On 4/21/05, michael higgins <[EMAIL PROTECTED]> wrote: > On Thu, 21 Apr 2005 12:06:12 -0700 > Mark Knecht <[EMAIL PROTECTED]> wrote: > > > On 4/21/05, michael higgins <[EMAIL PROTECTED]> wrote: > > > On Thu, 21 Apr 2005 09:49:50 -0700 > > > Mark Knecht <[EMAIL PROTECTED]> wrote: > > > > > > > using the patch from here: http://www.joq.us/realtime/ at the suggestion > > > of someone on this list... and getting some good low latency in, say, > > > jack rack. > > > > What are you seeing? > > > > Hearing, actually. I've only compared to the playability of my rig on 'doze. > No benchmarking. It's just slightly better. > > I feel it could be better still. '-) I look forward to any new clues on this > list. >
Well, what I meant was what sort of Jack settings are you using and what are you seeing in terms of xruns? I can run some of my systems with patched kernels and realtime-lsm (be they Gentoo or FC2/PLanet) at 128/2 or even 64/2. My laptop onboard sound chip will actually run 16/2 (well under 1mS latency) with a custom patched kernel. I don't remember doing much better than 512/2 or possibly 256/2 for short periods of time with an unpatched kernel. That said the info above is definitely NOT apples to apples since these are different systems, different sound cards and different distros and different setups. I'd like to get some consistency to all of this. I do have a very ck-sources kernel that seems to work pretty well but I haven't actually done a session with it so I really don't want to say it's ready. It does compile and run well under low loads though. - Mark -- gentoo-user@gentoo.org mailing list