Age | Commit message (Collapse) | Author | |
---|---|---|---|
2019-08-22 | Framework for new unit cell comparison function | Thomas White | |
2019-08-16 | cellcompare_check: Progress bars | Thomas White | |
2019-08-16 | Don't use H centering for cell comparison check | Thomas White | |
2019-08-16 | Reduce range of reindexing matrices | Thomas White | |
2019-08-16 | Set centering on input cell | Thomas White | |
2019-08-16 | Pass transformation matrix for debugging | Thomas White | |
2019-08-16 | Implement random reindexing matrix | Thomas White | |
2019-08-16 | "Final" cellcompare_check | Thomas White | |
2019-08-16 | Add cellcompare_check | Thomas White | |
2019-07-30 | Spectrum: Define Gaussian using area, not height | Thomas White | |
2019-06-04 | gpu_sim_check: Increase tolerance from 1.1% to 1.2% | Thomas White | |
1.13% deviation now observed in the wild, with no other problems. | |||
2019-05-29 | Fix SASE simulation | Thomas White | |
2019-05-29 | Spectrum generation | Thomas White | |
2019-05-29 | Use Spectrum API for simulation | Thomas White | |
2019-05-15 | Complete implementation of Spectrum API | Thomas White | |
2019-05-14 | Move Histogram out of API | Thomas White | |
It's only used by two test programs | |||
2019-05-14 | gpu_sim_check: Increase tolerance from 1% to 1.1% | Thomas White | |
1.03% deviation was observed in the wild, with no apparent problems. | |||
2019-05-13 | gpu_sim_check: Succeed if GPU not available | Thomas White | |
On recent OSes, OpenCL (the API) is always available, but a GPU device not always. Therefore, this test will always be built, and shouldn't fail just because a GPU isn't present. | |||
2019-05-07 | Remove intmat_set_all_3x3() | Thomas White | |
2019-03-13 | Rename rtnl_mtx_solve and rtnl_mtx_mult to make their purpose clearer | Thomas White | |
2019-03-11 | Change matrix notation to match ITA chapter 5.1 | Thomas White | |
2019-03-11 | tests/transformation_check: Tidy up and be louder | Thomas White | |
2019-03-11 | Keep track of the "un-centering" matrix, as well as the "centering" | Thomas White | |
This makes it easy to reverse the transformation, if required, which it is when comparing centered cells. | |||
2019-03-11 | Be clear about whether functions need rational or integer transformations | Thomas White | |
2019-03-11 | Add new rational number library | Thomas White | |
2019-03-11 | Use IntegerMatrix for all unit cell transformations | Thomas White | |
Get rid of UnitCellTransformation, a thin wrapper which didn't do anything. | |||
2018-11-14 | Merge branch 'tom/logmerge' | Thomas White | |
2018-09-10 | Change image.spectrum to image.spectrum0 to use a different structure for ↵ | Alexandra Tolstikova | |
spectrum. | |||
2018-06-28 | Fix tests for log merging | Thomas White | |
2018-06-25 | Rename first/second/third/fourth_merge_check to process_hkl_check_1/2/3/4 | Thomas White | |
2018-06-25 | Test the compiled binaries, not the installed ones | Thomas White | |
2018-06-25 | Switch remaining tests from "ex" to "sed" | Thomas White | |
2018-06-22 | tests/first_merge_check: Fix bracket | Thomas White | |
2018-06-11 | Use different temporary filenames for partialator merge checks | Thomas White | |
...because they might be run in parallel. | |||
2018-05-24 | adapt cmake to link to fdip includes in tests | Yaroslav Gevorkov | |
2018-05-17 | Remove .gitignore completely | Thomas White | |
2018-05-17 | Remove old test from CMakeLists.txt | Thomas White | |
2018-05-17 | Merge branch 'tom/cmake' | Thomas White | |
2018-05-11 | Add --no-logs to partialator tests | Thomas White | |
2018-05-11 | Much better way of stripping the audit information out of hkl files | Thomas White | |
2018-05-11 | New scaling test | Thomas White | |
Not just for linear scaling | |||
2018-05-07 | linear_scale_check: Set redundancies for reflection list | Thomas White | |
2018-05-07 | Simplify scaling | Thomas White | |
2018-05-07 | Preparation for adjusting B factor during post-refinement | Thomas White | |
Add --no-Bscale option to partialator, and pass down as far as needed residual() no longer does scaling: call scale_one_crystal() first if necessary scale_one() replaces old linear_scale() function to scale a pair of RefLists (but so far does the same as the old function) | |||
2018-03-22 | OpenCL stuff | Thomas White | |
2018-03-22 | Compile and run tests | Thomas White | |
2018-02-27 | Don't complain loudly about scaling failures during refinement | Thomas White | |
As long as it doesn't end up somewhere bad, it can probe where it likes. | |||
2018-02-27 | Add test for linear scaling | Thomas White | |
2018-02-27 | pattern_sim: Add --flat | Thomas White | |
2018-02-27 | Remove tests/pr_p_gradient_check | Thomas White | |
Will not be needed for new gradient-free minimiser |