Evidence · AFNI

3dTshift

3dTshift shifts each slice’s time series to one common time. LARMORX has two ports of it, both in larmorx itself. Of the 317 cases of its record, the replica gives AFNI 26.3.05’s bytes in 282; both programs refuse 32, and 3 are expected divergences. Of its own 317, the clean room gives AFNI’s bytes in 180 and differs in 102 only by the last bits of its Fourier shift, within the record’s thresholds. On fMRIPrep’s call the replica gives AFNI’s and fMRIPrep’s output bit for bit. Both ports correct AFNI’s negative scale factor defect (AFNI-01) by default.

Ports
  • replica larmorx Apache-2.0, with CC BY 4.0's attribution for the Medical College of Wisconsin's code: translated from AFNI's source; in pip install larmorx since 2026-10-29, until then the separate GPL package larmorx-gpl
  • clean room larmorx Apache-2.0: written from specs/3dTshift.md
Status
replica validated; clean room validated (marked before the statuses' definition)
Records
the replica's: standard comparison (AFNI 26.3.05) · the clean room's: standard comparison (AFNI 26.3.05) · benchmark afni-tshift
Reference
3dTshift · API page afni-tshift

What is compared

3dTshift shifts each slice’s time series to a common time, so that the slices of a volume acquired at different moments can be treated as simultaneous. fMRIPrep calls it with the slice times of the BIDS sidecar and -tzero at the middle of the acquisition, with AFNI’s default Fourier interpolation.

The two ports run beside AFNI 26.3.05’s 3dTshift and AFNI 25.2.09’s on the same files with the same arguments, both AFNI releases the repository’s builds; fMRIPrep’s own output is the file it wrote in its run.

bash
larmorx afni 3dTshift -ignore 0 -tpattern @slice_timing.1D -TR 2s -tzero 0.975 -prefix bold_stc.nii.gz bold.nii.gz
python
import json
import larmorx as lx

meta = json.load(open("sub-01_task-rest_bold.json"))
times, t0 = lx.afni.fmriprep_slice_timing(meta["SliceTiming"], meta.get("SliceEncodingDirection"))
stc = lx.afni.tshift("sub-01_task-rest_bold.nii.gz", slice_times=times, tzero=t0,
                     tr=meta["RepetitionTime"], n_threads=8)

The call above leaves implementation at "auto". The API page maps every option. How to read what follows: how to read the evidence.

Its two ports

Since 2026-10-29 both ports are part of LARMORX itself (pip install larmorx). AFNI 26.3.05 releases the Medical College of Wisconsin’s code, which these programs’ results depend on, under CC BY 4.0. So the replica moved from the separate GPL package larmorx-gpl into LARMORX, under the Apache License with CC BY 4.0’s attribution (NOTICE, licensing), and runs in the same process. The replica is translated from AFNI’s source. The clean room is written without it, from a description of what the program does (specs/3dTshift.md).

implementation="auto" | "replica" | "clean-room" chooses between them ("original" in 0.1.0rc1). On the command line, larmorx --implementation clean-room afni 3dTshift ... does the same from the next release; in 0.1.0rc1 it is the environment variable LARMORX_IMPLEMENTATION. auto takes the replica (before 2026-10-29 only where larmorx-gpl was installed, the clean room otherwise). The examples below were computed with the replica as the separate program it was then; the same files come out of it in the same process (its records, made again in the same process on 2026-10-29, agree case for case).

The C library’s last digits (since 2026-11-04). With -wsinc5 and -wsinc9, 3dTshift weights its neighbours by a windowed sinc whose sinf and cosf AFNI takes from glibc, the C library of Linux. glibc’s results are now and then one unit in the last place away 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 both its fix and its reproduce name GLIBC-07 exactly where a weight differs (about 1 % of the weights; register). Two cases show it, methods/glibc-sinf-wsinc5 and methods/glibc-sinf-wsinc9: slice times where the default’s output differs from AFNI’s in 2 and 14 values. In reproduce they are AFNI’s bytes, and the replica’s record has them too. fMRIPrep’s calls interpolate by Fourier, which calls neither function: the examples below are unchanged, the clean room’s runs giving the same bytes as before.

The verdicts at a glance

Agreement, replica against the original: AFNI 3dTshift 26.3.05 (standard comparison, 317 cases of the record), by category and verdict
synthetic series33
interpolation methods86
options42
slice patterns (-tpattern)20
slice times from the header13
defects of AFNI8
geometry3
copies of the input8
errors21
global options26
environment variables34
expected divergences2
real images20
random testing1

Select a segment to list its cases below.

Agreement, clean room against the original: AFNI 3dTshift 26.3.05 (standard comparison, 317 cases of the record), by category and verdict
synthetic series33
interpolation methods86
options42
slice patterns (-tpattern)20
slice times from the header13
defects of AFNI8
geometry3
copies of the input8
errors21
global options26
environment variables34
expected divergences2
real images20
random testing1

Select a segment to list its cases below.

Each record runs 317 cases against AFNI 26.3.05: the clean room’s two cases of the weighted sinc’s functions (above) joined its record on 2026-11-04, and the replica’s record has had them since the same day. Since 2026-10-29 both ports follow AFNI 26.3.05, AFNI’s newest release, and their records run against it, with AFNI 25.2.09 run beside it on every case (AFNI’s own step): the replica since the version update, the clean room since its alignment the same day. Between the two releases this program changed in exactly two places, both corrections LARMORX already made by default: header slice code 2 read as NIfTI defines it (AFNI-03), and -no_detrend de-meaning every voxel (AFNI-05). In reproduce both ports now give 26.3.05’s result there and name neither. The examples below show both versions’ outputs: on fMRIPrep’s calls they are the same bytes.

The replica gives AFNI’s bytes in every case that produces an output, and AFNI’s exit status and messages in the others. The clean room is bit-identical wherever it does not interpolate by Fourier, and within the record’s thresholds where it does. AFNI computes the Fourier shift in float32 with its own FFT, the clean room in double precision, so float outputs differ in the last bits and integer outputs now and then round the other way, by one. Those differences are marked open: explained, small, and still closable.

In fMRIPrep: the slice-timing correction of a real BOLD run

Worked example

fMRIPrep's slice-timing correction of a real BOLD run

What this shows: On fMRIPrep’s call the replica gives AFNI’s values, every one, in both releases, on 1 and 12 threads, and fMRIPrep’s own output. The clean room differs from AFNI in 189 stored values, each by one unit of the scale factor, within the record’s thresholds: AFNI shifts in float32 with its own FFT, the clean room in double precision. No correction by default changes this call.

The 3dTshift node fMRIPrep 25.2.6 ran on OpenNeuro ds005040 sub-002 (task clip, run 1): its arguments and the slice times it wrote, given to AFNI 25.2.09, to AFNI 26.3.05 (both ports’ reference since 2026-10-29) and to both ports, in fix and in reproduce and at 1 and 12 threads. fMRIPrep’s own output is the column it produced in its run. The clean room is compared with both releases; its record’s reference is AFNI 26.3.05.

Data
OpenNeuro ds005040, task clip run 1, sub-002 (CC0-1.0): 64 × 64 × 40 × 63 int16 values, 63 volumes; shown inside fMRIPrep’s BOLD brain mask, dilated by one voxel
Node
fMRIPrep 25.2.6, bold_stc_wf/slice_timing_correction
Command
3dTshift -ignore 0 -prefix sub-002_task-clip_run-1_bold_tshift.nii.gz -tpattern @slice_timing.1D -TR 2s -tzero 0.975 …/sub-002_task-clip_run-1_bold.nii.gz
Slice times
40 slices, ascending, 0 to 1.95 s into each TR of 2 s, shifted to 0.975 s
Programs
AFNI 25.2.09 and AFNI 26.3.05 (the repository’s builds); larmorx-gpl 0.0.1 (the replica); the clean room at commit a58978f4dfd1
AFNI versions
AFNI 26.3.05 (both ports’ reference since 2026-10-29) on the same arguments as AFNI 25.2.09: the same bytes in every output of fMRIPrep’s call. What changed between the two releases in these programs (header slice code 2, AFNI-03; -no_detrend, AFNI-05; both corrected by LARMORX before AFNI fixed them) does not meet this call.
Inputds005040 sub-002 BOLD, task clip run 1int16 × scale factor; 40 ascending slices over 1.95 s, TR 2 s
OriginalAFNI 3dTshift 25.2.09the repository's build
Replicareplicalarmorx, Apache-2.0 with CC BY 4.0's attribution · fix (the default), 12 threads · against AFNI 3dTshift 26.3.05, which it follows
Clean roomclean roomlarmorx, Apache-2.0 · fix (the default), 12 threads · against AFNI 3dTshift 26.3.05, which it follows

Bit-identical10,321,920 of 10,321,920 voxels equal

the original against itself: fMRIPrep's run and the repository's buildfMRIPrep's output − AFNI 3dTshift 25.2.09

Bit-identical10,321,920 of 10,321,920 voxels equal

the replica against the originalreplica − AFNI 3dTshift 25.2.09
the clean room against the originalclean room − AFNI 3dTshift 25.2.09scale ±47.7

Axial slice 11 of 40, volume 15 of 63, in the window of the brain's 0.5–99.5 %. The interactive viewer needs JavaScript; the metrics below are complete without it.

Who gives which bytes on this run

    • fMRIPrep 25.2.6, as it ranAFNI 25.2.09 in its image
    • AFNI 3dTshift 25.2.09the repository's build
    • AFNI 3dTshift 26.3.05the repository's build
    • replica1 and 12 threads, both modes
    • clean room1 and 12 threads, both modes

Each box holds runs that give the same values. The clean room differs from AFNI in 189 of 10,321,920 values, each by one stored unit.

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
AFNI 3dTshift 26.3.05 − 25.2.09referenceAFNI 26.3.05 gives AFNI 25.2.09’s bytes on this call: the version step changes nothing here (its two changes, AFNI-03 and AFNI-05, need header slice code 2 or -no_detrend).identical: 0 of 10,321,920 values differ
fMRIPrep's output − AFNI 3dTshift 25.2.09referenceTwo builds of AFNI 25.2.09 (fMRIPrep’s image and the repository’s) give the same values: AFNI’s own variability on this run is zero.identical: 0 of 10,321,920 values differ
The replica against the original and fMRIPrep's output
replica − AFNI 3dTshift 25.2.09Bit-identicalThe same values, every one.identical: 0 of 10,321,920 values differ
replica − AFNI 3dTshift 26.3.05Bit-identicalThe same values, every one.identical: 0 of 10,321,920 values differ
replica − fMRIPrep's outputBit-identicalThe same values, every one.identical: 0 of 10,321,920 values differ
The replica against itself: thread counts, fix and reproduce
replica, 12 threads − 1 threadBit-identicalThe same bytes on any number of threads.identical: 0 of 10,321,920 values differ
replica, fix − reproduceBit-identicalNo correction of the register applies to this run: fix and reproduce give the same bytes.identical: 0 of 10,321,920 values differ
The clean room against the original and fMRIPrep's output
clean room − AFNI 3dTshift 25.2.09Open difference189 stored values differ by one unit (the scale factor), within the record’s thresholds. AFNI computes the Fourier shift in float32 with its own FFT, the clean room in double precision: float outputs differ in their last bits, integer outputs now and then round the other way, by one unit. AFNI’s own builds agree exactly (reference row), so this is not within AFNI’s variability; the replica shows it can be closed.47.7 (1 stored unit)189 of 10,321,920 (0.00183 %)0.2040.0002442.56e−11
clean room − AFNI 3dTshift 26.3.05Open difference189 stored values differ by one unit (the scale factor), within the record’s thresholds. AFNI computes the Fourier shift in float32 with its own FFT, the clean room in double precision: float outputs differ in their last bits, integer outputs now and then round the other way, by one unit. AFNI’s own builds agree exactly (reference row), so this is not within AFNI’s variability; the replica shows it can be closed.47.7 (1 stored unit)189 of 10,321,920 (0.00183 %)0.2040.0002442.56e−11
clean room − fMRIPrep's outputOpen differenceThe same values as against AFNI above: fMRIPrep’s AFNI gives the repository AFNI’s bytes.47.7 (1 stored unit)189 of 10,321,920 (0.00183 %)0.2040.0002442.56e−11
The clean room against itself: thread counts, fix and reproduce
clean room, 12 threads − 1 threadBit-identicalThe same bytes on any number of threads.identical: 0 of 10,321,920 values differ
clean room, fix − reproduceBit-identicalNo correction of the register applies to this run: fix and reproduce give the same bytes.identical: 0 of 10,321,920 values differ

Δ is A − B, in the image’s units (stored value × scale factor); the stored units are given in brackets. Computed in float64 over all 10,321,920 values of the 64 × 64 × 40 × 63 series, not only the brain shown. The records’ thresholds: integer outputs within one stored unit, with the share that differs reported; bit-identity reported.

Provenance of this example
  • Data: OpenNeuro ds005040, task clip run 1, sub-002, CC0-1.0; fMRIPrep 25.2.6’s run ds005040-sub-002-noreconall for its own output and the brain mask
  • Inputs: BOLD (sha256 590cda2ebbff3701…); fMRIPrep’s output (sha256 ab54f976e126f608…); BOLD with a negative scale factor (made from BOLD) (sha256 d54b58e7c8b2d4c8…)
  • Runs: figures cache entry afni-tshift/fmriprep-ds005040-clip-1, computed 2026-10-21T09:26:38+00:00 with larmorx a58978f4dfd1, AFNI 25.2.09, larmorx-gpl 0.0.1; every command, its environment, threads and exit status are in its RUN_INFO.json

Every number above is in afni-tshift.json, exported by python -m larmorx_validation evidence afni-tshift.

A deliberate correction: a negative scale factor (AFNI-01)

The same run stored with a negative scale factor is the same data: every value is unchanged. AFNI applies that factor on output only, so its result comes out sign-flipped and scaled (AFNI-01). Both ports correct it by default, fix, and warn; reproduce gives AFNI’s result.

Worked example

AFNI-01: the same run, stored with a negative scale factor

What this shows: Stored with a negative scale factor, the same values come out of AFNI sign-flipped and scaled, in both releases (AFNI-01). In fix, their default, both ports correct it with a warning: the replica gives AFNI’s result on the usual encoding, the clean room its own. In reproduce the replica gives AFNI’s result, defect included, and the clean room does too, but for 2 values off by one stored unit.

NIfTI lets integer data carry a negative scale factor: the run below holds the same values as above, its stored integers negated and its scale factor −47.67. fMRIPrep’s command, given to AFNI 25.2.09, to AFNI 26.3.05 and to both ports in fix and in reproduce.

Input
the run above, re-stored by larmorx_validation.evidence.afni_tshift.negative_slope: only scl_slope and the data bytes change
Defect
AFNI-01: A negative scl_slope flips the data; corrected by default in both ports, with a warning
Command
fMRIPrep’s, as in the example above
AFNI versions
AFNI 26.3.05 (both ports’ reference since 2026-10-29) on the same arguments as AFNI 25.2.09: the same bytes in every output of the call on either encoding. What changed between the two releases in these programs (header slice code 2, AFNI-03; -no_detrend, AFNI-05; both corrected by LARMORX before AFNI fixed them) does not meet this call.
InputThe same run, negative scale factorstored integers negated, scale factor −47.67: the same values
OriginalAFNI 3dTshift 25.2.09on the negative scale factor
Replicareplicalarmorx, Apache-2.0 with CC BY 4.0's attribution · fix (the default), corrected · against AFNI 3dTshift 25.2.09
Replicareplicalarmorx, Apache-2.0 with CC BY 4.0's attribution · reproduce · against AFNI 3dTshift 25.2.09

Bit-identical10,321,920 of 10,321,920 voxels equal

the replica against the original: fix, on the usual encodingreplica (fix) − AFNI 3dTshift 25.2.09 on the usual encoding
the replica against the originalreplica (fix) − AFNI 3dTshift 25.2.09, same inputscale ±1.99e5

Bit-identical10,321,920 of 10,321,920 voxels equal

the replica against the originalreplica (reproduce) − AFNI 3dTshift 25.2.09, same input

Axial slice 11 of 40, volume 15 of 63, in the window of the brain's 0.5–99.5 %. The interactive viewer needs JavaScript; the metrics below are complete without it.

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 input against itself: two encodings
the two inputs' valuesreferenceThe two encodings hold the same values, every one. AFNI-01identical: 0 of 10,321,920 values differ
The original against itself: the defect
AFNI 3dTshift 25.2.09 on the negative scale factor − on the usual encodingreferenceAFNI’s result changes with the encoding of the same data: AFNI applies the negative scale factor on output only, so the shifted slices come out sign-flipped and scaled (AFNI-01). AFNI-011.99e59,161,285 of 10,321,920 (88.8 %)3.54e41.022
The original against itself: the yardstick
AFNI 3dTshift 26.3.05 − 25.2.09, on the negative scale factorreferenceAFNI 26.3.05 gives AFNI 25.2.09’s bytes on this call: the version step changes nothing here (its two changes, AFNI-03 and AFNI-05, need header slice code 2 or -no_detrend). AFNI-01 is in both releases. AFNI-01identical: 0 of 10,321,920 values differ
The replica against the original
replica (fix) − AFNI 3dTshift 25.2.09 on the usual encodingBit-identicalThe corrected result is AFNI’s own result on the same values stored the usual way. AFNI-01identical: 0 of 10,321,920 values differ
replica (fix) − AFNI 3dTshift 25.2.09, same inputDeliberate correctionAFNI-01, corrected by default with a warning: the replica applies the scale factor to the shifted slices, and gives the values the data call for, those of the usual encoding (the rows on the usual encoding); reproduce gives AFNI’s. AFNI-011.99e5 (4.18e3 stored units)9,161,285 of 10,321,920 (88.8 %)3.54e448.62
replica (reproduce) − AFNI 3dTshift 25.2.09, same inputBit-identicalIn reproduce, the replica gives AFNI’s result, defect included. AFNI-01identical: 0 of 10,321,920 values differ
replica (reproduce) − AFNI 3dTshift 26.3.05, same inputBit-identicalIn reproduce, the replica gives AFNI 26.3.05’s result, defect included (AFNI-01 is unchanged in 26.3.05). AFNI-01identical: 0 of 10,321,920 values differ
The clean room against the original
clean room (fix) − AFNI 3dTshift 25.2.09, same inputDeliberate correctionAFNI-01, corrected by default with a warning: the clean room applies the scale factor to the shifted slices, and gives the values the data call for, those of the usual encoding (the rows on the usual encoding); reproduce gives AFNI’s. AFNI-011.99e5 (4.18e3 stored units)9,161,285 of 10,321,920 (88.8 %)3.54e448.62
clean room (reproduce) − AFNI 3dTshift 25.2.09, same inputOpen differenceIn reproduce, the clean room gives AFNI’s result, defect included, but for 2 values off by one stored unit: AFNI computes the Fourier shift in float32 with its own FFT, the clean room in double precision: float outputs differ in their last bits, integer outputs now and then round the other way, by one unit. AFNI-0147.7 (1 stored unit)2 of 10,321,920 (1.94e−5 %)0.0210.01166.13e−10
clean room (reproduce) − AFNI 3dTshift 26.3.05, same inputOpen differenceIn reproduce, the clean room gives AFNI 26.3.05’s result, defect included, but for 2 values off by one stored unit: AFNI computes the Fourier shift in float32 with its own FFT, the clean room in double precision: float outputs differ in their last bits, integer outputs now and then round the other way, by one unit. AFNI-0147.7 (1 stored unit)2 of 10,321,920 (1.94e−5 %)0.0210.01166.13e−10
The clean room against itself: two encodings
clean room (fix) − itself on the usual encodingBit-identicalThe corrected result is the clean room’s own result on the same values stored the usual way. AFNI-01identical: 0 of 10,321,920 values differ

Δ is A − B, in the image’s units (stored value × scale factor); the stored units are given in brackets. Computed in float64 over all 10,321,920 values of the 64 × 64 × 40 × 63 series, not only the brain shown. The records’ thresholds: integer outputs within one stored unit, with the share that differs reported; bit-identity reported.

Provenance of this example
  • Data: OpenNeuro ds005040, task clip run 1, sub-002, CC0-1.0; fMRIPrep 25.2.6’s run ds005040-sub-002-noreconall for its own output and the brain mask
  • Inputs: BOLD (sha256 590cda2ebbff3701…); fMRIPrep’s output (sha256 ab54f976e126f608…); BOLD with a negative scale factor (made from BOLD) (sha256 d54b58e7c8b2d4c8…)
  • Runs: figures cache entry afni-tshift/fmriprep-ds005040-clip-1, computed 2026-10-21T09:26:38+00:00 with larmorx a58978f4dfd1, AFNI 25.2.09, larmorx-gpl 0.0.1; every command, its environment, threads and exit status are in its RUN_INFO.json

Every number above is in afni-tshift.json, exported by python -m larmorx_validation evidence afni-tshift.

Every case

Each case of the two records, with what it tests, its command line, each port’s verdict and the reason. Search, filter by category or verdict (or select a segment of a chart above), sort by a column; #case=<name> links to one case.

318 cases

CaseCategoryWhat it testsreplicaclean roomChecks
synthetic/content/clippingsynthetic seriesdefault Fourier, -tpattern alt+z: Series swinging between -32760 and 32760: shifted values clip to ±32767 (AFNI never writes -32768).
command line3dTshift -tpattern alt+z -prefix out.nii synthetic/tshift/content/clipping.nii.gz
Bit-identicalOpen difference

8 values differ from AFNI’s, within the record’s thresholds: AFNI’s Fourier shift is float32, the clean room’s double precision.

15/15
synthetic/content/constantsynthetic seriesdefault Fourier, -tpattern alt+z: Constant series (every voxel 500): detrending leaves zeros, the range is a point.
command line3dTshift -tpattern alt+z -prefix out.nii synthetic/tshift/content/constant.nii.gz
Bit-identicalBit-identical15/15
synthetic/content/half-integerssynthetic seriesdefault Fourier, -tpattern alt+z: float32 series made of k + 0.5 values (exact rounding ties when written as int16 by other tools).
command line3dTshift -tpattern alt+z -prefix out.nii synthetic/tshift/content/half-integers.nii.gz
Bit-identicalOpen difference

2,713 values differ from AFNI’s, within the record’s thresholds: AFNI’s Fourier shift is float32, the clean room’s double precision.

15/15
synthetic/content/stepssynthetic seriesdefault Fourier, -tpattern alt+z: Step functions: the Fourier shift rings (Gibbs), so the clipping to the input range matters.
command line3dTshift -tpattern alt+z -prefix out.nii synthetic/tshift/content/steps.nii.gz
Bit-identicalOpen difference

1,218 values differ from AFNI’s, within the record’s thresholds: AFNI’s Fourier shift is float32, the clean room’s double precision.

15/15
synthetic/dtypes/float32-nonfinitesynthetic seriesdefault Fourier, -tpattern alt+z: float32 with NaN, +inf and -inf samples: AFNI reads them as 0.
command line3dTshift -tpattern alt+z -prefix out.nii synthetic/tshift/dtypes/float32-nonfinite.nii.gz
Bit-identicalOpen difference

230 values differ from AFNI’s, within the record’s thresholds: AFNI’s Fourier shift is float32, the clean room’s double precision.

15/15
synthetic/dtypes/float32-slopesynthetic seriesdefault Fourier, -tpattern alt+z: float32 with scl_slope 2 and no intercept: AFNI keeps it as a brick factor even for float data.
command line3dTshift -tpattern alt+z -prefix out.nii synthetic/tshift/dtypes/float32-slope.nii.gz
Bit-identicalOpen difference

167 values differ from AFNI’s, within the record’s thresholds: AFNI’s Fourier shift is float32, the clean room’s double precision.

14/14
synthetic/dtypes/float32synthetic seriesdefault Fourier, -tpattern alt+z: float32: shifted values are stored as computed.
command line3dTshift -tpattern alt+z -prefix out.nii synthetic/tshift/dtypes/float32.nii.gz
Bit-identicalOpen difference

168 values differ from AFNI’s, within the record’s thresholds: AFNI’s Fourier shift is float32, the clean room’s double precision.

15/15
synthetic/dtypes/float64synthetic seriesdefault Fourier, -tpattern alt+z: float64: AFNI converts to float32 on read and writes float32.
command line3dTshift -tpattern alt+z -prefix out.nii synthetic/tshift/dtypes/float64.nii.gz
Bit-identicalOpen difference

168 values differ from AFNI’s, within the record’s thresholds: AFNI’s Fourier shift is float32, the clean room’s double precision.

14/14
synthetic/dtypes/int16-negative-slopesynthetic seriesdefault Fourier, -tpattern alt+z: int16 with scl_slope -0.5: AFNI applies a negative brick factor on output but not on input.
command line3dTshift -tpattern alt+z -prefix out.nii synthetic/tshift/dtypes/int16-negative-slope.nii.gz
Bit-identicalBit-identical15/15 · 14/14
synthetic/dtypes/int16-slope-intersynthetic seriesdefault Fourier, -tpattern alt+z: int16 with scl_slope 0.37 and scl_inter 12.5: AFNI converts to float32.
command line3dTshift -tpattern alt+z -prefix out.nii synthetic/tshift/dtypes/int16-slope-inter.nii.gz
Bit-identicalOpen difference

162 values differ from AFNI’s, within the record’s thresholds: AFNI’s Fourier shift is float32, the clean room’s double precision.

14/14
synthetic/dtypes/int16-slopesynthetic seriesdefault Fourier, -tpattern alt+z: int16 with scl_slope 0.25 and no intercept: a brick factor (output stays int16).
command line3dTshift -tpattern alt+z -prefix out.nii synthetic/tshift/dtypes/int16-slope.nii.gz
Bit-identicalBit-identical14/14
synthetic/dtypes/int16synthetic seriesdefault Fourier, -tpattern alt+z: int16, no scaling: AFNI keeps short and rounds the output with SHORTIZE.
command line3dTshift -tpattern alt+z -prefix out.nii synthetic/tshift/dtypes/int16.nii.gz
Bit-identicalBit-identical15/15
synthetic/dtypes/int32synthetic seriesdefault Fourier, -tpattern alt+z: int32: AFNI converts to float32 on read and writes float32.
command line3dTshift -tpattern alt+z -prefix out.nii synthetic/tshift/dtypes/int32.nii.gz
Bit-identicalOpen difference

167 values differ from AFNI’s, within the record’s thresholds: AFNI’s Fourier shift is float32, the clean room’s double precision.

14/14
synthetic/dtypes/uint8synthetic seriesdefault Fourier, -tpattern alt+z: uint8: AFNI keeps byte and rounds the output with BYTEIZE.
command line3dTshift -tpattern alt+z -prefix out.nii synthetic/tshift/dtypes/uint8.nii.gz
Bit-identicalBit-identical15/15
synthetic/lengths/nt0012-fft16synthetic seriesdefault Fourier, -tpattern alt+z: int16 series of 12 time points: AFNI's FFT length is 16 (fft16).
command line3dTshift -tpattern alt+z -prefix out.nii synthetic/tshift/lengths/nt0012-fft16.nii.gz
Bit-identicalBit-identical15/15
synthetic/lengths/nt0016-fft20synthetic seriesdefault Fourier, -tpattern alt+z: int16 series of 16 time points: AFNI's FFT length is 20 (radix-5).
command line3dTshift -tpattern alt+z -prefix out.nii synthetic/tshift/lengths/nt0016-fft20.nii.gz
Bit-identicalBit-identical15/15
synthetic/lengths/nt0020-fft24synthetic seriesdefault Fourier, -tpattern alt+z: int16 series of 20 time points: AFNI's FFT length is 24 (radix-3).
command line3dTshift -tpattern alt+z -prefix out.nii synthetic/tshift/lengths/nt0020-fft24.nii.gz
Bit-identicalBit-identical15/15
synthetic/lengths/nt0028-fft32synthetic seriesdefault Fourier, -tpattern alt+z: int16 series of 28 time points: AFNI's FFT length is 32 (fft32).
command line3dTshift -tpattern alt+z -prefix out.nii synthetic/tshift/lengths/nt0028-fft32.nii.gz
Bit-identicalBit-identical15/15
synthetic/lengths/nt0056-fft60synthetic seriesdefault Fourier, -tpattern alt+z: int16 series of 56 time points: AFNI's FFT length is 60 (radix-15).
command line3dTshift -tpattern alt+z -prefix out.nii synthetic/tshift/lengths/nt0056-fft60.nii.gz
Bit-identicalBit-identical15/15
synthetic/lengths/nt0060-fft64synthetic seriesdefault Fourier, -tpattern alt+z: int16 series of 60 time points: AFNI's FFT length is 64 (fft64).
command line3dTshift -tpattern alt+z -prefix out.nii synthetic/tshift/lengths/nt0060-fft64.nii.gz
Bit-identicalBit-identical15/15
synthetic/lengths/nt0116-fft120synthetic seriesdefault Fourier, -tpattern alt+z: int16 series of 116 time points: AFNI's FFT length is 120 (radix-15).
command line3dTshift -tpattern alt+z -prefix out.nii synthetic/tshift/lengths/nt0116-fft120.nii.gz
Bit-identicalBit-identical15/15
synthetic/lengths/nt0124-fft128synthetic seriesdefault Fourier, -tpattern alt+z: int16 series of 124 time points: AFNI's FFT length is 128 (fft128).
command line3dTshift -tpattern alt+z -prefix out.nii synthetic/tshift/lengths/nt0124-fft128.nii.gz
Bit-identicalBit-identical15/15
synthetic/lengths/nt0156-fft160synthetic seriesdefault Fourier, -tpattern alt+z: int16 series of 156 time points: AFNI's FFT length is 160 (radix-5).
command line3dTshift -tpattern alt+z -prefix out.nii synthetic/tshift/lengths/nt0156-fft160.nii.gz
Bit-identicalBit-identical15/15
synthetic/lengths/nt0188-fft192synthetic seriesdefault Fourier, -tpattern alt+z: int16 series of 188 time points: AFNI's FFT length is 192 (radix-3).
command line3dTshift -tpattern alt+z -prefix out.nii synthetic/tshift/lengths/nt0188-fft192.nii.gz
Bit-identicalBit-identical15/15
synthetic/lengths/nt0236-fft240synthetic seriesdefault Fourier, -tpattern alt+z: int16 series of 236 time points: AFNI's FFT length is 240 (radix-15).
command line3dTshift -tpattern alt+z -prefix out.nii synthetic/tshift/lengths/nt0236-fft240.nii.gz
Bit-identicalBit-identical15/15
synthetic/lengths/nt0252-fft256synthetic seriesdefault Fourier, -tpattern alt+z: int16 series of 252 time points: AFNI's FFT length is 256 (fft256).
command line3dTshift -tpattern alt+z -prefix out.nii synthetic/tshift/lengths/nt0252-fft256.nii.gz
Bit-identicalBit-identical15/15
synthetic/lengths/nt0476-fft480synthetic seriesdefault Fourier, -tpattern alt+z: int16 series of 476 time points: AFNI's FFT length is 480 (radix-15).
command line3dTshift -tpattern alt+z -prefix out.nii synthetic/tshift/lengths/nt0476-fft480.nii.gz
Bit-identicalOpen difference

1 values differ from AFNI’s, within the record’s thresholds: AFNI’s Fourier shift is float32, the clean room’s double precision.

15/15
synthetic/lengths/nt0508-fft512synthetic seriesdefault Fourier, -tpattern alt+z: int16 series of 508 time points: AFNI's FFT length is 512 (fft512).
command line3dTshift -tpattern alt+z -prefix out.nii synthetic/tshift/lengths/nt0508-fft512.nii.gz
Bit-identicalBit-identical15/15
synthetic/lengths/nt0764-fft768synthetic seriesdefault Fourier, -tpattern alt+z: int16 series of 764 time points: AFNI's FFT length is 768 (radix-3).
command line3dTshift -tpattern alt+z -prefix out.nii synthetic/tshift/lengths/nt0764-fft768.nii.gz
Bit-identicalBit-identical15/15
synthetic/lengths/nt1020-fft1024synthetic seriesdefault Fourier, -tpattern alt+z: int16 series of 1020 time points: AFNI's FFT length is 1024 (fft1024).
command line3dTshift -tpattern alt+z -prefix out.nii synthetic/tshift/lengths/nt1020-fft1024.nii.gz
Bit-identicalOpen difference

2 values differ from AFNI’s, within the record’s thresholds: AFNI’s Fourier shift is float32, the clean room’s double precision.

15/15
synthetic/lengths/nt1276-fft1280synthetic seriesdefault Fourier, -tpattern alt+z: int16 series of 1276 time points: AFNI's FFT length is 1280 (radix-5).
command line3dTshift -tpattern alt+z -prefix out.nii synthetic/tshift/lengths/nt1276-fft1280.nii.gz
Bit-identicalBit-identical15/15
synthetic/lengths/nt2044-fft2048synthetic seriesdefault Fourier, -tpattern alt+z: int16 series of 2044 time points: AFNI's FFT length is 2048 (fft2048).
command line3dTshift -tpattern alt+z -prefix out.nii synthetic/tshift/lengths/nt2044-fft2048.nii.gz
Bit-identicalOpen difference

1 values differ from AFNI’s, within the record’s thresholds: AFNI’s Fourier shift is float32, the clean room’s double precision.

15/15
synthetic/lengths/nt4092-fft4096synthetic seriesdefault Fourier, -tpattern alt+z: int16 series of 4092 time points: AFNI's FFT length is 4096 (fft_4dec).
command line3dTshift -tpattern alt+z -prefix out.nii synthetic/tshift/lengths/nt4092-fft4096.nii.gz
Bit-identicalOpen difference

2 values differ from AFNI’s, within the record’s thresholds: AFNI’s Fourier shift is float32, the clean room’s double precision.

15/15
methods/content/clipping-linearinterpolation methods-linear, -tpattern alt-z, on content/clipping
command line3dTshift -tpattern alt-z -linear -prefix out.nii synthetic/tshift/content/clipping.nii.gz
Bit-identicalBit-identical15/15
methods/content/clipping-cubicinterpolation methods-cubic, -tpattern alt-z, on content/clipping
command line3dTshift -tpattern alt-z -cubic -prefix out.nii synthetic/tshift/content/clipping.nii.gz
Bit-identicalBit-identical15/15
methods/content/clipping-quinticinterpolation methods-quintic, -tpattern alt-z, on content/clipping
command line3dTshift -tpattern alt-z -quintic -prefix out.nii synthetic/tshift/content/clipping.nii.gz
Bit-identicalBit-identical15/15
methods/content/clipping-hepticinterpolation methods-heptic, -tpattern alt-z, on content/clipping
command line3dTshift -tpattern alt-z -heptic -prefix out.nii synthetic/tshift/content/clipping.nii.gz
Bit-identicalBit-identical15/15
methods/content/clipping-wsinc5interpolation methods-wsinc5, -tpattern alt-z, on content/clipping
command line3dTshift -tpattern alt-z -wsinc5 -prefix out.nii synthetic/tshift/content/clipping.nii.gz
Bit-identicalBit-identical15/15
methods/content/clipping-wsinc9interpolation methods-wsinc9, -tpattern alt-z, on content/clipping
command line3dTshift -tpattern alt-z -wsinc9 -prefix out.nii synthetic/tshift/content/clipping.nii.gz
Bit-identicalBit-identical15/15
methods/content/constant-linearinterpolation methods-linear, -tpattern alt-z, on content/constant
command line3dTshift -tpattern alt-z -linear -prefix out.nii synthetic/tshift/content/constant.nii.gz
Bit-identicalBit-identical15/15
methods/content/constant-cubicinterpolation methods-cubic, -tpattern alt-z, on content/constant
command line3dTshift -tpattern alt-z -cubic -prefix out.nii synthetic/tshift/content/constant.nii.gz
Bit-identicalBit-identical15/15
methods/content/constant-quinticinterpolation methods-quintic, -tpattern alt-z, on content/constant
command line3dTshift -tpattern alt-z -quintic -prefix out.nii synthetic/tshift/content/constant.nii.gz
Bit-identicalBit-identical15/15
methods/content/constant-hepticinterpolation methods-heptic, -tpattern alt-z, on content/constant
command line3dTshift -tpattern alt-z -heptic -prefix out.nii synthetic/tshift/content/constant.nii.gz
Bit-identicalBit-identical15/15
methods/content/constant-wsinc5interpolation methods-wsinc5, -tpattern alt-z, on content/constant
command line3dTshift -tpattern alt-z -wsinc5 -prefix out.nii synthetic/tshift/content/constant.nii.gz
Bit-identicalBit-identical15/15
methods/content/constant-wsinc9interpolation methods-wsinc9, -tpattern alt-z, on content/constant
command line3dTshift -tpattern alt-z -wsinc9 -prefix out.nii synthetic/tshift/content/constant.nii.gz
Bit-identicalBit-identical15/15
methods/content/half-integers-linearinterpolation methods-linear, -tpattern alt-z, on content/half-integers
command line3dTshift -tpattern alt-z -linear -prefix out.nii synthetic/tshift/content/half-integers.nii.gz
Bit-identicalBit-identical15/15
methods/content/half-integers-cubicinterpolation methods-cubic, -tpattern alt-z, on content/half-integers
command line3dTshift -tpattern alt-z -cubic -prefix out.nii synthetic/tshift/content/half-integers.nii.gz
Bit-identicalBit-identical15/15
methods/content/half-integers-quinticinterpolation methods-quintic, -tpattern alt-z, on content/half-integers
command line3dTshift -tpattern alt-z -quintic -prefix out.nii synthetic/tshift/content/half-integers.nii.gz
Bit-identicalBit-identical15/15
methods/content/half-integers-hepticinterpolation methods-heptic, -tpattern alt-z, on content/half-integers
command line3dTshift -tpattern alt-z -heptic -prefix out.nii synthetic/tshift/content/half-integers.nii.gz
Bit-identicalBit-identical15/15
methods/content/half-integers-wsinc5interpolation methods-wsinc5, -tpattern alt-z, on content/half-integers
command line3dTshift -tpattern alt-z -wsinc5 -prefix out.nii synthetic/tshift/content/half-integers.nii.gz
Bit-identicalBit-identical15/15
methods/content/half-integers-wsinc9interpolation methods-wsinc9, -tpattern alt-z, on content/half-integers
command line3dTshift -tpattern alt-z -wsinc9 -prefix out.nii synthetic/tshift/content/half-integers.nii.gz
Bit-identicalBit-identical15/15
methods/content/steps-linearinterpolation methods-linear, -tpattern alt-z, on content/steps
command line3dTshift -tpattern alt-z -linear -prefix out.nii synthetic/tshift/content/steps.nii.gz
Bit-identicalBit-identical15/15
methods/content/steps-cubicinterpolation methods-cubic, -tpattern alt-z, on content/steps
command line3dTshift -tpattern alt-z -cubic -prefix out.nii synthetic/tshift/content/steps.nii.gz
Bit-identicalBit-identical15/15
methods/content/steps-quinticinterpolation methods-quintic, -tpattern alt-z, on content/steps
command line3dTshift -tpattern alt-z -quintic -prefix out.nii synthetic/tshift/content/steps.nii.gz
Bit-identicalBit-identical15/15
methods/content/steps-hepticinterpolation methods-heptic, -tpattern alt-z, on content/steps
command line3dTshift -tpattern alt-z -heptic -prefix out.nii synthetic/tshift/content/steps.nii.gz
Bit-identicalBit-identical15/15
methods/content/steps-wsinc5interpolation methods-wsinc5, -tpattern alt-z, on content/steps
command line3dTshift -tpattern alt-z -wsinc5 -prefix out.nii synthetic/tshift/content/steps.nii.gz
Bit-identicalBit-identical15/15
methods/content/steps-wsinc9interpolation methods-wsinc9, -tpattern alt-z, on content/steps
command line3dTshift -tpattern alt-z -wsinc9 -prefix out.nii synthetic/tshift/content/steps.nii.gz
Bit-identicalBit-identical15/15
methods/dtypes/float32-nonfinite-linearinterpolation methods-linear, -tpattern alt-z, on dtypes/float32-nonfinite
command line3dTshift -tpattern alt-z -linear -prefix out.nii synthetic/tshift/dtypes/float32-nonfinite.nii.gz
Bit-identicalBit-identical15/15
methods/dtypes/float32-nonfinite-cubicinterpolation methods-cubic, -tpattern alt-z, on dtypes/float32-nonfinite
command line3dTshift -tpattern alt-z -cubic -prefix out.nii synthetic/tshift/dtypes/float32-nonfinite.nii.gz
Bit-identicalBit-identical15/15
methods/dtypes/float32-nonfinite-quinticinterpolation methods-quintic, -tpattern alt-z, on dtypes/float32-nonfinite
command line3dTshift -tpattern alt-z -quintic -prefix out.nii synthetic/tshift/dtypes/float32-nonfinite.nii.gz
Bit-identicalBit-identical15/15
methods/dtypes/float32-nonfinite-hepticinterpolation methods-heptic, -tpattern alt-z, on dtypes/float32-nonfinite
command line3dTshift -tpattern alt-z -heptic -prefix out.nii synthetic/tshift/dtypes/float32-nonfinite.nii.gz
Bit-identicalBit-identical15/15
methods/dtypes/float32-nonfinite-wsinc5interpolation methods-wsinc5, -tpattern alt-z, on dtypes/float32-nonfinite
command line3dTshift -tpattern alt-z -wsinc5 -prefix out.nii synthetic/tshift/dtypes/float32-nonfinite.nii.gz
Bit-identicalBit-identical15/15
methods/dtypes/float32-nonfinite-wsinc9interpolation methods-wsinc9, -tpattern alt-z, on dtypes/float32-nonfinite
command line3dTshift -tpattern alt-z -wsinc9 -prefix out.nii synthetic/tshift/dtypes/float32-nonfinite.nii.gz
Bit-identicalBit-identical15/15
methods/dtypes/float32-slope-linearinterpolation methods-linear, -tpattern alt-z, on dtypes/float32-slope
command line3dTshift -tpattern alt-z -linear -prefix out.nii synthetic/tshift/dtypes/float32-slope.nii.gz
Bit-identicalBit-identical14/14
methods/dtypes/float32-slope-cubicinterpolation methods-cubic, -tpattern alt-z, on dtypes/float32-slope
command line3dTshift -tpattern alt-z -cubic -prefix out.nii synthetic/tshift/dtypes/float32-slope.nii.gz
Bit-identicalBit-identical14/14
methods/dtypes/float32-slope-quinticinterpolation methods-quintic, -tpattern alt-z, on dtypes/float32-slope
command line3dTshift -tpattern alt-z -quintic -prefix out.nii synthetic/tshift/dtypes/float32-slope.nii.gz
Bit-identicalBit-identical14/14
methods/dtypes/float32-slope-hepticinterpolation methods-heptic, -tpattern alt-z, on dtypes/float32-slope
command line3dTshift -tpattern alt-z -heptic -prefix out.nii synthetic/tshift/dtypes/float32-slope.nii.gz
Bit-identicalBit-identical14/14
methods/dtypes/float32-slope-wsinc5interpolation methods-wsinc5, -tpattern alt-z, on dtypes/float32-slope
command line3dTshift -tpattern alt-z -wsinc5 -prefix out.nii synthetic/tshift/dtypes/float32-slope.nii.gz
Bit-identicalBit-identical14/14
methods/dtypes/float32-slope-wsinc9interpolation methods-wsinc9, -tpattern alt-z, on dtypes/float32-slope
command line3dTshift -tpattern alt-z -wsinc9 -prefix out.nii synthetic/tshift/dtypes/float32-slope.nii.gz
Bit-identicalBit-identical14/14
methods/dtypes/float32-linearinterpolation methods-linear, -tpattern alt-z, on dtypes/float32
command line3dTshift -tpattern alt-z -linear -prefix out.nii synthetic/tshift/dtypes/float32.nii.gz
Bit-identicalBit-identical15/15
methods/dtypes/float32-cubicinterpolation methods-cubic, -tpattern alt-z, on dtypes/float32
command line3dTshift -tpattern alt-z -cubic -prefix out.nii synthetic/tshift/dtypes/float32.nii.gz
Bit-identicalBit-identical15/15
methods/dtypes/float32-quinticinterpolation methods-quintic, -tpattern alt-z, on dtypes/float32
command line3dTshift -tpattern alt-z -quintic -prefix out.nii synthetic/tshift/dtypes/float32.nii.gz
Bit-identicalBit-identical15/15
methods/dtypes/float32-hepticinterpolation methods-heptic, -tpattern alt-z, on dtypes/float32
command line3dTshift -tpattern alt-z -heptic -prefix out.nii synthetic/tshift/dtypes/float32.nii.gz
Bit-identicalBit-identical15/15
methods/dtypes/float32-wsinc5interpolation methods-wsinc5, -tpattern alt-z, on dtypes/float32
command line3dTshift -tpattern alt-z -wsinc5 -prefix out.nii synthetic/tshift/dtypes/float32.nii.gz
Bit-identicalBit-identical15/15
methods/dtypes/float32-wsinc9interpolation methods-wsinc9, -tpattern alt-z, on dtypes/float32
command line3dTshift -tpattern alt-z -wsinc9 -prefix out.nii synthetic/tshift/dtypes/float32.nii.gz
Bit-identicalBit-identical15/15
methods/dtypes/float64-linearinterpolation methods-linear, -tpattern alt-z, on dtypes/float64
command line3dTshift -tpattern alt-z -linear -prefix out.nii synthetic/tshift/dtypes/float64.nii.gz
Bit-identicalBit-identical14/14
methods/dtypes/float64-cubicinterpolation methods-cubic, -tpattern alt-z, on dtypes/float64
command line3dTshift -tpattern alt-z -cubic -prefix out.nii synthetic/tshift/dtypes/float64.nii.gz
Bit-identicalBit-identical14/14
methods/dtypes/float64-quinticinterpolation methods-quintic, -tpattern alt-z, on dtypes/float64
command line3dTshift -tpattern alt-z -quintic -prefix out.nii synthetic/tshift/dtypes/float64.nii.gz
Bit-identicalBit-identical14/14
methods/dtypes/float64-hepticinterpolation methods-heptic, -tpattern alt-z, on dtypes/float64
command line3dTshift -tpattern alt-z -heptic -prefix out.nii synthetic/tshift/dtypes/float64.nii.gz
Bit-identicalBit-identical14/14
methods/dtypes/float64-wsinc5interpolation methods-wsinc5, -tpattern alt-z, on dtypes/float64
command line3dTshift -tpattern alt-z -wsinc5 -prefix out.nii synthetic/tshift/dtypes/float64.nii.gz
Bit-identicalBit-identical14/14
methods/dtypes/float64-wsinc9interpolation methods-wsinc9, -tpattern alt-z, on dtypes/float64
command line3dTshift -tpattern alt-z -wsinc9 -prefix out.nii synthetic/tshift/dtypes/float64.nii.gz
Bit-identicalBit-identical14/14
methods/dtypes/int16-negative-slope-linearinterpolation methods-linear, -tpattern alt-z, on dtypes/int16-negative-slope
command line3dTshift -tpattern alt-z -linear -prefix out.nii synthetic/tshift/dtypes/int16-negative-slope.nii.gz
Bit-identicalBit-identical15/15 · 14/14
methods/dtypes/int16-negative-slope-cubicinterpolation methods-cubic, -tpattern alt-z, on dtypes/int16-negative-slope
command line3dTshift -tpattern alt-z -cubic -prefix out.nii synthetic/tshift/dtypes/int16-negative-slope.nii.gz
Bit-identicalBit-identical15/15 · 14/14
methods/dtypes/int16-negative-slope-quinticinterpolation methods-quintic, -tpattern alt-z, on dtypes/int16-negative-slope
command line3dTshift -tpattern alt-z -quintic -prefix out.nii synthetic/tshift/dtypes/int16-negative-slope.nii.gz
Bit-identicalBit-identical15/15 · 14/14
methods/dtypes/int16-negative-slope-hepticinterpolation methods-heptic, -tpattern alt-z, on dtypes/int16-negative-slope
command line3dTshift -tpattern alt-z -heptic -prefix out.nii synthetic/tshift/dtypes/int16-negative-slope.nii.gz
Bit-identicalBit-identical15/15 · 14/14
methods/dtypes/int16-negative-slope-wsinc5interpolation methods-wsinc5, -tpattern alt-z, on dtypes/int16-negative-slope
command line3dTshift -tpattern alt-z -wsinc5 -prefix out.nii synthetic/tshift/dtypes/int16-negative-slope.nii.gz
Bit-identicalBit-identical15/15 · 14/14
methods/dtypes/int16-negative-slope-wsinc9interpolation methods-wsinc9, -tpattern alt-z, on dtypes/int16-negative-slope
command line3dTshift -tpattern alt-z -wsinc9 -prefix out.nii synthetic/tshift/dtypes/int16-negative-slope.nii.gz
Bit-identicalBit-identical15/15 · 14/14
methods/dtypes/int16-slope-inter-linearinterpolation methods-linear, -tpattern alt-z, on dtypes/int16-slope-inter
command line3dTshift -tpattern alt-z -linear -prefix out.nii synthetic/tshift/dtypes/int16-slope-inter.nii.gz
Bit-identicalBit-identical14/14
methods/dtypes/int16-slope-inter-cubicinterpolation methods-cubic, -tpattern alt-z, on dtypes/int16-slope-inter
command line3dTshift -tpattern alt-z -cubic -prefix out.nii synthetic/tshift/dtypes/int16-slope-inter.nii.gz
Bit-identicalBit-identical14/14
methods/dtypes/int16-slope-inter-quinticinterpolation methods-quintic, -tpattern alt-z, on dtypes/int16-slope-inter
command line3dTshift -tpattern alt-z -quintic -prefix out.nii synthetic/tshift/dtypes/int16-slope-inter.nii.gz
Bit-identicalBit-identical14/14
methods/dtypes/int16-slope-inter-hepticinterpolation methods-heptic, -tpattern alt-z, on dtypes/int16-slope-inter
command line3dTshift -tpattern alt-z -heptic -prefix out.nii synthetic/tshift/dtypes/int16-slope-inter.nii.gz
Bit-identicalBit-identical14/14
methods/dtypes/int16-slope-inter-wsinc5interpolation methods-wsinc5, -tpattern alt-z, on dtypes/int16-slope-inter
command line3dTshift -tpattern alt-z -wsinc5 -prefix out.nii synthetic/tshift/dtypes/int16-slope-inter.nii.gz
Bit-identicalBit-identical14/14
methods/dtypes/int16-slope-inter-wsinc9interpolation methods-wsinc9, -tpattern alt-z, on dtypes/int16-slope-inter
command line3dTshift -tpattern alt-z -wsinc9 -prefix out.nii synthetic/tshift/dtypes/int16-slope-inter.nii.gz
Bit-identicalBit-identical14/14
methods/dtypes/int16-slope-linearinterpolation methods-linear, -tpattern alt-z, on dtypes/int16-slope
command line3dTshift -tpattern alt-z -linear -prefix out.nii synthetic/tshift/dtypes/int16-slope.nii.gz
Bit-identicalBit-identical14/14
methods/dtypes/int16-slope-cubicinterpolation methods-cubic, -tpattern alt-z, on dtypes/int16-slope
command line3dTshift -tpattern alt-z -cubic -prefix out.nii synthetic/tshift/dtypes/int16-slope.nii.gz
Bit-identicalBit-identical14/14
methods/dtypes/int16-slope-quinticinterpolation methods-quintic, -tpattern alt-z, on dtypes/int16-slope
command line3dTshift -tpattern alt-z -quintic -prefix out.nii synthetic/tshift/dtypes/int16-slope.nii.gz
Bit-identicalBit-identical14/14
methods/dtypes/int16-slope-hepticinterpolation methods-heptic, -tpattern alt-z, on dtypes/int16-slope
command line3dTshift -tpattern alt-z -heptic -prefix out.nii synthetic/tshift/dtypes/int16-slope.nii.gz
Bit-identicalBit-identical14/14
methods/dtypes/int16-slope-wsinc5interpolation methods-wsinc5, -tpattern alt-z, on dtypes/int16-slope
command line3dTshift -tpattern alt-z -wsinc5 -prefix out.nii synthetic/tshift/dtypes/int16-slope.nii.gz
Bit-identicalBit-identical14/14
methods/dtypes/int16-slope-wsinc9interpolation methods-wsinc9, -tpattern alt-z, on dtypes/int16-slope
command line3dTshift -tpattern alt-z -wsinc9 -prefix out.nii synthetic/tshift/dtypes/int16-slope.nii.gz
Bit-identicalBit-identical14/14
methods/dtypes/int16-linearinterpolation methods-linear, -tpattern alt-z, on dtypes/int16
command line3dTshift -tpattern alt-z -linear -prefix out.nii synthetic/tshift/dtypes/int16.nii.gz
Bit-identicalBit-identical15/15
methods/dtypes/int16-cubicinterpolation methods-cubic, -tpattern alt-z, on dtypes/int16
command line3dTshift -tpattern alt-z -cubic -prefix out.nii synthetic/tshift/dtypes/int16.nii.gz
Bit-identicalBit-identical15/15
methods/dtypes/int16-quinticinterpolation methods-quintic, -tpattern alt-z, on dtypes/int16
command line3dTshift -tpattern alt-z -quintic -prefix out.nii synthetic/tshift/dtypes/int16.nii.gz
Bit-identicalBit-identical15/15
methods/dtypes/int16-hepticinterpolation methods-heptic, -tpattern alt-z, on dtypes/int16
command line3dTshift -tpattern alt-z -heptic -prefix out.nii synthetic/tshift/dtypes/int16.nii.gz
Bit-identicalBit-identical15/15
methods/dtypes/int16-wsinc5interpolation methods-wsinc5, -tpattern alt-z, on dtypes/int16
command line3dTshift -tpattern alt-z -wsinc5 -prefix out.nii synthetic/tshift/dtypes/int16.nii.gz
Bit-identicalBit-identical15/15
methods/dtypes/int16-wsinc9interpolation methods-wsinc9, -tpattern alt-z, on dtypes/int16
command line3dTshift -tpattern alt-z -wsinc9 -prefix out.nii synthetic/tshift/dtypes/int16.nii.gz
Bit-identicalBit-identical15/15
methods/dtypes/int32-linearinterpolation methods-linear, -tpattern alt-z, on dtypes/int32
command line3dTshift -tpattern alt-z -linear -prefix out.nii synthetic/tshift/dtypes/int32.nii.gz
Bit-identicalBit-identical14/14
methods/dtypes/int32-cubicinterpolation methods-cubic, -tpattern alt-z, on dtypes/int32
command line3dTshift -tpattern alt-z -cubic -prefix out.nii synthetic/tshift/dtypes/int32.nii.gz
Bit-identicalBit-identical14/14
methods/dtypes/int32-quinticinterpolation methods-quintic, -tpattern alt-z, on dtypes/int32
command line3dTshift -tpattern alt-z -quintic -prefix out.nii synthetic/tshift/dtypes/int32.nii.gz
Bit-identicalBit-identical14/14
methods/dtypes/int32-hepticinterpolation methods-heptic, -tpattern alt-z, on dtypes/int32
command line3dTshift -tpattern alt-z -heptic -prefix out.nii synthetic/tshift/dtypes/int32.nii.gz
Bit-identicalBit-identical14/14
methods/dtypes/int32-wsinc5interpolation methods-wsinc5, -tpattern alt-z, on dtypes/int32
command line3dTshift -tpattern alt-z -wsinc5 -prefix out.nii synthetic/tshift/dtypes/int32.nii.gz
Bit-identicalBit-identical14/14
methods/dtypes/int32-wsinc9interpolation methods-wsinc9, -tpattern alt-z, on dtypes/int32
command line3dTshift -tpattern alt-z -wsinc9 -prefix out.nii synthetic/tshift/dtypes/int32.nii.gz
Bit-identicalBit-identical14/14
methods/dtypes/uint8-linearinterpolation methods-linear, -tpattern alt-z, on dtypes/uint8
command line3dTshift -tpattern alt-z -linear -prefix out.nii synthetic/tshift/dtypes/uint8.nii.gz
Bit-identicalBit-identical15/15
methods/dtypes/uint8-cubicinterpolation methods-cubic, -tpattern alt-z, on dtypes/uint8
command line3dTshift -tpattern alt-z -cubic -prefix out.nii synthetic/tshift/dtypes/uint8.nii.gz
Bit-identicalBit-identical15/15
methods/dtypes/uint8-quinticinterpolation methods-quintic, -tpattern alt-z, on dtypes/uint8
command line3dTshift -tpattern alt-z -quintic -prefix out.nii synthetic/tshift/dtypes/uint8.nii.gz
Bit-identicalBit-identical15/15
methods/dtypes/uint8-hepticinterpolation methods-heptic, -tpattern alt-z, on dtypes/uint8
command line3dTshift -tpattern alt-z -heptic -prefix out.nii synthetic/tshift/dtypes/uint8.nii.gz
Bit-identicalBit-identical15/15
methods/dtypes/uint8-wsinc5interpolation methods-wsinc5, -tpattern alt-z, on dtypes/uint8
command line3dTshift -tpattern alt-z -wsinc5 -prefix out.nii synthetic/tshift/dtypes/uint8.nii.gz
Bit-identicalBit-identical15/15
methods/dtypes/uint8-wsinc9interpolation methods-wsinc9, -tpattern alt-z, on dtypes/uint8
command line3dTshift -tpattern alt-z -wsinc9 -prefix out.nii synthetic/tshift/dtypes/uint8.nii.gz
Bit-identicalBit-identical15/15
options/float32-ignoreoptions-ignore 3: three leading points kept out of fit and shift (float32)
command line3dTshift -tpattern alt+z -ignore 3 -prefix out.nii synthetic/tshift/dtypes/float32.nii.gz
Bit-identicalOpen difference

170 values differ from AFNI’s, within the record’s thresholds: AFNI’s Fourier shift is float32, the clean room’s double precision.

15/15
options/float32-tzero-0options-tzero 0: every slice moved to the start of the TR (float32)
command line3dTshift -tpattern alt+z -tzero 0 -prefix out.nii synthetic/tshift/dtypes/float32.nii.gz
Bit-identicalOpen difference

117 values differ from AFNI’s, within the record’s thresholds: AFNI’s Fourier shift is float32, the clean room’s double precision.

15/15
options/float32-tzero-0.5options-tzero 0.5 (float32)
command line3dTshift -tpattern alt+z -tzero 0.5 -prefix out.nii synthetic/tshift/dtypes/float32.nii.gz
Bit-identicalOpen difference

113 values differ from AFNI’s, within the record’s thresholds: AFNI’s Fourier shift is float32, the clean room’s double precision.

15/15
options/float32-tzero-beyond-troptions-tzero 5: beyond the TR (allowed) (float32)
command line3dTshift -tpattern alt+z -tzero 5 -prefix out.nii synthetic/tshift/dtypes/float32.nii.gz
Bit-identicalOpen difference

117 values differ from AFNI’s, within the record’s thresholds: AFNI’s Fourier shift is float32, the clean room’s double precision.

15/15
options/float32-slice-2options-slice 2: align to slice 2's time (float32)
command line3dTshift -tpattern alt+z -slice 2 -prefix out.nii synthetic/tshift/dtypes/float32.nii.gz
Bit-identicalOpen difference

113 values differ from AFNI’s, within the record’s thresholds: AFNI’s Fourier shift is float32, the clean room’s double precision.

15/15
options/float32-slice-winsoptions-tzero and -slice: -slice wins (float32)
command line3dTshift -tpattern alt+z -tzero 0.5 -slice 1 -prefix out.nii synthetic/tshift/dtypes/float32.nii.gz
Bit-identicalOpen difference

139 values differ from AFNI’s, within the record’s thresholds: AFNI’s Fourier shift is float32, the clean room’s double precision.

15/15
options/float32-rltoptions-rlt: no trend added back (float32)
command line3dTshift -tpattern alt+z -rlt -prefix out.nii synthetic/tshift/dtypes/float32.nii.gz
Bit-identicalOpen difference

2,970 values differ from AFNI’s, within the record’s thresholds: AFNI’s Fourier shift is float32, the clean room’s double precision.

15/15
options/float32-rlt+options-rlt+: only the intercept added back (float32)
command line3dTshift -tpattern alt+z -rlt+ -prefix out.nii synthetic/tshift/dtypes/float32.nii.gz
Bit-identicalOpen difference

168 values differ from AFNI’s, within the record’s thresholds: AFNI’s Fourier shift is float32, the clean room’s double precision.

15/15
options/float32-rlt-last-winsoptions-rlt then -rlt+: the last wins (float32)
command line3dTshift -tpattern alt+z -rlt -rlt+ -prefix out.nii synthetic/tshift/dtypes/float32.nii.gz
Bit-identicalOpen difference

168 values differ from AFNI’s, within the record’s thresholds: AFNI’s Fourier shift is float32, the clean room’s double precision.

15/15
options/float32-no-detrendoptions-heptic -no_detrend: mean removed only (float32)
command line3dTshift -tpattern alt+z -heptic -no_detrend -prefix out.nii synthetic/tshift/dtypes/float32.nii.gz
Bit-identicalBit-identical15/15
options/float32-no-detrend-fourieroptions-no_detrend then -Fourier (allowed, with a warning) (float32)
command line3dTshift -tpattern alt+z -quintic -no_detrend -Fourier -prefix out.nii synthetic/tshift/dtypes/float32.nii.gz
Bit-identicalOpen difference

158 values differ from AFNI’s, within the record’s thresholds: AFNI’s Fourier shift is float32, the clean room’s double precision.

15/15
options/float32-TR-secondsoptions-TR 2.5s (float32)
command line3dTshift -tpattern alt+z -TR 2.5s -prefix out.nii synthetic/tshift/dtypes/float32.nii.gz
Bit-identicalOpen difference

168 values differ from AFNI’s, within the record’s thresholds: AFNI’s Fourier shift is float32, the clean room’s double precision.

15/15
options/float32-TR-plainoptions-TR 1.7 (no unit) (float32)
command line3dTshift -tpattern alt+z -TR 1.7 -prefix out.nii synthetic/tshift/dtypes/float32.nii.gz
Bit-identicalOpen difference

159 values differ from AFNI’s, within the record’s thresholds: AFNI’s Fourier shift is float32, the clean room’s double precision.

15/15
options/float32-TR-msoptions-TR 2000ms: every time in milliseconds (float32)
command line3dTshift -tpattern alt+z -TR 2000ms -prefix out.nii synthetic/tshift/dtypes/float32.nii.gz
Bit-identicalOpen difference

168 values differ from AFNI’s, within the record’s thresholds: AFNI’s Fourier shift is float32, the clean room’s double precision.

13/13
options/float32-TR-msecoptions-TR 2000msec -tzero 500 (float32)
command line3dTshift -tpattern alt+z -TR 2000msec -tzero 500 -prefix out.nii synthetic/tshift/dtypes/float32.nii.gz
Bit-identicalOpen difference

113 values differ from AFNI’s, within the record’s thresholds: AFNI’s Fourier shift is float32, the clean room’s double precision.

13/13
options/float32-verboseoptions-verbose (float32)
command line3dTshift -tpattern alt+z -verbose -prefix out.nii synthetic/tshift/dtypes/float32.nii.gz
Bit-identicalOpen difference

168 values differ from AFNI’s, within the record’s thresholds: AFNI’s Fourier shift is float32, the clean room’s double precision.

15/15
options/float32-abbreviationsoptions-cub and -verb: AFNI's abbreviations (float32)
command line3dTshift -tpattern alt+z -cub -verb -prefix out.nii synthetic/tshift/dtypes/float32.nii.gz
Bit-identicalBit-identical15/15
options/float32-wsinc-caseoptions-WSINC5: case-insensitive (float32)
command line3dTshift -tpattern alt+z -WSINC5 -prefix out.nii synthetic/tshift/dtypes/float32.nii.gz
Bit-identicalBit-identical15/15
options/float32-ignore-hepticoptions-ignore 5 with -heptic (float32)
command line3dTshift -tpattern alt+z -ignore 5 -heptic -prefix out.nii synthetic/tshift/dtypes/float32.nii.gz
Bit-identicalBit-identical15/15
options/float32-rlt-wsinc9options-rlt with -wsinc9 (float32)
command line3dTshift -tpattern alt+z -rlt -wsinc9 -prefix out.nii synthetic/tshift/dtypes/float32.nii.gz
Bit-identicalBit-identical15/15
options/float32-rlt+-linearoptions-rlt+ with -linear (float32)
command line3dTshift -tpattern alt+z -rlt+ -linear -prefix out.nii synthetic/tshift/dtypes/float32.nii.gz
Bit-identicalBit-identical15/15
options/int16-ignoreoptions-ignore 3: three leading points kept out of fit and shift (int16)
command line3dTshift -tpattern alt+z -ignore 3 -prefix out.nii synthetic/tshift/dtypes/int16.nii.gz
Bit-identicalBit-identical15/15
options/int16-tzero-0options-tzero 0: every slice moved to the start of the TR (int16)
command line3dTshift -tpattern alt+z -tzero 0 -prefix out.nii synthetic/tshift/dtypes/int16.nii.gz
Bit-identicalBit-identical15/15
options/int16-tzero-0.5options-tzero 0.5 (int16)
command line3dTshift -tpattern alt+z -tzero 0.5 -prefix out.nii synthetic/tshift/dtypes/int16.nii.gz
Bit-identicalBit-identical15/15
options/int16-tzero-beyond-troptions-tzero 5: beyond the TR (allowed) (int16)
command line3dTshift -tpattern alt+z -tzero 5 -prefix out.nii synthetic/tshift/dtypes/int16.nii.gz
Bit-identicalBit-identical15/15
options/int16-slice-2options-slice 2: align to slice 2's time (int16)
command line3dTshift -tpattern alt+z -slice 2 -prefix out.nii synthetic/tshift/dtypes/int16.nii.gz
Bit-identicalBit-identical15/15
options/int16-slice-winsoptions-tzero and -slice: -slice wins (int16)
command line3dTshift -tpattern alt+z -tzero 0.5 -slice 1 -prefix out.nii synthetic/tshift/dtypes/int16.nii.gz
Bit-identicalBit-identical15/15
options/int16-rltoptions-rlt: no trend added back (int16)
command line3dTshift -tpattern alt+z -rlt -prefix out.nii synthetic/tshift/dtypes/int16.nii.gz
Bit-identicalBit-identical15/15
options/int16-rlt+options-rlt+: only the intercept added back (int16)
command line3dTshift -tpattern alt+z -rlt+ -prefix out.nii synthetic/tshift/dtypes/int16.nii.gz
Bit-identicalBit-identical15/15
options/int16-rlt-last-winsoptions-rlt then -rlt+: the last wins (int16)
command line3dTshift -tpattern alt+z -rlt -rlt+ -prefix out.nii synthetic/tshift/dtypes/int16.nii.gz
Bit-identicalBit-identical15/15
options/int16-no-detrendoptions-heptic -no_detrend: mean removed only (int16)
command line3dTshift -tpattern alt+z -heptic -no_detrend -prefix out.nii synthetic/tshift/dtypes/int16.nii.gz
Bit-identicalBit-identical15/15
options/int16-no-detrend-fourieroptions-no_detrend then -Fourier (allowed, with a warning) (int16)
command line3dTshift -tpattern alt+z -quintic -no_detrend -Fourier -prefix out.nii synthetic/tshift/dtypes/int16.nii.gz
Bit-identicalBit-identical15/15
options/int16-TR-secondsoptions-TR 2.5s (int16)
command line3dTshift -tpattern alt+z -TR 2.5s -prefix out.nii synthetic/tshift/dtypes/int16.nii.gz
Bit-identicalBit-identical15/15
options/int16-TR-plainoptions-TR 1.7 (no unit) (int16)
command line3dTshift -tpattern alt+z -TR 1.7 -prefix out.nii synthetic/tshift/dtypes/int16.nii.gz
Bit-identicalBit-identical15/15
options/int16-TR-msoptions-TR 2000ms: every time in milliseconds (int16)
command line3dTshift -tpattern alt+z -TR 2000ms -prefix out.nii synthetic/tshift/dtypes/int16.nii.gz
Bit-identicalBit-identical13/13
options/int16-TR-msecoptions-TR 2000msec -tzero 500 (int16)
command line3dTshift -tpattern alt+z -TR 2000msec -tzero 500 -prefix out.nii synthetic/tshift/dtypes/int16.nii.gz
Bit-identicalBit-identical13/13
options/int16-verboseoptions-verbose (int16)
command line3dTshift -tpattern alt+z -verbose -prefix out.nii synthetic/tshift/dtypes/int16.nii.gz
Bit-identicalBit-identical15/15
options/int16-abbreviationsoptions-cub and -verb: AFNI's abbreviations (int16)
command line3dTshift -tpattern alt+z -cub -verb -prefix out.nii synthetic/tshift/dtypes/int16.nii.gz
Bit-identicalBit-identical15/15
options/int16-wsinc-caseoptions-WSINC5: case-insensitive (int16)
command line3dTshift -tpattern alt+z -WSINC5 -prefix out.nii synthetic/tshift/dtypes/int16.nii.gz
Bit-identicalBit-identical15/15
options/int16-ignore-hepticoptions-ignore 5 with -heptic (int16)
command line3dTshift -tpattern alt+z -ignore 5 -heptic -prefix out.nii synthetic/tshift/dtypes/int16.nii.gz
Bit-identicalBit-identical15/15
options/int16-rlt-wsinc9options-rlt with -wsinc9 (int16)
command line3dTshift -tpattern alt+z -rlt -wsinc9 -prefix out.nii synthetic/tshift/dtypes/int16.nii.gz
Bit-identicalBit-identical15/15
options/int16-rlt+-linearoptions-rlt+ with -linear (int16)
command line3dTshift -tpattern alt+z -rlt+ -linear -prefix out.nii synthetic/tshift/dtypes/int16.nii.gz
Bit-identicalBit-identical15/15
tpattern/alt+zslice patterns (-tpattern)-tpattern alt+z
command line3dTshift -tpattern alt+z -prefix out.nii synthetic/tshift/dtypes/float32.nii.gz
Bit-identicalOpen difference

168 values differ from AFNI’s, within the record’s thresholds: AFNI’s Fourier shift is float32, the clean room’s double precision.

15/15
tpattern/altplusslice patterns (-tpattern)-tpattern altplus
command line3dTshift -tpattern altplus -prefix out.nii synthetic/tshift/dtypes/float32.nii.gz
Bit-identicalOpen difference

168 values differ from AFNI’s, within the record’s thresholds: AFNI’s Fourier shift is float32, the clean room’s double precision.

15/15
tpattern/alt+z2slice patterns (-tpattern)-tpattern alt+z2
command line3dTshift -tpattern alt+z2 -prefix out.nii synthetic/tshift/dtypes/float32.nii.gz
Bit-identicalOpen difference

154 values differ from AFNI’s, within the record’s thresholds: AFNI’s Fourier shift is float32, the clean room’s double precision.

15/15
tpattern/alt-zslice patterns (-tpattern)-tpattern alt-z
command line3dTshift -tpattern alt-z -prefix out.nii synthetic/tshift/dtypes/float32.nii.gz
Bit-identicalOpen difference

148 values differ from AFNI’s, within the record’s thresholds: AFNI’s Fourier shift is float32, the clean room’s double precision.

15/15
tpattern/altminusslice patterns (-tpattern)-tpattern altminus
command line3dTshift -tpattern altminus -prefix out.nii synthetic/tshift/dtypes/float32.nii.gz
Bit-identicalOpen difference

148 values differ from AFNI’s, within the record’s thresholds: AFNI’s Fourier shift is float32, the clean room’s double precision.

15/15
tpattern/alt-z2slice patterns (-tpattern)-tpattern alt-z2
command line3dTshift -tpattern alt-z2 -prefix out.nii synthetic/tshift/dtypes/float32.nii.gz
Bit-identicalOpen difference

153 values differ from AFNI’s, within the record’s thresholds: AFNI’s Fourier shift is float32, the clean room’s double precision.

15/15
tpattern/seq+zslice patterns (-tpattern)-tpattern seq+z
command line3dTshift -tpattern seq+z -prefix out.nii synthetic/tshift/dtypes/float32.nii.gz
Bit-identicalOpen difference

165 values differ from AFNI’s, within the record’s thresholds: AFNI’s Fourier shift is float32, the clean room’s double precision.

15/15
tpattern/seqplusslice patterns (-tpattern)-tpattern seqplus
command line3dTshift -tpattern seqplus -prefix out.nii synthetic/tshift/dtypes/float32.nii.gz
Bit-identicalOpen difference

165 values differ from AFNI’s, within the record’s thresholds: AFNI’s Fourier shift is float32, the clean room’s double precision.

15/15
tpattern/seq-zslice patterns (-tpattern)-tpattern seq-z
command line3dTshift -tpattern seq-z -prefix out.nii synthetic/tshift/dtypes/float32.nii.gz
Bit-identicalOpen difference

151 values differ from AFNI’s, within the record’s thresholds: AFNI’s Fourier shift is float32, the clean room’s double precision.

15/15
tpattern/seqminusslice patterns (-tpattern)-tpattern seqminus
command line3dTshift -tpattern seqminus -prefix out.nii synthetic/tshift/dtypes/float32.nii.gz
Bit-identicalOpen difference

151 values differ from AFNI’s, within the record’s thresholds: AFNI’s Fourier shift is float32, the clean room’s double precision.

15/15
tpattern/zeroslice patterns (-tpattern)-tpattern zero
command line3dTshift -tpattern zero -prefix out.nii synthetic/tshift/dtypes/float32.nii.gz
Bit-identicalBit-identical15/15
tpattern/simultslice patterns (-tpattern)-tpattern simult
command line3dTshift -tpattern simult -prefix out.nii synthetic/tshift/dtypes/float32.nii.gz
Bit-identicalBit-identical15/15
tpattern/file-tabsslice patterns (-tpattern)-tpattern @file: one line, tab-separated (as fMRIPrep writes it)
command line3dTshift -tpattern @st.1D -prefix out.nii synthetic/tshift/dtypes/float32.nii.gz
Bit-identicalOpen difference

168 values differ from AFNI’s, within the record’s thresholds: AFNI’s Fourier shift is float32, the clean room’s double precision.

13/13
tpattern/file-linesslice patterns (-tpattern)-tpattern @file: one value per line
command line3dTshift -tpattern @st.1D -prefix out.nii synthetic/tshift/dtypes/float32.nii.gz
Bit-identicalOpen difference

148 values differ from AFNI’s, within the record’s thresholds: AFNI’s Fourier shift is float32, the clean room’s double precision.

13/13
tpattern/file-matrixslice patterns (-tpattern)-tpattern @file: two columns: read column by column
command line3dTshift -tpattern @st.1D -prefix out.nii synthetic/tshift/dtypes/float32.nii.gz
Bit-identicalOpen difference

148 values differ from AFNI’s, within the record’s thresholds: AFNI’s Fourier shift is float32, the clean room’s double precision.

13/13
tpattern/file-commentsslice patterns (-tpattern)-tpattern @file: comments and commas
command line3dTshift -tpattern @st.1D -prefix out.nii synthetic/tshift/dtypes/float32.nii.gz
Bit-identicalOpen difference

171 values differ from AFNI’s, within the record’s thresholds: AFNI’s Fourier shift is float32, the clean room’s double precision.

13/13
tpattern/file-extraslice patterns (-tpattern)-tpattern @file: more values than slices
command line3dTshift -tpattern @st.1D -prefix out.nii synthetic/tshift/dtypes/float32.nii.gz
Bit-identicalOpen difference

168 values differ from AFNI’s, within the record’s thresholds: AFNI’s Fourier shift is float32, the clean room’s double precision.

13/13
tpattern/file-repeatslice patterns (-tpattern)-tpattern @file: N@value repetition
command line3dTshift -tpattern @st.1D -prefix out.nii synthetic/tshift/dtypes/float32.nii.gz
Bit-identicalOpen difference

165 values differ from AFNI’s, within the record’s thresholds: AFNI’s Fourier shift is float32, the clean room’s double precision.

13/13
tpattern/file-trslice patterns (-tpattern)-tpattern @file: a value equal to the TR
command line3dTshift -tpattern @st.1D -prefix out.nii synthetic/tshift/dtypes/float32.nii.gz
Bit-identicalOpen difference

182 values differ from AFNI’s, within the record’s thresholds: AFNI’s Fourier shift is float32, the clean room’s double precision.

13/13
tpattern/inline-1Dslice patterns (-tpattern)-tpattern '@1D: 0 1 0.5 1.5'
command line3dTshift -tpattern '@1D: 0 1 0.5 1.5' -prefix out.nii synthetic/tshift/dtypes/float32.nii.gz
Bit-identicalOpen difference

168 values differ from AFNI’s, within the record’s thresholds: AFNI’s Fourier shift is float32, the clean room’s double precision.

13/13
methods/glibc-sinf-wsinc5interpolation methods-wsinc5 -tpattern '@1D: 0 0.543 0.102 0.728': weights where glibc's sinf and cosf differ from the correctly rounded ones (GLIBC-07)
command line3dTshift -tpattern '@1D: 0 0.543 0.102 0.728' -wsinc5 -prefix out.nii synthetic/tshift/dtypes/float32.nii.gz
Bit-identicalBit-identical13/13
methods/glibc-sinf-wsinc9interpolation methods-wsinc9 -tpattern '@1D: 0 0.911 1.821 0.603': weights where glibc's sinf and cosf differ from the correctly rounded ones (GLIBC-07)
command line3dTshift -tpattern '@1D: 0 0.911 1.821 0.603' -wsinc9 -prefix out.nii synthetic/tshift/dtypes/float32.nii.gz
Bit-identicalBit-identical13/13
header-timing/code1-seq-incslice times from the headerno -tpattern, slice_code 1 (sequential increasing)
command line3dTshift -cubic -prefix out.nii '<generated: code1-seq-inc>'
Bit-identicalBit-identical13/13
header-timing/code2-seq-decslice times from the headerno -tpattern, slice_code 2: AFNI gives every slice time 0
command line3dTshift -cubic -prefix out.nii '<generated: code2-seq-dec>'
Bit-identicalBit-identical13/13
header-timing/code3-alt-incslice times from the headerno -tpattern, slice_code 3 (alternating increasing)
command line3dTshift -cubic -prefix out.nii '<generated: code3-alt-inc>'
Bit-identicalBit-identical13/13
header-timing/code4-alt-decslice times from the headerno -tpattern, slice_code 4 (alternating decreasing)
command line3dTshift -cubic -prefix out.nii '<generated: code4-alt-dec>'
Bit-identicalBit-identical13/13
header-timing/code5-alt-inc2slice times from the headerno -tpattern, slice_code 5 (alternating increasing, from slice 1)
command line3dTshift -cubic -prefix out.nii '<generated: code5-alt-inc2>'
Bit-identicalBit-identical13/13
header-timing/code6-alt-dec2slice times from the headerno -tpattern, slice_code 6 (alternating decreasing, from nz-2)
command line3dTshift -cubic -prefix out.nii '<generated: code6-alt-dec2>'
Bit-identicalBit-identical13/13
header-timing/code3-partialslice times from the headerno -tpattern, slice_code 3 over slices 1..2 only
command line3dTshift -cubic -prefix out.nii '<generated: code3-partial>'
Bit-identicalBit-identical13/13
header-timing/code3-msslice times from the headerno -tpattern, slice_code 3 with the TR and duration in milliseconds
command line3dTshift -cubic -prefix out.nii '<generated: code3-ms>'
Bit-identicalBit-identical13/13
header-timing/code3-beyond-trslice times from the headerno -tpattern, times beyond the TR: a copy of the input, timing kept
command line3dTshift -cubic -prefix out.nii '<generated: code3-beyond-tr>'
Bit-identicalBit-identical13/13
header-timing/code2-beyond-trslice times from the headerno -tpattern, slice_code 2: AFNI gives every slice time 0 and shifts nothing; corrected, the times exceed the TR and the output is a copy (AFNI-03)
command line3dTshift -cubic -prefix out.nii '<generated: code2-beyond-tr>'
Bit-identicalBit-identical13/13
header-timing/code3-no-slice-dimslice times from the headerno -tpattern, no slice axis in dim_info: no timing, a copy
command line3dTshift -cubic -prefix out.nii '<generated: code3-no-slice-dim>'
Bit-identicalBit-identical13/13
header-timing/code3-zero-durationslice times from the headerno -tpattern, slice_duration 0: no timing, a copy
command line3dTshift -cubic -prefix out.nii '<generated: code3-zero-duration>'
Bit-identicalBit-identical13/13
header-timing/tpattern-overridesslice times from the header-tpattern overrides the header's timing
command line3dTshift -tpattern seq-z -prefix out.nii '<generated: code3-alt-inc>'
Bit-identicalOpen difference

151 values differ from AFNI’s, within the record’s thresholds: AFNI’s Fourier shift is float32, the clean room’s double precision.

15/15
defects/f32-slope-interdefects of AFNIfloat32 with scl_slope 2 and scl_inter 10, shifted (AFNI-02)
command line3dTshift -tpattern alt+z -linear -prefix out.nii '<generated: d13-f32-slope-inter>'
Bit-identicalBit-identical15/15 · 14/14
defects/f32-slope-inter-copydefects of AFNIfloat32 with scl_slope 2 and scl_inter 10, a copy (AFNI-02)
command line3dTshift -prefix out.nii '<generated: d13-f32-slope-inter>'
Bit-identicalBit-identical14/14 · 13/13
defects/metredefects of AFNIunits in metres with an origin (AFNI-07)
command line3dTshift -tpattern alt+z -prefix out.nii '<generated: d13-metre>'
Bit-identicalOpen difference

168 values differ from AFNI’s, within the record’s thresholds: AFNI’s Fourier shift is float32, the clean room’s double precision.

16/16 · 15/15
defects/micron-copydefects of AFNIunits in micrometres with an origin, a copy (AFNI-07)
command line3dTshift -prefix out.nii '<generated: d13-micron>'
Bit-identicalBit-identical14/14 · 13/13
defects/negative-slope-zerodefects of AFNIint16 with scl_slope -0.5 and -tpattern zero: nothing is shifted, so AFNI-01 changes nothing
command line3dTshift -tpattern zero -prefix out.nii synthetic/tshift/dtypes/int16-negative-slope.nii.gz
Bit-identicalBit-identical15/15 · 14/14
defects/oblique-metredefects of AFNIunits in metres, an oblique sform with an origin (AFNI-07)
command line3dTshift -tpattern alt+z -prefix out.nii '<generated: d13-oblique-metre>'
Bit-identicalOpen difference

168 values differ from AFNI’s, within the record’s thresholds: AFNI’s Fourier shift is float32, the clean room’s double precision.

16/16 · 15/15
defects/tiny-offsets-copydefects of AFNIheader slice times 0.5 ms apart beyond a 1 ms TR: a copy of the input that keeps them, all within 3 ms of 0 (AFNI-08)
command line3dTshift -prefix out.nii '<generated: d13-tiny-offsets>'
Bit-identicalBit-identical14/14 · 13/13
geometry/slightly-obliquegeometryan sform 0.006° from plumb (AFNI writes the plumb matrix below about 0.014°, AFNI-16)
command line3dTshift -tpattern alt+z -prefix out.nii '<generated: slightly-oblique>'
Bit-identicalOpen difference

168 values differ from AFNI’s, within the record’s thresholds: AFNI’s Fourier shift is float32, the clean room’s double precision.

16/16 · 15/15
geometry/slightly-oblique-qformgeometrya qform only, 0.006° from plumb: AFNI writes the plumb matrix with the header's voxel sizes as steps (AFNI-16)
command line3dTshift -tpattern alt+z -prefix out.nii '<generated: slightly-oblique-qform>'
Bit-identicalOpen difference

168 values differ from AFNI’s, within the record’s thresholds: AFNI’s Fourier shift is float32, the clean room’s double precision.

16/16 · 15/15
geometry/slightly-oblique-metregeometryan sform in metres 0.006° from plumb: AFNI writes the plumb matrix, its steps in millimetres and its origin in metres (AFNI-16, AFNI-07)
command line3dTshift -tpattern alt+z -prefix out.nii '<generated: d13-slightly-oblique-metre>'
Bit-identicalOpen difference

168 values differ from AFNI’s, within the record’s thresholds: AFNI’s Fourier shift is float32, the clean room’s double precision.

16/16 · 15/15
defects/oblique-metre-zero-origindefects of AFNIunits in metres, an oblique sform at the origin: AFNI keeps the whole sform in metres (AFNI-07)
command line3dTshift -tpattern alt+z -prefix out.nii '<generated: d13-oblique-metre-origin0>'
Bit-identicalOpen difference

168 values differ from AFNI’s, within the record’s thresholds: AFNI’s Fourier shift is float32, the clean room’s double precision.

16/16 · 15/15
copy/short-slice-spacingcopies of the inputheader slice times 2 ms apart beyond a 5 ms TR: a copy, whose slice fields AFNI derives from the times within 3 ms (slice_start 1)
command line3dTshift -prefix out.nii '<generated: d13-short-slices>'
Bit-identicalBit-identical13/13
copy/alt-two-slicescopies of the inputheader slice code 3 over slices 1..2, beyond the TR: a copy, written with slice code 1
command line3dTshift -prefix out.nii '<generated: d13-alt-two-slices>'
Bit-identicalBit-identical13/13
copy/no-timing-int16copies of the inputno slice timing anywhere: the output is a copy of the input (int16)
command line3dTshift -prefix out.nii synthetic/tshift/dtypes/int16.nii.gz
Bit-identicalBit-identical13/13
copy/no-timing-int16-slopecopies of the inputno slice timing anywhere: the output is a copy of the input (int16 with a brick factor)
command line3dTshift -prefix out.nii synthetic/tshift/dtypes/int16-slope.nii.gz
Bit-identicalBit-identical13/13
copy/no-timing-int16-slope-intercopies of the inputno slice timing anywhere: the output is a copy of the input (int16 scaled to float32)
command line3dTshift -prefix out.nii synthetic/tshift/dtypes/int16-slope-inter.nii.gz
Bit-identicalBit-identical13/13
copy/no-timing-float64copies of the inputno slice timing anywhere: the output is a copy of the input (float64 converted to float32)
command line3dTshift -prefix out.nii synthetic/tshift/dtypes/float64.nii.gz
Bit-identicalBit-identical13/13
copy/no-timing-float32-nonfinitecopies of the inputno slice timing anywhere: the output is a copy of the input (non-finite floats zeroed)
command line3dTshift -prefix out.nii synthetic/tshift/dtypes/float32-nonfinite.nii.gz
Bit-identicalBit-identical13/13
copy/3d-inputcopies of the inputa 3D image: one value per voxel, written back as 3D
command line3dTshift -tpattern alt+z -prefix out.nii synthetic/resampling/images/axial-1mm.nii.gz
Bit-identicalBit-identical13/13
errors/unknown-optionerrorsan unknown option
command line3dTshift -bogus -prefix out.nii synthetic/tshift/dtypes/float32.nii.gz
Rejected by both

AFNI: ** FATAL ERROR: Unknown option: -bogus; the replica: ** FATAL ERROR: Unknown option: -bogus

Rejected by both

AFNI: ** FATAL ERROR: Unknown option: -bogus; the clean room: ** FATAL ERROR: Unknown option: -bogus

1/1
errors/no-detrend-fouriererrors-no_detrend while the method is Fourier
command line3dTshift -no_detrend -prefix out.nii synthetic/tshift/dtypes/float32.nii.gz
Rejected by both

AFNI: ** FATAL ERROR: found -no_detrend, changing default to -heptic; the replica: ** FATAL ERROR: found -no_detrend, changing default to -heptic

Rejected by both

AFNI: ** FATAL ERROR: found -no_detrend, changing default to -heptic; the clean room: ** FATAL ERROR: found -no_detrend, changing default to -heptic

1/1
errors/rlt-no-detrenderrors-rlt with -no_detrend
command line3dTshift -linear -rlt -no_detrend -prefix out.nii synthetic/tshift/dtypes/float32.nii.gz
Rejected by both

AFNI: ** FATAL ERROR: cannot use both -rlt and -no_detrend; the replica: ** FATAL ERROR: cannot use both -rlt and -no_detrend

Rejected by both

AFNI: ** FATAL ERROR: cannot use both -rlt and -no_detrend; the clean room: ** FATAL ERROR: cannot use both -rlt and -no_detrend

1/1
errors/unknown-patternerrorspattern names are case-sensitive
command line3dTshift -tpattern ALT+Z -prefix out.nii synthetic/tshift/dtypes/float32.nii.gz
Rejected by both

AFNI: ** ERROR: Unknown tpattern = ALT+Z; the replica: ** ERROR: Unknown tpattern = ALT+Z

Rejected by both

AFNI: ** ERROR: Unknown tpattern = ALT+Z; the clean room: ** ERROR: Unknown tpattern = ALT+Z

1/1
errors/slice-too-largeerrors-slice beyond the last
command line3dTshift -tpattern alt+z -slice 4 -prefix out.nii synthetic/tshift/dtypes/float32.nii.gz
Rejected by both

AFNI: ** FATAL ERROR: -slice value is too large (4 >= 4); the replica: ** FATAL ERROR: -slice value is too large (4 >= 4)

Rejected by both

AFNI: ** FATAL ERROR: -slice value is too large (4 >= 4); the clean room: ** FATAL ERROR: -slice value is too large (4 >= 4)

1/1
errors/ignore-too-largeerrors-ignore > nt - 5
command line3dTshift -tpattern alt+z -ignore 96 -prefix out.nii synthetic/tshift/dtypes/float32.nii.gz
Rejected by both

AFNI: ** FATAL ERROR: -ignore value 96 is too large; the replica: ** FATAL ERROR: -ignore value 96 is too large

Rejected by both

AFNI: ** FATAL ERROR: -ignore value 96 is too large; the clean room: ** FATAL ERROR: -ignore value 96 is too large

1/1
errors/negative-tzeroerrors-tzero < 0
command line3dTshift -tpattern alt+z -tzero -1 -prefix out.nii synthetic/tshift/dtypes/float32.nii.gz
Rejected by both

AFNI: ** FATAL ERROR: illegal value ‘-1’ after -tzero!; the replica: ** FATAL ERROR: illegal value ‘-1’ after -tzero!

Rejected by both

AFNI: ** FATAL ERROR: illegal value ‘-1’ after -tzero!; the clean room: ** FATAL ERROR: illegal value ‘-1’ after -tzero!

1/1
errors/negative-ignoreerrors-ignore < 0
command line3dTshift -tpattern alt+z -ignore -1 -prefix out.nii synthetic/tshift/dtypes/float32.nii.gz
Rejected by both

AFNI: ** FATAL ERROR: -ignore value -1 is negative!; the replica: ** FATAL ERROR: -ignore value -1 is negative!

Rejected by both

AFNI: ** FATAL ERROR: -ignore value -1 is negative!; the clean room: ** FATAL ERROR: -ignore value -1 is negative!

1/1
errors/zero-TRerrors-TR 0
command line3dTshift -tpattern alt+z -TR 0 -prefix out.nii synthetic/tshift/dtypes/float32.nii.gz
Rejected by both

AFNI: ** FATAL ERROR: illegal value ‘0’ after -TR!; the replica: ** FATAL ERROR: illegal value ‘0’ after -TR!

Rejected by both

AFNI: ** FATAL ERROR: illegal value ‘0’ after -TR!; the clean room: ** FATAL ERROR: illegal value ‘0’ after -TR!

1/1
errors/bad-TRerrors-TR abc
command line3dTshift -tpattern alt+z -TR abc -prefix out.nii synthetic/tshift/dtypes/float32.nii.gz
Rejected by both

AFNI: ** FATAL ERROR: illegal value ‘abc’ after -TR!; the replica: ** FATAL ERROR: illegal value ‘abc’ after -TR!

Rejected by both

AFNI: ** FATAL ERROR: illegal value ‘abc’ after -TR!; the clean room: ** FATAL ERROR: illegal value ‘abc’ after -TR!

1/1
errors/missing-fileerrorsa missing -tpattern file
command line3dTshift -tpattern @none.1D -prefix out.nii synthetic/tshift/dtypes/float32.nii.gz
Rejected by both

AFNI: ** FATAL ERROR: Can’t read tpattern file /none.1D; the replica: ** FATAL ERROR: Can’t read tpattern file /none.1D

Rejected by both

AFNI: ** FATAL ERROR: Can’t read tpattern file /none.1D; the clean room: ** FATAL ERROR: Can’t read tpattern file /none.1D

1/1
errors/file-too-shorterrorsa -tpattern file with fewer values than slices
command line3dTshift -tpattern @st.1D -prefix out.nii synthetic/tshift/dtypes/float32.nii.gz
Rejected by both

AFNI: ** FATAL ERROR: tpattern file /st.1D has 3 values but have 4 slices; the replica: ** FATAL ERROR: tpattern file /st.1D has 3 values but have 4 slices

Rejected by both

AFNI: ** FATAL ERROR: tpattern file /st.1D has 3 values but have 4 slices; the clean room: ** FATAL ERROR: tpattern file /st.1D has 3 values but have 4 slices

1/1
errors/file-beyond-trerrorsa -tpattern value beyond the TR
command line3dTshift -tpattern @st.1D -prefix out.nii synthetic/tshift/dtypes/float32.nii.gz
Rejected by both

AFNI: ** FATAL ERROR: Illegal value 2.5 in tpattern file /st.1D; the replica: ** FATAL ERROR: Illegal value 2.5 in tpattern file /st.1D

Rejected by both

AFNI: ** FATAL ERROR: Illegal value 2.5 in tpattern file /st.1D; the clean room: ** FATAL ERROR: Illegal value 2.5 in tpattern file /st.1D

1/1
errors/file-negativeerrorsa negative -tpattern value
command line3dTshift -tpattern @st.1D -prefix out.nii synthetic/tshift/dtypes/float32.nii.gz
Rejected by both

AFNI: ** FATAL ERROR: Illegal value -0.5 in tpattern file /st.1D; the replica: ** FATAL ERROR: Illegal value -0.5 in tpattern file /st.1D

Rejected by both

AFNI: ** FATAL ERROR: Illegal value -0.5 in tpattern file /st.1D; the clean room: ** FATAL ERROR: Illegal value -0.5 in tpattern file /st.1D

1/1
errors/file-texterrorsa -tpattern file with text
command line3dTshift -tpattern @st.1D -prefix out.nii synthetic/tshift/dtypes/float32.nii.gz
Rejected by both

AFNI: ** FATAL ERROR: Can’t read tpattern file /st.1D; the replica: ** FATAL ERROR: Can’t read tpattern file /st.1D

Rejected by both

AFNI: ** FATAL ERROR: Can’t read tpattern file /st.1D; the clean room: ** FATAL ERROR: Can’t read tpattern file /st.1D

1/1
errors/missing-dataseterrorsan input file that does not exist
command line3dTshift -tpattern alt+z -prefix out.nii missing
Rejected by both

AFNI: ** FATAL ERROR: Can’t open input dataset ‘/does-not-exist.nii’; the replica: ** FATAL ERROR: Can’t open input dataset ‘/does-not-exist.nii’

Rejected by both

AFNI: ** FATAL ERROR: Can’t open input dataset ‘/does-not-exist.nii’; the clean room: ** FATAL ERROR: Can’t open input dataset ‘/does-not-exist.nii’

1/1
errors/existing-outputerrorsan output that exists already (AFNI exits 0 without writing; the replica exits 1, AFNI-11)
command line3dTshift -tpattern alt+z -prefix out.nii synthetic/tshift/dtypes/float32.nii.gz
Rejected by both

AFNI: ** ERROR: dataset NOT written to disk!; the replica: ** ERROR: dataset NOT written to disk!

Rejected by both

AFNI: ** ERROR: dataset NOT written to disk!; the clean room: ** ERROR: dataset NOT written to disk!

1/1
errors/prefix-with-blankerrorsa -prefix with a blank (AFNI refuses it; the replica writes it, AFNI-12)
command line3dTshift -tpattern alt+z -prefix 'out put.nii' synthetic/tshift/dtypes/float32.nii.gz
Rejected by both

AFNI: ** FATAL ERROR: illegal value ‘/afni-out put.nii’ after -prefix; the replica: ** FATAL ERROR: illegal value ‘/larmorx-out put.nii’ after -prefix

Rejected by both

AFNI: ** FATAL ERROR: illegal value ‘/afni-out put.nii’ after -prefix; the clean room: ** FATAL ERROR: illegal value ‘/larmorx-out put.nii’ after -prefix

1/1
global/overwrite-after-datasetglobal options-overwrite after the dataset replaces an existing output
command line3dTshift -tpattern alt+z -prefix out.nii synthetic/tshift/dtypes/float32.nii.gz -overwrite
Bit-identicalOpen difference

168 values differ from AFNI’s, within the record’s thresholds: AFNI’s Fourier shift is float32, the clean room’s double precision.

16/16
global/overwrite-as-option-valueglobal options-overwrite where -tpattern's value stands: taken out first, -tpattern reads alt+z
command line3dTshift -tpattern -overwrite alt+z -prefix out.nii synthetic/tshift/dtypes/float32.nii.gz
Bit-identicalOpen difference

168 values differ from AFNI’s, within the record’s thresholds: AFNI’s Fourier shift is float32, the clean room’s double precision.

16/16
global/overwrite-copyglobal options-overwrite on a copy of the input (no slice timing)
command line3dTshift -overwrite -prefix out.nii synthetic/tshift/dtypes/float32.nii.gz
Bit-identicalBit-identical16/16
global/overwrite-no-existing-fileglobal options-overwrite with no output to replace
command line3dTshift -overwrite -tpattern alt+z -prefix out.nii synthetic/tshift/dtypes/float32.nii.gz
Bit-identicalOpen difference

168 values differ from AFNI’s, within the record’s thresholds: AFNI’s Fourier shift is float32, the clean room’s double precision.

16/16
global/overwrite-then-Dglobal options-overwrite, then -DAFNI_DECONFLICT=YES: the later one wins (out_AA1.nii)
command line3dTshift -overwrite -DAFNI_DECONFLICT=YES -tpattern alt+z -prefix out.nii synthetic/tshift/dtypes/float32.nii.gz
Bit-identicalOpen difference

168 values differ from AFNI’s, within the record’s thresholds: AFNI’s Fourier shift is float32, the clean room’s double precision.

16/16
global/D-deconflict-yesglobal options-DAFNI_DECONFLICT=YES: an existing output makes the run write out_AA1.nii
command line3dTshift -DAFNI_DECONFLICT=YES -tpattern alt+z -prefix out.nii synthetic/tshift/dtypes/float32.nii.gz
Bit-identicalOpen difference

168 values differ from AFNI’s, within the record’s thresholds: AFNI’s Fourier shift is float32, the clean room’s double precision.

17/17
global/D-deconflict-noglobal options-DAFNI_DECONFLICT=NO: an existing output stops the shift before the work (exit 1)
command line3dTshift -DAFNI_DECONFLICT=NO -tpattern alt+z -prefix out.nii synthetic/tshift/dtypes/float32.nii.gz
Rejected by both

AFNI exit 1: ** FATAL ERROR: output dataset already exists ‘./out.nii’; the replica exit 1: ** FATAL ERROR: output dataset already exists ‘./out.nii’ (the same messages)

Rejected by both

AFNI exit 1: ** FATAL ERROR: output dataset already exists ‘./out.nii’; the clean room exit 1: ** FATAL ERROR: output dataset already exists ‘./out.nii’ (the same messages)

3/3
global/D-wins-over-environmentglobal options-DAFNI_DECONFLICT=OVERWRITE over AFNI_DECONFLICT=YES in the environment
command lineAFNI_DECONFLICT=YES 3dTshift -DAFNI_DECONFLICT=OVERWRITE -tpattern alt+z -prefix out.nii synthetic/tshift/dtypes/float32.nii.gz
Bit-identicalOpen difference

168 values differ from AFNI’s, within the record’s thresholds: AFNI’s Fourier shift is float32, the clean room’s double precision.

17/17
global/D-arbitrary-filenamesglobal options-DAFNI_ALLOW_ARBITRARY_FILENAMES=YES: a -prefix with a blank is written
command line3dTshift -DAFNI_ALLOW_ARBITRARY_FILENAMES=YES -tpattern alt+z -prefix 'out put.nii' synthetic/tshift/dtypes/float32.nii.gz
Bit-identicalOpen difference

168 values differ from AFNI’s, within the record’s thresholds: AFNI’s Fourier shift is float32, the clean room’s double precision.

17/17
global/D-after-datasetglobal options-D after the dataset applies too
command line3dTshift -tpattern alt+z -prefix 'out put.nii' synthetic/tshift/dtypes/float32.nii.gz -DAFNI_ALLOW_ARBITRARY_FILENAMES=YES
Bit-identicalOpen difference

168 values differ from AFNI’s, within the record’s thresholds: AFNI’s Fourier shift is float32, the clean room’s double precision.

16/16
global/D-as-option-valueglobal options-D where -tpattern's value stands: taken out first
command line3dTshift -tpattern -DAFNI_ALLOW_ARBITRARY_FILENAMES=YES alt+z -prefix 'out put.nii' synthetic/tshift/dtypes/float32.nii.gz
Bit-identicalOpen difference

168 values differ from AFNI’s, within the record’s thresholds: AFNI’s Fourier shift is float32, the clean room’s double precision.

16/16
global/D-value-with-blankglobal options-DAFNI_ALLOW_ARBITRARY_FILENAMES=Y ES: AFNI sets 'Y=ES', still YES
command line3dTshift '-DAFNI_ALLOW_ARBITRARY_FILENAMES=Y ES' -tpattern alt+z -prefix 'out put.nii' synthetic/tshift/dtypes/float32.nii.gz
Bit-identicalOpen difference

168 values differ from AFNI’s, within the record’s thresholds: AFNI’s Fourier shift is float32, the clean room’s double precision.

17/17
global/D-blank-before-valueglobal options-DAFNI_ALLOW_ARBITRARY_FILENAMES= YES: AFNI sets '=YES', not YES: refused
command line3dTshift '-DAFNI_ALLOW_ARBITRARY_FILENAMES= YES' -tpattern alt+z -prefix 'out put.nii' synthetic/tshift/dtypes/float32.nii.gz
Rejected by both

AFNI exit 1: ** FATAL ERROR: illegal value ‘out put.nii’ after -prefix; the replica exit 1: ** FATAL ERROR: illegal value ‘out put.nii’ after -prefix (the same messages)

Rejected by both

AFNI exit 1: ** FATAL ERROR: illegal value ‘out put.nii’ after -prefix; the clean room exit 1: ** FATAL ERROR: illegal value ‘out put.nii’ after -prefix (the same messages)

3/3
global/D-empty-valueglobal options-DAFNI_ALLOW_ARBITRARY_FILENAMES= (no value) sets nothing: refused
command line3dTshift -DAFNI_ALLOW_ARBITRARY_FILENAMES= -tpattern alt+z -prefix 'out put.nii' synthetic/tshift/dtypes/float32.nii.gz
Rejected by both

AFNI exit 1: ** FATAL ERROR: illegal value ‘out put.nii’ after -prefix; the replica exit 1: ** FATAL ERROR: illegal value ‘out put.nii’ after -prefix (the same messages)

Rejected by both

AFNI exit 1: ** FATAL ERROR: illegal value ‘out put.nii’ after -prefix; the clean room exit 1: ** FATAL ERROR: illegal value ‘out put.nii’ after -prefix (the same messages)

3/3
global/D-empty-nameglobal options-D=YES (no name) sets nothing
command line3dTshift -D=YES -tpattern alt+z -prefix 'out put.nii' synthetic/tshift/dtypes/float32.nii.gz
Rejected by both

AFNI exit 1: ** FATAL ERROR: illegal value ‘out put.nii’ after -prefix; the replica exit 1: ** FATAL ERROR: illegal value ‘out put.nii’ after -prefix (the same messages)

Rejected by both

AFNI exit 1: ** FATAL ERROR: illegal value ‘out put.nii’ after -prefix; the clean room exit 1: ** FATAL ERROR: illegal value ‘out put.nii’ after -prefix (the same messages)

3/3
global/D-without-equalsglobal options-DAFNI_ALLOW_ARBITRARY_FILENAMES (no '='): an unknown option
command line3dTshift -DAFNI_ALLOW_ARBITRARY_FILENAMES -tpattern alt+z -prefix out.nii synthetic/tshift/dtypes/float32.nii.gz
Rejected by both

AFNI exit 1: ** FATAL ERROR: Unknown option: -DAFNI_ALLOW_ARBITRARY_FILENAMES; the replica exit 1: ** FATAL ERROR: Unknown option: -DAFNI_ALLOW_ARBITRARY_FILENAMES (the same messages)

Rejected by both

AFNI exit 1: ** FATAL ERROR: Unknown option: -DAFNI_ALLOW_ARBITRARY_FILENAMES; the clean room exit 1: ** FATAL ERROR: Unknown option: -DAFNI_ALLOW_ARBITRARY_FILENAMES (the same messages)

3/3
global/D-aloneglobal options-D alone: an unknown option
command line3dTshift -D -tpattern alt+z -prefix out.nii synthetic/tshift/dtypes/float32.nii.gz
Rejected by both

AFNI exit 1: ** FATAL ERROR: Unknown option: -D; the replica exit 1: ** FATAL ERROR: Unknown option: -D (the same messages)

Rejected by both

AFNI exit 1: ** FATAL ERROR: Unknown option: -D; the clean room exit 1: ** FATAL ERROR: Unknown option: -D (the same messages)

3/3
global/D-lowercaseglobal options-d...=YES: not a global option, an unknown option
command line3dTshift -dAFNI_ALLOW_ARBITRARY_FILENAMES=YES -tpattern alt+z -prefix out.nii synthetic/tshift/dtypes/float32.nii.gz
Rejected by both

AFNI exit 1: ** FATAL ERROR: Unknown option: -dAFNI_ALLOW_ARBITRARY_FILENAMES=YES; the replica exit 1: ** FATAL ERROR: Unknown option: -dAFNI_ALLOW_ARBITRARY_FILENAMES=YES (the same messages)

Rejected by both

AFNI exit 1: ** FATAL ERROR: Unknown option: -dAFNI_ALLOW_ARBITRARY_FILENAMES=YES; the clean room exit 1: ** FATAL ERROR: Unknown option: -dAFNI_ALLOW_ARBITRARY_FILENAMES=YES (the same messages)

3/3
global/D-message-prefix-with-blankglobal options-DAFNI_MESSAGE_PREFIX=a b: AFNI's messages start with 'a=b::' (AFNI-22)
command line3dTshift '-DAFNI_MESSAGE_PREFIX=a b' -prefix out.nii synthetic/tshift/dtypes/float32.nii.gz
Bit-identicalBit-identical16/16
global/D-unknown-variableglobal options-D of a variable 3dTshift does not read: no effect
command line3dTshift -DAFNI_NOT_A_VARIABLE=1 -tpattern alt+z -prefix out.nii synthetic/tshift/dtypes/float32.nii.gz
Bit-identicalOpen difference

168 values differ from AFNI’s, within the record’s thresholds: AFNI’s Fourier shift is float32, the clean room’s double precision.

16/16
global/ok-1D-textglobal options-ok_1D_text: text in the -tpattern file is read as 0
command line3dTshift -ok_1D_text -tpattern @st.1D -prefix out.nii synthetic/tshift/dtypes/float32.nii.gz
Bit-identicalOpen difference

169 values differ from AFNI’s, within the record’s thresholds: AFNI’s Fourier shift is float32, the clean room’s double precision.

16/16
global/skip-afnircglobal options-skip_afnirc: accepted
command line3dTshift -skip_afnirc -tpattern alt+z -prefix out.nii synthetic/tshift/dtypes/float32.nii.gz
Bit-identicalOpen difference

168 values differ from AFNI’s, within the record’s thresholds: AFNI’s Fourier shift is float32, the clean room’s double precision.

16/16
environment/deconflict-overwriteenvironment variablesAFNI_DECONFLICT=OVERWRITE: an existing output is replaced, with AFNI's warning
command lineAFNI_DECONFLICT=OVERWRITE 3dTshift -tpattern alt+z -prefix out.nii synthetic/tshift/dtypes/float32.nii.gz
Bit-identicalOpen difference

168 values differ from AFNI’s, within the record’s thresholds: AFNI’s Fourier shift is float32, the clean room’s double precision.

17/17
environment/deconflict-yesenvironment variablesAFNI_DECONFLICT=YES: the output goes to out_AA1.nii
command lineAFNI_DECONFLICT=YES 3dTshift -tpattern alt+z -prefix out.nii synthetic/tshift/dtypes/float32.nii.gz
Bit-identicalOpen difference

168 values differ from AFNI’s, within the record’s thresholds: AFNI’s Fourier shift is float32, the clean room’s double precision.

17/17
environment/deconflict-no-copyenvironment variablesAFNI_DECONFLICT=NO on a copy: AFNI finds the output after the work, exit 0 (AFNI-11)
command lineAFNI_DECONFLICT=NO 3dTshift -prefix out.nii synthetic/tshift/dtypes/float32.nii.gz
Rejected by both

AFNI exit 0: ** ERROR: dataset NOT written to disk!; the replica exit 0: ** ERROR: dataset NOT written to disk! (the same messages)

Rejected by both

AFNI exit 0: ** ERROR: dataset NOT written to disk!; the clean room exit 0: ** ERROR: dataset NOT written to disk! (the same messages)

3/3
environment/arbitrary-filenamesenvironment variablesAFNI_ALLOW_ARBITRARY_FILENAMES=YES: a -prefix with a blank is written
command lineAFNI_ALLOW_ARBITRARY_FILENAMES=YES 3dTshift -tpattern alt+z -prefix 'out put.nii' synthetic/tshift/dtypes/float32.nii.gz
Bit-identicalOpen difference

168 values differ from AFNI’s, within the record’s thresholds: AFNI’s Fourier shift is float32, the clean room’s double precision.

17/17
environment/input-with-blankenvironment variablesan input in a directory with a blank: AFNI reads a list of datasets and fails (AFNI-18)
command line3dTshift -tpattern alt+z -prefix out.nii '<generated: blank-dir>'
Rejected by both

AFNI exit 1: ** FATAL ERROR: Can’t open input dataset ‘/with blank/in.nii’; the replica exit 1: ** FATAL ERROR: Can’t open input dataset ‘/with blank/in.nii’ (the same messages)

Rejected by both

AFNI exit 1: ** FATAL ERROR: Can’t open input dataset ‘/with blank/in.nii’; the clean room exit 1: ** FATAL ERROR: Can’t open input dataset ‘/with blank/in.nii’ (the same messages)

3/3
environment/path-spaces-okenvironment variablesAFNI_PATH_SPACES_OK=YES: an input in a directory with a blank is one file
command lineAFNI_PATH_SPACES_OK=YES 3dTshift -tpattern alt+z -prefix out.nii '<generated: blank-dir>'
Bit-identicalOpen difference

168 values differ from AFNI’s, within the record’s thresholds: AFNI’s Fourier shift is float32, the clean room’s double precision.

16/16
environment/nifti-priority-Qenvironment variablesAFNI_NIFTI_PRIORITY=Q: the qform rather than the sform (they differ)
command lineAFNI_NIFTI_PRIORITY=Q 3dTshift -tpattern alt+z -prefix out.nii '<generated: sq>'
Bit-identicalOpen difference

168 values differ from AFNI’s, within the record’s thresholds: AFNI’s Fourier shift is float32, the clean room’s double precision.

17/17
environment/nifti-priority-illegalenvironment variablesAFNI_NIFTI_PRIORITY=X: AFNI's warning, the sform
command lineAFNI_NIFTI_PRIORITY=X 3dTshift -tpattern alt+z -prefix out.nii '<generated: sq>'
Bit-identicalOpen difference

168 values differ from AFNI’s, within the record’s thresholds: AFNI’s Fourier shift is float32, the clean room’s double precision.

17/17
environment/nifti-scale-noenvironment variablesAFNI_NIFTI_SCALE=NO: int16 with a slope and an intercept stays int16 (slope kept)
command lineAFNI_NIFTI_SCALE=NO 3dTshift -tpattern alt+z -prefix out.nii synthetic/tshift/dtypes/int16-slope-inter.nii.gz
Bit-identicalBit-identical17/17
environment/nifti-scale-no-copyenvironment variablesAFNI_NIFTI_SCALE=NO on a copy of the input: the intercept dropped (AFNI-23)
command lineAFNI_NIFTI_SCALE=NO 3dTshift -prefix out.nii synthetic/tshift/dtypes/int16-slope-inter.nii.gz
Bit-identicalBit-identical16/16
environment/nifti-scale-no-rltenvironment variablesAFNI_NIFTI_SCALE=NO with -rlt: the trend takes the intercept away, nothing to correct
command lineAFNI_NIFTI_SCALE=NO 3dTshift -tpattern alt+z -rlt -prefix out.nii synthetic/tshift/dtypes/int16-slope-inter.nii.gz
Bit-identicalBit-identical17/17
environment/nifti-scale-no-rlt-ignoreenvironment variablesAFNI_NIFTI_SCALE=NO with -rlt -ignore 2: the ignored volumes lose the intercept (AFNI-23; corrected, read as without the variable)
command lineAFNI_NIFTI_SCALE=NO 3dTshift -tpattern alt+z -rlt -ignore 2 -prefix out.nii synthetic/tshift/dtypes/int16-slope-inter.nii.gz
Bit-identicalBit-identical17/17
environment/type-warn-defaultenvironment variablesAFNI_NIFTI_TYPE_WARN unset: AFNI's conversion message, with the muting lines
command line3dTshift -tpattern alt+z -prefix out.nii synthetic/tshift/dtypes/float64.nii.gz
Bit-identicalOpen difference

168 values differ from AFNI’s, within the record’s thresholds: AFNI’s Fourier shift is float32, the clean room’s double precision.

17/17
environment/type-warn-yesenvironment variablesAFNI_NIFTI_TYPE_WARN=YES: the conversion message alone
command lineAFNI_NIFTI_TYPE_WARN=YES 3dTshift -tpattern alt+z -prefix out.nii synthetic/tshift/dtypes/float64.nii.gz
Bit-identicalOpen difference

168 values differ from AFNI’s, within the record’s thresholds: AFNI’s Fourier shift is float32, the clean room’s double precision.

17/17
environment/nifti-view-tlrcenvironment variablesAFNI_NIFTI_VIEW=TLRC: sform code 2 is the tlrc view, so code 3
command lineAFNI_NIFTI_VIEW=TLRC 3dTshift -tpattern alt+z -prefix out.nii '<generated: code2>'
Bit-identicalOpen difference

168 values differ from AFNI’s, within the record’s thresholds: AFNI’s Fourier shift is float32, the clean room’s double precision.

17/17
environment/default-std-spaceenvironment variablesAFNI_DEFAULT_STD_SPACE=MNI: sform code 5 gives code 4
command lineAFNI_DEFAULT_STD_SPACE=MNI 3dTshift -tpattern alt+z -prefix out.nii '<generated: code5>'
Bit-identicalOpen difference

168 values differ from AFNI’s, within the record’s thresholds: AFNI’s Fourier shift is float32, the clean room’s double precision.

17/17
environment/sform-code-7environment variablessform code 7: AFNI's warning; AFNI_NIFTI_VIEW=tlrc makes it code 3
command lineAFNI_NIFTI_VIEW=tlrc 3dTshift -tpattern alt+z -prefix out.nii '<generated: code7>'
Bit-identicalOpen difference

168 values differ from AFNI’s, within the record’s thresholds: AFNI’s Fourier shift is float32, the clean room’s double precision.

17/17
environment/write-type-2environment variablesAFNI_NIFTI_WRITE_TYPE=2: a NIfTI-2 output
command lineAFNI_NIFTI_WRITE_TYPE=2 3dTshift -tpattern alt+z -prefix out.nii synthetic/tshift/dtypes/float32.nii.gz
Bit-identicalOpen difference

168 values differ from AFNI’s, within the record’s thresholds: AFNI’s Fourier shift is float32, the clean room’s double precision.

17/17
environment/allow-millisecondsenvironment variablesAFNI_ALLOW_MILLISECONDS=YES with -TR 2000ms: toffset written in ms (AFNI-19)
command lineAFNI_ALLOW_MILLISECONDS=YES 3dTshift -tpattern alt+z -TR 2000ms -prefix out.nii synthetic/tshift/dtypes/float32.nii.gz
Bit-identicalOpen difference

168 values differ from AFNI’s, within the record’s thresholds: AFNI’s Fourier shift is float32, the clean room’s double precision.

16/16
environment/1D-zero-textenvironment variablesAFNI_1D_ZERO_TEXT=YES: text in the -tpattern file is read as 0
command lineAFNI_1D_ZERO_TEXT=YES 3dTshift -tpattern @st.1D -prefix out.nii synthetic/tshift/dtypes/float32.nii.gz
Bit-identicalOpen difference

169 values differ from AFNI’s, within the record’s thresholds: AFNI’s Fourier shift is float32, the clean room’s double precision.

16/16
environment/1D-text-errorenvironment variablestext in the -tpattern file: AFNI's parse error
command line3dTshift -tpattern @st.1D -prefix out.nii synthetic/tshift/dtypes/float32.nii.gz
Rejected by both

AFNI exit 1: ** FATAL ERROR: Can’t read tpattern file st.1D; the replica exit 1: ** FATAL ERROR: Can’t read tpattern file st.1D (the same messages)

Rejected by both

AFNI exit 1: ** FATAL ERROR: Can’t read tpattern file st.1D; the clean room exit 1: ** FATAL ERROR: Can’t read tpattern file st.1D (the same messages)

3/3
environment/use-fgetsenvironment variablesAFNI_USE_FGETS=YES: a file with CR line ends is one row, of 8 values (9 > TR)
command lineAFNI_USE_FGETS=YES 3dTshift -tpattern @st.1D -prefix out.nii synthetic/tshift/dtypes/float32.nii.gz
Rejected by both

AFNI exit 1: ** FATAL ERROR: Illegal value 9 in tpattern file st.1D; the replica exit 1: ** FATAL ERROR: Illegal value 9 in tpattern file st.1D (the same messages)

Rejected by both

AFNI exit 1: ** FATAL ERROR: Illegal value 9 in tpattern file st.1D; the clean room exit 1: ** FATAL ERROR: Illegal value 9 in tpattern file st.1D (the same messages)

3/3
environment/cr-line-endsenvironment variablesa file with CR line ends: four rows of two columns, read column by column
command line3dTshift -tpattern @st.1D -prefix out.nii synthetic/tshift/dtypes/float32.nii.gz
Bit-identicalOpen difference

148 values differ from AFNI’s, within the record’s thresholds: AFNI’s Fourier shift is float32, the clean room’s double precision.

16/16
environment/realpathenvironment variablesa conversion message names the input by its real path ('..' resolved)
command lineAFNI_NIFTI_TYPE_WARN=YES 3dTshift -tpattern alt+z -prefix out.nii '<generated: dotdot-float64>'
Bit-identicalOpen difference

168 values differ from AFNI’s, within the record’s thresholds: AFNI’s Fourier shift is float32, the clean room’s double precision.

17/17
environment/norealpathenvironment variablesAFNI_NOREALPATH=YES: the message names the input as given
command lineAFNI_NIFTI_TYPE_WARN=YES AFNI_NOREALPATH=YES 3dTshift -tpattern alt+z -prefix out.nii '<generated: dotdot-float64>'
Bit-identicalOpen difference

168 values differ from AFNI’s, within the record’s thresholds: AFNI’s Fourier shift is float32, the clean room’s double precision.

17/17
environment/oblique-warningenvironment variablesan oblique input: AFNI's obliquity report
command line3dTshift -tpattern alt+z -prefix out.nii '<generated: oblique>'
Bit-identicalOpen difference

168 values differ from AFNI’s, within the record’s thresholds: AFNI’s Fourier shift is float32, the clean room’s double precision.

17/17
environment/no-oblique-warningenvironment variablesAFNI_NO_OBLIQUE_WARNING=YES: no obliquity report
command lineAFNI_NO_OBLIQUE_WARNING=YES 3dTshift -tpattern alt+z -prefix out.nii '<generated: oblique>'
Bit-identicalOpen difference

168 values differ from AFNI’s, within the record’s thresholds: AFNI’s Fourier shift is float32, the clean room’s double precision.

16/16
environment/message-prefixenvironment variablesAFNI_MESSAGE_PREFIX: on every labelled message
command lineAFNI_MESSAGE_PREFIX=run1 3dTshift -verbose -prefix out.nii synthetic/tshift/dtypes/float32.nii.gz
Bit-identicalBit-identical16/16
environment/message-colorize-noenvironment variablesAFNI_MESSAGE_COLORIZE=NO: labels without terminal escape codes
command lineAFNI_MESSAGE_COLORIZE=NO 3dTshift -prefix out.nii synthetic/tshift/dtypes/float32.nii.gz
Bit-identicalBit-identical16/16
environment/echo-commandlineenvironment variablesAFNI_ECHO_COMMANDLINE=YES (and a log file): the command line is echoed
command lineAFNI_ECHO_COMMANDLINE=YES 3dTshift -tpattern alt+z -prefix out.nii synthetic/tshift/dtypes/float32.nii.gz
Bit-identicalOpen difference

168 values differ from AFNI’s, within the record’s thresholds: AFNI’s Fourier shift is float32, the clean room’s double precision.

16/16
environment/quiet-startupenvironment variablesAFNI_QUIET_STARTUP=YES: no version banner (the replica prints none)
command lineAFNI_QUIET_STARTUP=YES 3dTshift -tpattern alt+z -prefix out.nii synthetic/tshift/dtypes/float32.nii.gz
Bit-identicalOpen difference

168 values differ from AFNI’s, within the record’s thresholds: AFNI’s Fourier shift is float32, the clean room’s double precision.

17/17
environment/nifti-noextenvironment variablesAFNI_NIFTI_NOEXT=YES: AFNI writes no header extension, as the replica: the whole file
command lineAFNI_NIFTI_NOEXT=YES 3dTshift -tpattern alt+z -prefix out.nii synthetic/tshift/dtypes/float32.nii.gz
Bit-identicalOpen difference

168 values differ from AFNI’s, within the record’s thresholds: AFNI’s Fourier shift is float32, the clean room’s double precision.

18/18
environment/one-row-datasetenvironment variablesa dataset of one row: AFNI writes a 1D text file under the NIfTI name (AFNI-20)
command line3dTshift -tpattern alt+z -prefix out.nii '<generated: one-row>'
Expected divergence

AFNI writes a dataset of one row as a 1D text file under the NIfTI name, with a random identifier (AFNI-20, unless AFNI_WRITE_1D_AS_PREFIX is YES); the replica writes NIfTI in fix and in reproduce, as AFNI does with that variable set (environment/write-1D-as-prefix)

Expected divergence

AFNI writes a dataset of one row as a 1D text file under the NIfTI name, with a random identifier (AFNI-20, unless AFNI_WRITE_1D_AS_PREFIX is YES); the clean room writes NIfTI in fix and in reproduce, as AFNI does with that variable set (environment/write-1D-as-prefix)

4/5
environment/write-1D-as-prefixenvironment variablesAFNI_WRITE_1D_AS_PREFIX=YES: a dataset of one row is written as NIfTI
command lineAFNI_WRITE_1D_AS_PREFIX=YES 3dTshift -tpattern alt+z -prefix out.nii '<generated: one-row>'
Bit-identicalBit-identical17/17
errors/missing-directoryerrorsan output in a missing directory: nothing written, exit 0 (AFNI-26)
command line3dTshift -tpattern alt+z -prefix nodir/out.nii synthetic/tshift/dtypes/float32.nii.gz
Rejected by both

AFNI exit 0: ** ERROR (nifti_image_write_engine): cannot open output file ‘nodir/out.nii’; the replica exit 0: ** ERROR (nifti_image_write_engine): cannot open output file ‘nodir/out.nii’ (the same messages)

Rejected by both

AFNI exit 0: ** ERROR (nifti_image_write_engine): cannot open output file ‘nodir/out.nii’; the clean room exit 0: ** ERROR (nifti_image_write_engine): cannot open output file ‘nodir/out.nii’ (the same messages)

3/3
errors/missing-directory-copyerrorsthe same for a copy of the input (AFNI-26)
command line3dTshift -prefix nodir/out.nii synthetic/tshift/dtypes/float32.nii.gz
Rejected by both

AFNI exit 0: ** ERROR (nifti_image_write_engine): cannot open output file ‘nodir/out.nii’; the replica exit 0: ** ERROR (nifti_image_write_engine): cannot open output file ‘nodir/out.nii’ (the same messages)

Rejected by both

AFNI exit 0: ** ERROR (nifti_image_write_engine): cannot open output file ‘nodir/out.nii’; the clean room exit 0: ** ERROR (nifti_image_write_engine): cannot open output file ‘nodir/out.nii’ (the same messages)

3/3
errors/missing-directory-verboseerrorsthe same with -verbose, whose last line names the file not written (AFNI-26)
command line3dTshift -verbose -tpattern alt+z -prefix nodir/out.nii synthetic/tshift/dtypes/float32.nii.gz
Rejected by both

AFNI exit 0: ** ERROR (nifti_image_write_engine): cannot open output file ‘nodir/out.nii’; the replica exit 0: ** ERROR (nifti_image_write_engine): cannot open output file ‘nodir/out.nii’ (the same messages)

Rejected by both

AFNI exit 0: ** ERROR (nifti_image_write_engine): cannot open output file ‘nodir/out.nii’; the clean room exit 0: ** ERROR (nifti_image_write_engine): cannot open output file ‘nodir/out.nii’ (the same messages)

3/3
global/D-nifti-priority-Qglobal options-DAFNI_NIFTI_PRIORITY=Q: as the environment variable
command line3dTshift -DAFNI_NIFTI_PRIORITY=Q -tpattern alt+z -prefix out.nii '<generated: sq>'
Bit-identicalOpen difference

168 values differ from AFNI’s, within the record’s thresholds: AFNI’s Fourier shift is float32, the clean room’s double precision.

17/17
global/D-nifti-view-tlrcglobal options-DAFNI_NIFTI_VIEW=TLRC: as the environment variable
command line3dTshift -DAFNI_NIFTI_VIEW=TLRC -tpattern alt+z -prefix out.nii '<generated: code2>'
Bit-identicalOpen difference

168 values differ from AFNI’s, within the record’s thresholds: AFNI’s Fourier shift is float32, the clean room’s double precision.

17/17
global/D-write-type-2global options-DAFNI_NIFTI_WRITE_TYPE=2: as the environment variable
command line3dTshift -DAFNI_NIFTI_WRITE_TYPE=2 -tpattern alt+z -prefix out.nii synthetic/tshift/dtypes/float32.nii.gz
Bit-identicalOpen difference

168 values differ from AFNI’s, within the record’s thresholds: AFNI’s Fourier shift is float32, the clean room’s double precision.

17/17
global/D-nifti-scale-noglobal options-DAFNI_NIFTI_SCALE=NO: as the environment variable
command line3dTshift -DAFNI_NIFTI_SCALE=NO -tpattern alt+z -prefix out.nii synthetic/tshift/dtypes/int16-slope-inter.nii.gz
Bit-identicalBit-identical17/17
divergence/afni-format-outputexpected divergences-prefix without .nii: AFNI writes its own BRIK/HEAD format
command line3dTshift -tpattern alt+z -prefix out.nii synthetic/tshift/dtypes/float32.nii.gz
Expected divergence

The replica writes NIfTI only; it rejects a -prefix that does not end in .nii or .nii.gz, where AFNI writes its own format.

Expected divergence

The clean room writes NIfTI only; it rejects a -prefix that does not end in .nii or .nii.gz, where AFNI writes its own format.

2/2
divergence/voxshiftexpected divergences-voxshift: per-voxel shifts from a dataset
command line3dTshift -voxshift gen:voxshift -prefix out.nii synthetic/tshift/dtypes/float32.nii.gz
Expected divergence

-voxshift is not supported by the replica (fMRIPrep does not use it).

Expected divergence

-voxshift is not supported by the clean room (fMRIPrep does not use it).

2/2
real/ds001600-acq-v4real imagesfMRIPrep's call (-ignore 0 -tzero -TR -tpattern @file) on sub-1_task-rest_acq-v4_bold
command line3dTshift -ignore 0 -tzero <middle> -TR <TR>s -tpattern @slice_timing.1D -prefix out.nii openneuro/ds001600/sub-1/func/sub-1_task-rest_acq-v4_bold.nii.gz (fMRIPrep's slice times from openneuro/ds001600/sub-1/func/sub-1_task-rest_acq-v4_bold.json)
Bit-identicalBit-identical15/15
real/ds001600-acq-v1real imagesfMRIPrep's call (-ignore 0 -tzero -TR -tpattern @file) on sub-1_task-rest_acq-v1_bold
command line3dTshift -ignore 0 -tzero <middle> -TR <TR>s -tpattern @slice_timing.1D -prefix out.nii openneuro/ds001600/sub-1/func/sub-1_task-rest_acq-v1_bold.nii.gz (fMRIPrep's slice times from openneuro/ds001600/sub-1/func/sub-1_task-rest_acq-v1_bold.json)
Bit-identicalBit-identical15/15
real/ds001600-acq-v2real imagesfMRIPrep's call (-ignore 0 -tzero -TR -tpattern @file) on sub-1_task-rest_acq-v2_bold
command line3dTshift -ignore 0 -tzero <middle> -TR <TR>s -tpattern @slice_timing.1D -prefix out.nii openneuro/ds001600/sub-1/func/sub-1_task-rest_acq-v2_bold.nii.gz (fMRIPrep's slice times from openneuro/ds001600/sub-1/func/sub-1_task-rest_acq-v2_bold.json)
Bit-identicalBit-identical15/15
real/ds001600-acq-PAreal imagesfMRIPrep's call (-ignore 0 -tzero -TR -tpattern @file) on sub-1_task-rest_acq-PA_bold
command line3dTshift -ignore 0 -tzero <middle> -TR <TR>s -tpattern @slice_timing.1D -prefix out.nii openneuro/ds001600/sub-1/func/sub-1_task-rest_acq-PA_bold.nii.gz (fMRIPrep's slice times from openneuro/ds001600/sub-1/func/sub-1_task-rest_acq-PA_bold.json)
Bit-identicalBit-identical15/15
real/ds000210-echo-1real imagesfMRIPrep's call (-ignore 0 -tzero -TR -tpattern @file) on sub-02_task-cuedSGT_run-01_echo-1_bold
command line3dTshift -ignore 0 -tzero <middle> -TR <TR>s -tpattern @slice_timing.1D -prefix out.nii openneuro/ds000210/sub-02/func/sub-02_task-cuedSGT_run-01_echo-1_bold.nii.gz (fMRIPrep's slice times from openneuro/ds000210/task-cuedSGT_echo-1_bold.json)
Bit-identicalOpen difference

240 values differ from AFNI’s, within the record’s thresholds: AFNI’s Fourier shift is float32, the clean room’s double precision.

15/15
real/ds000210-echo-2real imagesfMRIPrep's call (-ignore 0 -tzero -TR -tpattern @file) on sub-02_task-cuedSGT_run-01_echo-2_bold
command line3dTshift -ignore 0 -tzero <middle> -TR <TR>s -tpattern @slice_timing.1D -prefix out.nii openneuro/ds000210/sub-02/func/sub-02_task-cuedSGT_run-01_echo-2_bold.nii.gz (fMRIPrep's slice times from openneuro/ds000210/task-cuedSGT_echo-2_bold.json)
Bit-identicalOpen difference

173 values differ from AFNI’s, within the record’s thresholds: AFNI’s Fourier shift is float32, the clean room’s double precision.

15/15
real/ds000210-echo-3real imagesfMRIPrep's call (-ignore 0 -tzero -TR -tpattern @file) on sub-02_task-cuedSGT_run-01_echo-3_bold
command line3dTshift -ignore 0 -tzero <middle> -TR <TR>s -tpattern @slice_timing.1D -prefix out.nii openneuro/ds000210/sub-02/func/sub-02_task-cuedSGT_run-01_echo-3_bold.nii.gz (fMRIPrep's slice times from openneuro/ds000210/task-cuedSGT_echo-3_bold.json)
Bit-identicalOpen difference

146 values differ from AFNI’s, within the record’s thresholds: AFNI’s Fourier shift is float32, the clean room’s double precision.

15/15
real/ds006736real imagesfMRIPrep's call (-ignore 0 -tzero -TR -tpattern @file) on sub-004_task-freeRecall_bold
command line3dTshift -ignore 0 -tzero <middle> -TR <TR>s -tpattern @slice_timing.1D -prefix out.nii openneuro/ds006736/sub-004/func/sub-004_task-freeRecall_bold.nii.gz (fMRIPrep's slice times from openneuro/ds006736/sub-004/func/sub-004_task-freeRecall_bold.json)
Bit-identicalOpen difference

31 values differ from AFNI’s, within the record’s thresholds: AFNI’s Fourier shift is float32, the clean room’s double precision.

15/15
real/ds006010-uint16real imagesfMRIPrep's call (-ignore 0 -tzero -TR -tpattern @file) on sub-206_task-category_run-01_bold
command line3dTshift -ignore 0 -tzero <middle> -TR <TR>s -tpattern @slice_timing.1D -prefix out.nii openneuro/ds006010/sub-206/func/sub-206_task-category_run-01_bold.nii.gz (fMRIPrep's slice times from openneuro/ds006010/sub-206/func/sub-206_task-category_run-01_bold.json)
Bit-identicalOpen difference

23,875,172 values differ from AFNI’s, within the record’s thresholds: AFNI’s Fourier shift is float32, the clean room’s double precision.

14/14
real/ds003345-ms-headerreal imagesfMRIPrep's call (-ignore 0 -tzero -TR -tpattern @file) on sub-22973_task-PenaltyKik_run-02_bold
command line3dTshift -ignore 0 -tzero <middle> -TR <TR>s -tpattern @slice_timing.1D -prefix out.nii openneuro/ds003345/sub-22973/func/sub-22973_task-PenaltyKik_run-02_bold.nii.gz (fMRIPrep's slice times from openneuro/ds003345/sub-22973/func/sub-22973_task-PenaltyKik_run-02_bold.json)
Bit-identicalOpen difference

177 values differ from AFNI’s, within the record’s thresholds: AFNI’s Fourier shift is float32, the clean room’s double precision.

15/15
real/ds005454-mb4-96-slicesreal imagesfMRIPrep's call (-ignore 0 -tzero -TR -tpattern @file) on sub-16_task-rest_bold
command line3dTshift -ignore 0 -tzero <middle> -TR <TR>s -tpattern @slice_timing.1D -prefix out.nii openneuro/ds005454/sub-16/func/sub-16_task-rest_bold.nii.gz (fMRIPrep's slice times from openneuro/ds005454/sub-16/func/sub-16_task-rest_bold.json)
Bit-identicalOpen difference

1,808 values differ from AFNI’s, within the record’s thresholds: AFNI’s Fourier shift is float32, the clean room’s double precision.

15/15
real/ds003345-linearreal imagesfMRIPrep's call with -linear
command line3dTshift -ignore 0 -linear -tzero <middle> -TR <TR>s -tpattern @slice_timing.1D -prefix out.nii openneuro/ds003345/sub-22973/func/sub-22973_task-PenaltyKik_run-02_bold.nii.gz (fMRIPrep's slice times from openneuro/ds003345/sub-22973/func/sub-22973_task-PenaltyKik_run-02_bold.json)
Bit-identicalBit-identical15/15
real/ds003345-cubicreal imagesfMRIPrep's call with -cubic
command line3dTshift -ignore 0 -cubic -tzero <middle> -TR <TR>s -tpattern @slice_timing.1D -prefix out.nii openneuro/ds003345/sub-22973/func/sub-22973_task-PenaltyKik_run-02_bold.nii.gz (fMRIPrep's slice times from openneuro/ds003345/sub-22973/func/sub-22973_task-PenaltyKik_run-02_bold.json)
Bit-identicalBit-identical15/15
real/ds003345-quinticreal imagesfMRIPrep's call with -quintic
command line3dTshift -ignore 0 -quintic -tzero <middle> -TR <TR>s -tpattern @slice_timing.1D -prefix out.nii openneuro/ds003345/sub-22973/func/sub-22973_task-PenaltyKik_run-02_bold.nii.gz (fMRIPrep's slice times from openneuro/ds003345/sub-22973/func/sub-22973_task-PenaltyKik_run-02_bold.json)
Bit-identicalOpen difference

6 values differ from AFNI’s, within the record’s thresholds: quintic interpolation, in the last float32 bit; the cause is not reproduced yet (the record’s notes).

15/15
real/ds003345-hepticreal imagesfMRIPrep's call with -heptic
command line3dTshift -ignore 0 -heptic -tzero <middle> -TR <TR>s -tpattern @slice_timing.1D -prefix out.nii openneuro/ds003345/sub-22973/func/sub-22973_task-PenaltyKik_run-02_bold.nii.gz (fMRIPrep's slice times from openneuro/ds003345/sub-22973/func/sub-22973_task-PenaltyKik_run-02_bold.json)
Bit-identicalOpen difference

6 values differ from AFNI’s, within the record’s thresholds: heptic interpolation, in the last float32 bit; the cause is not reproduced yet (the record’s notes).

15/15
real/ds003345-wsinc5real imagesfMRIPrep's call with -wsinc5
command line3dTshift -ignore 0 -wsinc5 -tzero <middle> -TR <TR>s -tpattern @slice_timing.1D -prefix out.nii openneuro/ds003345/sub-22973/func/sub-22973_task-PenaltyKik_run-02_bold.nii.gz (fMRIPrep's slice times from openneuro/ds003345/sub-22973/func/sub-22973_task-PenaltyKik_run-02_bold.json)
Bit-identicalBit-identical15/15
real/ds003345-wsinc9real imagesfMRIPrep's call with -wsinc9
command line3dTshift -ignore 0 -wsinc9 -tzero <middle> -TR <TR>s -tpattern @slice_timing.1D -prefix out.nii openneuro/ds003345/sub-22973/func/sub-22973_task-PenaltyKik_run-02_bold.nii.gz (fMRIPrep's slice times from openneuro/ds003345/sub-22973/func/sub-22973_task-PenaltyKik_run-02_bold.json)
Bit-identicalBit-identical15/15
real/ds003345-ignore-4real imagesfMRIPrep's call with -ignore 4 (non-steady-state volumes)
command line3dTshift -ignore 4 -tzero <middle> -TR <TR>s -tpattern @slice_timing.1D -prefix out.nii openneuro/ds003345/sub-22973/func/sub-22973_task-PenaltyKik_run-02_bold.nii.gz (fMRIPrep's slice times from openneuro/ds003345/sub-22973/func/sub-22973_task-PenaltyKik_run-02_bold.json)
Bit-identicalOpen difference

137 values differ from AFNI’s, within the record’s thresholds: AFNI’s Fourier shift is float32, the clean room’s double precision.

15/15
real/ds000210-header-timingreal imagesno -tpattern: the slice timing in the header (slice_code 3, ALT_INC)
command line3dTshift -prefix out.nii openneuro/ds000210/sub-02/func/sub-02_task-cuedSGT_run-01_echo-1_bold.nii.gz
Bit-identicalOpen difference

222 values differ from AFNI’s, within the record’s thresholds: AFNI’s Fourier shift is float32, the clean room’s double precision.

13/13
real/ds001600-header-zero-durationreal imagesno -tpattern: slice_code 5 with slice_duration 0, so a copy of the input
command line3dTshift -prefix out.nii openneuro/ds001600/sub-1/func/sub-1_task-rest_acq-v4_bold.nii.gz
Bit-identicalBit-identical13/13
difftest/metre-oblique-default-stepsrandom testingBit-identical–14/14
difftest/header-milliseconds-toffsetrandom testing–Bit-identical13/13

Speed

Speed, both ports against the original: AFNI 3dTshift

Seconds for the whole command (reading, shifting, writing), the median of repeated runs, on a log scale. The factor is how many times faster the clean room is than AFNI, both on one thread.

  • AFNI 3dTshift, 1 thread
  • replica, 1 thread
  • replica, 12 threads
  • clean room, 1 thread
  • clean room, 12 threads
  1. ds000210: 64×64×33, 260 vol.AFNI 3dTshift, 1 thread 2.53 sreplica, 1 thread 1.53 sreplica, 12 threads 0.701 sclean room, 1 thread 1.23 sclean room, 12 threads 0.43 s2.1×
  2. ds006010: 96×96×69, 113 vol.AFNI 3dTshift, 1 thread 6.68 sreplica, 1 thread 3.03 sreplica, 12 threads 1.48 sclean room, 1 thread 2.46 sclean room, 12 threads 1.13 s2.7×
  3. ds005454: 160×160×96, 280 vol.AFNI 3dTshift, 1 thread 53 sreplica, 1 thread 29.9 sreplica, 12 threads 15.6 sclean room, 1 thread 23.1 sclean room, 12 threads 8.41 s2.3×
0.1 s1 s10 s100 s

From the benchmark report docs/benchmarks/afni-tshift.json (Intel® Core™ i7-8750H CPU @ 2.20GHz, 12 logical CPUs).

Corrections by default

Where AFNI has a deterministic defect that changes a result, both ports correct it by default, fix, and warn. upstream_defects="reproduce" (on the command line --upstream-defects reproduce from the next release, LARMORX_UPSTREAM_DEFECTS=reproduce in 0.1.0rc1) gives AFNI’s behaviour, and the records compare with it. The records’ lists of what fix changed meet AFNI-01, AFNI-02, AFNI-07, AFNI-08, AFNI-16, AFNI-19, AFNI-22 and AFNI-23 (register). AFNI-03 and AFNI-05 were AFNI 25.2.09’s: AFNI 26.3.05 behaves as LARMORX corrected them.

Random testing

Random testing runs random valid inputs and option combinations through the original and both ports, 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: AFNI 3dTshift 26.3.05, 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
3dTshiftreplica, clean room2,2660182,117·442184···
  • fixed in LARMORX 3dTshift (clean room), 1 case(s): the fix pass, header fields (scaling, timing, slice fields, units, xform codes): AFNI 26.3.05 fixed AFNI-03 (header slice code 2 read end…start); the clean room’s reproduce still reads 25.2.09’s zeros and names AFNI-03 (the random-testing run against 26.3.05: 73 findings, 4 with AFNI-05 too); its alignment is the clean room’s own, from its specification (specs/3dTshift.md §3.2, §6.4); fixed 2026-10-29 by the clean-room alignment (next-steps 3b): both modes follow AFNI 26.3.05 and never name the defect; none in that part’s random-testing run of half an hour against 26.3.05 (its run of 2026-10-29) (details)
  • fixed in LARMORX 3dTshift (clean room), 1 case(s): header fields (scaling, timing, slice fields, units, xform codes): header slice times in milliseconds or microseconds: the step in seconds stays a double and each time is a float32 running sum of it (observed: -slice k reads every time back); both findings replay as agree (2026-10-29) (details)
  • fixed in LARMORX 3dTshift (replica), 83 case(s): the replica’s default equal to the clean room’s: the clean room’s AFNI-07 default now writes the replica’s steps (the file matrix’s column lengths, then converted) and, for a qform, the qform matrix computed with millimetre voxel sizes and offsets, rounded once; the AFNI-07 and AFNI-16 notices in the replica’s words; all 83 findings replay as agree (2026-10-29) (AFNI-07)
  • fixed in LARMORX 3dTshift (replica), 1 case(s): the replica’s default equal to the clean room’s, the fix pass: the clean room’s AFNI-07 default now writes the replica’s steps (the file matrix’s column lengths, then converted) and, for a qform, the qform matrix computed with millimetre voxel sizes and offsets, rounded once; the AFNI-07 and AFNI-16 notices in the replica’s words; all 83 findings replay as agree (2026-10-29) (AFNI-07)

Notices: a correction named in a case where it changed nothing the input shows (defect, cases): 3dTshift: AFNI-01 (16), AFNI-03 (28), AFNI-05 (38), AFNI-07 (2), AFNI-16 (12).

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.

Known limits

LARMORX writes NIfTI only (a -prefix in AFNI’s own format is refused), -voxshift is not supported, and a dataset of one row is written as NIfTI where AFNI writes a 1D text file (AFNI-20): the three expected divergences. Writing an existing output in place in a directory that cannot be written, as AFNI does, is planned.

History

  • until 2026-10-29: the replica followed AFNI 25.2.09, bit-identical to it in every compared case of its record.
  • 2026-10-29: AFNI 26.3.05 releases the Medical College of Wisconsin’s code under CC BY 4.0; the replica moves from larmorx-gpl into larmorx, and both ports follow AFNI 26.3.05.
  • 2026-11-04: the clean room’s reproduce computes sinf and cosf as glibc 2.35 does (GLIBC-07); the two cases of the weighted sinc’s functions join both records.
  • 2026-11-13: the replica’s record made again from a commit with no uncommitted changes, as every record must be (the record of 2026-11-04 was not): the same result in every one of its 317 cases.

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.