February 2005 - MPlayer-dev-eng - lists.mplayerhq.hu

[PATCH] makes --enable-*-faad behave correctly
by Aurelien Jacobs 03 Feb '05

03 Feb '05

Hi, Currently when using --enable-*-faad, configure still try to compile and execute some code to verify if the right version of faad is available. This patch makes enable really mean enable without any test, to be consistent (and also to ease cross-compilation). To simplify things, it also drop support for old version of faad2 (which had an incompatible API). Those old version of faad2 are outdated since more than a year and a half. So I personally think that it is safe to drop this old version support but I would like to know what others thinks about that ? Anyway, as we now have an internal version of faad2 people with a very old version installed can still use the internal one. Aurel

3 6

Dithering 16-bit output for libmad
by Vladimir Mosgalin 03 Feb '05

03 Feb '05

Since a lot of people use mplayer as mp3 player (personally, xmms crashed a lot lately, I disliked UI in other gui players, and console ones were lacking good controls), I made a patch that adds dithered 16-bit output for libmad decoder, to increase mp3 playback quality. It works because output from libmad is 24-bit, and producing dithered 16-bit samples out if gives more quality results than rounding to 16 bits. The code is ported from madplay. Of course, if you don't have mp3s with high bitrates and good sound hardware, probably you won't be able to notice the difference. It's easier to notice it with headphones. Some more steps are required in order to get higher quality sound: of course, you need to use libmad decoder (afm = libmad in config), and if you have sound card from Creative or some other card that doesn't natively support 44.1khz sound, you need to use software resampling to 48k, as hardware one is lame. In mplayer, af = lavcresample=48000:32:1:12 works good. Some options may be returned to default or linear interpolation turned off, but not both. At first I wanted to add ability to switch back to rounding instead of dithering, but I can't find out how to pass suboptions to audio decoders... Anyway, there is only one drawback: because of dithering, the output now can give worse analytic results. For example, it is less likely that auCDtect will detect dithered output as mp3 file. Since there are only few integer operations, the performance drop is very small, less than 0.1% of total cpu usage. It is about 10-50 times less than performance penalty from using libmad instead of mp3lib or from using lavcresample. -- Vladimir

6 10

[PATCH] test if X11 libs can be used for linking when autodetecting
by Reimar D�ffinger 03 Feb '05

03 Feb '05

Hi, the attached patch extends configure to not only check which lib dirs exist but also which contains libs that can be used for linking. See also http://bugzilla.mplayerhq.hu/show_bug.cgi?id=206 Does anything speak against this modification or can I apply? Greetings, Reimar D?ffinger

2 2

Re: Postproc Hell
by Andrea Palmatè 03 Feb '05

03 Feb '05

>The later fsf-gcc at least on linux seems to produce correct altivec >code, given you code it the right way (that could be not necessary the >apple-gcc right way) > >Not really sure if the AmigaOS4 problem is related to the gcc (seems >not..) or other. Maybe. It is strange. Surely i have resolved some problem updating the las libav but more problems with altivec remains. First of all wrong colours on postproc that derive by yuv2rgb. if i use the pre4 the colours are correct. this could derive by yuv2rgb_altivec_template.c but i don't know. Now all is played correctly with postproc. Other problems are on libavcodec when compiled with altivec. I have a crash on a function (i don't remember the name and i haven't the sources on hand). With liba52 all is compiled correctly but the sound is broken when play DVD. Remember. I don't think that is mplayer that goes wrong but now i have no idea of what go wrong. I update my cvs every single day. so maybe with future versions the problems disappear. Andrea

2 1

check for XFree86-VidModeExtension broken for xfree 3.3.x
by Trent Piepho 02 Feb '05

02 Feb '05

The XFree86-VidModeExtension extension isn't detected, when it used to be in older versions of mplayer. This is problematic, as xfree 3.3.x doesn't have DGA2 to switch modes. This patch fixes it.

2 5

tremor and libvorbis
by compn 02 Feb '05

02 Feb '05

there is a small problem with internal tremor in mplayer, it is named libvorbis. Opening audio decoder: [libvorbis] Ogg/Vorbis audio decoder AUDIO: 44100 Hz, 2 ch, 16 bit (0x9), ratio: 20000->176400 (160.0 kbit) Selected audio codec: [vorbis] afm:libvorbis (OggVorbis Audio Decoder) this also means that at compile time, one cannot have both libvorbis and tremor support at the same time, nor can he call -ac libvorbis and -ac tremor to test between the two. i would send a patch if i knew where to start, this is not a priority problem. the codecs.conf has to be changed or added to at least. probably add two codecs tremor and tremorlow, is it possible to tell the difference between the two from codecs.conf?

1 0

Postproc Hell..
by Andrea Palmatè 02 Feb '05

02 Feb '05

Hi, i have many problems with postproc. Actually only if I use the 1.0pre4 version it works, with newer versions the colors are wrong and there are many problems with avi files (DIVX, XVID and so on). If I compile the altivec versions the problems remains. What are the main differences between 1.0pre4 and 1.0pre6? Bye, Andrea Palmatè

6 7

Postproc Hell
by Andrea Palmatè 01 Feb '05

01 Feb '05

Damn.. the problem was the libavcodec!! I've downloaded the las ftom cvs and all problems disappear! (except for one but i'll test it tonight..) Sorry for all.. Another thing.. if af_tools.h (or c i don't remember well) it is possible to change the #include <af.h> with #include "af.h" bye

2 1

win32 loader exception handler
by Gianluigi Tiesi 01 Feb '05

01 Feb '05

I'm working on avisynth on linux, but I have two problems, corrupted stack when I call some functions and win32 misses exception handler. It can be merged from wine but it's not so easy, I'm attaching a preliminar patch to win32, also my avs patch is needed, if someone would to be involved I can work on this, but I'm not very familiar with exceptions, also I known it's a bit difficult but if someone can help me to setup the exception handler I can offer my time to make it works. Regards -- Gianluigi Tiesi <sherpya(a)netfarm.it> EDP Project Leader Netfarm S.r.l. - http://www.netfarm.it/ Free Software: http://oss.netfarm.it/

1 0