James Ritson
New Member
Hi there, I've tried searching with the term "audio buffer" but can't find any topics that match my question..
I'm using OBS to capture screen recordings for Affinity Photo on both macOS and Windows. For some half-decent audio I'm using a Rode wireless lav kit into an external M-Audio M-Track 2x2M interface. This setup all works fine, bar an issue with some very rare audio dropout/stutter.
The dropout or stutter can sometimes be predicted with certain operations that cause the CPU usage to spike, but at other times it can seem totally random (e.g. when the image canvas has rendered, the mouse cursor is still and there shouldn't be any CPU activity). There's no knowing when it might happen, basically. I've made sure the sample rate is at 48000Hz and not 192000Hz, which the device uses by default.
This was initially worse on Windows but I seem to have ironed it out entirely by increasing the audio buffer size to 4096 from 512. This was done through the little external application that comes with the M-Audio drivers. Unfortunately, though, from my reading, it looks like the audio buffer size is controlled by each application individually on Mac - I can't find anything that relates to it in OBS.
Does anyone know if the audio buffer size can be controlled within OBS, or if there's another way to do it?
I've tried various encoder combinations to try and nail down the issue - x264 encoding with CRF 1 and ultrafast, Apple VT Hardware on Mac, NVENC on Windows, but all of them can exhibit the issue.
The only guaranteed fix I can see at this point is to use the onboard headphone/mic jack on the Macbook Pro with the correct adapter - I'd really prefer not to do this though as the sound quality is pretty bad compared to an external interface...
Thanks,
James
I'm using OBS to capture screen recordings for Affinity Photo on both macOS and Windows. For some half-decent audio I'm using a Rode wireless lav kit into an external M-Audio M-Track 2x2M interface. This setup all works fine, bar an issue with some very rare audio dropout/stutter.
The dropout or stutter can sometimes be predicted with certain operations that cause the CPU usage to spike, but at other times it can seem totally random (e.g. when the image canvas has rendered, the mouse cursor is still and there shouldn't be any CPU activity). There's no knowing when it might happen, basically. I've made sure the sample rate is at 48000Hz and not 192000Hz, which the device uses by default.
This was initially worse on Windows but I seem to have ironed it out entirely by increasing the audio buffer size to 4096 from 512. This was done through the little external application that comes with the M-Audio drivers. Unfortunately, though, from my reading, it looks like the audio buffer size is controlled by each application individually on Mac - I can't find anything that relates to it in OBS.
Does anyone know if the audio buffer size can be controlled within OBS, or if there's another way to do it?
I've tried various encoder combinations to try and nail down the issue - x264 encoding with CRF 1 and ultrafast, Apple VT Hardware on Mac, NVENC on Windows, but all of them can exhibit the issue.
The only guaranteed fix I can see at this point is to use the onboard headphone/mic jack on the Macbook Pro with the correct adapter - I'd really prefer not to do this though as the sound quality is pretty bad compared to an external interface...
Thanks,
James