Linux is not ready for Steam

[blacknova]

As I've written my bottom-line is OSS. It is highly portable and the only *nix so far which declined OSS usage and developed absolutely new API is Linux. And it have taken many years for developers to change to ALSA and not completely at that as most of them considered using ESD, aRTs, NAS,... but not to use ALSA directly.
That is for low-level API.

[darkphoenix22]

As I've written my bottom-line is OSS. It is highly portable and the only *nix so far which declined OSS usage and developed absolutely new API is Linux. And it have taken many years for developers to change to ALSA and not completely at that as most of them considered using ESD, aRTs, NAS,... but not to use ALSA directly.
That is for low-level API.

Too bad the community doesn't want to go back to OSS.

So fixing ALSA will have to do.

[kazetsukai]

As I've written my bottom-line is OSS. It is highly portable and the only *nix so far which declined OSS usage and developed absolutely new API is Linux.

I wonder what your take would be on this if ALSA was 'AOSA' (Advanced Open Sound Architecture) instead, and could run on other platforms.

I think the best case scenario is a fork of ALSA into an 'AOSA' like project to achieve the multiplatform compliance while applications default to OpenAL for audio output. That way application developers wont be screwed by standards that come and go as they please.

No matter how I look at it, OSS is legacy on Linux. Noone wants to backtrack to an old standard just because it opened back up on a whim.

[blacknova]

I wonder what your take would be on this if ALSA was 'AOSA' (Advanced Open Sound Architecture) instead, and could run on other platforms.

I think the best case scenario is a fork of ALSA into an 'AOSA' like project to achieve the multiplatform compliance while applications default to OpenAL for audio output. That way application developers wont be screwed by standards that come and go as they please.

No matter how I look at it, OSS is legacy on Linux. Noone wants to backtrack to an old standard just because it opened back up on a whim.

I doubt that ALSA's API is lot more complex than OSS's.
And I also doubt anyone would want to fork it in order to port to other platforms *BSDs, Solaris, etc... all of these system already have working sound systems. Working without much problems. Why would they want to reinvent the wheel?

[drag]

I doubt that ALSA's API is lot more complex than OSS's.
And I also doubt anyone would want to fork it in order to port to other platforms *BSDs, Solaris, etc... all of these system already have working sound systems. Working without much problems. Why would they want to reinvent the wheel?

Yeah.

Remember that the reason why Alsa was created in the first place on Linux was because OSS had gotten completely obsolete and people wanted to use a audio card that OSS (at the time) could not possibly support.

It was the 'Gravis Ultrasound'.

Remember also that newer versions of OSS had gone completely proprietary and you had to pay money to get access to newer Linux audio drivers coming from them.


If OSS never went proprietary then we would not be having this discusion...

However we would of still need a sound server like Pulse Audio irregardless.

[deanjo]

It was the 'Gravis Ultrasound'.

You just had to go and bring up all those painful memories didn't you?

RIP GUS.

[drag]

Even if you get everyone on Linux running PulseAudio (which I don't believe), you won't have everyone on BSD, Solaris, and a number of other platforms. All libraries worth anything will have a Win backend, an OSX backend, an ALSA backend and an OSS backend. Now you're simply adding PulseAudio as yet another target.

All the good ones will be running PA. The ones that won't are simply going to be irrelevant.


The problem is not so much that 'Alsa sucks and lets use PA to fix it'. That's just stupid.

It's just that the _ARCHITECTURE_ sucks. Having a low-level driver + Application libraries is completely and totally insufficient solution to the problem of audio on the desktop.

You CANNOT solve the problems that way. It's just not going to work and has never worked and is never going to work.

Get it?

People complaning about 'Pulse Audio' sucks and thinking that replacing it with OSS is a solution are not really understanding the issues going on here.

It's like taking a car with no wheels, no brakes, no body, no seats, and saying 'Hey, If I just swap out the motor, that will fix it!'

And then you have a bunch of people arguing over whether they should use a gasoline motor or a diesel motor.

No. You need a steering mechanism. You need controls and ways to sync audio, manage audio, deal with network issues, volume controls, and so on and so forth. You need a centralized way to manage it and standard ways for applications and users to access it.

Giving applications read/write access to some character devices in /dev and giving them the ability to do it simultaneously is a non-solution.

If you think that GTK and Qt are going to drop ALSA and OSS, you're insane.

GTK and QT are widget libraries. They don't have anything to do with audio.

Gnome-based desktops are all largely switched over to PA.

[pingufunkybeat]

You just had to go and bring up all those painful memories didn't you?

RIP GUS.

Gravis Ultrasound had the best support from the early 1990s demo scene because Gravis would send them free cards.

It was the best marketing ever, all the coolest demos had support for the card, and it sounded awesome.

[pingufunkybeat]

All the good ones will be running PA. The ones that won't are simply going to be irrelevant.

Yeah, Slack, Gentoo and Arch are going to become irrelevant. I've heard that one before...

GTK and QT are widget libraries. They don't have anything to do with audio.

Try again.

http://labs.trolltech.com/blogs/2009/09/03/multimedia/

[deanjo]

Gravis Ultrasound had the best support from the early 1990s demo scene because Gravis would send them free cards.

It was the best marketing ever, all the coolest demos had support for the card, and it sounded awesome.

Absolutely, feature wise as well it was waaay ahead of it's time. Too bad their windows drivers sucked ass and that is what killed it in the long run.