Bufferunderruns and echo effects at windows.

[mixxxbot]

Reported by: daschuer
Date: 2014-12-05T07:17:33Z
Status: Fix Released
Importance: Critical
Launchpad Issue: lp1399548

---

The windows build of Mixxx 1.12-alpha has a really bad performance compared to Mixxx 1.11.
I had the chance to test Mixxx on a Pentium 4 windows Vista Notebook.

It was not possible to use my Maya 44 usb card with direct sound, even though it was working as windows system sound card.
After installing the ASIO driver, it works, but very bad. It suffers interrupts of > 1 s when storing the analyses waveforms. Sometimes pressing pause produces a loop like echo effect. The whole system behaves somehow sluggish.
Changing the buffer settings does not have any positive effect. But there are no underruns reported

After installing 1.11 Mixxx everything works as expected again.

Do we probably suffer a portaudio bug?
http://comments.gmane.org/gmane.comp.audio.portaudio.devel/11875

[mixxxbot]

Commented by: rryan
Date: 2014-12-09T20:46:26Z

---

Figuring this out sounds like a critical bug.

BTW when you file new bugs (i.e. not feature bugs) that affect master please put them in the 1.12.0 milestone by default. I'll triage them out if applicable but otherwise some bugs get left in the New/Undecided lurch.

[mixxxbot]

Commented by: ywwg
Date: 2015-09-25T01:27:32Z

---

daniel are you still seeing such major performance issues? I think we have done a lot of work to fix this and this bug is not really diagnosable because it's just generally "things are bad"

[mixxxbot]

Commented by: daschuer
Date: 2015-09-25T06:31:37Z

---

I have currently no access to the effected Windows device.

Can one else compare Mixxx 1.11 and Mixxx 1.12 on a legacy windows device?
If there are no notable issues, I think we can close the bug.

[mixxxbot]

Commented by: ywwg
Date: 2015-09-25T21:30:00Z

---

ok, I guess we'll pretend this is fixed

[mixxxbot]

Issue closed with status Fix Released.