now this gets a bit ridiculous, 8 frames for usb audio buffer size.
gets latency quite low, but easily xrun if playing anything complex in FLStudio.
seems stable/sustainable while there is very little stuff running though.
I like the small deviation in those "distance" numbers (it is how far userspace is from the kernel position in the ringbuffer).
@x42 system latency is quite low, I posted some cyclictests some posts ago, but it is roughly around 10us.
not sure what you mean by roundtrip latency here, but this device has 1.3ms total IO latency, so worst case we combine the USB audio latency to this one.
@falktx I meant systemic (not system) latency. ie measure with jack-iodelay or similar tool using a cable to connect audio playback to an audio input port of the USB soundcard.
@x42 ah got it. but that is what the RTL tool is measuring.
back on Linux I can't run jackd with as low buffer sizes as I can with Windows + ASIO, but 64 frames + 3 periods works fine.
8.2ms with this setup.
The same jackd args on a "real" soundcard (Roland Rubix22) gives me 7.6ms latency.
So it's quite a good result considering the Anagram side runs a full JACK audio graph inside.
@falktx wow nice hack.
I assume it is dominated by systemic latency. Have you measured that? I expect that total roundtrip latency is an oder of magnitude higher.