You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
As the title says. When using 22050 or lower samples rates, something blows up after processing a few samples and the reverb will keep outputting NaN values from which it never recovers.
We're trying to find the cause with the DOSBox Staging team but it's not so easy... (using 22050 sample rate is a valid use case for Raspberry Pi4 users in order to improve performance on that underpowered machine).
The text was updated successfully, but these errors were encountered:
I've not had chance to test it but I'd presume it to be due to the damping / bandwidth filters inside mverb struggling at that low a samplerate. You may need to reduce the maximum frequency they filter at, iirc they go up to 18500Hz by default which is likely too high at a samplerate of 22050.
Thanks for your reply @martineastwood, I came to the same conclusion myself and managed to fix it in the meantime with the exact same approach you proposed, which got rid of the filter instability problem completely. I'll raise a PR for the fix, I've just been busy lately...
As the title says. When using 22050 or lower samples rates, something blows up after processing a few samples and the reverb will keep outputting NaN values from which it never recovers.
We're trying to find the cause with the DOSBox Staging team but it's not so easy... (using 22050 sample rate is a valid use case for Raspberry Pi4 users in order to improve performance on that underpowered machine).
The text was updated successfully, but these errors were encountered: