Skip to content

docs: status.md says what is live at 0.5.0 #49

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Merged
Paullo9 merged 1 commit into from
Nov 6, 2026
Merged

docs: status.md says what is live at 0.5.0 #49

Changes from all commits
Commits
Show all changes
1 commit
Select commit
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
10 changes: 6 additions & 4 deletions docs/status.md
View file
Open in desktop
Original file line number Diff line number Diff line change
@@ -1,9 +1,11 @@
# Status

SimulatoKit 0.4.0 (2026-10-22), against abTEM 1.0.10. Live: crystal cells and the Lobato projected potential, any grid length of at least 2 (vDSP where it has
the length, a mixed-radix FFT for most other lengths, Bluestein where a large prime factor makes it the cheaper), multislice on `.cpu` (float64) and `.metal` (float32, held to the CPU), beam tilt, the probe with
polar aberrations C10 to C56, `GridScan`, frozen phonons, annular, pixelated, segmented and centre-of-mass detectors, `STEM.scan`, the contrast transfer function,
source size and Poisson noise, `LineScan` and line profiles, integrated centre of mass (`integrate_gradient`, iCOM), and `WavesDetector` (the exit wave at every scan position). What is checked against what: `docs/parity.md` (37 present, 36 partial, 55 missing of 128 rows); timings against Python:
SimulatoKit 0.5.0 (2026-11-06), against abTEM 1.0.10. Live: crystal cells and the Lobato projected potential (`project`, tiled; slices that follow the crystal planes), any grid length of at least 2 (vDSP where it has
the length, a mixed-radix FFT for 2, 3, 5 and 7 and a generic odd-radix pass for the other lengths, Bluestein where a large prime factor makes it the cheaper), multislice on `.cpu` (float64) and `.metal` (float32, held to the CPU), beam tilt, the probe with
polar aberrations C10 to C56, `GridScan`, frozen phonons (exit waves too), annular, pixelated, segmented (DPC: integrate, integrateRadial, differentials on a polar scan) and centre-of-mass detectors, `STEM.scan`, the contrast transfer function,
source size and Poisson noise (samples too), `LineScan` (matched to a probe: end and Nyquist sampling defaults) and line profiles, integrated centre of mass (`integrate_gradient`, iCOM), the azimuthal average of a centred pattern, Fourier-space image interpolation and `integrateDisc(interpolate:)`, and `WavesDetector` (the exit wave at every scan position). What is checked against what: `docs/parity.md` (38 present, 40 partial, 51 missing of 129 rows); timings against Python:
`docs/benchmarks.md`; what is open: the GitHub issues; known limits: `docs/parity.md`.

Since 0.4.0: 0.5.0 (2026-10-31) added the items above and the mixed-radix and odd-radix FFT passes (250² 2.2x, 500² 2.3x off vDSP). Details: `CHANGELOG.md`.

Tests are split in two. `SimulatoKitParityTests` holds the pinned abTEM comparisons, the truth tests, Metal against CPU and the benchmarks (111 tests; fixtures live in its `Fixtures/`); the other targets hold the fast unit tests (35 tests); `SimulatoKitTestSupport` holds the helpers both use. `swift test -c release --parallel --skip SimulatoKitParityTests` runs the fast half; the plain `swift test -c release --parallel` runs all 146.
Loading