From: Jonathan W. <jw...@ph...> - 2007-08-20 23:24:28
|
Hi Francois > >>> A minor observation. I've had communication from someone who's tested > >>> svn515 with a MOTU 828MkII. They noticed the following: > >>> > >>> > Latency is reported to Jack's clients as if the number of period per > >>> > buffer was 1. > > The attached patch fix the latency reported to Jackd's clients and restores > the -I and -O options to set extra latency frames. Sounds good. I'll let Pieter review this since it probably affects his code more than anyone's. > The best would be to take in account the hardware latency (ADC, DSP, DAC, > etc...) on a per port basis. The problem is to know the actual hardware > input and output latency (we only can measure the some of both). I can certainly measure it but it might be a while yet - I have other things to do which are of higher priority. > Anybody knows if this information is available for other interfaces than > MOTU? I doubt it - I suspect you'd have to measure it. Manufacturers would know but whether they'll share it with us is anyone's guess. > I will try to get the information from MOTU but nothing is sure. Best of luck. Just don't tell them you're working on the Linux driver or you'll get the standard "we don't support Linux" answer. :-( Regards jonathan |