3750 Commits

Author SHA1 Message Date
Flössie
beaea22779 Vectorize color space conversion for HaldCLUT
Vectorize color space conversion for HaldCLUT depending on the
definition of `VECTLENSP`. It's not fully AVX compatible because `F2V`,
`LVF`, and `STVF` are SSE only.
2016-04-29 20:35:37 +02:00
Flössie
29fe23e517 Move film_simulation_strength calculation into HaldCLUT::getRGB()
- Moved `film_simulation_strength` calculation into `HaldCLUT::getRGB()`
- Removed unneeded base class `CLUT`
- Used `_MM_SHUFFLE`
2016-04-29 17:26:56 +02:00
Desmis
e9beb1b69c Add rec2020.icm 2016-04-29 07:53:41 +02:00
Desmis
b97ed08987 Add working profile Rec2020 2016-04-29 07:41:17 +02:00
Flössie
9dee6dddf1 Hoist out_rgbx allocation out of the loops 2016-04-26 22:16:23 +02:00
Flössie
bf499055e1 Apply HaldCLUT::getRGB() per tile line
`getRGB()` now takes a whole tile line instead of a single pixel.
2016-04-26 21:57:58 +02:00
Flössie
78c08e9e5c Add Ingo's optimizations
Add Ingo's SSE optimizations and clean up the non-SSE part of `getRGB()`
with `intp()`.
2016-04-26 20:48:11 +02:00
Flössie
56f8ea086c Correct whitespace
Last commit messed up some whitespace, this one fixes it.
2016-04-23 23:02:02 +02:00
Flössie
b1a9e96836 Store HaldCLUT as flat RGBx array
Instead of using an `Image16`, which is organized in planes, store the
HaldCLUT in an `AlignedBuffer<std::uint16_t>` with sequential RGBx
values. This gives a speedup of roughly 23% here.
2016-04-23 22:55:28 +02:00
Flössie
f639cd6b82 Use Image16 instead of Imagefloat for CLUT
Gain speed and reduce memory by using Image16 instead of Imagefloat for
the CLUT.
2016-04-23 22:55:28 +02:00
Flössie
e495093b18 Clean up clutstore.* and add LRU cache
This commit adds a true LRU cache to `rtengine` which is used in the new
`CLUTStore` class. The code in `clutstore.*` was cleaned up with C++11
features and small optimizations taken from my `clutbench` project.
The `CLUTStore` class was converted to a true singleton.
2016-04-23 22:55:28 +02:00
heckflosse
e61e488346 small speedup for lut += operator 2016-04-23 22:45:40 +02:00
heckflosse
d153b55493 Reduce default strength of raw ca-auto-correction 2016-04-23 20:31:35 +02:00
Ingo Weyrich
2382341f41 Merge pull request #3260 from Floessie/fix-babehav
Fix #3259, RT fails with ASan
2016-04-23 15:59:13 +02:00
Beep6581
50e6dfc95a Recreated GNU GPL v3 license file, hopefully fixes #3218 2016-04-23 03:17:22 +02:00
Beep6581
35267fe8fd Added DCPs: Sony ILCE-6000 #3220, Nikon D7200 #3236, Fujifilm X-T1 #3243. 2016-04-23 02:50:49 +02:00
Beep6581
43329b89b1 Updated Deutsch by TooWaBoo, #3247 2016-04-23 00:04:39 +02:00
heckflosse
a79e4bc24b small correction to last commit 2016-04-22 13:25:52 +02:00
Ingo Weyrich
609814afae Merge pull request #3262 from Beep6581/cacorrectoddheightfix
Unstable output w/ same PP3 and RAW CA auto-correction when image height % 2 == 1
2016-04-22 12:57:02 +02:00
heckflosse
e0e52fbffc replace all __attribute__ ((aligned (16))) with ALIGNED16 2016-04-22 12:54:53 +02:00
heckflosse
58fb7af2bb Unstable output w/ same PP3 and RAW CA auto-correction when image height % 2 == 1, fixes #3258 2016-04-21 20:51:47 +02:00
Flössie
296919eb2b Fix #3259, RT fails with ASan
When compiling RT with `-fsanitize=address` it fails early in
`Options::setDefaults()`. To fix it, initialize `baBehav` directly as
permitted with C++11. As a bonus, I converted `addsetids.h` from
defines to an enum.
2016-04-20 20:29:51 +02:00
heckflosse
f7af57d11e Allow lcp vignetting correction for lenses without focal length information (e.g. lenses without chip) 2016-04-12 13:38:45 +02:00
heckflosse
8e4c4be6d6 Allow lcp distortion correction for lenses without focal length information (e.g. lenses without chip) 2016-04-12 13:38:27 +02:00
heckflosse
44b9fdaae8 Fix crash and freeze opening files with GMCY cfa (e.g. Canon Powershot G1) 2016-04-11 17:06:54 +02:00
Hombre
3ac59ad467 Solving issue #3249: "Fit crop to screen doesn't work as expected
anymore"
2016-04-11 01:37:23 +02:00
Desmis
324b54bb46 extend the limit of the tint slider to 10 for IR 2016-04-10 07:26:51 +02:00
heckflosse
86d7d9063f correction for LUT += operator, fixes #3245 2016-04-08 12:46:08 +02:00
Ingo Weyrich
f64e3deea4 Merge pull request #3242 from Beep6581/ris_speedup
Speedup for getAutoExpHistogram and getAutoWBMultipliers. += operator for lookup tables (handy to sum up per thread built histograms)
2016-04-07 18:45:04 +02:00
heckflosse
1eb37d7730 LUT.h: new += operator to sum up per thread histograms; RawImageSource::getRAWHistogram: very small speedup 2016-04-07 18:34:39 +02:00
Ingo Weyrich
05aed401da Merge pull request #3239 from Beep6581/xtransnew
Merge xtrans speedups into master
2016-04-03 18:28:08 +02:00
heckflosse
4858315e24 xtrans_interpolate: removed benchmark code and astyled 2016-04-03 18:24:40 +02:00
heckflosse
d629006252 Speedup for getAutoExpHistogram and getAutoWBMultipliers 2016-04-03 17:40:03 +02:00
Beep6581
8464ecaec8 Added DCP profiles for Fujifilm X-E1, X-E2, X-T1 and X-T10 with permission from user cack 2016-03-31 13:59:34 +02:00
Adam Reichold
4b9bc5be88 Try to fix #3132 by making sure every file chooser dialog is given an appropriate parent window. 2016-03-28 13:32:30 +02:00
heckflosse
a053f38059 11% speedup for xtrans demosaic 2016-03-24 13:40:16 +01:00
heckflosse
f6cb2579d0 fix possible buffer overrun in raw ca correction 2016-03-22 17:46:30 +01:00
heckflosse
bfc0505320 xtrans demosaic: 4% speedup and disabled CLIP 2016-03-22 17:29:48 +01:00
Ingo Weyrich
71dd6a6e59 Merge pull request #3228 from Beep6581/xtransnew
Speedup for xtrans demosaic
2016-03-20 01:45:01 +01:00
Hombre
4dddb349b4 Fix #2388: "Partial Paste still buggy, quite impartial" 2016-03-20 01:41:56 +01:00
heckflosse
db275b1f91 removed benchamrk code from xtrans demosaic 2016-03-20 01:32:10 +01:00
heckflosse
0c786ee1bc RT won't read width of toolpanel on startup 2016-03-20 00:35:05 +01:00
Beep6581
d2d436ea6a Deutsch updated by TooWaBoo, closes #3227 2016-03-19 21:29:43 +01:00
Hombre
f36565cba9 Fix issue #3182: "Preview image jumps around on canvas while zooming" 2016-03-18 21:49:23 +01:00
heckflosse
8e35deb9c9 Fix uninitialized variable in toolpanelcoord.cc 2016-03-16 19:34:31 +01:00
heckflosse
636d0be314 about 4% speedup for xtrans demosaic 2016-03-15 19:21:07 +01:00
heckflosse
703b3edb0a Don't demosaic again when changing 'False Colour Suppression Steps 2016-03-14 17:40:19 +01:00
Ingo Weyrich
bc38a58297 Merge pull request #3215 from Beep6581/falsecoloursuppression
Speedups for raw false colour suppression
2016-03-13 23:25:36 +01:00
heckflosse
6434460efb removed benchmark code from raw false colour suppression 2016-03-13 23:23:32 +01:00
Beep6581
b0bfe48723 Increased Fast Export's max resize width and height to 10000, closes #3212 2016-03-13 19:59:04 +01:00