log postproc/swscale.c @ 5908:31159f453cf9

age author description
Mon, 01 Apr 2002 14:01:22 +0000 michael overread in the mmx2 horizontal scaler fixed
Thu, 21 Mar 2002 18:53:32 +0000 michael fixing bgr15&16 on big_endian
Fri, 22 Feb 2002 15:24:19 +0000 michael use assert.h
Thu, 21 Feb 2002 19:41:50 +0000 michael 101++
Thu, 21 Feb 2002 17:48:00 +0000 michael fixing bgr32 output on big-endian systems
Thu, 21 Feb 2002 16:57:35 +0000 michael unscaled yuv2rgb fix for big-endian systems
Sun, 10 Feb 2002 23:57:47 +0000 michael unscaled memcpy variant bugfix (negative strides)
Sun, 10 Feb 2002 00:43:31 +0000 michael unscaled BGR24 -> YV12
Sat, 09 Feb 2002 23:43:48 +0000 michael bgr24toUV in MMX
Sat, 09 Feb 2002 17:03:53 +0000 michael bgr24toY in MMX
Fri, 08 Feb 2002 15:54:56 +0000 michael fastmemcpy
Fri, 08 Feb 2002 04:18:01 +0000 michael unscaled BGR15->BGR16
Fri, 08 Feb 2002 03:09:46 +0000 michael bgr15 input support
Fri, 08 Feb 2002 00:42:00 +0000 michael bgr16 input support
Thu, 07 Feb 2002 15:29:15 +0000 michael unscaled converters
Thu, 07 Feb 2002 03:39:30 +0000 michael fixing green line at right side bug (reported by Nick Kurshev <nickols_k@mail.ru>)
Wed, 06 Feb 2002 22:55:57 +0000 michael rgb32 & rgb24 input support
Wed, 06 Feb 2002 21:59:23 +0000 michael 101 (yuv2rgb with mpeg1,...) ;)
Wed, 06 Feb 2002 20:52:14 +0000 michael swScale internally uses yuv2rgb now if possible
Mon, 04 Feb 2002 18:53:01 +0000 michael printing error messages if something is wrong instead of just return NULL;
Sun, 03 Feb 2002 02:08:05 +0000 michael filterPos bugfix (only nearest neighbor scaling was affected)
Sat, 02 Feb 2002 18:18:58 +0000 michael MMX(2) optimized YUY2 input
Fri, 01 Feb 2002 19:39:37 +0000 michael printing correct format info
Fri, 01 Feb 2002 19:25:09 +0000 michael YUY2, BGR24, BGR32 input support (no mmx yet)
Wed, 30 Jan 2002 15:32:43 +0000 michael -sws 2 is default now
Wed, 30 Jan 2002 15:21:08 +0000 michael YV12 like formats support (I420/IYUV)
Tue, 29 Jan 2002 01:43:15 +0000 michael area averageing scaling support (-sws 5) (is identical to bilinear for upscale)
Mon, 28 Jan 2002 23:57:18 +0000 michael nearest neighbor / sdl emulation ;) scaling (-sws 4)
Tue, 22 Jan 2002 15:54:44 +0000 michael minor bugfix in the -v output
Mon, 21 Jan 2002 22:11:09 +0000 michael 15/16 bit dithering in C (5% slower, can be disabled by comenting #define DITHER1XBPP out)
Mon, 21 Jan 2002 18:32:31 +0000 michael copyright(swscaler) = GPL
Mon, 21 Jan 2002 15:22:28 +0000 michael a few filters (should be removed/merged when arpis videofilter stuff is finished)
Mon, 21 Jan 2002 03:56:47 +0000 michael general convolution filtering of the source picture
Sun, 20 Jan 2002 13:45:19 +0000 michael 101
Sun, 20 Jan 2002 05:30:23 +0000 michael removed global vars so that multiple swscalers can be used
Sat, 19 Jan 2002 05:14:47 +0000 atmos4 mangle for win32 in postproc
Fri, 21 Dec 2001 02:09:34 +0000 michael minor bugfixes (noone noticed them)
Thu, 06 Dec 2001 19:07:25 +0000 michael print more info if -v
Thu, 06 Dec 2001 00:10:42 +0000 michael vertical cubic/linear scaling
Sun, 02 Dec 2001 20:03:26 +0000 michael horizontal up/downscale linear & cubic
Thu, 29 Nov 2001 20:19:17 +0000 michael swscaler cleanup
Tue, 27 Nov 2001 01:19:56 +0000 michael runtime cpu detection optional (compiles faster)
Mon, 26 Nov 2001 02:20:51 +0000 michael oops
Mon, 26 Nov 2001 00:31:43 +0000 michael runtime cpu detection
Sat, 10 Nov 2001 20:39:23 +0000 michael .balign or weĦ­ll align by 64kb on some architectures
Sat, 10 Nov 2001 19:46:04 +0000 michael .align 16
Tue, 06 Nov 2001 19:26:22 +0000 michael gcc does optimize writes to non volatile variables away if it didnt know that they were read in between
Tue, 06 Nov 2001 18:42:22 +0000 michael better dithering
Tue, 06 Nov 2001 04:06:55 +0000 michael faster bgr24 output
Mon, 05 Nov 2001 23:49:15 +0000 michael only 6 registers used
Sun, 04 Nov 2001 13:50:20 +0000 michael green line fix for dstw%8!=0
Sun, 04 Nov 2001 03:07:02 +0000 michael fixed warnings
Sun, 04 Nov 2001 02:28:23 +0000 michael faster bgr15/16
Fri, 02 Nov 2001 19:21:02 +0000 michael right green line bugfix for width not %8==0 (untested -vo vesa doesnt work)
Wed, 31 Oct 2001 01:30:28 +0000 michael c speedup
Wed, 31 Oct 2001 00:26:29 +0000 michael 15/16bit in C speedup
Tue, 30 Oct 2001 22:24:38 +0000 michael c optimizations
Tue, 30 Oct 2001 22:03:25 +0000 michael c optimizations
Tue, 30 Oct 2001 21:33:01 +0000 michael c optimizations (array is faster than pointer) (16bpp variants tested and 2% faster)
Tue, 30 Oct 2001 21:04:42 +0000 michael interpolate chrominance for every second line in the 1:1 vertical scale function