Small problem with Mixbus 32C with AvLinux MXE [Solved]
Posted: Mon Jul 31, 2023 8:09 am
I'll post this here as well. I also wrote on the linuxmusicians forum:
Happens dsp spikes randomly. I wrote to Mixbus support:
"What could be the cause of Random DSP spikes. Sometimes the DSP jumps to 92% and then settles down to around 20%. It might take hours, but sometimes there will be a few for a while. There aren't many plugins in use because I'm preparing and editing the session for mixing
This is a pretty powerful intel i9 based pc, AvLinux MXE. The audio interface is RME UFX+, in class compliant mode."
I got a response:
"It can be very hard to troubleshoot the cause of a DSP spike.
Each time the soundcard 'interrupt' occurs, the operating system must 'wake up' Mixbus, and then we have a few tens of milliseconds (depending on your buffersize) to wake up the DSP threads, process all the incoming audio, and then pass it back to the operating system so it can hand it back to the soundcard, before the soundcard streams it to your ears.
This is a very strict time schedule, and there are uncountable things that can interrupt it.
The fact that you aren't using plugins, and the fact that it happens with long intervals between, suggests that it is not a mixbus/software bug. And even if it is, a bug like that would be almost impossible to troubleshoot without your exact setup and some very advanced troubleshooting tools. Since we don't have multiple reports of this problem, we cannot consider this a bug in Mixbus but rather a system issue.
As a first step, I suggest that you use the ALSA backend directly so Mixbus talks directly the soundcard and does not utilize JACK, pulseaudio, or any other layer.
Additionally, you might consider whether one of these issues is occurring:
your cpu's have entered a low-power state, momentarily
your motherboard has a built-in NMI /remote-management function that triggers a high-priority interrupt (we've had this problem before)
I hope this helps!"
What should be done next? I have used ALSA directly until now.
motherboard: https://rog.asus.com/fi/motherboards/ro ... ing-model/
INTEL Core i9-9900K
4 SSD drives in total
Separate graphics card
Anyone have any more thoughts on this?
I hope the harware is not faulty. Perhaps replacing the machine with AMD iron would come up. And it will be expensive
Happens dsp spikes randomly. I wrote to Mixbus support:
"What could be the cause of Random DSP spikes. Sometimes the DSP jumps to 92% and then settles down to around 20%. It might take hours, but sometimes there will be a few for a while. There aren't many plugins in use because I'm preparing and editing the session for mixing
This is a pretty powerful intel i9 based pc, AvLinux MXE. The audio interface is RME UFX+, in class compliant mode."
I got a response:
"It can be very hard to troubleshoot the cause of a DSP spike.
Each time the soundcard 'interrupt' occurs, the operating system must 'wake up' Mixbus, and then we have a few tens of milliseconds (depending on your buffersize) to wake up the DSP threads, process all the incoming audio, and then pass it back to the operating system so it can hand it back to the soundcard, before the soundcard streams it to your ears.
This is a very strict time schedule, and there are uncountable things that can interrupt it.
The fact that you aren't using plugins, and the fact that it happens with long intervals between, suggests that it is not a mixbus/software bug. And even if it is, a bug like that would be almost impossible to troubleshoot without your exact setup and some very advanced troubleshooting tools. Since we don't have multiple reports of this problem, we cannot consider this a bug in Mixbus but rather a system issue.
As a first step, I suggest that you use the ALSA backend directly so Mixbus talks directly the soundcard and does not utilize JACK, pulseaudio, or any other layer.
Additionally, you might consider whether one of these issues is occurring:
your cpu's have entered a low-power state, momentarily
your motherboard has a built-in NMI /remote-management function that triggers a high-priority interrupt (we've had this problem before)
I hope this helps!"
What should be done next? I have used ALSA directly until now.
motherboard: https://rog.asus.com/fi/motherboards/ro ... ing-model/
INTEL Core i9-9900K
4 SSD drives in total
Separate graphics card
Anyone have any more thoughts on this?
I hope the harware is not faulty. Perhaps replacing the machine with AMD iron would come up. And it will be expensive
