Commit Graph

1018 Commits

Author SHA1 Message Date
Marek Nečada 6cf1f667de Default tolerance constant.
Former-commit-id: 19951825d21f94455da0228430a674eec37e7653
2020-01-10 15:20:45 +02:00
Marek Nečada c86b881088 Forgotten function renames
Former-commit-id: 89fa50cd8cfbdcf055e57f54093464f1e028c5bc
2020-01-10 15:20:10 +02:00
Marek Nečada 5a98b91b3d Rewriting scatsystem: compiles without errors now.
Former-commit-id: cd68b0feaef7181874d94dc535fd2cc9bc89e518
2020-01-10 11:44:15 +02:00
Marek Nečada 541af5a984 WIP Rewriting scatsystem.
Former-commit-id: 17f0f48ab54b84c4701b17846f941dd0142eb668
2020-01-09 16:57:30 +02:00
Marek Nečada 5dd93235f0 New qpms_scatsys_apply_symmetry kinda done?
Former-commit-id: 49a7a7984af6ad6e0e5ec1b5cc7b61ac06f81b8d
2020-01-09 10:42:49 +02:00
Marek Nečada 7e57c3cc81 WIP new scatsystem
(keskeytetty n. scatsystem.c:214)


Former-commit-id: 5d205f46f38f7b9e988bd03b8a9bff70b3986808
2020-01-08 16:00:09 +02:00
Marek Nečada 3bf263c4f3 Copying of T-matrix operations
Former-commit-id: dea91f97e5e72146039868ab5f0c8ac5e7ea7a57
2020-01-08 15:18:25 +02:00
Marek Nečada e1a6389232 Some new convenience functions and types.
Former-commit-id: 7701cd8ee779e06ba18d6e19bfe650bd9465487f
2020-01-08 14:39:29 +02:00
Marek Nečada d31d8737b8 WIP scatsystem update for "abstract" T-matrices.
Former-commit-id: 7f723a0f459f263e12282edfb1e8deb440650880
2020-01-07 16:57:59 +02:00
Marek Nečada c2b4787cd5 Implementation of qpms_tmatrix_apply_operation().
Former-commit-id: 6773f0e1d02d5f929c2039f99338f08c25d0ccab
2020-01-07 08:14:06 +02:00
Marek Nečada dff8293e6d T-matrix general operation type definitions and destructor.
Former-commit-id: 491a4d8ad602a7252aa9f4446b55c7c905102de9
2020-01-06 02:17:09 +02:00
Marek Nečada d17a5e5eea WIP data structure redefinitions.
Former-commit-id: 5e2baffb4a47657233e792407630507ba611b129
2019-12-21 11:36:01 +02:00
Marek Nečada d53f2964f0 WIP abstract t-matrices
Former-commit-id: 8c573ac3a62bf92195246d6eb95f95df240c48a1
2019-12-19 13:50:12 +02:00
Marek Nečada b6e6554323 Fix imports in symmetries.py
Former-commit-id: 54104859c5858f92fdef0250991802629003e144
2019-12-19 05:30:46 +02:00
Marek Nečada dc5d2cde0b Upgrades to argproc.py, finite rectangular lattice scatter script.
Former-commit-id: 36aba53dc445752cf50e1638883f5a280ccab753
2019-12-14 13:26:40 +02:00
Marek Nečada ef1c699861 (Temporary) Makefile for calculating benchmark T-matrices
Former-commit-id: 2a30fcd6597c317d332224c27d59e60de1b5e5b7
2019-12-14 10:22:48 +02:00
Marek Nečada fb3e5467d6 Preparations for SCUFF-EM benchmarks
Former-commit-id: 9ab0dfb39833e0da0db78680a90984160e24e3e9
2019-12-14 10:22:48 +02:00
Marek Nečada 2f9e5670da Don't import legacy code to speed up qpms import.
Former-commit-id: b34cdc751f6076d0a02ebc6b122abbd8fbe9cde8
2019-12-14 09:17:29 +02:00
Marek Nečada 4c7dd1ee61 cytmatrix interpolator: access frequency table from python
Former-commit-id: 69a85dba08347f0c0543ecb7913e0b8e2c20473c
2019-12-14 08:57:19 +02:00
Marek Nečada f1f2c821df New CLI argument processing
Former-commit-id: d8fba975ccf08a11e0a4515e5af92edb7856f643
2019-12-14 08:54:43 +02:00
Marek Nečada 1dcebe4fee Add nogil
Former-commit-id: 8ef30e7002dace1691ce32815dc24975b5d7131e
2019-11-17 10:05:29 +02:00
Marek Nečada a712789386 modeproblem matrix parallel implementation to become default
Former-commit-id: ddf4ab3b83490de67034107b96272d725dde6a89
2019-11-17 09:59:06 +02:00
Marek Nečada d1068419f4 Finite square lattice scattering script
Former-commit-id: d44c0ecb929378e6ede63548bbc47825dacd6088
2019-11-14 17:23:19 +02:00
Marek Nečada c9a5661519 Fix invalid pointer in qpms_apply_tmatrix.
Former-commit-id: ba9400c3e1a39d472cfdebf7e61ab175c5c8fb6e
2019-11-14 13:36:45 +02:00
Marek Nečada 1821a0d8f5 Try to recover from failed quadrature.
Former-commit-id: 746709b224d825baf4a41ddfcd040a0c80fc45c8
2019-11-12 07:18:20 +02:00
Marek Nečada e4f368f47a Multi-threaded integration in axial symmetric T-matrix.
Former-commit-id: 71dc4ae5978f0dc10d82f4d8450fb2117ba64ce8
2019-11-08 17:43:07 +02:00
Marek Nečada cfda120a20 Fix sign in Drude-Lorentz model.
In our convention, where outgoing wave correspond to
Hankel functions of FIRST kind, the time-harmonic waves carry
(contrary to the more usual electrotechnical convention)
the factor exp(-iωt) and lossy materials are represented
by Im ε > 0, i.e. Im n > 0.


Former-commit-id: 19108c43a31899c2f9196c1a0fc44c5ecae69b5a
2019-11-08 09:36:12 +02:00
Marek Nečada 28f4e7f3d2 Complexify wavenumber in (finite) translation operators.
Former-commit-id: f84e1588cdda916d8feda6d807c12bca69512e5f
2019-11-06 18:13:50 +02:00
Marek Nečada cc9dbb6cc5 Make the additional Drude-Lorentz parameters available from Python
Former-commit-id: f08c96459c7cfd07b2adbf4670de59b1ce0b6ea1
2019-11-06 12:04:52 +02:00
Marek Nečada 493ba079e2 Some more Drude-Lorentz data from drudelorentz.com
Former-commit-id: 62f7f550e4e72d176690ac2b3532cec9bedb889f
2019-11-04 21:42:19 +02:00
Marek Nečada d79a35110c Merge branch 'beyn_rework'
Former-commit-id: 3366b1421056a2ac097b79f62bfbccd82d87de1b
2019-11-04 09:43:44 +02:00
Marek Nečada 84acb1ada2 Allow to select only either of long/short range Ewald sum parts
(cherry picked from commit 97b7782291380193835c23a4f0ea04ff5f44273e [formerly 41b1a76d5ed571b507abc0515b3cba60e8c0ccca])


Former-commit-id: 4d51186e653babfdc84fc45a4ab70a3ffdc02eee
2019-11-04 09:38:23 +02:00
Marek Nečada 4dd3234b0a Remove the branch cut at negative real axis in Ewald sum.
Move the cut to the negative imaginary axis instead.

Also update comments.


Former-commit-id: a70e6a4db0ac33836d672d9ee2356ea19a899a30
2019-10-09 13:20:20 +03:00
Marek Nečada acff55f396 Extend hardcoded T-matrix similarity tolerance.
Also some cython functions->properties.


Former-commit-id: 4a0605b1cf8b857c681f194a475ff3f10e7dfe71
2019-10-07 15:15:34 +03:00
Marek Nečada e7f0e0131f Basic branch selection functionality for incomplete Γ.
No modifications with effects on Ewald sums done yet.


Former-commit-id: e362341713ce306482386b9e6f2a48c336fad7a1
2019-10-06 01:30:29 +03:00
Marek Nečada 2dd9fe2f34 Minor stuff.
Former-commit-id: c0de35f6a52ce4cc4f34cf4b417364d92e8dac4d
2019-10-05 20:38:04 +03:00
Marek Nečada aa86fcfb08 Expose low-level scalar Ewald sums via cython.
Former-commit-id: 9910decff1e8e91c802f0f8d10573ec430dda468
2019-10-03 00:52:14 +03:00
Marek Nečada 2d891be12f ewald.h cleanup
Former-commit-id: c02360cfa9df1dac06897883115cafc98a912107
2019-10-02 20:11:25 +03:00
Marek Nečada 0239a3c9a1 Fix memset size in Mie T-matrix generator.
Former-commit-id: 244dad82dec42fdc1e08cc12205e373711465622
2019-10-02 17:53:01 +03:00
Marek Nečada 79e6c47f94 Beyn integration contours to pxd
Former-commit-id: d33de9a694ea7d5600a627062c7119e8483d753e
2019-10-01 11:52:29 +03:00
Marek Nečada c7f2e32ee4 Free unfinished PGens from Ewald sums.
Also replace some bare `if (xxx) abort();` with
`QPMS_ENSURE_SUCCESS(xxx)`.


Former-commit-id: d4712292fe0f9fb397cf3a490131bcb37d542fa6
2019-10-01 11:29:15 +03:00
Marek Nečada 7e74cb100c Experimental speedup by earlier cutoff for Ewald summation.
TODO check whether PGen is properly destroyed


Former-commit-id: db2c64292b9c1bd80e97255bf71e570385b409a4
2019-09-30 17:57:06 +03:00
Marek Nečada dad170b8fe Kidney contour basic test
Former-commit-id: 861cf5f9a8409db360e48db74274f20069b94391
2019-09-30 10:54:10 +03:00
Marek Nečada f940e62e52 Test and fix the rounded half-ellipse contour..
Former-commit-id: 448c30d375b3f9d0abab9442ae00bfcd123e5cb9
2019-09-30 08:53:57 +03:00
Marek Nečada d6084f3264 WIP Alternative integration contour (untested)
Former-commit-id: c82da58ac33364797d7175a69feed475184937d9
2019-09-29 22:32:53 +03:00
Marek Nečada 2e7b925be2 WIP trying to fix the half-ellipse contour.
Former-commit-id: cbba6fffacbb38322a590478899f38c7d7dafe8e
2019-09-26 11:36:41 +03:00
Marek Nečada b3c3eeb2a2 Fix derivatives in integration contours.
Former-commit-id: 7575a69a82eb19126aaac9aede9f170815b6015a
2019-09-26 10:09:35 +03:00
Marek Nečada 2fcc6282fa "Half-ellipse" integration contour.
Former-commit-id: 800079fba837d65f84af62d5adfad177a6aa3cdf
2019-09-26 10:09:35 +03:00
Marek Nečada 67f93e461c Ewald translation calculator function using base specs.
Former-commit-id: d49880caa280cc089fa688d0cdeccb95db7cb64b
2019-09-23 08:59:18 +03:00
Marek Nečada 6445c3523e Experiments with Beyn's algorithm and singularities.
It seems that everything is fine in the end except
essential singularities inside the contour.


Former-commit-id: 9fe7135cb30d1fff1dc3ff9bf8a9152c6b0ef9b4
2019-09-17 15:14:09 +03:00