Announcement

Collapse
No announcement yet.

Google Chrome/Chromium Now Supports PulseAudio

Collapse
X
 
  • Filter
  • Time
  • Show
Clear All
new posts

  • Google Chrome/Chromium Now Supports PulseAudio

    Phoronix: Google Chrome/Chromium Now Supports PulseAudio

    Making news in the browser world this morning is word of Firefox 7 Beta. Firefox 7 has optimized memory usage, improved memory management, enhances Firefox Sync, and other enhancements. But there's also some other interesting news in the browser world for the more technical users and it's concerning Google Chrome/Chromium...

    Phoronix, Linux Hardware Reviews, Linux hardware benchmarks, Linux server benchmarks, Linux benchmarking, Desktop Linux, Linux performance, Open Source graphics, Linux How To, Ubuntu benchmarks, Ubuntu hardware, Phoronix Test Suite

  • #2
    What is wrong with Chrome not using PulseAudio? I hear sound fine from the browser, or is PA going to make it sound better? Use less CPU?

    Comment


    • #3
      Originally posted by numasan View Post
      What is wrong with Chrome not using PulseAudio? I hear sound fine from the browser, or is PA going to make it sound better? Use less CPU?
      What's wrong with it supporting it? Besides irrational dislike for the technology?

      Comment


      • #4
        Originally posted by loonyphoenix View Post
        What's wrong with it supporting it? Besides irrational dislike for the technology?
        Probably a fear that developers will drop the ALSA backend or stop adding to it. It happened with OSS, so it's not a stretch of the imagination.

        Comment


        • #5
          Well it probably can help when it comes to audio management.

          With PA I notice that apps connect and stay connected. In the audio management stuff they have nice names and stick around. That way you can set audio levels for them and have is persistant.

          With non-PA applications, especially browsers, I notice that when a sound event happens or something starts playing back then the connections come and goes.

          From a application standpoint I think that PA is simpler interface to do things 'right' and it can help with performance.

          Comment


          • #6
            Well, as long as it stays out of my way, I don't mind PA. My question was based on this:

            It's pleasant to see the support of PulseAudio finally, which will hopefully lead to a better web-browser audio experience for most modern Linux distributions that use this open-source audio library, compared to Chrome talking directly to ALSA.
            I'm just wondering how the experience will be better?

            Edit:
            Thanks drag. I haven't noticed problems with audio in the browser before, but maybe others have?
            Last edited by numasan; 19 August 2011, 02:40 PM.

            Comment


            • #7
              Originally posted by DanL View Post
              Probably a fear that developers will drop the ALSA backend or stop adding to it. It happened with OSS, so it's not a stretch of the imagination.
              There is a active movement to get rid of OSS support in Linux. The reason, of course, is that OSS is terrible.

              I wouldn't mind seeing Alsa go away, of course, for many applications.

              I noticed quite a bit with applications that are Alsa-only that they do a lot of unnecessary junk. Stuff like trying to probe for audio devices and stuff like that. I had audacity break bluetooth audio because it was trying to detect things, but doing it poorly and trying to configure stuff. I had a similar thing happen with Ekiga. Every time I use it it jacks up my microphone levels or breaks something stupid to were I have to drop back into the command line and run 'alsamixer' to fix what ever stupid thing it did.

              In a Alsa-only world it is understandable a bit. Setting up microphones and such with Alsa is such a pain in the ass for end users that these applications try to make it 'easy' by detecting and auto-configuring as much as possible. This, of course, is full of fail. The reason is that they have no way of knowing what other applications are doing and what else is going on in the system.

              With PA management of inputs and outputs is all centralized and you don't have to muck around with settings in each application. They apps don't have to care about what devices you have or how to setup them up... PA does all the work and the users will have a consistent way to get stuff working for all applications.

              Comment


              • #8
                Originally posted by numasan View Post
                Well, as long as it stays out of my way, I don't mind PA. My question was based on this:



                I'm just wondering how the experience will be better?

                Edit:
                Thanks drag. I haven't noticed problems with audio in the browser before, but maybe others have?
                If you want to use VoIP applications especially. Especially x 2 if your using bluetooth or USB headphones or something.

                Getting stuff like that working properly in Linux without PA is a nightmare. With PA it is far easier then any other OS, even OS X.

                Comment


                • #9
                  Originally posted by drag View Post
                  From a application standpoint I think that PA is simpler interface to do things 'right' and it can help with performance.
                  PA surely could be a simpler interface for applications, but can't help with performance since it is an added abstraction layer.
                  PA source code (expecially resamplers) are filled with notices about poor performance because they are straight implementations of filters without particular optimizations in place. Not to mention they don't use SIMDs at all.

                  Also I noticed PA has the bad behaviour to resample even those streams that have the same properties (frequency, depth) of the output stream.

                  Comment


                  • #10
                    Originally posted by blackshard View Post
                    PA surely could be a simpler interface for applications, but can't help with performance since it is an added abstraction layer.
                    No. It's not a abstraction layer. It's a sound server.

                    And yes it can actually improve performance. Everything PA does is needed to be done. There is no free lunch here. You need to perform sound mixing somewhere, you need to have accurate timings. You need something to manage your sound devices, etc etc. You can do it through a combination of application logic and driver logic, or you can separate that stuff off and handle it centrally by a program that is designed specifically to solve these problems.

                    (meaning it can improve performance because it does a better job)

                    PA source code (expecially resamplers) are filled with notices about poor performance because they are straight implementations of filters without particular optimizations in place. Not to mention they don't use SIMDs at all.
                    Well maybe that is something you can fix if you think that SIMD can help with those filters.

                    Anyways it only performs remixing if it has to.

                    Also I noticed PA has the bad behaviour to resample even those streams that have the same properties (frequency, depth) of the output stream.
                    It's not suppose to do that.

                    Got a example?
                    Last edited by drag; 19 August 2011, 06:17 PM.

                    Comment

                    Working...
                    X