f149f0a024
RESULT-CHANGING: Refinement: the bulk-solvent mask grid is never coarser than the density grid
leonarski_fandClaude Opus 5.52026-10-27 20:16:17 +01:00
ae5b52bfeb
Merge twin-refine into refine-merge: the twin decision between two whole refinements
leonarski_fandClaude Opus 5.52026-10-27 19:07:56 +01:00
1afe439ca2
Merge tls-aniso into refine-merge: TLS / anisotropic U decided after the solvent build
leonarski_fandClaude Opus 5.52026-10-27 19:04:42 +01:00
d63ae4b951
Merge ncs-refine into refine-merge: NCS local distance restraints with the solvent build in the trajectory
leonarski_fandClaude Opus 5.52026-10-27 18:59:16 +01:00
2a0e91bd6a
RESULT-CHANGING: Refinement: the solvent build within the weight search, two rounds each judged by refining the solvent
leonarski_fandClaude Opus 5.52026-10-27 18:39:47 +01:00
26469ca8e8
RESULT-CHANGING: Refinement: twin-law refinement, decided against the untwinned one and a null
leonarski_fandClaude Opus 5.52026-10-27 17:58:20 +01:00
47f4a603a6
RESULT-CHANGING: Solvent: waters and ions relabelled by their shell and their density, metal records decided once
leonarski_fandClaude Opus 5.52026-10-27 17:38:10 +01:00
6472ce3ff1
RESULT-CHANGING: Refinement: anisotropic atoms and TLS groups, the displacement model decided on the free set
leonarski_fandClaude Opus 5.52026-10-27 17:34:27 +01:00
696a066442
RESULT-CHANGING: RefinementNcs: copies need 80% identity over the longer chain, not the shorter
leonarski_fandClaude Opus 5.52026-10-27 16:33:31 +01:00
613b3eda2c
RESULT-CHANGING: Refinement: copies of a molecule found by sequence alignment and restrained locally, decided by R-free
leonarski_fandClaude Opus 5.52026-10-27 16:09:40 +01:00
0adc7e0f3e
RESULT-CHANGING: Refinement: atoms on special positions held on them, their contacts counted once
leonarski_fandClaude Opus 5.52026-10-27 15:40:25 +01:00
51cbe0ef7d
RESULT-CHANGING: Refinement: waters built at difference-map peaks and taken out of negative density, between refinements
leonarski_fandClaude Opus 5.52026-10-27 14:57:12 +01:00
5b6e567d8d
RESULT-CHANGING: RefinementRestraints: a metal's LINK record becomes a bond only within its inner sphere
leonarski_fandClaude Opus 5.52026-10-27 14:03:33 +01:00
873fd25837
battery: a refine arm - public depositions through rugnux --mode refine, first set 1HR2
leonarski_fandClaude Opus 5.52026-10-27 13:21:39 +01:00
85e0d2da95
Refinement: the solvent audit of the refined model, as REFINE_SOLVENT_ keys and <prefix>_solvent.json
leonarski_fandClaude Opus 5.52026-10-27 13:20:01 +01:00
ca660416a5
SolventSites: score every water and ion of a model as water, Mg, Na, K, Ca or Cl by its surroundings
leonarski_fandClaude Opus 5.52026-10-27 13:19:49 +01:00
ef9c48a641
RefinementRestraints: AtomsNear - the atoms near a point, symmetry mates and lattice translates included
leonarski_fandClaude Opus 5.52026-10-27 13:19:49 +01:00
a645d0a9fd
docs: front page names the reimplemented algorithms and generative AI in the licence section, and says the licences' sharing condition plainly
leonarski_fandClaude Opus 5.52026-10-27 11:04:46 +01:00
3006f2bb33
docs: front page footer says the software reimplements existing algorithms, moved to the GPU and made faster, and was developed with the help of generative AI
leonarski_fandClaude Opus 5.52026-10-27 10:55:57 +01:00
4fd2bf231d
docs: front page says "structures" rather than "refined structures", names the programs' user operation at the SLS 2.0 MX beamlines, and carries the PSI logo
leonarski_fandClaude Opus 5.52026-10-27 09:23:47 +01:00
546566cff4
Refinement: the gather's pairs built in fixed-shape buffers, and each evaluation replayed as a CUDA graph
leonarski_fandClaude Fable 5.12026-10-27 04:00:42 +01:00
d7c15c629e
Rotation tail: the final merge made ahead on a spare card, in the groups the run is expected to adopt
leonarski_fandClaude Fable 5.12026-10-27 04:07:12 +01:00
037c32d9fb
Error model: ISa is reported only where b is the solution of a fit - not beside a clamped a, not when b was held at zero, and on stills under the same rule
leonarski_fandClaude Fable 5.12026-10-26 23:06:54 +01:00
b16873036c
Refinement: density records sent from pinned memory, and the L-BFGS direction in fused sweeps
leonarski_fandClaude Fable 5.12026-10-26 22:14:35 +01:00
3189f71e8f
Refinement: the engine polls for its evaluations and the pair count comes back pinned; the slots beyond one per card go to the cards the ladder's long trials leave free
leonarski_fandClaude Fable 5.12026-10-26 22:30:31 +01:00
68e567bd02
Tail: the molecular replacement of a prediction or --mr starts beside the analyses; a pass's integrated observations are handed back on a thread of their own
leonarski_fandClaude Opus 5.52026-10-26 22:20:24 +01:00
b0c04d0070
broker: rugnux live publishes its result before the run's teardown, which goes on beside the end message
leonarski_fandClaude Opus 5.52026-10-26 20:54:44 +01:00
604b270114
ModelDensityGPU: the gather stages each atom's box as the brick sees it, and only what the brick needs of the atom
leonarski_fandClaude Fable 5.12026-10-26 20:42:32 +01:00
f724bd1117
ModelScaleGPU: the host polls for a scale fit's launches instead of sleeping on them
leonarski_fandClaude Fable 5.12026-10-26 20:37:07 +01:00
128e3bb9ae
RefinementGPU: the symmetry phases from a table instead of sincospi in the compose and scatter kernels
leonarski_fandClaude Fable 5.12026-10-26 20:26:20 +01:00
dfb1536fe2
Sweep ends, overloads and ice flags: one walk over the reflections instead of two
leonarski_fandClaude Fable 5.12026-10-26 20:25:56 +01:00
00f8c8ba4b
Refinement: the final ladder takes the rungs the step from the best state tried already; the longest trials to the slots first
leonarski_fandClaude Fable 5.12026-10-26 19:24:48 +01:00
e4d64e67c9
rugnux: French-Wilson takes the anisotropy tensor from the merge alone, the rest of the analysis is joined after molecular replacement
leonarski_fandClaude Opus 5.52026-10-26 18:18:28 +01:00
5338e1b535
Centre-of-symmetry statistics and anisotropy: hashed lookups and the simulation and cell assignment on all threads, output unchanged
leonarski_fandClaude Opus 5.52026-10-26 18:16:38 +01:00
7c49fefb54
Tail: the cell e.s.d.s and the unmerged MTZ table beside the refinement; the refined difference and anomalous maps side by side
leonarski_fandClaude Opus 5.52026-10-26 18:03:56 +01:00
067f3a7d7b
docs landing: Rugnux card names synchrotron rotation and XFEL serial data and links the public test-data lists; footer motto removed
leonarski_f2026-10-26 16:53:04 +01:00
2736f4c85e
Docs landing page: the hero is a live Rugnux run - a frame, the placed model in its density, the timeline
leonarski_f2026-10-26 16:54:19 +01:00
2e296e6bed
jfjoch_writer: the live refinement's model, maps and blobs are written beside the master file
leonarski_f2026-10-26 15:48:40 +01:00
2461d67e14
NXmx: the unit cell's e.s.d.s are saved as /entry/sample/unit_cell_esd
leonarski_f2026-10-26 15:47:45 +01:00
d50b3d7034
CI: the documentation goes to rugnux/gitea-pages only when a release is created
leonarski_fandClaude Opus 5.52026-10-26 15:46:17 +01:00
9830fe9dd0
broker: redoc regenerated from the merged API (blob distance optional, live refinement on the CPU too)
leonarski_fandClaude Opus 5.52026-10-26 15:39:21 +01:00
13bd9db55c
Refinement: a B factor is held at its bound by the whole target's gradient, restraints included
leonarski_f2026-10-26 15:37:00 +01:00
2be88e11e3
Refinement restraints: contacts with lattice translates are kept, and an atom's own symmetry contact once
leonarski_f2026-10-26 15:35:20 +01:00
470fc4e479
ffbidx (vendored): INDEXER_GPU_DEVICE is not read, the indexer stays on the thread's pinned GPU
leonarski_f2026-10-26 15:22:22 +01:00
257bf2fe2b
Refinement: trial slots counted with their scale pool and both restraint sets; on the CPU no more than -N
leonarski_f2026-10-26 15:22:22 +01:00
ac58921445
rugnux live: a score that is not finite is left out of the status, not sent as null
leonarski_f2026-10-26 15:21:45 +01:00
3d382ce0a0
Refinement: an exception in a trial slot or in RefineModel no longer ends the run
leonarski_f2026-10-26 15:21:45 +01:00
0143d57eb0
CellEsdTest: synthetic cells in place of cells that looked measured
leonarski_f2026-10-26 15:21:31 +01:00
48bf9ebe3a
CellEsd: PI from JFJochMath.h instead of M_PI, and the Herbstein (2000) credit at the block e.s.d.s
leonarski_f2026-10-26 15:21:31 +01:00
1b48f3873f
docs: front page and quick start - Rugnux goes on to a refined structure with a model
leonarski_f2026-10-26 15:22:38 +01:00
cd0b994027
API: rugnux_live_refinement no longer says live refinement needs a GPU
leonarski_f2026-10-26 15:19:29 +01:00
66cc0bd563
docs: acknowledge Herbstein (2000) for the unit-cell e.s.d.s
leonarski_f2026-10-26 14:49:02 +01:00
af28cf430b
docs: technical pages caught up with the stills rocking curve, beam-shadow integration and refinement
leonarski_f2026-10-26 14:49:02 +01:00
7b1bc3a3ea
docs: report headline keys, model section for every way a model is given, still battery
leonarski_f2026-10-26 14:45:12 +01:00
07c8b2ddd4
docs: the Rugnux card and the naming table mention refinement and live collections
leonarski_f2026-10-26 14:43:18 +01:00
c3c5a32a0d
docs: rugnux getting-started pages - refinement, the full output list, a troubleshooting note
leonarski_f2026-10-26 14:43:13 +01:00
8edd0e5f3e
docs: rugnux options, model and export pages brought in line with the usage text
leonarski_f2026-10-26 14:42:44 +01:00
d8c14046b2
rugnux: unit-cell e.s.d.s - ZERR in a new SHELX .ins beside the .hkl, _cell.*_esd in the mmCIF, CELL_ESD_* in the report
leonarski_f2026-10-26 14:10:26 +01:00
22debb1be6
docs: the stills rocking-curve datasets in the _process.h5 master
leonarski_f2026-10-26 11:35:15 +01:00
e6b550fe42
Tests: the stills rocking width - the window at a reflection's resolution, _process.h5 round trip
leonarski_f2026-10-26 11:34:56 +01:00
8a123e794d
Stills: the run's rocking curve goes to _process.h5 and --mode scale merges with it; MeasureRunWidth takes the window at each reflection's resolution
leonarski_f2026-10-26 11:34:56 +01:00
00662c4110
Refinement: common cofactors and modified RNA nucleotides in the embedded dictionary; heme iron restrained from LINK records
leonarski_f2026-10-26 13:50:58 +01:00
d90bc68085
CI: the documentation also pushed to the gitea-pages branch of rugnux/gitea-pages, on every push for now (token test)
leonarski_fandClaude Opus 5.52026-10-26 11:39:10 +01:00
771e669f75
rugnux usage: -S on serial stills also settles the indexing ambiguity from the data
leonarski_fandClaude Opus 5.52026-10-26 11:16:34 +01:00
6491cc5a7e
docs: serial-stills indexing ambiguity described the same way everywhere
leonarski_f2026-10-26 11:15:28 +01:00
0cc3129650
docs: refinement's GPU engine in GPU data analysis, refinement in the pipeline figures
leonarski_f2026-10-26 11:13:23 +01:00
f960bd7a80
Docs landing page: a real diffraction frame in the hero, and the command with a model
leonarski_f2026-10-26 10:51:15 +01:00
4dbcb35708
Stills partiality: the rocking width measured in shells of equal width in d*, used as a table
leonarski_f2026-10-26 10:47:54 +01:00
0829b9a72d
Stills prediction: every frame predicts through two rocking widths of the run, measured first and centred on its radial scale
leonarski_f2026-10-25 06:01:28 +01:00
13da79b788
Refinement GPU: the atom gradient kernel walks its box without integer division
leonarski_f2026-10-26 08:40:13 +01:00
4a8e13ddfa
Refinement GPU engine: its waits on the stream asleep (blocking-sync event) rather than spinning
leonarski_f2026-10-26 08:31:43 +01:00
f34d2f4cca
Refinement: the weight trials' own slots made while the placement runs
leonarski_f2026-10-26 08:13:06 +01:00
1f8b5b1986
Refinement: a Refit's solvent grid starts around where the context's last one ended
leonarski_f2026-10-26 08:08:32 +01:00
1f1aa18b67
ModelScaleGPU: the robust fit's median of the residuals on the device
leonarski_f2026-10-26 08:00:27 +01:00
ca0a6dd343
Refinement: each slot's likelihood fit on all of the run's threads
leonarski_f2026-10-26 07:57:30 +01:00
2c1df0ebfa
Refinement: the likelihood's D and beta searches by Newton's method; D on the working set from the last fit's
leonarski_f2026-10-26 07:52:50 +01:00
a8d3a8dbc2
Refinement: the mask grid sized without allocating it; the L-BFGS pairs keep their s.y
leonarski_f2026-10-26 07:33:16 +01:00
6da51dc5d7
rugnux: no --refine flag - a placed model is refined by default, --no-refine skips it
leonarski_fandClaude Opus 5.52026-10-26 06:54:15 +01:00
3882b0f062
Serial stills with a reference or model: crystals their own cells anchor keep that indexing
leonarski_f2026-10-26 01:36:51 +01:00
ba52aadbce
rugnux: the geometry refinement algorithm is -r, --geom-refine; --refine was declared twice and its no-argument form shadowed it
leonarski_fandClaude Opus 5.52026-10-26 06:49:20 +01:00
a78bfe951d
rugnux: a model and a reference among the inputs, in any order, told by their contents
leonarski_f2026-10-26 00:40:00 +01:00
794db0ca34
Refinement cut by the run's cancel; the hero shows the refined R-work / R-free
leonarski_f2026-10-26 00:40:00 +01:00
baec34685d
Rugnux live: the uploaded model is refined after placement; refined scores, model, maps and blobs in the status, the viewer and the hero
leonarski_f2026-10-25 17:34:00 +01:00
7bf906e4a2
rugnux: a placed model is refined by default - giving a model is the opt-in
leonarski_f2026-10-25 17:33:46 +01:00
3b06bcee78
Refinement: the trajectory ends at its best step, steps try the positional weight halved too, the final ladder must win
leonarski_f2026-10-25 23:57:56 +01:00