This website requires JavaScript.
f36cd88795
Give a 16-bit image a saturation code when its limit is 65534
2608-performance
jungfrau and Claude Opus 5
2026-08-18 18:41:53 -04:00
a97d763098
Do not assume a DECTRIS detector's bit depths before asking it
jungfrau and Claude Opus 5
2026-08-18 17:48:19 -04:00
9f49e5abb7
Keep 16-bit images 16-bit through the GPU pipeline
jungfrau and Claude Opus 5
2026-08-18 01:52:27 -04:00
642cbc7271
Stop the rotation pre-pass once it has what the second pass needs, and say where time went
jungfrau and Claude Opus 5
2026-08-18 01:11:47 -04:00
debd73c0bd
Find the first pass's spots on every worker
jungfrau and Claude Opus 5
2026-08-17 18:06:03 -04:00
a22a51372e
Run the parallel helpers on a pool instead of starting threads per chunk
jungfrau and Claude Opus 5
2026-08-17 18:05:35 -04:00
65eb86ab66
Size the post-refine gather by its reflections, not by its frames
jungfrau and Claude Opus 5
2026-08-17 18:05:19 -04:00
2308bbad8c
Stop scaling and merging what the resolution range excludes
jungfrau and Claude Opus 5
2026-08-17 18:04:55 -04:00
20f1827550
Fit the modulation surface on a grid that spans the detector
jungfrau and Claude Opus 5
2026-08-17 18:04:23 -04:00
088ba1cff8
Stop paying for workers and threads that do no work
jungfrau and Claude Opus 5
2026-08-16 10:53:47 -04:00
f8b449a944
Score the first-pass validation frames together
jungfrau and Claude Opus 5
2026-08-16 09:28:21 -04:00
057fff98b2
Give each FFT direction a block and its histogram shared memory
jungfrau and Claude Opus 5
2026-08-16 06:29:48 -04:00
8ea3076f96
Take the serial passes out of the combine phase
jungfrau and Claude Opus 5
2026-08-16 05:59:16 -04:00
a1325637d2
Spread the scaling corrections and the space-group search over the cores
jungfrau and Claude Opus 5
2026-08-16 02:38:20 -04:00
f1dcfbe525
Give the process file its own thread
jungfrau and Claude Opus 5
2026-08-16 01:09:02 -04:00
f5b3193253
Ask HDF5 where an image is, then read it without the lock
jungfrau and Claude Opus 5
2026-08-16 01:04:56 -04:00
09cb9e2be9
Take the double precision out of the box integrator's inner loops
jungfrau and Claude Opus 5
2026-08-15 23:22:29 -04:00
3aa239fce8
Parallelise the incident-flux divide, drop a redundant sync
jungfrau and Claude Opus 5
2026-08-15 22:57:24 -04:00
f36349060a
Sort the partials and the post-refine events in buckets, in parallel
jungfrau and Claude Opus 5
2026-08-15 22:31:05 -04:00
621019140e
Reduce within the warp before the Bragg integration atomics
jungfrau and Claude Opus 5
2026-08-15 22:15:00 -04:00
2cbb3fc8b4
Build the GPU engines a worker never uses on first use, not always
jungfrau and Claude Opus 5
2026-08-15 21:42:41 -04:00
eb634400ea
Read four pixels at a time in the ring reduction
jungfrau and Claude Opus 5
2026-08-15 19:21:24 -04:00
56512414aa
Make the partials order total, and hoist 1/sigma out of the IRLS loop
jungfrau and Claude Opus 5
2026-08-15 18:56:26 -04:00
e11c2a2b20
Fit the goniometer rotation scale in closed form
jungfrau and Claude Opus 5
2026-08-15 18:56:09 -04:00
6368c00173
Decode and accumulate the beam-stop projection on the GPU
jungfrau and Claude Opus 5
2026-08-15 18:38:26 -04:00
2f586d2267
Stop allocating GPU and pinned memory nothing reads
jungfrau and Claude Opus 5
2026-08-15 17:50:19 -04:00
50941f8df0
Make the beam-stop mask O(pixels) and parallel
jungfrau and Claude Opus 5
2026-08-15 17:50:06 -04:00
f12e3b6252
Parallelize the beam-stop pre-scan
jungfrau and Claude Opus 5
2026-08-15 17:13:53 -04:00
fad395dc60
Deploy site
gitea-pages
leonarski_f
2026-08-13 18:43:50 +00:00
538f3504d3
v1.0.0.rc-161 (#71 )
main
1.0.0-rc.161
leonarski_f
2026-08-13 17:03:10 +02:00
93093c5f21
Viewer: every panel section starts folded, and calibration carries the azimuthal settings
adaptive-spot-finding
leonarski_f and Claude Opus 5
2026-08-13 15:21:41 +02:00
cac669495d
rugnux: the pre-scan says what it is doing
leonarski_f and Claude Opus 5
2026-08-13 04:18:27 +02:00
9f86384210
docs: summarise the rc.161 changelog, and give CPU_DATA_ANALYSIS its proportions back
leonarski_f and Claude Opus 5
2026-08-13 03:19:53 +02:00
bce8ff8a1f
Bragg integration: say why the peak rule cuts on any unreadable pixel
leonarski_f and Claude Opus 5
2026-08-13 03:19:40 +02:00
52a8ac5afa
rugnux: refuse a stride on rotation data
leonarski_f and Claude Opus 5
2026-08-13 03:19:32 +02:00
914be292ca
Viewer: offer the beam-centre measurement in the new-job dialog
leonarski_f
2026-08-13 00:30:06 +02:00
3e556c4452
docs: changelog for the beam-stop outward-background fix
leonarski_f
2026-08-12 20:49:39 +02:00
08b9baa0e0
Beam stop: a ring too small to judge is too small to judge others by
leonarski_f
2026-08-12 20:49:03 +02:00
29a56fc3ed
docs: why the spots half a turn apart are Friedel mates, and how many there are
leonarski_f and Claude Opus 5
2026-08-13 00:09:11 +02:00
abfdb89b79
Beam centre: measure the direct beam from the symmetry of the spots, before indexing
leonarski_f and Claude Opus 5
2026-08-12 23:54:09 +02:00
b34aabe39f
docs: changelog for the lattice-search and two-pass symmetry fixes
leonarski_f and Claude Opus 5
2026-08-12 18:38:03 +02:00
c95951e872
Lattice search: gate the type-boundary retry on the boundary, not on the metric tolerance
leonarski_f
2026-08-12 17:17:06 +02:00
996d8e7d41
Lattice search: a reduced cell sitting on the Niggli type boundary keeps its centring
leonarski_f
2026-08-12 17:02:36 +02:00
8874a788e6
Two-pass: pass 2 reuses the space group the centred-lattice test found, like any other
leonarski_f
2026-08-12 16:55:40 +02:00
1f2dad530c
Space-group search: the merge with all the observations decides
leonarski_f
2026-08-12 15:07:34 +02:00
7ce47bd4d1
Radiation damage: report a measurement, or nothing
leonarski_f and Claude Opus 5
2026-08-12 05:40:32 +02:00
2222a4085c
Scaling: correct the absorption that changes as the crystal turns
leonarski_f and Claude Opus 5
2026-08-12 02:49:06 +02:00
b8e7e9c8bf
Scaling: fit a correction surface with the observation as the response, not the regressor
leonarski_f and Claude Opus 5
2026-08-12 02:51:09 +02:00
d74ff7fc29
Viewer: the settings panel says how the sample moved, and how polarized the beam was
leonarski_f and Claude Opus 5
2026-08-12 05:13:57 +02:00
e4ce573075
Bragg integration: keep a reflection that lost a wing, not one that lost its peak
leonarski_f and Claude Opus 5
2026-08-12 00:57:18 +02:00
cafc312681
docs: SHADOW_FINDER.md described a wiring plan that had already been carried out
leonarski_f and Claude Opus 5
2026-08-12 03:16:13 +02:00
3224ca9d8f
docs: beam_stop is built and runs, not an unbuilt prototype
leonarski_f and Claude Opus 5
2026-08-12 03:07:46 +02:00
12ae6c5228
Bragg integration: fit a reflection over the pixels it has, not only over all of them
leonarski_f
2026-08-11 16:40:45 +02:00
e0e429ae2f
Rotation: pass 2 divides partiality by the width pass 2 measured
leonarski_f and Claude Opus 5
2026-08-11 21:42:54 +02:00
a0604ff244
Rotation: a diagnostic that opens the prediction window on its own width
leonarski_f
2026-08-11 21:30:57 +02:00
677ece7b59
Post-refinement: correct a goniometer that turned further than it was told
leonarski_f and Claude Opus 5
2026-08-11 23:07:12 +02:00
ce3199748d
Scaling: report the stretches of a sweep the crystal did not deliver
leonarski_f and Claude Opus 5
2026-08-11 17:51:19 +02:00
e5c0066129
rugnux: write a results report next to the reflections
leonarski_f and Claude Opus 5
2026-08-11 18:03:07 +02:00
b9078d9a59
Scaling: drop a frame whose scale collapses, do not merge it unscaled
leonarski_f and Claude Opus 5
2026-08-11 17:59:45 +02:00
9885d1fc28
Rotation: integrate every frame the sweep's lattice explains
leonarski_f and Claude Opus 5
2026-08-11 15:53:06 +02:00
d8029524e7
Scaling: never let an observation's own fluctuation set its weight
leonarski_f and Claude Opus 5
2026-08-11 16:06:39 +02:00
1eb9d5a2fa
Notices: list traccc in the canonical third-party manifest
leonarski_f and Claude Opus 5
2026-08-11 17:08:34 +02:00
43627e22dc
docs: a rule for licences and academic credit, and apply it
leonarski_f and Claude Opus 5
2026-08-11 16:33:13 +02:00
6d39a4e1ab
Space-group search: decide a screw from the evidence, not from a count of absences
leonarski_f and Claude Opus 5
2026-08-11 15:52:37 +02:00
7bbf072ad2
Scaling: do not report an ISa that was never measured
leonarski_f and Claude Opus 5
2026-08-11 06:14:46 +02:00
b40abe31cf
Scaling: one exact-Bragg angle per rocking event
leonarski_f and Claude Opus 5
2026-08-11 05:25:42 +02:00
6f7b136ec2
Bragg integration: a shared signal pixel belongs to the nearer reflection
leonarski_f and Claude Opus 5
2026-08-11 05:02:31 +02:00
7627c282ba
Battery: one unreadable column must not discard a resolution shell
leonarski_f and Claude Opus 5
2026-08-11 01:49:13 +02:00
06a5a118bf
Bragg integration: widen the background ring to r3 = 13
leonarski_f and Claude Opus 5
2026-08-11 00:07:30 +02:00
e0bd208666
Prediction: measure the rocking curve against the frame's edge, not its centre
leonarski_f and Claude Opus 5
2026-08-10 22:20:00 +02:00
27d0b1db74
docs: state the reflection-file conventions
leonarski_f and Claude Opus 5
2026-08-10 19:08:58 +02:00
adf87e8675
Merging: export the XDS-comparable ISa under jfjoch_diffrn_ISa
leonarski_f and Claude Opus 5
2026-08-10 18:55:46 +02:00
ac06b5c64f
Merging: report the error model in XDS's convention
leonarski_f and Claude Opus 5
2026-08-10 18:50:22 +02:00
bf80935b80
Bragg integration: do not let the radial correction outlive its kernel table
leonarski_f and Claude Opus 5
2026-08-10 16:38:37 +02:00
61d24db59f
Bragg integration: elongate the background ring per reflection
leonarski_f and Claude Opus 5
2026-08-10 15:19:28 +02:00
52ea727650
Reader: reconstruct the background variance a legacy file does not store
leonarski_f and Claude Opus 5
2026-08-10 05:52:56 +02:00
a7c5d7a89b
CBOR: carry the reflection's background variance
leonarski_f and Claude Opus 5
2026-08-10 05:52:44 +02:00
4869dbd984
Space-group search: do not judge centering on a non-positive present mean
leonarski_f and Claude Opus 5
2026-08-10 05:52:33 +02:00
01626a54e0
Space-group search: keep the lowest shell when no shell passes the cut
leonarski_f and Claude Opus 5
2026-08-10 05:52:23 +02:00
f766a342bb
Scaling: take the detector modulation surface to a 24x24 grid
leonarski_f and Claude Opus 5
2026-08-10 04:18:44 +02:00
7523c67655
Space-group search: set the operator-H bound from the measured gap
leonarski_f and Claude Opus 5
2026-08-10 04:18:30 +02:00
bc1c4c6800
Rotation: land the rest of the bandwidth term
leonarski_f and Claude Opus 5
2026-08-09 22:26:26 +02:00
0b5fb4fb92
Merge branch 'fix56-work' into integration-variance-fixes
leonarski_f
2026-08-09 21:08:39 +02:00
1239c49731
Bragg integration: separate the three things a bandwidth used to switch
leonarski_f and Claude Opus 5
2026-08-09 21:08:29 +02:00
3d3fb0e58b
Bragg integration: stop rectifying the fitted intensity into its own variance
leonarski_f and Claude Opus 5
2026-08-09 21:08:29 +02:00
1283e04ada
Merge branch 'fix23-work' into integration-variance-fixes
leonarski_f
2026-08-09 20:58:41 +02:00
97dbbc50b4
Merging: fit the error model on the reflections the cutoff keeps
leonarski_f and Claude Opus 5
2026-08-09 20:58:35 +02:00
72efb75a8c
Merging: do not floor the merged sigma at the systematic term
leonarski_f and Claude Opus 5
2026-08-09 20:58:35 +02:00
f4e281b2f5
Rotation: give prediction and partiality the energy bandwidth
leonarski_f and Claude Opus 5
2026-08-09 20:16:39 +02:00
e352227a2d
Scaling: divide out the incident flux before fitting the per-frame scale
leonarski_f and Claude Opus 5
2026-08-09 20:01:03 +02:00
f60768d49c
Bragg integration: drop the 2% sigma floor and carry the background variance
leonarski_f and Claude Opus 5
2026-08-09 19:10:57 +02:00
d565b66916
Post-refine: drop the rocking-width diagnostic, report frames per event
leonarski_f and Claude Opus 5
2026-08-09 19:10:18 +02:00
4d3434e2a5
Beam stop: compare each pixel only against its own ring
leonarski_f and Claude Opus 5
2026-08-09 05:31:57 +02:00
a29c36600f
Beam-stop shadow detection, and a low-resolution limit for scaling
leonarski_f and Claude Opus 5
2026-08-09 01:05:31 +02:00
c673521b76
Space-group search: do not veto on systematic-b where the H test confirms
leonarski_f and Claude Opus 5
2026-08-09 01:05:14 +02:00
df9a9c2a2c
Fix the defects found reviewing the branch before merge
leonarski_f and Claude Opus 5
2026-08-08 18:18:46 +02:00
3ccb97e31b
Adaptive spot finder: pin the per-ring host buffers
leonarski_f and Claude Opus 5
2026-08-08 15:33:24 +02:00
5830f78d57
Revert the azimuthal-integration sigma clip
leonarski_f and Claude Opus 5
2026-08-08 15:31:19 +02:00
b52729f12c
CLAUDE.md: changelog rules, and never push
leonarski_f and Claude Opus 5
2026-08-08 03:55:47 +02:00
a8d289e7cf
Powder calibration: write Poni1/Poni2 in pyFAI's frame, not ours
leonarski_f and Claude Opus 5
2026-08-08 03:55:42 +02:00