Arch Linux

ckristi

Member

From: Bucharest, Romania

Registered: 2006-11-21

Posts: 225

Re: Time to get rid of XMMS?

I see this is a kinda old thread but I wanted to post my XMMS vote here. XMMS is at the moment the most responsive player I've used. I like Audacious but besides the fact that is slow I had problems with some audio files (even mp3s) that XMMS played and Audacious only sounded very strangely.. quirks and very odd sounds that almost blew my ears. :-(

Long live XMMS!

P.S.: I wonder why nobody wants to take this wonderful project and make a good and maintained GTK2/QT application out of it.

---

In love I believe and in Linux I trust

iphitus

Forum Fellow

From: Melbourne, Australia

Registered: 2004-10-09

Posts: 4,927

Re: Time to get rid of XMMS?

I see this is a kinda old thread but I wanted to post my XMMS vote here. XMMS is at the moment the most responsive player I've used. I like Audacious but besides the fact that is slow I had problems with some audio files (even mp3s) that XMMS played and Audacious only sounded very strangely.. quirks and very odd sounds that almost blew my ears. :-(

Long live XMMS!

P.S.: I wonder why nobody wants to take this wonderful project and make a good and maintained GTK2/QT application out of it.

they have. see older bmp, audacious, and im sure there's another app. Audacious is an example of what happens. XMMS's codebase isnt really the nicest thing from what i've gathered, hence the problems both conversions suffered.

James

predatorfreak

Member

Registered: 2006-12-08

Posts: 13

Re: Time to get rid of XMMS?

I see this is a kinda old thread but I wanted to post my XMMS vote here. XMMS is at the moment the most responsive player I've used. I like Audacious but besides the fact that is slow I had problems with some audio files (even mp3s) that XMMS played and Audacious only sounded very strangely.. quirks and very odd sounds that almost blew my ears. :-(

Long live XMMS!

P.S.: I wonder why nobody wants to take this wonderful project and make a good and maintained GTK2/QT application out of it.

they have. see older bmp, audacious, and im sure there's another app. Audacious is an example of what happens. XMMS's codebase isnt really the nicest thing from what i've gathered, hence the problems both conversions suffered.

James

Another problem that I think might affect this is that Audacious uses forward ported plugins from XMMS, most based on VERY old code (for example, the MP3 plugin is based on mpg123) and have seen very little serious work done to either replace them with more modern equivalents (say a MAD based MP3 plugin) or fix issues in the code. The code could probably do with some optimisation here and there as well.

iphitus

Forum Fellow

From: Melbourne, Australia

Registered: 2004-10-09

Posts: 4,927

Re: Time to get rid of XMMS?

Another problem that I think might affect this is that Audacious uses forward ported plugins from XMMS, most based on VERY old code (for example, the MP3 plugin is based on mpg123) and have seen very little serious work done to either replace them with more modern equivalents (say a MAD based MP3 plugin) or fix issues in the code. The code could probably do with some optimisation here and there as well.

Retrofitting new stuff into old things isnt all too easy either, so it's problematic either way, as seen by the various problems shown by the GTK2 versions of XMMS.

It seems that attempts to do "some optimisation", nowadays results in rewriting a large amount of the codebase.

James

predatorfreak

Member

Registered: 2006-12-08

Posts: 13

Re: Time to get rid of XMMS?

Another problem that I think might affect this is that Audacious uses forward ported plugins from XMMS, most based on VERY old code (for example, the MP3 plugin is based on mpg123) and have seen very little serious work done to either replace them with more modern equivalents (say a MAD based MP3 plugin) or fix issues in the code. The code could probably do with some optimisation here and there as well.

Retrofitting new stuff into old things isnt all too easy either, so it's problematic either way, as seen by the various problems shown by the GTK2 versions of XMMS.

It seems that attempts to do "some optimisation", nowadays results in rewriting a large amount of the codebase.

James

True, people are getting too invasive, when I say "some optimisation", I only mean some, pinpointing serious bottlenecks and improving those areas to make the code perform better.

I do think that realistically, we do need a good universal backend for video/audio. I can't fully support gstreamer because it's horrible on postprocessing and scaling, yet I can't fully support xine as a backend because it's not very easy to plug things into it. Ideally, we need a cross between xine's features and gstreamer's extensibility.

Anyway, I'll stop rambling.