The calibration never stabilizes in my experience, it usually jumps to what I assume is the right value of about 150ms, and then gradually goes up the longer I run the calibrator. I thought it was supposed to average on a single value as it doesn't make sense to me that the delay will effectively increase the longer I run the calibration.
This has happened to me on both Chrome and Firefox, Windows 10 and Linux, and two different USB soundcards.
Comments (8)
yes, I'll do it sometime today
Do you mean a slow gradual increase or does it rather double, triple, etc.?
Is there max value?
Didn't get to record a screencast yesterday. It seems to add just a few milliseconds each cycle,
a show gradual increase. It does seem to go on forever but I'll leave it for a few minutes and see if it eventually stabilizes.
Sorry for the late reply, couldn't replicate this on Windows so it looks like a Linux specific issue. I'm on the latest version of Manjaro.
Here's a screencast. The audio goes through my field recorder which is acting as my soundcard and microphone. It starts off very unstable and then levels out at what I assume is double the correct value. From there it rises slowly, sometimes jumping back 10ms or so.
Including os and kernel version
OS: Manjaro 19.0.2 Kyria
Kernel: x86_64 Linux 5.4.24-1-MANJARO
Here's the output from dmesg, I'm not using any special drivers I think. If it's still relevant I can retest my other soundcard to doublecheck that it's not an issue related to this specific soundcard. I believe I was using a different one back home but I'll be traveling back to my dorm room in a few days.
[xnstad@archibald ~]$ sudo dmesg | grep ZOOM
[sudo] password for xnstad:
[ 4261.116037] usb 2-1.2: Manufacturer: ZOOM Corporation
[ 4417.439197] usb 2-1.2: Manufacturer: ZOOM Corporation
The noise at the beginning is me moving my microphone away from my laptop's exhaust port, the noise was not loud enough to reach the test range. Tested a few times in a less noisy environment and got weird results each time, I can upload another test where I'm sure to have no external noise.
I'll try to find out what system module the sound card is using as I believe it's utilizing the built in open source USB audio drivers on Linux. The audio is not being processed in any way.
Are there any tools or other websites where one can test if the problem lies in our implementation?
Cool, I'll test this soon
Managed to replicate this on windows with the same audio interface. Seems to not be a Linux specific issue after all. Will try a different usb Interface in the coming days to see if that's where the issue lies
https://youtu.be/msAldDhL0vM