On Fri, Mar 24, 2006 at 11:11:37PM +0100, Atte André Jensen atte.jensen@xxxxxxxxx wrote: > 1) I have a csound instrument that reads the interval between taps and > writes a percentage of that to the delay time of a global delay (code > attached). That was totally screwed up, and taps would translate to > something 2-3 times faster. > > 2) At the end of the rehearsal the clock of my laptop was late (I think, > maybe it was early) by something like 30-45 minutes. Normally the clock > is just fine and I set if from rdate every hour, so It had dropped over > 30 minutes while running about two hours on battery. This sounds supiciously like a problem I reported here not long ago, but didn't get an answer for, especially the second problem. This wouldn't happen to be a Pentium III, would it? After reporting the problem on the LKML, I was given a temporary workaround by John Stultz that involved booting with "clockwource=pit", which fixed the problem for me. The problem also exists for the -mm kernel, and he provided a patch for that that also worked, but none for the -rt kernel (I didn't try the patch he provided on the -rt kernel; maybe I should). Anyway, I don't think the fix has appeared in the -rt patches yet, but I haven't been able to get sound to work in the newest kernels, so I'm still trying to solve that problem before I can say for sure. Chuck
