[babl] update NEWS
- From: Øyvind "pippin" Kolås <ok src gnome org>
- To: commits-list gnome org
- Cc:
- Subject: [babl] update NEWS
- Date: Wed, 24 Jul 2019 14:15:19 +0000 (UTC)
commit dff11553ae4bce9e4009ea2489dddc0f1821ae10
Author: Øyvind Kolås <pippin gimp org>
Date: Wed Jul 24 16:09:48 2019 +0200
update NEWS
NEWS | 205 +++----------------------------------------------------------------
1 file changed, 7 insertions(+), 198 deletions(-)
---
diff --git a/NEWS b/NEWS
index 828d864..5ff22eb 100644
--- a/NEWS
+++ b/NEWS
@@ -3,6 +3,13 @@ The NEWS file in the babl source tree is the source location for
the news section both in the README and the webpage.
-->
+2019-07-2x babl-0.1.68 </dt><dd>
+Added Yu'v' (CIE 1976 UCS) color model. Fixed a severe non-initialised memory
+issue - that kept us from using fast paths for many babl spaces loaded from
+profile. More fully symmetric conversions between associated and separate
+alpha. Added more converters, both SSE2 and non-SSE2 for matrix-conversion
+between different RGB spaces. Improvements to website/documentation; babl is now using the terms associated
alpha and separate alpha, all of: nonpremultiplied-, premultiplied- and nonassociated- alpha are now
considered deprecated API. HCY luminance is now relative to the RGB space used - no change for sRGB related
pixel formats.
+
2019-06-12 babl-0.1.66 </dt><dd>
Added API call, babl_space_get_rgb_luminance,
Make most SIMD code-paths optional again on i686.
@@ -20,201 +27,3 @@ scRGB space, works like sRGB but always with linear TRCs.
Model introspection API permitting low overhead checks whether a format/model
is RGB, CMYK type of alpha and similar.
</dd><dt>
-2018-10-05 babl-0.1.60 </dt><dd>
-Improved thread safety, acceleration for R'G'B'A u8 -> cairo-ARGB32 conversion.
- </dd><dt>
-2018-10-05 babl-0.1.58 </dt><dd>
-Preserve color of transparent pixels in conversion to premultiplied alpha,
-Added single precision code-paths for faster fallback/reference conversions.
-New BABL_ICC_INTENT_PERFORMANCE bitflag for combining with intent as bitflags,
-use of matrix+trc when relative colorimetric CLUT is present. New color model
-and formats, CIE xyY.
-
- </dd><dt>
-2018-08-14 babl-0.1.56 </dt><dd>
-Improvements to the caching of profiled conversion chains between invocations
-by ignoring unknown bits in cache file and remember which conversions yielded
-reference fishes.
-
-2018-07-23 babl-0.1.54 </dt><dd>
- </dd><dt>
-Export babl_space_get_icc, babl_space_get, babl_model_with_space,
-babl_space_with_trc, babl_format_get_encoding, babl_model_is, SSE2 versions of
-YA float and Y float to CIE L float.
- </dd><dt>
-2018-06-02 babl-0.1.52 </dt><dd>
-Concurrency fixes and fixes to handling of 0/1 entry palettes, do not
-pre-equalize XYZ conversion matrices for sRGB, internal clean-ups, pre-define
-ACES2065-1 and ACEScg BablSpaces, add R~G~B~ set of spaces, which for all
-BablSpaces mean use sRGB TRC.
- </dd><dt>
-2018-05-20 babl-0.1.50 </dt><dd>
-Improvements to speed and precision of indexed code, improvements to meson
-build.
- </dd><dt>
-2018-05-15 babl-0.1.48 </dt><dd>
-fix u8 <-> double conversions for chroma, SSE2 version of RGBA float to
-CIE L / Lab. Build with -Ofast by default.
- </dd><dt>
-2018-04-10 babl-0.1.46 </dt><dd>
-added extensions with more coverage for u32, half and other utilit fast paths
-improving fast path coverage.
- </dd><dt>
-2018-02-18 babl-0.1.44 </dt><dd>
-Fix bug in custom primaries/ICC fast paths, improve meson build<br/>
- </dd><dt>
-2018-01-23 babl-0.1.42 </dt><dd>
-Fully initialize fishes when loading from cache. <br/>
- </dd><dt>
-2018-01-16 babl-0.1.40 </dt><dd>
-Added format "CIE XYZ alpha" color model and formats.
-New API babl_process_rows for reduced overhead in some scenarios; though bigger
-gains seen also for regular babl_process with reimplemented branch-free
-dispatch, and faster cbrt. Added meson build; being tested in parallel with
-auotmake, Improved reference conversions for formats skipping some of models
-components, Fixed gamma handling in indexed/palettized formats and improved
-gamma precision consistenct in sse2 conversions.<br/>
- </dd><dt>
-2017-11-15 babl-0.1.38 </dt><dd>
-Added format "CIE L float", a couple of protections against division by 0.0
- </dd><dt>
-2017-11-10 babl-0.1.36 </dt><dd>
-Optimized customized primary aware code paths for CIE Lab<->RGB conversions,
-improved accuracy of gamma approximations. New API babl_format_exists() for
-checking validity of babl format name, crash proofing of cache handling and use
-of environment variables.
- </dd><dt>
-2017-10-06 babl-0.1.34 </dt><dd>
-Brown paper bag release, Fix indexed / custom primaries conflict, and re-export
-a symbol used by old GEGL/GIMPs.
- </dd><dt>
-2017-10-03 babl-0.1.32 </dt><dd>
-Added custom primaries and TRC support through ICC parsing, improved
-float<->half performance, rewrite of all conversions functions to adhere to new
-signature.
- </dd><dt>
-2017-07-15 babl-0.1.30 </dt><dd>
-Thread stability for palette modes, stricter alpha preservation, now
-cross-compilable on android.
- </dd><dt>
-2017-05-30 babl-0.1.28 </dt><dd>
-Fast paths for Lav <-> Lch, release triggered by pending GIMP release.
- </dd><dt>
-2017-05-09 babl-0.1.26 </dt><dd>
-Build and install HCY color space, platform independences fixes to fish cache.
- </dd><dt>
-2017-02-01 babl-0.1.24 </dt><dd>
-Improvements to profile cache persistance, fast paths that improve actual GIMP
-use on various precisions, use single precision constants for some more of CIE
-computions.
- </dd><dt>
-2016-11-05 babl-0.1.22 </dt><dd>
-Added cache of profiled conversions, added HCY color model, some precision and
-performance updated for fast paths.
- </dd><dt>
-2016-11-05 babl-0.1.20 </dt><dd>
-Fix run-time errors, leaks, and race conditions. Add conditional fast paths.
- </dd><dt>
-2016-06-13 babl-0.1.18 </dt><dd>
-Bugfix: take alpha threshold in consideration when processing floats in SSE2.
- </dd><dt>
-2016-02-12 babl-0.1.16 </dt><dd>
-Improvements to half float reference, SIMD, and fast-paths, and CIE float fast
-paths; cleanups of fast path extensions.
- </dd><dt>
-2015-11-19 babl-0.1.14 </dt><dd>
-sRGB precision tuning, stability fixes, locale fix for setting error tolerance.
- </dd><dt>
-2015-02-03 babl-0.1.12 </dt><dd>
-optimizations for floating point conversions, HSV and HSL color models,
-removal of dead code, fixed CIE Lab conversions.
- </dd><dt>
-2012-03-30 babl-0.1.10 </dt><dd>
-LUT based speedups for gamma correction / uncorrection.
- </dd><dt>
-2012-03-30 babl-0.1.8 </dt><dd>
-Added support for indexed/pallette based formats, constified API.
- </dd><dt>
-2011-11-18 babl-0.1.6 </dt><dd>
-Build improvements, remove blatantly wrong conversions from extensions,
-made it possible to distinguish format_n formats from others, improvements
-to vala/gobject introspection support.
-
- </dd><dt>
-2011-01-20 babl-0.1.4 </dt><dd>
-Improved cross platform build ability, added code to handle n-component formats,
-remove extraenous runtime profiling that impacted performance.
-
- </dd><dt>
-2010-01-15 babl-0.1.2 </dt><dd>
-Made babl mostly threadsafe (sufficient for GIMP/GEGLs needs). Streamlined core functionality, reducing
number of function calls and removing overeager instrumentation.
- </dd><dt>
-2009-05-20 babl-0.1.0 </dt><dd>
-Support for RGBA representation in the frequency domain,
-un-pre-multiply close-to-zero alpha values to complete black instead
-of slightly brighter than black, add a BABL_ALPHA_THRESHOLD constant
-to the API, do a complete overhaul of the babl API and do some
-changes to increase portability of the library.
- </dd><dt>
-2008-06-12 babl-0.0.22 </dt><dd>
-Speed improvements due to: coalesced hashing, early bail out upon creation of
-duplicate formats, caching of non existing paths. Large amounts of the core has
-also been reviewed and refactored. Improved conversion coverage in the matrix.
- </dd><dt>
-2008-02-27 babl-0.0.20 </dt><dd>
-Builds on OSX
-Check <0.0 and >1.0 values conversions for accuracy as well.
-
- </dd></dl>
-
-<!--
- </dd><dt>
-2007-12-30 babl-0.0.18 </dt><dd>
-Constified use of babl formats.
-
- </dd><dt>
-2007-11-25 babl-0.0.16 </dt><dd>
-Reduced constant overhead for linear conversions.
-Misc build environment cleanups.
-
- </dd><dt>
-2007-01-07 babl-0.0.14 </dt><dd>
-Cache values parsed from BABL_foo environment variables.
- </dd><dt>
-2007-01-07 babl-0.0.12 </dt><dd>
-Speedups for GEGL
-Build environment fixes.
- </dd><dt>
-2006-xx-xx babl-0.0.10 </dt><dd>
-Build environment fixes.
-
- </dd><dt>
-2006-12-29 babl-0.0.8 </dt><dd>
-Made code adhere more strictly to C standard.
-Set the BABL_PATH based on build prefix
-
- </dd><dt>
-2006-10-09 babl-0.0.6 </dt><dd>
-Build environment fixes.
-Fixed timing code for picking correct conversions.
-
- </dd><dt>
-2006-08-31 babl-0.0.4 </dt><dd>
-Build environment fixes
-Reduced used/exposed API.
-Added GEGL speedup extension.
- </dd><dt>
-2005-0x-xx babl-0.0.2 <em>Green fish release</em> </dt><dd>
-
-Initial release.
-
- </dd><dt>
-2005-08-xx babl-0.0.1 </dt><dd>
-
-Development release of babl, minimal working configuration for
-RGB, CMYK and YCbCr for 8bit 16bit float and double. Emphasis is on
-precision and clarity of code not speed.
-
- </dd></dl>
-
--->
[
Date Prev][
Date Next] [
Thread Prev][
Thread Next]
[
Thread Index]
[
Date Index]
[
Author Index]
