Language Selection

English French German Italian Portuguese Spanish

Should Pulseaudio Die? What do you think?

Filed under
Linux

When I was installing and testing Mandriva 2010 Cooker (new development) releases this last Fall season, I kept having persistent problems with sound. Eventually, the advice in the Mandriva Cooker forum for KDE users became: "Disable Pulseaudio, and set Xine as the preferred back end over GStreamer (in the KDE multimedia settings).

Unlike the latest 'buntu, Mandriva does supply a checkbox in their Control Center Sound configuration utility to enable/disable Pulseaudio.

Then Mandriva 2010 final came out, and Pulseaudio more or less works, (and GStreamer works fine for the KDE back end). Some users report that flash videos have no sound with Pulseaudio enabled. Certainly there's some sound lag with Pulseaudio.

Today I was listening to the latest tllts podcast (The Linux Link Tech Show). Tllts member Patrick Davila started his "rant" segment, and said:

Quote:
Pulseaudio sucks. The experiment [with Pulseaudio] has gone far too long. It's time to admit that it's a f*ing failure and every implementation sucks. The guy who runs it [Pulseaudio]--any time anybody tries to talk to him, is very negative and tells you you're an idiot. Canonical should hire Paul Davis, the developer of Jack and Ardour and Jack should become the default sound daemon."

What do you think? I tend to agree with Davila, however Linux sound is not my area of expertise. But I've certainly been plagued and frustrated over the years with Linux sound issues, and Pulseaudio seems to make things worse.

Comment viewing options

Select your preferred way to display the comments and click "Save settings" to activate your changes.

I use Sidux AMD64 with KDE

I use Sidux AMD64 with KDE 4.3 and I have a proper sound card - an M-Audio Revolution 5.1. I do not have any Pulseaudio installed. My sound works perfectly. In part this is because my sound card supports hardware mixing. KDE uses Phonon not Pulse.

Now, this week I built a second machine from old bits for the purpose of video editing. I wanted to try OpenShot so I installed Karmic Koala, as OpenShot requires Python 2.6 and so it will not run on anything Debian (yet). After installation I had no sound at all. The Pulseaudio controlled sound applet showed everything was working, but nothing emanated from the speakers. I installed Gnome Alsamixer and found that the output was not selected, a simple click of the mouse and music poured forth. To me the biggest fault here is Gnome using this new mixer applet instead of the traditional Alsa one. Ubuntu, or Gnome in general, have also removed the older Gnome Sound properties dialog with a Pulse based one. You can no longer change the default output easily. If it does not work there is now nowhere to go unless you are an experienced Linux user. I removed Pulseaudio as I had other issues with input from Aux (but that one was an ALSA issue), but in doing so lost the speaker applet. I disabled the onboard sound and put in a PCI sound card and reinstalled Pulseaudio and now it all works.

Is this the fault of Pulseaudo, Gnome or Ubuntu - or all of them. When Pulseaudio works it is a good tool, when it does not it is an utter pain. Unfortunately it seems to fail way too often for my liking. The failures are often simple and can be overcome with simple access to good Alsa based mixer. This then is the fault of Gnome and the distro builders. Bring back the old sound properties dialog and install Gnome Alsamixer by default. Give the noobs a fighting chance.

My last peeve is definitley distro based. Why is my sound constantly muted by updates? Not Sidux - the others. Why on earth would you set mute as the default? Ever?

enough ranting

It is junk.

I agree too. Pulse audio is junk just like udev, hal, dbus and polkit. Useless crap that don't work right.

Depends on the distro

I have an older Creative card that uses the snd-emu* drivers. It and PA (PulseAudio) work fine with certain distros, like Mandriva 2010 and Ubuntu 9.10. It and PA don't work well with others, like Fedora 12 and openSUSE 11.2. In those distros, system sounds come out layered in static. Flash and other audio playback can sometimes be staticky as well. Why one distro gives me no problems while another does, I have no idea.

The reaction of the PulseAudio dev is as follows: "Regarding snd-emu*: Creative doesn't like Open Source -- there are no docs available. If you buy Creative it is hence a bit your own fault."

Fair enough, except the card works fine with ALSA. Why the difference? The PulseAudio dev again: "As long as you only play MP3 music directly on the device and use the traditional sound card interrupt based wakeups the bogus timing information the sound drivers export doesn't matter. That's why you won't notice this issue without PA but it becomes very visible (or audible...) when PA is used."

Whoopee.

PulseAudio can do all kinds of cool things, but I'd just like to have an alternative to using it. I've tried but I can't replace it with any other backend in openSUSE 11.2 and have working sound, even in KDE 3/4.

re: depends distro

Yeah, I had trouble with a creative card and opensuse 11.2 too. Sound would only come out of one of the four speakers. That didn't go over too well with me.

I haven't had too much trouble with most of others distros except that Flash sound wouldn't work in the latest Mandriva. The only solution really offered was to disable pulseaudio. That did indeed do the trick.

Distros' fault

Pay attention to what PulseAudio's dev has said many times. Ubuntu in particular, he argues, does a HUGE disservice to the reputation of PulseAudio.

re: Distros fault

Yes, Linux Excuses 101.

Whenever something sucks, blame everyone (the distro, the user, the girlscouts) but the monkey who codes it.

its every distro?

I've only read the horror stories of PA but I've no experience myself. However, its constantly repeated that it is the distros' fault in their implementations. Perhaps the developer needs to provide more-clear examples or documentation so that his product doesn't get such a bad reputation?

On the flip side, which distros are implementing this correctly according to the PA developer(s)?

PulseAudio...

I've been using Linux on and off for the past four years and since my first stab at it, everyone always suggest that I stay away from PulseAudio. Because of that, I've always stuck with Alsa. It works and it works great. I'm personally a fan of building minimal Debian systems and I've always relied Alsa for sound. Most of it is because I've heard nothing but bad stories about PulseAudio. Lots of promise but just as much letdown. Does this project really lack that much direction/ it clearly isn't moving forward if this debate is still going on four years later.

yep

Pulse audio is garbage. I actually do blame the distros a little bit. If the crap isn't tested and working on common hardware, why put it in a so called stable release.The developer can test it all he wants, can ask for testers, or keep it on his hard drive for all I care. When the major distros release new versions, something as basic as sound should work. Period. I suppose the community working together on improving what we already have working would be too much to ask.

PulseAudio

I'm going to be murdered for posting this, but I think we should all go back to OSS, with version 4 and later version 5. I'm on Arch, so I can pick and choose from the beginning what I want to use. I have tried PulseAudio, which seems to lag, ALSA works but it sounds weak, compared to sound in OSS4 and Windows 7. I dare say it sounds better in OSS4 than Windows 7. I even have a Creative card and it works better in OSS than the rest with BETA drivers. Granted I don't do any recording or anything complicated that doesn't involve listening to FLAC audio, movies and youtube, with a 5.1 surround system. Adding to this, OSS is so easy to configure, everything residing in /usr/lib/oss and while the wiki is not updated as often, it has the usual problems that I encountered listed and in the forums, there are friendly people who have helped me, up to the extent of modifying source code and recompiling OSS on Arch in order to get a USB sound card working. I know what the developer did in the past still makes some people angry or distrustful but I don't think that's going to happen again, everyone can make mistakes and I think if we show more support with development and testing, OSS can be the greatest thing for sound in Linux and all of *nix systems. Freebsd uses it, maybe a modified version but to me OSS is both user and developer friendly.

I have not tried JACK, but if that works good, I don't have any complaints. I just want to move to something that is not ALSA or Pulseaudio and more importantly, that it works as intended.

http://www.4front-tech.com/forum/index.php
http://www.4front-tech.com
http://www.opensound.com/wiki/index.php

I agree

I find PC-BSD with OSS has the best sound on my system. I should try this with Chakra. JACK sounds interesting as well.

I also agree with Anonymo

I read "State of Sound in Linux not so Sorry After All"

http://insanecoding.blogspot.com/2009/06/state-of-sound-in-linux-not-so-sorry.html

I switched Ubuntu to OSS and everything was better / cooler / easier. i.e. the sound quality was better, the user interface was nicer too. (Although there was no OSS HDMI driver at the time.)

I also have used OSS on Solaris. It was the exact same experience. I plugged in a USB DAC and viola! It just worked.

Pulse made me busy all this year

Hi. I don't know if pulse have to die, but I sure hate its guts. I work for an OEM company who sell various machines with Linux/KDE installed. Pulse failed on basically all the HW configs that I saw.

At home, here's what I have to say : Pulse is not ready. When I manage to get it to work, it hangs.

I totally agree with the guy who proposed to replace pulse with jack, maybe with a simple frontend ? Jack is the only software that I know of that actually aknowledge that a user can have more than ONE soundcard.

More in Tux Machines

Linspire 9.0 Released

Today our development team is excited to announce the release of Linspire 9.0; packed with a TON of improvements and security updates, this is a major update that we’ve been working hard to get out to our faithful users. The global pandemic has delayed its release, but the development team has worked diligently and meticulously behind-the-scenes over the past few months, fine-tuning every detail of what is widely considered to be the premier Linux desktop on the market today. The Linspire 9.0 series will be the last one featuring the 18.04 LTS codebase; upcoming Linspire X will be based on the 20.04 LTS code and kernel. Read more Also: Linspire 9.0 Officially Released, Based on Ubuntu 18.04 LTS and Linux 5.4 LTS

today's leftovers

  • Fast Bare Metal provisioning and infrastructure automation with MAAS
  • [Updated] Michael Stapelberg: Optional dependencies don’t work

    In the i3 projects, we have always tried hard to avoid optional dependencies. There are a number of reasons behind it, and as I have recently encountered some of the downsides of optional dependencies firsthand, I summarized my thoughts in this article.

  • Benchmarking NetBSD, second evaluation report

    This report was written by Apurva Nandan as part of Google Summer of Code 2020. This blog post is in continuation of GSoC Reports: Benchmarking NetBSD, first evaluation report blog and describes my progress in the second phase of GSoC 2020 under The NetBSD Foundation. In this phase, I worked on the automation of the regression suite made using Phoronix Test Suite (PTS) and its integration with Anita. The automation framework consists of two components Phoromatic server, provided by Phoronix Test Suite in pkgsrc, and Anita, a Python tool for automating NetBSD installation.

  • Interest in Kodi Declines After a Turmultuous Few Years of Piracy Headlines

    After many years of being mentioned in the same breath as movie and TV show piracy, interest in the Kodi media player appears to have peaked and is now on the decline. That's according to Google Trends data which suggests that after reaching a high in early 2017, interest via search is now on a continuous downward trend.

Programming Leftovers

  • RcppSimdJson 0.1.1: More Features

    A first update following for the exciting RcppSimdJson 0.1.0 release last month is now on CRAN. Version 0.1.1 brings further enhancements such direct parsing of raw chars, working with compressed files as well as much expanded querying ability all thanks to Brendan, some improvements to our demos thanks to Daniel as well as a small fix via a one-liner borrowed from upstream for a reported UBSAN issue. RcppSimdJson wraps the fantastic and genuinely impressive simdjson library by Daniel Lemire and collaborators. Via very clever algorithmic engineering to obtain largely branch-free code, coupled with modern C++ and newer compiler instructions, it results in parsing gigabytes of JSON parsed per second which is quite mindboggling. The best-case performance is ‘faster than CPU speed’ as use of parallel SIMD instructions and careful branch avoidance can lead to less than one cpu cycle use per byte parsed; see the video of the talk by Daniel Lemire at QCon (also voted best talk).

  • Jonathan Dowland: Generic Haskell

    When I did the work described earlier in template haskell, I also explored generic programming in Haskell to solve a particular problem. StrIoT is a program generator: it outputs source code, which may depend upon other modules, which need to be imported via declarations at the top of the source code files. The data structure that StrIoT manipulates contains information about what modules are loaded to resolve the names that have been used in the input code, so we can walk that structure to automatically derive an import list. The generic programming tools I used for this are from Structure Your Boilerplate (SYB), a module written to complement a paper of the same name.

  • 9 reasons I upgraded from AngularJS to Angular

    In 2010, Google released AngularJS, an open source, JavaScript-based frontend structure for developing single-page applications (SPAs) for the internet. With its move to version 2.0 in 2016, the framework's name was shortened to Angular. AngularJS is still being developed and used, but Angular's advantages mean it's a smart idea to migrate to the newer version.

  • [Old/Odd] 5 news feautures of PHP-7.2

    Before PHP 7.2 the object keyword was used to convert one data type to another (boxing and unboxing), for example, an array to an object of the sdtClass class and/or vice versa, as of PHP 7.2 the object data type can be used as parameter type or as function return type.

  • This Week In Rust: This Week in Rust 351

Proprietary Software and Linux Foundation

  • [PCLinuxOS] Opera Browser updated to 70.0.3728.106

    Opera is a Chromium-based browser using the Blink layout engine. It differentiates itself because of a distinct user interface and other features.

  • Vivaldi Explains Why They Make "Proprietary Garbage"

    It is unfair to say that Vivaldi is not open source at all as someone like Distrotube has done, the way the company behind Vivaldi has decided to handle this application is by using a dual licensing system where the open source portion of the application is licensed under an open source BSD license but that's not the point of today, the point is to explain why they have decided to license their software in such a way.

  • Scientists Forced To Change Names Of Human Genes Because Of Microsoft's Failure To Patch Excel

    Six years ago, Techdirt wrote about a curious issue with Microsoft's Excel. A default date conversion feature was altering the names of genes, because they looked like dates. For example, the tumor suppressor gene DEC1 (Deleted in Esophageal Cancer 1) was being converted to "1-DEC". Hardly a widespread problem, you might think. Not so: research in 2016 found that nearly 20% of 3500 papers taken from leading genomic journals contained gene lists that had been corrupted by Excel's re-interpretation of names as dates. Although there don't seem to be any instances where this led to serious errors, there is a natural concern that it could distort research results. The good news is this problem has now been fixed. The rather surprising news is that it wasn't Microsoft that fixed it, even though Excel was at fault. As an article in The Verge reports:

  • The Linux Foundation Wants Open-Source Tech to Address Future Pandemics

    The Linux Foundation, which supports open-source innovation in blockchain tech, launched the Linux Foundation Public Health Initiative (LFPHI) at the end of July. The LFPHI’s goal is to promote the use of open source by public health authorities, which can be scrutinized by anyone, to fight not just COVID-19 but future pandemics as well.

  • LF Edge’s Akraino Project Release 3 Now Available, Unifying Open Source Blueprints Across MEC, AI, Cloud and Telecom Edge

    LF Edge, an umbrella organization within the Linux Foundation that aims to establish an open, interoperable framework for edge computing independent of hardware, silicon, cloud, or operating system, today announced the availability of Akraino Release 3 (“Akraino R3”). Akraino’s third and most mature release to date delivers fully functional edge solutions– implemented across global organizations– to enable a diversity of edge deployments across the globe. New blueprints include a focus on MEC, AI/ML, and Cloud edge. In addition, the community authored the first iteration of a new white paper to bring common open edge API standards to align the industry.

  • Linux Foundation Launches Jenkins X Training Course

    Linux Foundation has launched a new training course, LFS268 – CI/CD with Jenkins X. Developed in conjunction with the Continuous Delivery Foundation, the course will introduce the fundamentals of Jenkins X.