Help - Search - Members - Calendar
Full Version: Sound Quality In Linux
Linuxhelp > Support > Technical Support
Joey
Has anyone noticed a difference in sound quality in Linux as apposed to playing the same mp3 in windows?

Everything sounds fine in windows but if I use XMMS with RedHat 9 it sounds like shit. I just tried playing the same mp3 with mpg123 in console and the sound quality is similar to windows. Maybe this is related to how X handles sound?

Any ideas or am I just crazy? smile.gif
LinuxHatesMe
I would think that it would be relevant to the way the program itself initializes and uses sound. I've played the same mp3 in different mp3 players on windows and I noticed a sound quality difference. While it was slight and not a dramatic difference, this fact leads me to believe what I previously stated about programs handling sound quality. I could be wrong. I can't use sound in linux because my lovely sound card manufacturer doesn't seam to support it at this current time. So my statement IS inconclusive and is obviously opinionated due to this fact. smile.gif
Corey
Overall, the sound quality in Linux is less then satisfactory. This is due to many reasons. First off, Open Sound System, the current standard as of the 2.4.x kernel series cannot handel multiple input to the sound card. For example, you cannot play Quake3 and listen to mp3's at the same time. What happens is that a program like mpg123 will lock /dev/dsp and not let anything else touch it. In X, different desktop environments use different "Sound Servers" which take all incomming sound, mixes it into one and streams it to /dev/dsp. KDE uses aRts, Gnome uses Esound, etc. While using a sound server sounds (pardon the pun) like the optimal idea, in exchange you give up sound quailty and lag. For example, in order to play Quake3 while using aRts, you have to use artsdsp which gives you a delay in sound occurance and when you actually hear it. While a program like XMMS, this wouldn't matter, in games, this is of great annoyance.

As of the 2.5.x development kernel tree, and eventually the 2.6.x stable kernel tree, OSS will be replaced with ALSA, which will (hopefully) give greater results. In the meantime, you can install ALSA for your current sound system if desired.

So, I would conclude that when you use mpg123 in the console and play your mp3, it pipes it directly to your /dev/dsp with no delay and no degregation in quality. While, if you're using KDE or Gnome, the sound gets passed to the sound server for processing causin ga delay and degregation in quality (slightly). XFree86 is currently developing a sound server built specifically for X-windows that may or may not be a better solution to current sound servers.

In the meantime, do what I do, disable all sound servers associated with your Window Manager, and deal with the fact that you only want to hear one sound at one time.

I, by no means, have researched any of the above information, this has been from personal experiences.
This is a "lo-fi" version of our main content. To view the full version with more information, formatting and images, please click here.
Invision Power Board © 2001-2017 Invision Power Services, Inc.