mcflirt writes (8/8), its exit status and both streams identical to mcflirt’s (FSL 6.0.7.23 as built), with the replica in reproduce.

16/16 · 6/6 · 7/7
syn/motion/identical-copies/defaultsynthetic: motiondefaults
command linemcflirt -in identical-copies.nii.gz -out out_mcf.nii.gz -mats -plots -rmsrel -rmsabs
Open difference

95th-percentile RMS deviation 0 mm, median Δ 0 mm / 0°; 6/6 matrices identical as text: within the record’s thresholds (0.1 mm; 0.05 mm / 0.05°); mcflirt’s own variability is not measured for this run, so the difference is not yet set against it.

Bit-identical

Every file mcflirt writes (12/12), its exit status and both streams identical to mcflirt’s (FSL 6.0.7.17 as built), with the replica in reproduce.

Bit-identical

Every file mcflirt writes (12/12), its exit status and both streams identical to mcflirt’s (FSL 6.0.7.23 as built), with the replica in reproduce.

16/16 · 6/6 · 7/7
syn/motion/identical-copies/reffilesynthetic: motion-reffile phantom/ref
command linemcflirt -in identical-copies.nii.gz -out out_mcf.nii.gz -reffile ref.nii.gz -mats -plots -rmsrel -rmsabs
Within the original's own variability

95th-percentile RMS deviation 0.0474 mm, median Δ 0.0032 mm / 0.0129°; 0/6 matrices identical as text; mcflirt’s own: 0.065 mm, 0.00401 mm / 0.0163°

Bit-identical

Every file mcflirt writes (12/12), its exit status and both streams identical to mcflirt’s (FSL 6.0.7.17 as built), with the replica in reproduce.

Bit-identical

Every file mcflirt writes (12/12), its exit status and both streams identical to mcflirt’s (FSL 6.0.7.23 as built), with the replica in reproduce.

17/17 · 6/6 · 7/7
syn/motion/identical-copies/reffile-fudgesynthetic: motion-fudge on identical copies: every volume starts from the identity
command linemcflirt -in identical-copies.nii.gz -out out_mcf.nii.gz -reffile ref.nii.gz -mats -plots -rmsrel -rmsabs -fudge
Open difference

95th-percentile RMS deviation 0.0218 mm, median Δ 0.0043 mm / 0.0116°; 0/6 matrices identical as text: within the record’s thresholds (0.1 mm; 0.05 mm / 0.05°); mcflirt’s own variability is not measured for this run, so the difference is not yet set against it.

Bit-identical

Every file mcflirt writes (12/12), its exit status and both streams identical to mcflirt’s (FSL 6.0.7.17 as built), with the replica in reproduce.

Bit-identical

Every file mcflirt writes (12/12), its exit status and both streams identical to mcflirt’s (FSL 6.0.7.23 as built), with the replica in reproduce.

17/17 · 6/6 · 7/7
syn/motion/large/defaultsynthetic: motiondefaults
command linemcflirt -in large.nii.gz -out out_mcf.nii.gz -mats -plots -rmsrel -rmsabs
Open difference

95th-percentile RMS deviation 0.0644 mm, median Δ 0.0035 mm / 0.0122°; 1/12 matrices identical as text: within the record’s thresholds (0.1 mm; 0.05 mm / 0.05°); mcflirt’s own variability is not measured for this run, so the difference is not yet set against it.

Bit-identical

Every file mcflirt writes (18/18), its exit status and both streams identical to mcflirt’s (FSL 6.0.7.17 as built), with the replica in reproduce.

Bit-identical

Every file mcflirt writes (18/18), its exit status and both streams identical to mcflirt’s (FSL 6.0.7.23 as built), with the replica in reproduce.

16/16 · 6/6 · 7/7
syn/motion/large/reffilesynthetic: motion-reffile phantom/ref
command linemcflirt -in large.nii.gz -out out_mcf.nii.gz -reffile ref.nii.gz -mats -plots -rmsrel -rmsabs
Open difference

95th-percentile RMS deviation 0.0729 mm, median Δ 0.0039 mm / 0.0192°; 0/12 matrices identical as text, FD r 0.9988: above mcflirt’s own variability (mcflirt’s own: 0.0887 mm, 0.0041 mm / 0.0157°, FD r 0.9991), within the record’s thresholds

Bit-identical

Every file mcflirt writes (18/18), its exit status and both streams identical to mcflirt’s (FSL 6.0.7.17 as built), with the replica in reproduce.

Bit-identical

Every file mcflirt writes (18/18), its exit status and both streams identical to mcflirt’s (FSL 6.0.7.23 as built), with the replica in reproduce.

17/17 · 6/6 · 7/7
syn/motion/small/2dsynthetic: motion-2d forces in-plane registration
command linemcflirt -in small.nii.gz -out out_mcf.nii.gz -reffile ref.nii.gz -mats -plots -rmsrel -rmsabs -2d -report
Open difference

95th-percentile RMS deviation 0.0514 mm, median Δ 0.0011 mm / 0°; 0/12 matrices identical as text: within the record’s thresholds (0.1 mm; 0.05 mm / 0.05°); mcflirt’s own variability is not measured for this run, so the difference is not yet set against it.

Bit-identical

Every file mcflirt writes (18/18), its exit status and both streams identical to mcflirt’s (FSL 6.0.7.17 as built), with the replica in reproduce.

Bit-identical

Every file mcflirt writes (18/18), its exit status and both streams identical to mcflirt’s (FSL 6.0.7.23 as built), with the replica in reproduce.

17/17 · 6/6 · 7/7
syn/motion/small/defaultsynthetic: motiondefaults
command linemcflirt -in small.nii.gz -out out_mcf.nii.gz -mats -plots -rmsrel -rmsabs
Open difference

95th-percentile RMS deviation 0.0601 mm, median Δ 0.0019 mm / 0.0097°; 1/12 matrices identical as text, FD r 0.9982: above mcflirt’s own variability (mcflirt’s own: 0.0558 mm, 0.00144 mm / 0.0094°, FD r 0.9974), within the record’s thresholds

Bit-identical

Every file mcflirt writes (18/18), its exit status and both streams identical to mcflirt’s (FSL 6.0.7.17 as built), with the replica in reproduce.

Bit-identical

Every file mcflirt writes (18/18), its exit status and both streams identical to mcflirt’s (FSL 6.0.7.23 as built), with the replica in reproduce.

16/16 · 6/6 · 7/7
syn/motion/small/default-reportsynthetic: motionprogress messages, middle-volume reference
command linemcflirt -in small.nii.gz -out out_mcf.nii.gz -mats -plots -rmsrel -rmsabs -report
Open difference

95th-percentile RMS deviation 0.0601 mm, median Δ 0.0019 mm / 0.0097°; 1/12 matrices identical as text: within the record’s thresholds (0.1 mm; 0.05 mm / 0.05°); mcflirt’s own variability is not measured for this run, so the difference is not yet set against it.

Bit-identical

Every file mcflirt writes (18/18), its exit status and both streams identical to mcflirt’s (FSL 6.0.7.17 as built), with the replica in reproduce.

Bit-identical

Every file mcflirt writes (18/18), its exit status and both streams identical to mcflirt’s (FSL 6.0.7.23 as built), with the replica in reproduce.

17/17 · 6/6 · 7/7
syn/motion/small/dof12synthetic: motionsyn/motion/small/dof12
command linemcflirt -in small.nii.gz -out out_mcf.nii.gz -reffile ref.nii.gz -mats -plots -rmsrel -rmsabs -dof 12
Within the original's own variability

95th-percentile RMS deviation 0.0699 mm, median Δ 0.0031 mm / 0.0152°; 0/12 matrices identical as text, FD r 0.9978; mcflirt’s own: 1.16 mm, 0.0278 mm / 0.0734°, FD r 0.8346

Bit-identical

Every file mcflirt writes (18/18), its exit status and both streams identical to mcflirt’s (FSL 6.0.7.17 as built), with the replica in reproduce.

Bit-identical

Every file mcflirt writes (18/18), its exit status and both streams identical to mcflirt’s (FSL 6.0.7.23 as built), with the replica in reproduce.

17/17 · 6/6 · 7/7
syn/motion/small/dof7synthetic: motionsyn/motion/small/dof7
command linemcflirt -in small.nii.gz -out out_mcf.nii.gz -reffile ref.nii.gz -mats -plots -rmsrel -rmsabs -dof 7
Open difference

95th-percentile RMS deviation 0.085 mm, median Δ 0.0027 mm / 0.0145°; 0/12 matrices identical as text, FD r 0.9991: above mcflirt’s own variability (mcflirt’s own: 0.0759 mm, 0.00405 mm / 0.0252°, FD r 0.9968), within the record’s thresholds

Bit-identical

Every file mcflirt writes (18/18), its exit status and both streams identical to mcflirt’s (FSL 6.0.7.17 as built), with the replica in reproduce.

Bit-identical

Every file mcflirt writes (18/18), its exit status and both streams identical to mcflirt’s (FSL 6.0.7.23 as built), with the replica in reproduce.

17/17 · 6/6 · 7/7
syn/motion/small/fsloutputtype-niftisynthetic: motionFSLOUTPUTTYPE=NIFTI with -out ending in .nii.gz
command linemcflirt -in small.nii.gz -out out_mcf.nii.gz -reffile ref.nii.gz -mats -plots -rmsrel -rmsabs
Open difference

95th-percentile RMS deviation 0.0667 mm, median Δ 0.0037 mm / 0.016°; 0/12 matrices identical as text: within the record’s thresholds (0.1 mm; 0.05 mm / 0.05°); mcflirt’s own variability is not measured for this run, so the difference is not yet set against it.

Bit-identical

Every file mcflirt writes (18/18), its exit status and both streams identical to mcflirt’s (FSL 6.0.7.17 as built), with the replica in reproduce.

Bit-identical

Every file mcflirt writes (18/18), its exit status and both streams identical to mcflirt’s (FSL 6.0.7.23 as built), with the replica in reproduce.

17/17 · 6/6 · 7/7
syn/motion/small/fudgesynthetic: motion-fudge: no initialisation from the previous volume
command linemcflirt -in small.nii.gz -out out_mcf.nii.gz -reffile ref.nii.gz -mats -plots -rmsrel -rmsabs -fudge
Open difference

95th-percentile RMS deviation 0.0613 mm, median Δ 0.0018 mm / 0.0109°; 0/12 matrices identical as text: within the record’s thresholds (0.1 mm; 0.05 mm / 0.05°); mcflirt’s own variability is not measured for this run, so the difference is not yet set against it.

Bit-identical

Every file mcflirt writes (18/18), its exit status and both streams identical to mcflirt’s (FSL 6.0.7.17 as built), with the replica in reproduce.

Bit-identical

Every file mcflirt writes (18/18), its exit status and both streams identical to mcflirt’s (FSL 6.0.7.23 as built), with the replica in reproduce.

17/17 · 6/6 · 7/7
syn/motion/small/initsynthetic: motion-init with a 5 mm x translation
command linemcflirt -in small.nii.gz -out out_mcf.nii.gz -reffile ref.nii.gz -mats -plots -rmsrel -rmsabs -init init.mat
Open difference

95th-percentile RMS deviation 0.0667 mm, median Δ 0.0037 mm / 0.016°; 0/12 matrices identical as text: within the record’s thresholds (0.1 mm; 0.05 mm / 0.05°); mcflirt’s own variability is not measured for this run, so the difference is not yet set against it.

Bit-identical

Every file mcflirt writes (18/18), its exit status and both streams identical to mcflirt’s (FSL 6.0.7.17 as built), with the replica in reproduce.

Bit-identical

Every file mcflirt writes (18/18), its exit status and both streams identical to mcflirt’s (FSL 6.0.7.23 as built), with the replica in reproduce.

17/17 · 6/6 · 7/7
syn/motion/small/meanvolsynthetic: motionsyn/motion/small/meanvol
command linemcflirt -in small.nii.gz -out out_mcf.nii.gz -mats -plots -rmsrel -rmsabs -meanvol
Within the original's own variability

95th-percentile RMS deviation 0.0557 mm, median Δ 0.0024 mm / 0.0118°; 0/12 matrices identical as text, FD r 0.9943; mcflirt’s own: 0.418 mm, 0.00533 mm / 0.0157°, FD r 0.9015

Bit-identical

Every file mcflirt writes (19/19), its exit status and both streams identical to mcflirt’s (FSL 6.0.7.17 as built), with the replica in reproduce.

Bit-identical

Every file mcflirt writes (19/19), its exit status and both streams identical to mcflirt’s (FSL 6.0.7.23 as built), with the replica in reproduce.

18/18 · 6/6 · 7/7
syn/motion/small/nifti2synthetic: motionFSLOUTPUTTYPE=NIFTI2_GZ
command linemcflirt -in small.nii.gz -out out_mcf.nii.gz -reffile ref.nii.gz -mats -plots -rmsrel -rmsabs
Open difference

95th-percentile RMS deviation 0.0667 mm, median Δ 0.0037 mm / 0.016°; 0/12 matrices identical as text: within the record’s thresholds (0.1 mm; 0.05 mm / 0.05°); mcflirt’s own variability is not measured for this run, so the difference is not yet set against it.

Bit-identical

Every file mcflirt writes (18/18), its exit status and both streams identical to mcflirt’s (FSL 6.0.7.17 as built), with the replica in reproduce.

Bit-identical

Every file mcflirt writes (18/18), its exit status and both streams identical to mcflirt’s (FSL 6.0.7.23 as built), with the replica in reproduce.

17/17 · 6/6 · 7/7
syn/motion/small/nn_finalsynthetic: motionsyn/motion/small/nn_final
command linemcflirt -in small.nii.gz -out out_mcf.nii.gz -reffile ref.nii.gz -mats -plots -rmsrel -rmsabs -nn_final
Open difference

95th-percentile RMS deviation 0.0667 mm, median Δ 0.0037 mm / 0.016°; 0/12 matrices identical as text: within the record’s thresholds (0.1 mm; 0.05 mm / 0.05°); mcflirt’s own variability is not measured for this run, so the difference is not yet set against it.

Bit-identical

Every file mcflirt writes (18/18), its exit status and both streams identical to mcflirt’s (FSL 6.0.7.17 as built), with the replica in reproduce.

Bit-identical

Every file mcflirt writes (18/18), its exit status and both streams identical to mcflirt’s (FSL 6.0.7.23 as built), with the replica in reproduce.

17/17 · 6/6 · 7/7
syn/motion/small/no-fsloutputtypesynthetic: motionFSLOUTPUTTYPE unset
command linemcflirt -in small.nii.gz -out out_mcf.nii.gz -reffile ref.nii.gz -mats
Rejected by both

Without FSLOUTPUTTYPE, mcflirt writes the .mat files and stops at its first image with FSL’s message (exit 1), and so does the clean room in reproduce (the same matrices within the thresholds); the clean room’s default, fix, writes every output, the images as .nii.gz, and name…

Rejected by both

mcflirt and the replica fail alike, with the same messages and status: mcflirt exits [1]; the same from the replica.

Rejected by both

mcflirt and the replica fail alike, with the same messages and status: mcflirt exits [1]; the same from the replica.

13/13 · 5/5 · 6/6
syn/motion/small/no-inputsynthetic: motionno -in
command linemcflirt -out out_mcf.nii.gz -mats
Rejected by both

mcflirt exit [2]: Input filename not found; the clean room exit [2]: Input filename not found

Rejected by both

mcflirt and the replica fail alike, with the same messages and status: mcflirt exits [2]; the same from the replica.

Rejected by both

mcflirt and the replica fail alike, with the same messages and status: mcflirt exits [2]; the same from the replica.

0/0 · 5/5 · 6/6
syn/motion/small/no-outsynthetic: motionno -out: the output name is <input>_mcf
command linemcflirt -in small.nii.gz -mats -plots
Open difference

95th-percentile RMS deviation 0.0601 mm, median Δ 0.0019 mm / 0.0097°; 1/12 matrices identical as text: within the record’s thresholds (0.1 mm; 0.05 mm / 0.05°); mcflirt’s own variability is not measured for this run, so the difference is not yet set against it.

Bit-identical

Every file mcflirt writes (14/14), its exit status and both streams identical to mcflirt’s (FSL 6.0.7.17 as built), with the replica in reproduce.

Bit-identical

Every file mcflirt writes (14/14), its exit status and both streams identical to mcflirt’s (FSL 6.0.7.23 as built), with the replica in reproduce.

12/12 · 6/6 · 7/7
syn/motion/small/out-no-extsynthetic: motion-out without an extension
command linemcflirt -in small.nii.gz -out plain -reffile ref.nii.gz -mats -plots -rmsrel -rmsabs
Open difference

95th-percentile RMS deviation 0.0667 mm, median Δ 0.0037 mm / 0.016°; 0/12 matrices identical as text: within the record’s thresholds (0.1 mm; 0.05 mm / 0.05°); mcflirt’s own variability is not measured for this run, so the difference is not yet set against it.

Bit-identical

Every file mcflirt writes (18/18), its exit status and both streams identical to mcflirt’s (FSL 6.0.7.17 as built), with the replica in reproduce.

Bit-identical

Every file mcflirt writes (18/18), its exit status and both streams identical to mcflirt’s (FSL 6.0.7.23 as built), with the replica in reproduce.

17/17 · 6/6 · 7/7
syn/motion/small/out-twicesynthetic: motionthe same -out twice: the second .mat directory gets a '+'
command linemcflirt -in small.nii.gz -out out_mcf.nii.gz -reffile ref.nii.gz -mats -plots -rmsrel -rmsabs
Open difference

95th-percentile RMS deviation 0.0667 mm, median Δ 0.0037 mm / 0.016°; 0/12 matrices identical as text: within the record’s thresholds (0.1 mm; 0.05 mm / 0.05°); mcflirt’s own variability is not measured for this run, so the difference is not yet set against it.

Bit-identical

Every file mcflirt writes (30/30), its exit status and both streams identical to mcflirt’s (FSL 6.0.7.17 as built), with the replica in reproduce.

Bit-identical

Every file mcflirt writes (30/30), its exit status and both streams identical to mcflirt’s (FSL 6.0.7.23 as built), with the replica in reproduce.

24/24 · 6/6 · 7/7
syn/motion/small/reffilesynthetic: motion-reffile phantom/ref
command linemcflirt -in small.nii.gz -out out_mcf.nii.gz -reffile ref.nii.gz -mats -plots -rmsrel -rmsabs
Within the original's own variability

95th-percentile RMS deviation 0.0667 mm, median Δ 0.0037 mm / 0.016°; 0/12 matrices identical as text, FD r 0.9981; mcflirt’s own: 0.107 mm, 0.00439 mm / 0.018°, FD r 0.9942

Bit-identical

Every file mcflirt writes (18/18), its exit status and both streams identical to mcflirt’s (FSL 6.0.7.17 as built), with the replica in reproduce.

Bit-identical

Every file mcflirt writes (18/18), its exit status and both streams identical to mcflirt’s (FSL 6.0.7.23 as built), with the replica in reproduce.

17/17 · 6/6 · 7/7
syn/motion/small/reffile-4dsynthetic: motiona 4D file as -reffile (its first volume is used)
command linemcflirt -in small.nii.gz -out out_mcf.nii.gz -reffile small.nii.gz -mats -plots -rmsrel -rmsabs
Open difference

95th-percentile RMS deviation 0.0447 mm, median Δ 0.0037 mm / 0.0127°; 1/12 matrices identical as text: within the record’s thresholds (0.1 mm; 0.05 mm / 0.05°); mcflirt’s own variability is not measured for this run, so the difference is not yet set against it.

Bit-identical

Every file mcflirt writes (18/18), its exit status and both streams identical to mcflirt’s (FSL 6.0.7.17 as built), with the replica in reproduce.

Bit-identical

Every file mcflirt writes (18/18), its exit status and both streams identical to mcflirt’s (FSL 6.0.7.23 as built), with the replica in reproduce.

16/16 · 6/6 · 7/7
syn/motion/small/reffile-grid2synthetic: motionreference on a different grid
command linemcflirt -in small.nii.gz -out out_mcf.nii.gz -reffile ref-grid2.nii.gz -mats -plots -rmsrel -rmsabs
Rejected by both

mcflirt exit [-6]: Image Exception : #2 :: Attempted to copydata with non-matching sizes terminate called after throwing an instance of ‘std::runtime_error’ what(): Attempted to c; the clean room exit [1]: Error: the reference has 57600 voxels and the volumes of the…

Rejected by both

mcflirt and the replica fail alike, with the same messages and status: mcflirt exits [-6]; the same from the replica.

Rejected by both

mcflirt and the replica fail alike, with the same messages and status: mcflirt exits [-6]; the same from the replica.

0/0 · 5/5 · 6/6
syn/motion/small/reffile-rassynthetic: motionreference stored RAS (x-flipped data): same physical image as phantom/ref
command linemcflirt -in small.nii.gz -out out_mcf.nii.gz -reffile ref-ras.nii.gz -mats -plots -rmsrel -rmsabs
Open difference

95th-percentile RMS deviation 0.0667 mm, median Δ 0.0037 mm / 0.016°; 0/12 matrices identical as text: within the record’s thresholds (0.1 mm; 0.05 mm / 0.05°); mcflirt’s own variability is not measured for this run, so the difference is not yet set against it.

Bit-identical

Every file mcflirt writes (18/18), its exit status and both streams identical to mcflirt’s (FSL 6.0.7.17 as built), with the replica in reproduce.

Bit-identical

Every file mcflirt writes (18/18), its exit status and both streams identical to mcflirt’s (FSL 6.0.7.23 as built), with the replica in reproduce.

16/16 · 6/6 · 7/7
syn/motion/small/refvol-0synthetic: motionsyn/motion/small/refvol-0
command linemcflirt -in small.nii.gz -out out_mcf.nii.gz -mats -plots -rmsrel -rmsabs -refvol 0
Open difference

95th-percentile RMS deviation 0.0447 mm, median Δ 0.0037 mm / 0.0127°; 1/12 matrices identical as text: within the record’s thresholds (0.1 mm; 0.05 mm / 0.05°); mcflirt’s own variability is not measured for this run, so the difference is not yet set against it.

Bit-identical

Every file mcflirt writes (18/18), its exit status and both streams identical to mcflirt’s (FSL 6.0.7.17 as built), with the replica in reproduce.

Bit-identical

Every file mcflirt writes (18/18), its exit status and both streams identical to mcflirt’s (FSL 6.0.7.23 as built), with the replica in reproduce.

16/16 · 6/6 · 7/7
syn/motion/small/refvol-out-of-rangesynthetic: motion-refvol 12 with 12 volumes
command linemcflirt -in small.nii.gz -out out_mcf.nii.gz -mats -plots -rmsrel -rmsabs -refvol 12
Rejected by both

mcflirt exit [-6]: Image Exception : #61 :: Invalid t index in [] operator terminate called after throwing an instance of ‘std::runtime_error’ what(): Invalid t index in [] operat; the clean room exit [1]: Error: -refvol 12: invalid volume index for a series of 12 v…

Rejected by both

mcflirt and the replica fail alike, with the same messages and status: mcflirt exits [-6]; the same from the replica.

Rejected by both

mcflirt and the replica fail alike, with the same messages and status: mcflirt exits [-6]; the same from the replica.

0/0 · 5/5 · 6/6
syn/motion/small/reportsynthetic: motionprogress messages
command linemcflirt -in small.nii.gz -out out_mcf.nii.gz -reffile ref.nii.gz -mats -plots -rmsrel -rmsabs -report
Open difference

95th-percentile RMS deviation 0.0667 mm, median Δ 0.0037 mm / 0.016°; 0/12 matrices identical as text: within the record’s thresholds (0.1 mm; 0.05 mm / 0.05°); mcflirt’s own variability is not measured for this run, so the difference is not yet set against it.

Bit-identical

Every file mcflirt writes (18/18), its exit status and both streams identical to mcflirt’s (FSL 6.0.7.17 as built), with the replica in reproduce.

Bit-identical

Every file mcflirt writes (18/18), its exit status and both streams identical to mcflirt’s (FSL 6.0.7.23 as built), with the replica in reproduce.

17/17 · 6/6 · 7/7
syn/motion/small/sinc_finalsynthetic: motionsyn/motion/small/sinc_final
command linemcflirt -in small.nii.gz -out out_mcf.nii.gz -reffile ref.nii.gz -mats -plots -rmsrel -rmsabs -sinc_final
Open difference

95th-percentile RMS deviation 0.0667 mm, median Δ 0.0037 mm / 0.016°; 0/12 matrices identical as text: within the record’s thresholds (0.1 mm; 0.05 mm / 0.05°); mcflirt’s own variability is not measured for this run, so the difference is not yet set against it.

Bit-identical

Every file mcflirt writes (18/18), its exit status and both streams identical to mcflirt’s (FSL 6.0.7.17 as built), with the replica in reproduce.

Bit-identical

Every file mcflirt writes (18/18), its exit status and both streams identical to mcflirt’s (FSL 6.0.7.23 as built), with the replica in reproduce.

17/17 · 6/6 · 7/7
syn/motion/small/smooth0synthetic: motionsyn/motion/small/smooth0
command linemcflirt -in small.nii.gz -out out_mcf.nii.gz -reffile ref.nii.gz -mats -plots -rmsrel -rmsabs -smooth 0
Open difference

95th-percentile RMS deviation 0.0585 mm, median Δ 0.0029 mm / 0.0151°; 0/12 matrices identical as text: within the record’s thresholds (0.1 mm; 0.05 mm / 0.05°); mcflirt’s own variability is not measured for this run, so the difference is not yet set against it.

Bit-identical

Every file mcflirt writes (18/18), its exit status and both streams identical to mcflirt’s (FSL 6.0.7.17 as built), with the replica in reproduce.

Bit-identical

Every file mcflirt writes (18/18), its exit status and both streams identical to mcflirt’s (FSL 6.0.7.23 as built), with the replica in reproduce.

17/17 · 6/6 · 7/7
syn/motion/small/spline_finalsynthetic: motionsyn/motion/small/spline_final
command linemcflirt -in small.nii.gz -out out_mcf.nii.gz -reffile ref.nii.gz -mats -plots -rmsrel -rmsabs -spline_final
Open difference

95th-percentile RMS deviation 0.0667 mm, median Δ 0.0037 mm / 0.016°; 0/12 matrices identical as text: within the record’s thresholds (0.1 mm; 0.05 mm / 0.05°); mcflirt’s own variability is not measured for this run, so the difference is not yet set against it.

Bit-identical

Every file mcflirt writes (18/18), its exit status and both streams identical to mcflirt’s (FSL 6.0.7.17 as built), with the replica in reproduce.

Bit-identical

Every file mcflirt writes (18/18), its exit status and both streams identical to mcflirt’s (FSL 6.0.7.23 as built), with the replica in reproduce.

17/17 · 6/6 · 7/7
syn/motion/small/stages1synthetic: motionsyn/motion/small/stages1
command linemcflirt -in small.nii.gz -out out_mcf.nii.gz -reffile ref.nii.gz -mats -plots -rmsrel -rmsabs -stages 1
Open difference

95th-percentile RMS deviation 0.076 mm, median Δ 0.0011 mm / 0.0011°; 2/12 matrices identical as text: within the record’s thresholds (0.1 mm; 0.05 mm / 0.05°); mcflirt’s own variability is not measured for this run, so the difference is not yet set against it.

Bit-identical

Every file mcflirt writes (18/18), its exit status and both streams identical to mcflirt’s (FSL 6.0.7.17 as built), with the replica in reproduce.

Bit-identical

Every file mcflirt writes (18/18), its exit status and both streams identical to mcflirt’s (FSL 6.0.7.23 as built), with the replica in reproduce.

17/17 · 6/6 · 7/7
syn/motion/small/stages4synthetic: motionsyn/motion/small/stages4
command linemcflirt -in small.nii.gz -out out_mcf.nii.gz -reffile ref.nii.gz -mats -plots -rmsrel -rmsabs -stages 4
Open difference

95th-percentile RMS deviation 0.068 mm, median Δ 0.0037 mm / 0.0115°; 0/12 matrices identical as text: within the record’s thresholds (0.1 mm; 0.05 mm / 0.05°); mcflirt’s own variability is not measured for this run, so the difference is not yet set against it.

Bit-identical

Every file mcflirt writes (18/18), its exit status and both streams identical to mcflirt’s (FSL 6.0.7.17 as built), with the replica in reproduce.

Bit-identical

Every file mcflirt writes (18/18), its exit status and both streams identical to mcflirt’s (FSL 6.0.7.23 as built), with the replica in reproduce.

17/17 · 6/6 · 7/7
syn/motion/small/statssynthetic: motionsyn/motion/small/stats
command linemcflirt -in small.nii.gz -out out_mcf.nii.gz -reffile ref.nii.gz -mats -plots -rmsrel -rmsabs -stats
Open difference

95th-percentile RMS deviation 0.0667 mm, median Δ 0.0037 mm / 0.016°; 0/12 matrices identical as text: within the record’s thresholds (0.1 mm; 0.05 mm / 0.05°); mcflirt’s own variability is not measured for this run, so the difference is not yet set against it.

Bit-identical

Every file mcflirt writes (21/21), its exit status and both streams identical to mcflirt’s (FSL 6.0.7.17 as built), with the replica in reproduce.

Bit-identical

Every file mcflirt writes (21/21), its exit status and both streams identical to mcflirt’s (FSL 6.0.7.23 as built), with the replica in reproduce.

23/23 · 6/6 · 7/7
syn/motion/small/unknown-costsynthetic: motionan unknown cost name
command linemcflirt -in small.nii.gz -out out_mcf.nii.gz -reffile ref.nii.gz -mats -plots -rmsrel -rmsabs -cost bogus
Open difference

95th-percentile RMS deviation 0.0667 mm, median Δ 0.0037 mm / 0.016°; 0/12 matrices identical as text: within the record’s thresholds (0.1 mm; 0.05 mm / 0.05°); mcflirt’s own variability is not measured for this run, so the difference is not yet set against it.

Bit-identical

Every file mcflirt writes (18/18), its exit status and both streams identical to mcflirt’s (FSL 6.0.7.17 as built), with the replica in reproduce.

Bit-identical

Every file mcflirt writes (18/18), its exit status and both streams identical to mcflirt’s (FSL 6.0.7.23 as built), with the replica in reproduce.

17/17 · 6/6 · 7/7
syn/motion/small/unknown-optionsynthetic: motion-edge (offered by nipype) is not an mcflirt option
command linemcflirt -in small.nii.gz -out out_mcf.nii.gz -reffile ref.nii.gz -mats -edge
Rejected by both

mcflirt exit [255]: Lacking argument to option -edge; the clean room exit [255]: Lacking argument to option -edge

Rejected by both

mcflirt and the replica fail alike, with the same messages and status: mcflirt exits [255]; the same from the replica.

Rejected by both

mcflirt and the replica fail alike, with the same messages and status: mcflirt exits [255]; the same from the replica.

0/0 · 5/5 · 6/6
syn/orientation/oblique/defaultsynthetic: orientationdefaults
command linemcflirt -in oblique.nii.gz -out out_mcf.nii.gz -mats -plots -rmsrel -rmsabs
Open difference

95th-percentile RMS deviation 0.0601 mm, median Δ 0.0019 mm / 0.0097°; 1/12 matrices identical as text: within the record’s thresholds (0.1 mm; 0.05 mm / 0.05°); mcflirt’s own variability is not measured for this run, so the difference is not yet set against it.

Bit-identical

Every file mcflirt writes (18/18), its exit status and both streams identical to mcflirt’s (FSL 6.0.7.17 as built), with the replica in reproduce.

Bit-identical

Every file mcflirt writes (18/18), its exit status and both streams identical to mcflirt’s (FSL 6.0.7.23 as built), with the replica in reproduce.

16/16 · 6/6 · 7/7
syn/orientation/oblique/reffilesynthetic: orientation-reffile phantom/ref
command linemcflirt -in oblique.nii.gz -out out_mcf.nii.gz -reffile ref.nii.gz -mats -plots -rmsrel -rmsabs
Open difference

95th-percentile RMS deviation 0.0667 mm, median Δ 0.0037 mm / 0.016°; 0/12 matrices identical as text: within the record’s thresholds (0.1 mm; 0.05 mm / 0.05°); mcflirt’s own variability is not measured for this run, so the difference is not yet set against it.

Bit-identical

Every file mcflirt writes (18/18), its exit status and both streams identical to mcflirt’s (FSL 6.0.7.17 as built), with the replica in reproduce.

Bit-identical

Every file mcflirt writes (18/18), its exit status and both streams identical to mcflirt’s (FSL 6.0.7.23 as built), with the replica in reproduce.

16/16 · 6/6 · 7/7
syn/orientation/ras-unflipped/defaultsynthetic: orientationdefaults
command linemcflirt -in ras-unflipped.nii.gz -out out_mcf.nii.gz -mats -plots -rmsrel -rmsabs
Open difference

95th-percentile RMS deviation 0.0723 mm, median Δ 0.0029 mm / 0.0095°; 1/12 matrices identical as text: within the record’s thresholds (0.1 mm; 0.05 mm / 0.05°); mcflirt’s own variability is not measured for this run, so the difference is not yet set against it.

Bit-identical

Every file mcflirt writes (18/18), its exit status and both streams identical to mcflirt’s (FSL 6.0.7.17 as built), with the replica in reproduce.

Bit-identical

Every file mcflirt writes (18/18), its exit status and both streams identical to mcflirt’s (FSL 6.0.7.23 as built), with the replica in reproduce.

16/16 · 6/6 · 7/7
syn/orientation/ras-unflipped/reffilesynthetic: orientation-reffile phantom/ref
command linemcflirt -in ras-unflipped.nii.gz -out out_mcf.nii.gz -reffile ref.nii.gz -mats -plots -rmsrel -rmsabs
Expected divergence

The clean room’s evaluated matrices are mcflirt’s for the first 41 evaluations (count_trace.py, mcflirt’s -verbose 20), then part in a translation’s 6th decimal and end 0.110 mm from mcflirt’s at the 95th percentile, beyond the 0.1 mm made for real BOLD runs but wit…

Bit-identical

Every file mcflirt writes (18/18), its exit status and both streams identical to mcflirt’s (FSL 6.0.7.17 as built), with the replica in reproduce.

Bit-identical

Every file mcflirt writes (18/18), its exit status and both streams identical to mcflirt’s (FSL 6.0.7.23 as built), with the replica in reproduce.

15/16 · 6/6 · 7/7
syn/orientation/ras/defaultsynthetic: orientationdefaults
command linemcflirt -in ras.nii.gz -out out_mcf.nii.gz -mats -plots -rmsrel -rmsabs
Open difference

95th-percentile RMS deviation 0.0601 mm, median Δ 0.0019 mm / 0.0097°; 1/12 matrices identical as text: within the record’s thresholds (0.1 mm; 0.05 mm / 0.05°); mcflirt’s own variability is not measured for this run, so the difference is not yet set against it.

Bit-identical

Every file mcflirt writes (18/18), its exit status and both streams identical to mcflirt’s (FSL 6.0.7.17 as built), with the replica in reproduce.

Bit-identical

Every file mcflirt writes (18/18), its exit status and both streams identical to mcflirt’s (FSL 6.0.7.23 as built), with the replica in reproduce.

16/16 · 6/6 · 7/7
syn/orientation/ras/reffilesynthetic: orientation-reffile phantom/ref
command linemcflirt -in ras.nii.gz -out out_mcf.nii.gz -reffile ref.nii.gz -mats -plots -rmsrel -rmsabs
Open difference

95th-percentile RMS deviation 0.0667 mm, median Δ 0.0037 mm / 0.016°; 0/12 matrices identical as text: within the record’s thresholds (0.1 mm; 0.05 mm / 0.05°); mcflirt’s own variability is not measured for this run, so the difference is not yet set against it.

Bit-identical

Every file mcflirt writes (18/18), its exit status and both streams identical to mcflirt’s (FSL 6.0.7.17 as built), with the replica in reproduce.

Bit-identical

Every file mcflirt writes (18/18), its exit status and both streams identical to mcflirt’s (FSL 6.0.7.23 as built), with the replica in reproduce.

17/17 · 6/6 · 7/7
failures/fsl20-matsfailure cases (specs/mcflirt.md §14.1)-mats in a directory that does not exist (FSL-20)Open difference

Within the record’s thresholds (files and messages compared); no matrices to set against mcflirt’s own variability.

–Bit-identical

Every file mcflirt writes, its exit status and both streams identical to mcflirt’s (FSL 6.0.7.23 as built), with the replica in reproduce.

8/8
failures/fsl20-mats-plotsfailure cases (specs/mcflirt.md §14.1)-mats -plots in a directory that does not exist (FSL-20)Open difference

Within the record’s thresholds (files and messages compared); no matrices to set against mcflirt’s own variability.

–Bit-identical

Every file mcflirt writes, its exit status and both streams identical to mcflirt’s (FSL 6.0.7.23 as built), with the replica in reproduce.

8/8
failures/fsl20-reffilefailure cases (specs/mcflirt.md §14.1)-mats with -reffile in a missing directory (FSL-20)Open difference

Within the record’s thresholds (files and messages compared); no matrices to set against mcflirt’s own variability.

–Bit-identical

Every file mcflirt writes, its exit status and both streams identical to mcflirt’s (FSL 6.0.7.23 as built), with the replica in reproduce.

8/8
failures/fsl20-file-parentfailure cases (specs/mcflirt.md §14.1)-mats where the directory is a file (FSL-20)Open difference

Within the record’s thresholds (files and messages compared); no matrices to set against mcflirt’s own variability.

–Bit-identical

Every file mcflirt writes, its exit status and both streams identical to mcflirt’s (FSL 6.0.7.23 as built), with the replica in reproduce.

8/8
failures/fsl20-rmsrelfailure cases (specs/mcflirt.md §14.1)-rmsrel without -mats in a missing directory (FSL-13, FSL-20)Open difference

Within the record’s thresholds (files and messages compared); no matrices to set against mcflirt’s own variability.

–Bit-identical

Every file mcflirt writes, its exit status and both streams identical to mcflirt’s (FSL 6.0.7.23 as built), with the replica in reproduce.

8/8
failures/fsl20-rmsabs-reportfailure cases (specs/mcflirt.md §14.1)-rmsabs -report without -mats in a missing directory (FSL-13, FSL-20)Open difference

Within the record’s thresholds (files and messages compared); no matrices to set against mcflirt’s own variability.

–Bit-identical

Every file mcflirt writes, its exit status and both streams identical to mcflirt’s (FSL 6.0.7.23 as built), with the replica in reproduce.

8/8
failures/fsl20-permfailure cases (specs/mcflirt.md §14.1)-mats in a read-only directory: mcflirt's permission message, then it abortsRejected by both

mcflirt ends by signal 6 (Error! Could not create directory: //ro/o.mat. No write permission); the clean room exits 1 (reproduce) and 1 (default) with its error

–Rejected by both

mcflirt and the replica fail alike, with the same messages and status: mcflirt ends by signal 6 (Error! Could not create directory: //ro/o.mat. No write permission); the replica exits 1 (reproduce) and 1 (default) with its error

7/7
failures/fsl20-plots-onlyfailure cases (specs/mcflirt.md §14.1)-plots alone in a missing directory (FSL-65)Open difference

Within the record’s thresholds (files and messages compared); no matrices to set against mcflirt’s own variability.

–Bit-identical

Every file mcflirt writes, its exit status and both streams identical to mcflirt’s (FSL 6.0.7.23 as built), with the replica in reproduce.

8/8
failures/fsl20-nonefailure cases (specs/mcflirt.md §14.1)no output option, in a missing directory: the series cannot be writtenRejected by both

mcflirt ends by signal 6 (terminate called after throwing an instance of ‘NiftiIO::NiftiException’); the clean room exits 1 (reproduce) and 1 (default) with its error

–Rejected by both

mcflirt and the replica fail alike, with the same messages and status: mcflirt ends by signal 6 (terminate called after throwing an instance of ‘NiftiIO::NiftiException’); the replica exits 1 (reproduce) and 1 (default) with its error

7/7
failures/par-dir-plots-matsfailure cases (specs/mcflirt.md §14.1)a directory named o.par, with -mats -plots -rmsrel (FSL-65)Open difference

95th-percentile RMS deviation 0.005 mm, median Δ 0 mm / 0°: within the record’s thresholds (0.1 mm; 0.05 mm / 0.05°); mcflirt’s own variability is not measured for this run, so the difference is not yet set against it.

–Bit-identical

Every file mcflirt writes, its exit status and both streams identical to mcflirt’s (FSL 6.0.7.23 as built), with the replica in reproduce.

9/9
failures/par-dir-plotsfailure cases (specs/mcflirt.md §14.1)a directory named o.par, -plots (FSL-65)Open difference

Within the record’s thresholds (files and messages compared); no matrices to set against mcflirt’s own variability.

–Bit-identical

Every file mcflirt writes, its exit status and both streams identical to mcflirt’s (FSL 6.0.7.23 as built), with the replica in reproduce.

8/8
failures/par-dir-plots-mats-ref0failure cases (specs/mcflirt.md §14.1)a directory named o.par, -mats -plots -refvol 0 (FSL-65)Open difference

95th-percentile RMS deviation 0 mm, median Δ 0 mm / 0°: within the record’s thresholds (0.1 mm; 0.05 mm / 0.05°); mcflirt’s own variability is not measured for this run, so the difference is not yet set against it.

–Bit-identical

Every file mcflirt writes, its exit status and both streams identical to mcflirt’s (FSL 6.0.7.23 as built), with the replica in reproduce.

9/9
failures/rms-dirfailure cases (specs/mcflirt.md §14.1)directories named o_rel.rms and o_abs_mean.rms (FSL-65)Open difference

95th-percentile RMS deviation 0.004 mm, median Δ 0 mm / 0°: within the record’s thresholds (0.1 mm; 0.05 mm / 0.05°); mcflirt’s own variability is not measured for this run, so the difference is not yet set against it.

–Bit-identical

Every file mcflirt writes, its exit status and both streams identical to mcflirt’s (FSL 6.0.7.23 as built), with the replica in reproduce.

9/9
failures/bins7-mutualinfofailure cases (specs/mcflirt.md §14.1)mutualinfo with -bins 7 (FSL-21)Open difference

Within the record’s thresholds (files and messages compared); no matrices to set against mcflirt’s own variability.

–Bit-identical

Every file mcflirt writes, its exit status and both streams identical to mcflirt’s (FSL 6.0.7.23 as built), with the replica in reproduce.

9/9
failures/bins4-mutualinfo-s2failure cases (specs/mcflirt.md §14.1)mutualinfo, -bins 4, two stages (FSL-21)Open difference

Within the record’s thresholds (files and messages compared); no matrices to set against mcflirt’s own variability.

–Bit-identical

Every file mcflirt writes, its exit status and both streams identical to mcflirt’s (FSL 6.0.7.23 as built), with the replica in reproduce.

9/9
failures/bins4-mutualinfo-s3-plotsfailure cases (specs/mcflirt.md §14.1)mutualinfo, -bins 4, three stages, -plots (FSL-21)Open difference

Within the record’s thresholds (files and messages compared); no matrices to set against mcflirt’s own variability.

–Bit-identical

Every file mcflirt writes, its exit status and both streams identical to mcflirt’s (FSL 6.0.7.23 as built), with the replica in reproduce.

9/9
failures/bins0-normmifailure cases (specs/mcflirt.md §14.1)normmi with -bins 0 (FSL-21)Open difference

Within the record’s thresholds (files and messages compared); no matrices to set against mcflirt’s own variability.

–Bit-identical

Every file mcflirt writes, its exit status and both streams identical to mcflirt’s (FSL 6.0.7.23 as built), with the replica in reproduce.

9/9
failures/bins1-woodsfailure cases (specs/mcflirt.md §14.1)woods with -bins 1 (FSL-21)Open difference

Within the record’s thresholds (files and messages compared); no matrices to set against mcflirt’s own variability.

–Bit-identical

Every file mcflirt writes, its exit status and both streams identical to mcflirt’s (FSL 6.0.7.23 as built), with the replica in reproduce.

9/9
failures/binsabc-mutualinfofailure cases (specs/mcflirt.md §14.1)-bins abc, read as 0 (FSL-07; FSL-21 reproduced)Open difference

Within the record’s thresholds (files and messages compared); no matrices to set against mcflirt’s own variability.

–Bit-identical

Every file mcflirt writes, its exit status and both streams identical to mcflirt’s (FSL 6.0.7.23 as built), with the replica in reproduce.

9/9
failures/bins2-corratiofailure cases (specs/mcflirt.md §14.1)corratio with -bins 2: mcflirt aborts (FSL-21)Rejected by both

mcflirt ends by signal 6 (double free or corruption (out)); the clean room exits 1 (reproduce) and 1 (default) with its error

–Rejected by both

mcflirt and the replica fail alike, with the same messages and status: mcflirt ends by signal 6 (double free or corruption (out)); the replica exits 1 (reproduce) and 1 (default) with its error

8/8
failures/bins2-corratio-smooth0failure cases (specs/mcflirt.md §14.1)corratio, -bins 2, -smooth 0 (FSL-21)Rejected by both

mcflirt ends by signal 6 (double free or corruption (out)); the clean room exits 1 (reproduce) and 1 (default) with its error

–Rejected by both

mcflirt and the replica fail alike, with the same messages and status: mcflirt ends by signal 6 (double free or corruption (out)); the replica exits 1 (reproduce) and 1 (default) with its error

8/8
failures/bins2-corratio-reportfailure cases (specs/mcflirt.md §14.1)corratio, -bins 2, -report (FSL-21)Rejected by both

mcflirt ends by signal 6 (Processed data will be saved as /o); the clean room exits 1 (reproduce) and 1 (default) with its error

–Rejected by both

mcflirt and the replica fail alike, with the same messages and status: mcflirt ends by signal 6 (Processed data will be saved as /o); the replica exits 1 (reproduce) and 1 (default) with its error

8/8
failures/bins8-corratiofailure cases (specs/mcflirt.md §14.1)corratio with -bins 8: one bin in stage 1Open difference

95th-percentile RMS deviation 0.0416 mm, median Δ 0 mm / 0°: within the record’s thresholds (0.1 mm; 0.05 mm / 0.05°); mcflirt’s own variability is not measured for this run, so the difference is not yet set against it.

–Bit-identical

Every file mcflirt writes, its exit status and both streams identical to mcflirt’s (FSL 6.0.7.23 as built), with the replica in reproduce.

7/7
failures/bins8-mutualinfo-s3failure cases (specs/mcflirt.md §14.1)mutualinfo with -bins 8: one bin in stage 1 (a cost of rounding noise)Open difference

Within the record’s thresholds (files and messages compared); no matrices to set against mcflirt’s own variability.

–Bit-identical

Every file mcflirt writes, its exit status and both streams identical to mcflirt’s (FSL 6.0.7.23 as built), with the replica in reproduce.

7/7
failures/bins2-normcorrfailure cases (specs/mcflirt.md §14.1)normcorr with -bins 2 (it uses no bins)Open difference

95th-percentile RMS deviation 0.0425 mm, median Δ 0 mm / 0°: within the record’s thresholds (0.1 mm; 0.05 mm / 0.05°); mcflirt’s own variability is not measured for this run, so the difference is not yet set against it.

–Bit-identical

Every file mcflirt writes, its exit status and both streams identical to mcflirt’s (FSL 6.0.7.23 as built), with the replica in reproduce.

7/7
failures/bins2-leastsqfailure cases (specs/mcflirt.md §14.1)leastsquares with -bins 2 (it uses no bins)Open difference

95th-percentile RMS deviation 0.0001 mm, median Δ 0 mm / 0°: within the record’s thresholds (0.1 mm; 0.05 mm / 0.05°); mcflirt’s own variability is not measured for this run, so the difference is not yet set against it.

–Bit-identical

Every file mcflirt writes, its exit status and both streams identical to mcflirt’s (FSL 6.0.7.23 as built), with the replica in reproduce.

7/7
failures/binsm1-normcorrfailure cases (specs/mcflirt.md §14.1)normcorr with -bins -1: no bins anywhere (FSL-66)Open difference

95th-percentile RMS deviation 0.0425 mm, median Δ 0 mm / 0°: within the record’s thresholds (0.1 mm; 0.05 mm / 0.05°); mcflirt’s own variability is not measured for this run, so the difference is not yet set against it.

–Bit-identical

Every file mcflirt writes, its exit status and both streams identical to mcflirt’s (FSL 6.0.7.23 as built), with the replica in reproduce.

9/9
failures/bins4-mutualinfo-s0failure cases (specs/mcflirt.md §14.1)mutualinfo, -bins 4, -stages 0: no stage runsOpen difference

95th-percentile RMS deviation 0 mm, median Δ 0 mm / 0°: within the record’s thresholds (0.1 mm; 0.05 mm / 0.05°); mcflirt’s own variability is not measured for this run, so the difference is not yet set against it.

–Bit-identical

Every file mcflirt writes, its exit status and both streams identical to mcflirt’s (FSL 6.0.7.23 as built), with the replica in reproduce.

7/7
failures/bins4-mutualinfo-1volfailure cases (specs/mcflirt.md §14.1)mutualinfo, -bins 4, one volume: nothing is registeredOpen difference

95th-percentile RMS deviation 0 mm, median Δ 0 mm / 0°: within the record’s thresholds (0.1 mm; 0.05 mm / 0.05°); mcflirt’s own variability is not measured for this run, so the difference is not yet set against it.

–Bit-identical

Every file mcflirt writes, its exit status and both streams identical to mcflirt’s (FSL 6.0.7.23 as built), with the replica in reproduce.

9/9
failures/binsm4-normcorr-s2failure cases (specs/mcflirt.md §14.1)-bins -4, two stages: mcflirt is killed by SIGFPE (FSL-66)Rejected by both

mcflirt ends by signal 8 (no message); the clean room exits 1 (reproduce) and 1 (default) with its error

–Rejected by both

mcflirt and the replica fail alike, with the same messages and status: mcflirt ends by signal 8 (no message); the replica exits 1 (reproduce) and 1 (default) with its error

8/8
failures/binsm8-normcorrfailure cases (specs/mcflirt.md §14.1)-bins -8: SIGFPE in stage 1 (FSL-66)Rejected by both

mcflirt ends by signal 8 (no message); the clean room exits 1 (reproduce) and 1 (default) with its error

–Rejected by both

mcflirt and the replica fail alike, with the same messages and status: mcflirt ends by signal 8 (no message); the replica exits 1 (reproduce) and 1 (default) with its error

8/8
failures/binsm16-normcorrfailure cases (specs/mcflirt.md §14.1)-bins -16: std::bad_array_new_length (FSL-66)Rejected by both

mcflirt ends by signal 6 (terminate called after throwing an instance of ‘std::bad_array_new_length’); the clean room exits 1 (reproduce) and 1 (default) with its error

–Rejected by both

mcflirt and the replica fail alike, with the same messages and status: mcflirt ends by signal 6 (terminate called after throwing an instance of ‘std::bad_array_new_length’); the replica exits 1 (reproduce) and 1 (default) with its error

8/8
failures/binsm16-mutualinfofailure cases (specs/mcflirt.md §14.1)mutualinfo, -bins -16 (FSL-66)Rejected by both

mcflirt ends by signal 6 (terminate called after throwing an instance of ‘std::bad_array_new_length’); the clean room exits 1 (reproduce) and 1 (default) with its error

–Rejected by both

mcflirt and the replica fail alike, with the same messages and status: mcflirt ends by signal 6 (terminate called after throwing an instance of ‘std::bad_array_new_length’); the replica exits 1 (reproduce) and 1 (default) with its error

8/8
difftest/normmi-aligns-worserandom testingHead-motion correction with -cost normmi on a random phantom with large motion: once the subset that aligned worse (ite…Expected divergence

-cost normmi, computed exactly as specs/mcflirt.md appendix B describes it since 2026-10-29: the clean room now aligns this phantom better than mcflirt 6.0.7.23 (temporal SD 0.93 of mcflirt’s; 1.38 before), its estimates ending in other minima of a phantom moved by …

––7/13
difftest/phantom-several-minimarandom testingHead-motion correction on a random phantom with large motion: an explained divergence (item 13, 2026-10-29)Expected divergence

A small phantom moved by up to 15 degrees: the normalised correlation has several minima, and the clean room’s estimates end in other ones than mcflirt’s beyond the record’s thresholds, aligning the series as well (temporal SD 0.999 of mcflirt’s here; over the 782 f…

––13/16
difftest/time-step-negative-seriesrandom testingmcflirt's engine on a series of time step -2: FSL writes the time step as its image layer reads it (its magnitude, 1 fo…Open difference

95th-percentile RMS deviation 0 mm, median Δ 0 mm / 0°; 1/1 matrices identical as text: within the record’s thresholds (0.1 mm; 0.05 mm / 0.05°); mcflirt’s own variability is not measured for this run, so the difference is not yet set against it.

––13/13
real/ds000005/opt-sinc_finaloptions, on ds0000056/6–Bit-identical

Every file mcflirt writes (246/246), its exit status and both streams identical to mcflirt’s (FSL 6.0.7.17 as built), with the replica in reproduce.

–6/6
real/ds000005/opt-stages4options, on ds0000056/6–Bit-identical

Every file mcflirt writes (246/246), its exit status and both streams identical to mcflirt’s (FSL 6.0.7.17 as built), with the replica in reproduce.

–6/6
real/ds000005/opt-stages4-sinc_finaloptions, on ds0000056/6–Bit-identical

Every file mcflirt writes (246/246), its exit status and both streams identical to mcflirt’s (FSL 6.0.7.17 as built), with the replica in reproduce.

–6/6
edge/gdtthe replica's own: edge cases6/6–Bit-identical

Every file mcflirt writes (3/3), its exit status and both streams identical to mcflirt’s (FSL 6.0.7.17 as built), with the replica in reproduce.

Bit-identical

Every file mcflirt writes (3/3), its exit status and both streams identical to mcflirt’s (FSL 6.0.7.23 as built), with the replica in reproduce.

6/6 · 7/7
edge/bins4-mutualinfothe replica's own: edge cases6/6–Bit-identical

Every file mcflirt writes (1/1), its exit status and both streams identical to mcflirt’s (FSL 6.0.7.17 as built), with the replica in reproduce.

Bit-identical

Every file mcflirt writes (1/1), its exit status and both streams identical to mcflirt’s (FSL 6.0.7.23 as built), with the replica in reproduce.

6/6 · 7/7
edge/bins1-normmithe replica's own: edge cases6/6–Bit-identical

Every file mcflirt writes (1/1), its exit status and both streams identical to mcflirt’s (FSL 6.0.7.17 as built), with the replica in reproduce.

Bit-identical

Every file mcflirt writes (1/1), its exit status and both streams identical to mcflirt’s (FSL 6.0.7.23 as built), with the replica in reproduce.

6/6 · 7/7
edge/bins2-corratiothe replica's own: edge casesmcflirt exits [-6]; the same from the replica. double free or corruption (out)–Rejected by both

mcflirt and the replica fail alike, with the same messages and status: mcflirt exits [-6]; the same from the replica. double free or corruption (out)

Rejected by both

mcflirt and the replica fail alike, with the same messages and status: mcflirt exits [-6]; the same from the replica. double free or corruption (out)

5/5 · 6/6
edge/in-missingthe replica's own: edge casesmcflirt exits [-6]; the same from the replica. Image Exception : #63 :: No image files match: nosuch Image Exception : #22 :: Failed to read volume nosuch Erro…–Rejected by both

mcflirt and the replica fail alike, with the same messages and status: mcflirt exits [-6]; the same from the replica. Image Exception : #63 :: No image files match: nosuch Image Exception : #22 :: Failed to read volume nosuch Error : No image files match: nosuch terminate called after throwing an instance of 'std::ru

Rejected by both

mcflirt and the replica fail alike, with the same messages and status: mcflirt exits [-6]; the same from the replica. Image Exception : #63 :: No image files match: nosuch Image Exception : #22 :: Failed to read volume nosuch Error : No image files match: nosuch terminate called after throwing an instance of 'std::ru

5/5 · 6/6
edge/init-missingthe replica's own: edge casesmcflirt exits [-6]; the same from the replica. Could not open matrix file nosuch.mat terminate called after throwing an instance of 'std::logic_error' what(): …–Rejected by both

mcflirt and the replica fail alike, with the same messages and status: mcflirt exits [-6]; the same from the replica. Could not open matrix file nosuch.mat terminate called after throwing an instance of ‘std::logic_error’ what(): matrix multiplication: incompatible matrix dimensions: 4x4 and 0x0

Rejected by both

mcflirt and the replica fail alike, with the same messages and status: mcflirt exits [-6]; the same from the replica. Could not open matrix file nosuch.mat terminate called after throwing an instance of ‘std::logic_error’ what(): matrix multiplication: incompatible matrix dimensions: 4x4 and 0x0

5/5 · 6/6
edge/init-shortthe replica's own: edge casesmcflirt exits [-6]; the same from the replica. terminate called after throwing an instance of 'std::logic_error' what(): matrix multiplication: incompatible ma…–Rejected by both

mcflirt and the replica fail alike, with the same messages and status: mcflirt exits [-6]; the same from the replica. terminate called after throwing an instance of ‘std::logic_error’ what(): matrix multiplication: incompatible matrix dimensions: 4x4 and 1x3

Rejected by both

mcflirt and the replica fail alike, with the same messages and status: mcflirt exits [-6]; the same from the replica. terminate called after throwing an instance of ‘std::logic_error’ what(): matrix multiplication: incompatible matrix dimensions: 4x4 and 1x3

5/5 · 6/6
edge/init-3x3the replica's own: edge casesmcflirt exits [-6]; the same from the replica. terminate called after throwing an instance of 'std::logic_error' what(): matrix multiplication: incompatible ma…–Rejected by both

mcflirt and the replica fail alike, with the same messages and status: mcflirt exits [-6]; the same from the replica. terminate called after throwing an instance of ‘std::logic_error’ what(): matrix multiplication: incompatible matrix dimensions: 4x4 and 3x3

Rejected by both

mcflirt and the replica fail alike, with the same messages and status: mcflirt exits [-6]; the same from the replica. terminate called after throwing an instance of ‘std::logic_error’ what(): matrix multiplication: incompatible matrix dimensions: 4x4 and 3x3

5/5 · 6/6
edge/verbose20the replica's own: edge cases6/6–Bit-identical

Every file mcflirt writes (1/1), its exit status and both streams identical to mcflirt’s (FSL 6.0.7.17 as built), with the replica in reproduce.

Bit-identical

Every file mcflirt writes (1/1), its exit status and both streams identical to mcflirt’s (FSL 6.0.7.23 as built), with the replica in reproduce.

6/6 · 7/7
edge/verbose20-reportthe replica's own: edge cases6/6–Bit-identical

Every file mcflirt writes (2/2), its exit status and both streams identical to mcflirt’s (FSL 6.0.7.17 as built), with the replica in reproduce.

Bit-identical

Every file mcflirt writes (2/2), its exit status and both streams identical to mcflirt’s (FSL 6.0.7.23 as built), with the replica in reproduce.

6/6 · 7/7
edge/out-dir-missingthe replica's own: edge cases6/6–Bit-identical

Every file mcflirt writes (0/0), its exit status and both streams identical to mcflirt’s (FSL 6.0.7.17 as built), with the replica in reproduce.

Bit-identical

Every file mcflirt writes (0/0), its exit status and both streams identical to mcflirt’s (FSL 6.0.7.23 as built), with the replica in reproduce.

6/6 · 7/7
edge/out-dir-readonlythe replica's own: edge casesmcflirt exits [-6]; the same from the replica. Error! Could not create directory: <larmorx-testdata>/data/synthetic/hmc/motion/ro/out.mat. No write permission …–Rejected by both

mcflirt and the replica fail alike, with the same messages and status: mcflirt exits [-6]; the same from the replica. Error! Could not create directory: /data/synthetic/hmc/motion/ro/out.mat. No write permission terminate called after throwing an instance of

Rejected by both

mcflirt and the replica fail alike, with the same messages and status: mcflirt exits [-6]; the same from the replica. Error! Could not create directory: /data/synthetic/hmc/motion/ro/out.mat. No write permission terminate called after throwing an instance of

5/5 · 6/6
edge/mat-existsthe replica's own: edge cases6/6–Bit-identical

Every file mcflirt writes (2/2), its exit status and both streams identical to mcflirt’s (FSL 6.0.7.17 as built), with the replica in reproduce.

Bit-identical

Every file mcflirt writes (2/2), its exit status and both streams identical to mcflirt’s (FSL 6.0.7.23 as built), with the replica in reproduce.

6/6 · 7/7
edge/fsloutputtype-pairthe replica's own: edge cases6/6–Bit-identical

Every file mcflirt writes (2/2), its exit status and both streams identical to mcflirt’s (FSL 6.0.7.17 as built), with the replica in reproduce.

Bit-identical

Every file mcflirt writes (2/2), its exit status and both streams identical to mcflirt’s (FSL 6.0.7.23 as built), with the replica in reproduce.

6/6 · 7/7
edge/fsloutputtype-pair-gzthe replica's own: edge cases6/6–Bit-identical

Every file mcflirt writes (2/2), its exit status and both streams identical to mcflirt’s (FSL 6.0.7.17 as built), with the replica in reproduce.

Bit-identical

Every file mcflirt writes (2/2), its exit status and both streams identical to mcflirt’s (FSL 6.0.7.23 as built), with the replica in reproduce.

6/6 · 7/7
edge/fsloutputtype-nifti2-gzthe replica's own: edge cases6/6–Bit-identical

Every file mcflirt writes (1/1), its exit status and both streams identical to mcflirt’s (FSL 6.0.7.17 as built), with the replica in reproduce.

Bit-identical

Every file mcflirt writes (1/1), its exit status and both streams identical to mcflirt’s (FSL 6.0.7.23 as built), with the replica in reproduce.

6/6 · 7/7
edge/fsloutputtype-analyzethe replica's own: edge casesmcflirt exits [1]; the same from the replica. ERROR:: Unrecognised value (ANALYZE) of environment variable FSLOUTPUTTYPE Legal values are: NIFTI, NIFTI_PAIR, N…–Rejected by both

mcflirt and the replica fail alike, with the same messages and status: mcflirt exits [1]; the same from the replica. ERROR:: Unrecognised value (ANALYZE) of environment variable FSLOUTPUTTYPE Legal values are: NIFTI, NIFTI_PAIR, NIFTI_GZ, NIFTI_PAIR_GZ

Rejected by both

mcflirt and the replica fail alike, with the same messages and status: mcflirt exits [1]; the same from the replica. ERROR:: Unrecognised value (ANALYZE) of environment variable FSLOUTPUTTYPE Legal values are: NIFTI, NIFTI_PAIR, NIFTI2, NIFTI2_PAIR, NIFTI_GZ, NIFTI_PAIR_GZ, NIFTI2_GZ, NIFTI2_PAIR_GZ, NIFTI_ZST, NIFT

5/5 · 6/6
edge/fsloutputtype-bogusthe replica's own: edge casesmcflirt exits [1]; the same from the replica. ERROR:: Unrecognised value (BOGUS) of environment variable FSLOUTPUTTYPE Legal values are: NIFTI, NIFTI_PAIR, NIF…–Rejected by both

mcflirt and the replica fail alike, with the same messages and status: mcflirt exits [1]; the same from the replica. ERROR:: Unrecognised value (BOGUS) of environment variable FSLOUTPUTTYPE Legal values are: NIFTI, NIFTI_PAIR, NIFTI_GZ, NIFTI_PAIR_GZ

Rejected by both

mcflirt and the replica fail alike, with the same messages and status: mcflirt exits [1]; the same from the replica. ERROR:: Unrecognised value (BOGUS) of environment variable FSLOUTPUTTYPE Legal values are: NIFTI, NIFTI_PAIR, NIFTI2, NIFTI2_PAIR, NIFTI_GZ, NIFTI_PAIR_GZ, NIFTI2_GZ, NIFTI2_PAIR_GZ, NIFTI_ZST, NIFTI_

5/5 · 6/6
edge/stages0-meanvolthe replica's own: edge cases6/6–Bit-identical

Every file mcflirt writes (2/2), its exit status and both streams identical to mcflirt’s (FSL 6.0.7.17 as built), with the replica in reproduce.

Bit-identical

Every file mcflirt writes (2/2), its exit status and both streams identical to mcflirt’s (FSL 6.0.7.23 as built), with the replica in reproduce.

6/6 · 7/7
edge/stages0-refvolthe replica's own: edge cases6/6–Bit-identical

Every file mcflirt writes (3/3), its exit status and both streams identical to mcflirt’s (FSL 6.0.7.17 as built), with the replica in reproduce.

Bit-identical

Every file mcflirt writes (3/3), its exit status and both streams identical to mcflirt’s (FSL 6.0.7.23 as built), with the replica in reproduce.

6/6 · 7/7
edge/dof13the replica's own: edge cases6/6–Bit-identical

Every file mcflirt writes (1/1), its exit status and both streams identical to mcflirt’s (FSL 6.0.7.17 as built), with the replica in reproduce.

Bit-identical

Every file mcflirt writes (1/1), its exit status and both streams identical to mcflirt’s (FSL 6.0.7.23 as built), with the replica in reproduce.

6/6 · 7/7
edge/int8the replica's own: edge cases6/6–Bit-identical

Every file mcflirt writes (1/1), its exit status and both streams identical to mcflirt’s (FSL 6.0.7.17 as built), with the replica in reproduce.

Bit-identical

Every file mcflirt writes (1/1), its exit status and both streams identical to mcflirt’s (FSL 6.0.7.23 as built), with the replica in reproduce.

6/6 · 7/7
edge/int32the replica's own: edge cases6/6–Bit-identical

Every file mcflirt writes (1/1), its exit status and both streams identical to mcflirt’s (FSL 6.0.7.17 as built), with the replica in reproduce.

Bit-identical

Every file mcflirt writes (1/1), its exit status and both streams identical to mcflirt’s (FSL 6.0.7.23 as built), with the replica in reproduce.

6/6 · 7/7
edge/analyze-inputthe replica's own: edge cases6/6–Bit-identical

Every file mcflirt writes (1/1), its exit status and both streams identical to mcflirt’s (FSL 6.0.7.17 as built), with the replica in reproduce.

Bit-identical

Every file mcflirt writes (1/1), its exit status and both streams identical to mcflirt’s (FSL 6.0.7.23 as built), with the replica in reproduce.

6/6 · 7/7
edge/rgb-inputthe replica's own: edge casesmcflirt exits [-6]; the same from the replica. Image Exception : #8 :: Fslread: DT 128 not supported terminate called after throwing an instance of 'std::runti…–Rejected by both

mcflirt and the replica fail alike, with the same messages and status: mcflirt exits [-6]; the same from the replica. Image Exception : #8 :: Fslread: DT 128 not supported terminate called after throwing an instance of ‘std::runtime_error’ what(): Fslread: DT 128 not supported

Rejected by both

mcflirt and the replica fail alike, with the same messages and status: mcflirt exits [-6]; the same from the replica. Image Exception : #8 :: Fslread: DT 128 not supported terminate called after throwing an instance of ‘std::runtime_error’ what(): Fslread: DT 128 not supported

5/5 · 6/6
edge/nan-reffilethe replica's own: edge cases6/6–Bit-identical

Every file mcflirt writes (2/2), its exit status and both streams identical to mcflirt’s (FSL 6.0.7.17 as built), with the replica in reproduce.

Bit-identical

Every file mcflirt writes (2/2), its exit status and both streams identical to mcflirt’s (FSL 6.0.7.23 as built), with the replica in reproduce.

6/6 · 7/7
edge/refvol-lastthe replica's own: edge cases6/6–Bit-identical

Every file mcflirt writes (3/3), its exit status and both streams identical to mcflirt’s (FSL 6.0.7.17 as built), with the replica in reproduce.

Bit-identical

Every file mcflirt writes (3/3), its exit status and both streams identical to mcflirt’s (FSL 6.0.7.23 as built), with the replica in reproduce.

6/6 · 7/7
edge/meanvol-reffile-sinc-statsthe replica's own: edge cases6/6–Bit-identical

Every file mcflirt writes (17/17), its exit status and both streams identical to mcflirt’s (FSL 6.0.7.17 as built), with the replica in reproduce.

Bit-identical

Every file mcflirt writes (17/17), its exit status and both streams identical to mcflirt’s (FSL 6.0.7.23 as built), with the replica in reproduce.

6/6 · 7/7
edge/atanf-phantom-2the replica's own: edge cases6/6–Bit-identical

Every file mcflirt writes (2/2), its exit status and both streams identical to mcflirt’s (FSL 6.0.7.17 as built), with the replica in reproduce.

Bit-identical

Every file mcflirt writes (2/2), its exit status and both streams identical to mcflirt’s (FSL 6.0.7.23 as built), with the replica in reproduce.

6/6 · 7/7
edge/atanf-phantom-11the replica's own: edge cases6/6–Bit-identical

Every file mcflirt writes (2/2), its exit status and both streams identical to mcflirt’s (FSL 6.0.7.17 as built), with the replica in reproduce.

Bit-identical

Every file mcflirt writes (2/2), its exit status and both streams identical to mcflirt’s (FSL 6.0.7.23 as built), with the replica in reproduce.

6/6 · 7/7
edge/atanf-phantom-27the replica's own: edge cases6/6–Bit-identical

Every file mcflirt writes (2/2), its exit status and both streams identical to mcflirt’s (FSL 6.0.7.17 as built), with the replica in reproduce.

Bit-identical

Every file mcflirt writes (2/2), its exit status and both streams identical to mcflirt’s (FSL 6.0.7.23 as built), with the replica in reproduce.

6/6 · 7/7
edge/rotation0the replica's own: edge cases6/6–Bit-identical

Every file mcflirt writes (1/1), its exit status and both streams identical to mcflirt’s (FSL 6.0.7.17 as built), with the replica in reproduce.

Bit-identical

Every file mcflirt writes (1/1), its exit status and both streams identical to mcflirt’s (FSL 6.0.7.23 as built), with the replica in reproduce.

6/6 · 7/7
edge/smooth-negativethe replica's own: edge cases6/6–Bit-identical

Every file mcflirt writes (1/1), its exit status and both streams identical to mcflirt’s (FSL 6.0.7.17 as built), with the replica in reproduce.

Bit-identical

Every file mcflirt writes (1/1), its exit status and both streams identical to mcflirt’s (FSL 6.0.7.23 as built), with the replica in reproduce.

6/6 · 7/7
edge/scaling-histthe replica's own: edge cases6/6–Bit-identical

Every file mcflirt writes (1/1), its exit status and both streams identical to mcflirt’s (FSL 6.0.7.17 as built), with the replica in reproduce.

Bit-identical

Every file mcflirt writes (1/1), its exit status and both streams identical to mcflirt’s (FSL 6.0.7.23 as built), with the replica in reproduce.

6/6 · 7/7
edge/fov0-2dthe replica's own: edge cases6/6–Bit-identical

Every file mcflirt writes (2/2), its exit status and both streams identical to mcflirt’s (FSL 6.0.7.17 as built), with the replica in reproduce.

Bit-identical

Every file mcflirt writes (2/2), its exit status and both streams identical to mcflirt’s (FSL 6.0.7.23 as built), with the replica in reproduce.

6/6 · 7/7
fmriprep/ds000005-sub-01-noreconall/task_mixedgamblestask_run_01the replica's own: fMRIPrep's calls7/7–Bit-identical

Every file mcflirt writes (241/241), its exit status and both streams identical to mcflirt’s (FSL 6.0.7.17 as built), with the replica in reproduce.

Bit-identical

Every file mcflirt writes (241/241), its exit status and both streams identical to mcflirt’s (FSL 6.0.7.23 as built), with the replica in reproduce.

7/7 · 8/8
fmriprep/ds000005-sub-01-noreconall/task_mixedgamblestask_run_02the replica's own: fMRIPrep's calls7/7–Bit-identical

Every file mcflirt writes (241/241), its exit status and both streams identical to mcflirt’s (FSL 6.0.7.17 as built), with the replica in reproduce.

Bit-identical

Every file mcflirt writes (241/241), its exit status and both streams identical to mcflirt’s (FSL 6.0.7.23 as built), with the replica in reproduce.

7/7 · 8/8
fmriprep/ds000005-sub-01-noreconall/task_mixedgamblestask_run_03the replica's own: fMRIPrep's calls7/7–Bit-identical

Every file mcflirt writes (241/241), its exit status and both streams identical to mcflirt’s (FSL 6.0.7.17 as built), with the replica in reproduce.

Bit-identical

Every file mcflirt writes (241/241), its exit status and both streams identical to mcflirt’s (FSL 6.0.7.23 as built), with the replica in reproduce.

7/7 · 8/8
fmriprep/ds000228-sub-pixar008-noreconall/task_pixarthe replica's own: fMRIPrep's calls7/7–Bit-identical

Every file mcflirt writes (169/169), its exit status and both streams identical to mcflirt’s (FSL 6.0.7.17 as built), with the replica in reproduce.

Bit-identical

Every file mcflirt writes (169/169), its exit status and both streams identical to mcflirt’s (FSL 6.0.7.23 as built), with the replica in reproduce.

7/7 · 8/8
fmriprep/ds000256-sub-CTS203-noreconall/task_restbaselinethe replica's own: fMRIPrep's calls7/7–Bit-identical

Every file mcflirt writes (165/165), its exit status and both streams identical to mcflirt’s (FSL 6.0.7.17 as built), with the replica in reproduce.

Bit-identical

Every file mcflirt writes (165/165), its exit status and both streams identical to mcflirt’s (FSL 6.0.7.23 as built), with the replica in reproduce.

7/7 · 8/8
fmriprep/ds000256-sub-CTS203-noreconall/task_restfixationfixedthe replica's own: fMRIPrep's calls7/7–Bit-identical

Every file mcflirt writes (165/165), its exit status and both streams identical to mcflirt’s (FSL 6.0.7.17 as built), with the replica in reproduce.

Bit-identical

Every file mcflirt writes (165/165), its exit status and both streams identical to mcflirt’s (FSL 6.0.7.23 as built), with the replica in reproduce.

7/7 · 8/8
fmriprep/ds000256-sub-CTS203-noreconall/task_restfixationnofeedbackthe replica's own: fMRIPrep's calls7/7–Bit-identical

Every file mcflirt writes (165/165), its exit status and both streams identical to mcflirt’s (FSL 6.0.7.17 as built), with the replica in reproduce.

Bit-identical

Every file mcflirt writes (165/165), its exit status and both streams identical to mcflirt’s (FSL 6.0.7.23 as built), with the replica in reproduce.

7/7 · 8/8
fmriprep/ds002422-sub-02-noreconall/task_arithm_run_01the replica's own: fMRIPrep's calls7/7–Bit-identical

Every file mcflirt writes (211/211), its exit status and both streams identical to mcflirt’s (FSL 6.0.7.17 as built), with the replica in reproduce.

Bit-identical

Every file mcflirt writes (211/211), its exit status and both streams identical to mcflirt’s (FSL 6.0.7.23 as built), with the replica in reproduce.

7/7 · 8/8
fmriprep/ds002422-sub-02-noreconall/task_arithm_run_02the replica's own: fMRIPrep's calls7/7–Bit-identical

Every file mcflirt writes (211/211), its exit status and both streams identical to mcflirt’s (FSL 6.0.7.17 as built), with the replica in reproduce.

Bit-identical

Every file mcflirt writes (211/211), its exit status and both streams identical to mcflirt’s (FSL 6.0.7.23 as built), with the replica in reproduce.

7/7 · 8/8
fmriprep/ds002422-sub-02-noreconall/task_restthe replica's own: fMRIPrep's calls7/7–Bit-identical

Every file mcflirt writes (201/201), its exit status and both streams identical to mcflirt’s (FSL 6.0.7.17 as built), with the replica in reproduce.

Bit-identical

Every file mcflirt writes (201/201), its exit status and both streams identical to mcflirt’s (FSL 6.0.7.23 as built), with the replica in reproduce.

7/7 · 8/8
fmriprep/ds002603-sub-01-noreconall/task_wm_acq_Mb4Mm3Tr550the replica's own: fMRIPrep's calls7/7–Bit-identical

Every file mcflirt writes (589/589), its exit status and both streams identical to mcflirt’s (FSL 6.0.7.17 as built), with the replica in reproduce.

Bit-identical

Every file mcflirt writes (589/589), its exit status and both streams identical to mcflirt’s (FSL 6.0.7.23 as built), with the replica in reproduce.

7/7 · 8/8
fmriprep/ds002603-sub-01-noreconall/task_wm_acq_SeqMm3Tr2000the replica's own: fMRIPrep's calls7/7–Bit-identical

Every file mcflirt writes (161/161), its exit status and both streams identical to mcflirt’s (FSL 6.0.7.17 as built), with the replica in reproduce.

Bit-identical

Every file mcflirt writes (161/161), its exit status and both streams identical to mcflirt’s (FSL 6.0.7.23 as built), with the replica in reproduce.

7/7 · 8/8
fmriprep/ds003592-sub-280-noreconall/ses_1_task_rest_echo_1the replica's own: fMRIPrep's calls7/7–Bit-identical

Every file mcflirt writes (201/201), its exit status and both streams identical to mcflirt’s (FSL 6.0.7.17 as built), with the replica in reproduce.

Bit-identical

Every file mcflirt writes (201/201), its exit status and both streams identical to mcflirt’s (FSL 6.0.7.23 as built), with the replica in reproduce.

7/7 · 8/8
fmriprep/ds004884-sub-M2079-noreconall/ses_217_task_naming40_acq_epfid2_dir_AP_run_2the replica's own: fMRIPrep's calls7/7–Bit-identical

Every file mcflirt writes (61/61), its exit status and both streams identical to mcflirt’s (FSL 6.0.7.17 as built), with the replica in reproduce.

Bit-identical

Every file mcflirt writes (61/61), its exit status and both streams identical to mcflirt’s (FSL 6.0.7.23 as built), with the replica in reproduce.

7/7 · 8/8
fmriprep/ds004884-sub-M2079-noreconall/ses_217_task_rest_acq_epfidp2p2_dir_AP_run_9the replica's own: fMRIPrep's calls7/7–Bit-identical

Every file mcflirt writes (197/197), its exit status and both streams identical to mcflirt’s (FSL 6.0.7.17 as built), with the replica in reproduce.

Bit-identical

Every file mcflirt writes (197/197), its exit status and both streams identical to mcflirt’s (FSL 6.0.7.23 as built), with the replica in reproduce.

7/7 · 8/8
fmriprep/ds004958-sub-fsm64ps-noreconall/task_restthe replica's own: fMRIPrep's calls7/7–Bit-identical

Every file mcflirt writes (601/601), its exit status and both streams identical to mcflirt’s (FSL 6.0.7.17 as built), with the replica in reproduce.

Bit-identical

Every file mcflirt writes (601/601), its exit status and both streams identical to mcflirt’s (FSL 6.0.7.23 as built), with the replica in reproduce.

7/7 · 8/8
fmriprep/ds005003-sub-RS001-noreconall/ses_1_task_motorthe replica's own: fMRIPrep's calls7/7–Bit-identical

Every file mcflirt writes (81/81), its exit status and both streams identical to mcflirt’s (FSL 6.0.7.17 as built), with the replica in reproduce.

Bit-identical

Every file mcflirt writes (81/81), its exit status and both streams identical to mcflirt’s (FSL 6.0.7.23 as built), with the replica in reproduce.

7/7 · 8/8
fmriprep/ds005003-sub-RS001-noreconall/ses_1_task_restthe replica's own: fMRIPrep's calls7/7–Bit-identical

Every file mcflirt writes (81/81), its exit status and both streams identical to mcflirt’s (FSL 6.0.7.17 as built), with the replica in reproduce.

Bit-identical

Every file mcflirt writes (81/81), its exit status and both streams identical to mcflirt’s (FSL 6.0.7.23 as built), with the replica in reproduce.

7/7 · 8/8
fmriprep/ds005040-sub-002-noreconall/task_clip_run_1the replica's own: fMRIPrep's calls7/7–Bit-identical

Every file mcflirt writes (64/64), its exit status and both streams identical to mcflirt’s (FSL 6.0.7.17 as built), with the replica in reproduce.

Bit-identical

Every file mcflirt writes (64/64), its exit status and both streams identical to mcflirt’s (FSL 6.0.7.23 as built), with the replica in reproduce.

7/7 · 8/8
fmriprep/ds005454-sub-16-noreconall/task_restthe replica's own: fMRIPrep's calls7/7–Bit-identical

Every file mcflirt writes (281/281), its exit status and both streams identical to mcflirt’s (FSL 6.0.7.17 as built), with the replica in reproduce.

–7/7
fmriprep/ds005752-sub-ON01016-noreconall/ses_01_task_rest_dir_forwardthe replica's own: fMRIPrep's calls7/7–Bit-identical

Every file mcflirt writes (201/201), its exit status and both streams identical to mcflirt’s (FSL 6.0.7.17 as built), with the replica in reproduce.

Bit-identical

Every file mcflirt writes (201/201), its exit status and both streams identical to mcflirt’s (FSL 6.0.7.23 as built), with the replica in reproduce.

7/7 · 8/8
fmriprep/ds005752-sub-ON01016-noreconall/ses_01_task_rest_dir_reversethe replica's own: fMRIPrep's calls7/7–Bit-identical

Every file mcflirt writes (11/11), its exit status and both streams identical to mcflirt’s (FSL 6.0.7.17 as built), with the replica in reproduce.

Bit-identical

Every file mcflirt writes (11/11), its exit status and both streams identical to mcflirt’s (FSL 6.0.7.23 as built), with the replica in reproduce.

7/7 · 8/8
edge/fsl75-narrowthe replica's own: edge casesmcflirt exits [-6]; the same from the replica. double free or corruption (out)––Rejected by both

mcflirt and the replica fail alike, with the same messages and status: mcflirt exits [-6]; the same from the replica. double free or corruption (out)

3/3
edge/fsloutputtype-zstthe replica's own: edge cases7/7––Bit-identical

Every file mcflirt writes (4/4), its exit status and both streams identical to mcflirt’s (FSL 6.0.7.23 as built), with the replica in reproduce.

7/7
edge/fsloutputtype-pair-bz2the replica's own: edge cases7/7––Bit-identical

Every file mcflirt writes (2/2), its exit status and both streams identical to mcflirt’s (FSL 6.0.7.23 as built), with the replica in reproduce.

7/7
edge/fsloutputtype-nifti2-zstthe replica's own: edge cases7/7––Bit-identical

Every file mcflirt writes (1/1), its exit status and both streams identical to mcflirt’s (FSL 6.0.7.23 as built), with the replica in reproduce.

7/7
edge/input-zstthe replica's own: edge cases7/7––Bit-identical

Every file mcflirt writes (1/1), its exit status and both streams identical to mcflirt’s (FSL 6.0.7.23 as built), with the replica in reproduce.

7/7
edge/input-bz2-plain-namethe replica's own: edge cases7/7––Bit-identical

Every file mcflirt writes (1/1), its exit status and both streams identical to mcflirt’s (FSL 6.0.7.23 as built), with the replica in reproduce.

7/7
edge/img-in-directorythe replica's own: edge casesmcflirt exits [-6]; the same from the replica. Image Exception : #22 :: Failed to read volume a.img.d/in.nii.gz Error : Error: cant open file a.hdr.d/in.nii.gz…––Rejected by both

mcflirt and the replica fail alike, with the same messages and status: mcflirt exits [-6]; the same from the replica. Image Exception : #22 :: Failed to read volume a.img.d/in.nii.gz Error : Error: cant open file a.hdr.d/in.nii.gz terminate called after throwing an instance of ‘std::runtime_error’ what(): Failed to r

6/6
edge/img-in-output-directorythe replica's own: edge cases7/7––Bit-identical

Every file mcflirt writes (13/13), its exit status and both streams identical to mcflirt’s (FSL 6.0.7.23 as built), with the replica in reproduce.

7/7
difftest/atanf-default-callrandom testing6/6––Bit-identical

Every file mcflirt writes (1/1), its exit status and both streams identical to mcflirt’s (FSL 6.0.7.23 as built), with the replica in reproduce.

6/6
difftest/atanf-matrices-from-0random testing6/6––Bit-identical

Every file mcflirt writes (11/11), its exit status and both streams identical to mcflirt’s (FSL 6.0.7.23 as built), with the replica in reproduce.

6/6
difftest/atanf-matrices-from-4random testing6/6––Bit-identical

Every file mcflirt writes (9/9), its exit status and both streams identical to mcflirt’s (FSL 6.0.7.23 as built), with the replica in reproduce.

6/6
difftest/atanf-par-onlyrandom testing6/6––Bit-identical

Every file mcflirt writes (12/12), its exit status and both streams identical to mcflirt’s (FSL 6.0.7.23 as built), with the replica in reproduce.

6/6
difftest/atanf-printed-paramsrandom testing6/6––Bit-identical

Every file mcflirt writes (1/1), its exit status and both streams identical to mcflirt’s (FSL 6.0.7.23 as built), with the replica in reproduce.

6/6
difftest/meanvol-name-noticerandom testing6/6––Bit-identical

Every file mcflirt writes (2/2), its exit status and both streams identical to mcflirt’s (FSL 6.0.7.23 as built), with the replica in reproduce.

6/6

Speed

mcflirt runs on one thread; the clean room runs on as many as it is given, with the same result.

  • On ds000005’s run (240 volumes) the clean room takes 26.6 s on one thread against mcflirt’s 30.7 s, and 6.2 s on 12.
  • On ds006010’s multiband run (113 volumes of 96 × 96 × 69) it takes 23.6 s against 54.5 s, and 5.8 s on 12.
  • On ds000258’s run (239 volumes) it takes 33.2 s against 39.2 s, and 7.5 s on 12.

These timings come from the benchmark report of 2026-10-10: the whole command, the median of three runs after a warm-up (mcflirt one run), on an Intel Core i7-8750H with 12 logical CPUs, against FSL 6.0.7’s mcflirt.

Speed, the clean room against the original: FSL mcflirt 6.0.7

Seconds for the whole command (reading the run, the three-stage estimation, writing the corrected series and the matrices), the median of three runs after a warm-up (mcflirt one run), on a log scale. The factor is how many times faster the clean room is than mcflirt, both on one thread.

  • FSL mcflirt 6.0.7, 1 thread
  • clean room, 1 thread
  • clean room, 12 threads
  1. ds000005: 64×64×34, 240 vol.FSL mcflirt 6.0.7, 1 thread 30.7 sclean room, 1 thread 26.6 sclean room, 12 threads 6.21 s1.2×
  2. ds006010: 96×96×69, 113 vol.FSL mcflirt 6.0.7, 1 thread 54.5 sclean room, 1 thread 23.6 sclean room, 12 threads 5.79 s2.3×
  3. ds000258: 64×64×30, 239 vol.FSL mcflirt 6.0.7, 1 thread 39.2 sclean room, 1 thread 33.2 sclean room, 12 threads 7.48 s1.2×
1 s10 s100 s

From the benchmark report docs/benchmarks/mri-hmc.json (Intel® Core™ i7-8750H CPU @ 2.20GHz, 12 logical CPUs).

Corrections by default

The clean room’s record names its corrections where they change a result: FSL-01 (the last volume not initialised from the previous one), FSL-03, FSL-05, FSL-06, FSL-07, FSL-08, FSL-09, FSL-10, FSL-13, FSL-14, FSL-20, FSL-21, FSL-59, FSL-65 and FSL-66. Each is corrected by default, fix, with upstream_defects="reproduce" (on the command line --upstream-defects reproduce from the next release, LARMORX_UPSTREAM_DEFECTS=reproduce in 0.1.0rc1) giving mcflirt’s behaviour, the mode the comparison runs in (register). The worked example shows what they change on fMRIPrep’s run.

  • Since 2026-10-30 the clean room handles FSL-20, FSL-21, FSL-65 and FSL-66 as its specification says (outputs that cannot be written, impossible bin counts): the record’s 33 failure cases of specs/mcflirt.md §14.1.
  • Without FSLOUTPUTTYPE, mcflirt writes the .mat files and stops at its first image with FSL’s message (FSL-11), and so does the clean room in reproduce; its fix writes every output, the images as .nii.gz (syn/motion/small/no-fsloutputtype).

The replica corrects the same defects the same way by default, and three more that its translation found:

  • -gdt (FSL-04): refused; reproduce registers the gradient images as mcflirt does;
  • a .mat directory that cannot be created (FSL-20): mcflirt exits with status 0 having written nothing more; an error;
  • the histogram costs with fewer than 8 bins (FSL-21): the first stage has no bins, the matrices are arbitrary, and corratio corrupts its heap and aborts; refused.

reproduce gives mcflirt’s bytes, aborts included. FSL 6.0.7.23 adds one: -sinc_final and -nn_final are resampled trilinearly (FSL-59: each volume keeps the interpolation it was first read with); the replica uses the method asked for by default, and 6.0.7.23’s trilinear resampling in reproduce. An image name holding .img or .hdr before its extension makes 6.0.7.23 open another file (FSL-60): corrected the same way. The clean room corrects FSL-59 and FSL-60 alike since 2026-10-29 (FSL-60 on its command line; its Python API opens the name given).

Random testing

Random testing runs random valid inputs and option combinations through the original program and LARMORX, in long runs of several hours: inputs from seeded plain arithmetic (up to 32³ voxels; orientations, data types, scale factors and value ranges of every kind), with options drawn in random combinations from those the program accepts (how random testing works). A difference is shrunk to the smallest input and fewest options that still show it, then fixed and kept as a case of the record, registered as a defect of the original, or reported. The latest long run:

Random testing, both ports against the original: FSL mcflirt 6.0.7.17, the long run of 2026-10-25 (seed 20261025, 6.2 h for 21 programs): random valid inputs and option combinations, each run through all three

ProgramRandom casesExtremeOptions coveredAgreeglibcNoticesUndefinedDifferHeavyGuardUnclear
mcflirtreplica1,1250381,017···107··1
head-motion correction (larmorx mri hmc)clean room1,119038336···782··1
  • fixed in LARMORX mcflirt (replica), 2 case(s): the fix pass: the default renamed the -meanvol image (FSL-08’s correction, _mean_reg) without naming FSL-08: the replica’s own part names it there now (d196995, on main), and its shrunk case is that part’s permanent case meanvol-name-notice
  • fixed in LARMORX mcflirt (replica), 101 case(s): every file bit-identical: the mcflirt replica’s atanf held fdlibm’s commented value of aT[0], not glibc’s 0x3eaaaaab, and FSL-08’s notice missed the -meanvol image’s name (both fixed by the replica’s own part, e687cd1 and aa2c28f): with them every one of the first long random-testing run’s 100 cases so far is bit-identical
  • fixed in LARMORX mcflirt (replica), 4 case(s): every file bit-identical, stderr identical: the atanf constant (as above); bit-identical with the replica’s fixes
  • an expected difference, measured and explained (no mistake, no defect) head-motion correction (larmorx mri hmc) (clean room), 782 case(s): the notices in fix and in reproduce, FD (Power, N mm) correlation, out_mcf.nii.gz.mat: matrix RMS deviation, Nth percentile, out_mcf.nii.gz.mat: median parameter difference, out_mcf.nii.gz.par: median difference, out_mcf.nii.gz: voxel values, out_mcf.nii.gz_abs.rms: values, out_mcf.nii.gz_abs_mean.rms: values, out_mcf.nii.gz_mean_reg.nii.gz: voxel values, out_mcf.nii.gz_meanvol.nii.gz: voxel values, out_mcf.nii.gz_rel.rms: values, out_mcf.nii.gz_rel_mean.rms: values, out_mcf.nii.gz_sigma.nii.gz: voxel values, out_mcf.nii.gz_variance.nii.gz: voxel values: measured on all 782 findings of the first long random-testing run (2026-10-25) with validation/investigations/cleanroom-fixes/hmc_alignment.py (the corrected series’ temporal standard deviation, LARMORX over mcflirt): every cost but -cost normmi as well aligned as mcflirt (691 findings, median ratio 1.000): large motions on small phantoms with a cost of several minima (permanent case mri-hmc/phantom-several-minima, an expected divergence); -cost normmi aligned worse (median 1.033) until 2026-10-29, when the clean-room alignment (next-steps 3b) computed normmi and mutualinfo exactly as specs/mcflirt.md appendix B (the 91 normmi findings: median 1.051 -> 1.000 against mcflirt 6.0.7.23, 55 within 1 %; the 78 mutualinfo findings: 0.983 -> 1.000); that part’s random-testing run of half an hour: 1,732 phantoms beyond the record’s thresholds, as well aligned as mcflirt for every cost (median ratios 0.999 to 1.001) (details)

Every case is drawn from its tool, the run's seed and its number, and replays anywhere: how random testing works; each difference's triage: validation/difftest/triage.toml.

The clean room’s random-testing run after the catch-up (2026-11-05, after its change of the centres of mass), 30 minutes long against FSL 6.0.7.23’s mcflirt, each difference in one category:

Random testing after the catch-up, the clean room against the original: FSL mcflirt 6.0.7.23 (83,244 random calls, 2026-11-05)

Cases
Random cases, each compared with FSL 6.0.7.23’s mcflirt83,244
within the record’s thresholds, every exact check passed31,552
differing (each one in a category below)51,681
a register entry named in one mode only (a notice)11
-mats, -plots, -rmsrel, -rmsabs58,004 / 58,354 / 58,370 / 58,352
-meanvol, -stats, -reffile, -refvol12,517 / 12,396 / 33,268 / 14,864
-sinc_final, -spline_final, -nn_final7,022 / 6,918 / 6,885
each cost (corratio, mutualinfo, normmi, normcorr, leastsquares)6,703 / 6,651 / 6,757 / 6,500 / 6,614
6, 7, 9, 12 dof5,396 / 5,234 / 5,170 / 5,093
1, 2, 3, 4 stages5,122 / 5,289 / 5,190 / 5,105
distinct options and settings drawn38
length (minutes; 12 jobs at one thread each)30

A random-testing run of the clean room, 30 minutes long, after the catch-up’s change of its centres of mass (mcflirt’s blocked sums, specs/mcflirt.md §6.1) and of atan2 (larmorx_core::math), seed 20261030: the cases of the random-testing run of 2026-10-30 drawn again and further, mcflirt’s runs served by the cache of the original’s runs where they had run; each case compared as the record compares it, in reproduce. The build: larmorx 0b11e921f431. Every case in a memory box of 3 GiB (peaks below 0.93 GiB).

The clean room’s differences from the original in the run of 2026-11-05 (after the catch-up), by category

CategoryCasesVerdictReproducer
The registrations of small phantoms with large motions end elsewhere than mcflirt’s, as well aligned: 400 sampled findings measured (hmc_alignment.py: the corrected series’ temporal standard deviation in the phantom, LARMORX’s over mcflirt’s, in reproduce, as the run compares): median 1.000, 5 and 95 % quantiles 0.933 and 1.041, 279 within 1 %, 67 lower, 54 higher (mcflirt’s own residual against the input’s: median 1.41); in 4,065 of them mcflirt’s run also lacks its empty .mat directory (the class below). In fix, the default, the ratio has a longer tail (above 1.05 in 102 of the 400), for the head-motion correction’s next part49,383Within the original's own variability20261030-00000-original
Only the files written differ: runs with -rmsrel or -rmsabs without -mats, where mcflirt makes an empty .mat directory (FSL-13) and its run was served by the cache of the original’s runs, which keeps a run’s files and not its empty directories (the first random-testing run, with mcflirt run live, had none of them): the harness’s, fixed in the descriptor (such mcflirt runs go live, validation/tests/test_mri_hmc_descriptor.py); four replayed: the files agree2,298the harness’s, fixed: the files agree on replay20261030-00007-original

Every differing case of the run in one category by its failed checks (validation/investigations/cleanroom-fixes/trial_catchup_hmc_categories.py, its output validation/investigations/cleanroom-fixes/trial_catchup_hmc_triage.json); the alignment measured by hmc_alignment.py on 400 sampled findings; the triage’s entries: validation/difftest/triage.toml. The reproducer is a finding of the run (difftest --replay).

What is open

The clean room’s open cases, by why they are open

WhycasesWhat would close them
misses a threshold of the record0the record’s failure
above mcflirt’s own variability in at least one estimate, within the record’s thresholds6mcflirt’s band is measured for these runs
within the record’s thresholds; mcflirt’s variability not measured for the run87options and synthetic series: their band is not measured yet
within the record’s thresholds; no matrices to set against it19runs that write no matrices, or whose motion only one parameter can show

From the case explorer below: each case’s reason gives its numbers.

  • No case misses its threshold. Until the record of 2026-11-07 one did: real/ds003345/default (mcflirt’s defaults, the middle volume as reference), whose framewise displacement correlated with mcflirt’s at r = 0.9885 where mcflirt’s own band reaches 0.9893. With the correlation’s count per point it correlates at r = 0.9907, within the band.
  • mcflirt’s variability is measured for fMRIPrep’s command on the nine real runs, for mcflirt’s defaults on most of them and for a few synthetic series. The options and the other synthetic series are held to the record’s fixed thresholds only, so they stay open until their band is measured.
  • The replica’s full comparison against FSL 6.0.7.23 waits for the final validation phase; the cross-platform checks (how they work) hold the other five platforms to LARMORX’s own Linux output.

History

  • 2026-10-10: the clean room, experimental, and its benchmark against FSL 6.0.7’s mcflirt.
  • 2026-10-25: the replica, translated from FSL 6.0.7.17’s source, and its full comparison, 173 of 173 compared cases bit-identical; the first long random-testing run, whose differences were all one constant of atanf and a missing FSL-08 notice.
  • 2026-10-29: both ports follow FSL 6.0.7.23 (FSL-59 to FSL-63 registered); the replica follows FSL’s matrix library bit for bit; the clean room’s -cost mutualinfo and normmi computed exactly.
  • 2026-10-30: the clean room handles FSL-20, FSL-21, FSL-65 and FSL-66 as its specification says; the replica’s standard comparison gains the 33 failure cases of specs/mcflirt.md §14.1.
  • 2026-11-04: the clean room’s reproduce computes logf as glibc 2.35 does.
  • 2026-11-05: the clean-room catch-up: the centres of mass as mcflirt’s blocked sums, 145 of its 149 compared cases within the thresholds then, 3 expected divergences and one case, real/ds003345/default, just beyond its band; its random-testing run of 30 minutes.
  • 2026-11-07: the weighted correlation’s count per point and FLIRT’s leastsquares: no case misses its threshold since.
  • 2026-11-11: the clean room’s record made again: syn/motion/small/no-fsloutputtype is rejected by both (it was an expected divergence), every other case as before.

Data credits

The data this page shows or names: each dataset with its version, DOI and licence, from LARMORX's test-data catalog. Public-domain dedications (CC0, PDDL) ask nothing; the datasets are cited all the same.