It could easily be that it mattered in the past and/or matters on some part- icular hardware (I spent years fidding with ALSA trying to find magic to get its latency down on various devices).
cheers M On Tue, Oct 07, 2014 at 07:41:20PM -0400, Ivica Bukvic wrote: > ALSA appears to have it but I notice no observable changes in perceived > latency. > On Oct 7, 2014 6:54 PM, "Miller Puckette" <[email protected]> wrote: > > > I couldn't immediately find anything short of going through each > > s_audio_*.c > > implementation and checking whether the "blocksize" parameter in the open > > routine gets used... ugh. > > > > M > > > > On Tue, Oct 07, 2014 at 09:14:43PM +0000, Jonathan Wilkes via Pd-list > > wrote: > > > Thanks. > > > Is there a quick way to figure out which APIs use it? > > > -Jonathan > > > > > > > > > On Tuesday, October 7, 2014 5:07 PM, Miller Puckette <[email protected]> > > wrote: > > > > > > > > > Sorry - I had forgotten an important detail - not every API implements > > > blocksize. In cases where either you can't or it would never help > > anything, > > > it's simply left at 64. (and really, the control shouldn't appear on the > > > dialog when it isn't used :) > > > > > > I believe blocksize is most important in windows. > > > > > > cheers > > > Miller > > > > > > On Tue, Oct 07, 2014 at 02:15:09PM -0400, Ivica Ico Bukvic wrote: > > > > On 10/07/2014 01:41 PM, Miller Puckette wrote: > > > > >sys_blksize() reports the block size used by Pd's acheduler (I > > believe it's > > > > >always 64 in valnilla but will be settable someday :) > > > > > > > > > >the I/O blocksize, which is set in the audio settings dialog, is what > > > > >Pd uses to interface with whatever audio API is in use. > > > > > > > > Thanks for the clarification, Miller. What I am confused by is that > > when > > > > using ALSA and changing blocksize from 64 to 2048, there is no audible > > > > difference in the delay, even though it should be profound enough to be > > > > preceivable by ear. Is this because ALSA has another internal buffer > > that > > > > essentially supersedes either blocksize, making the final output > > (near?) > > > > identical? > > > > > > > > Also, response to Jonathan's follow-up question regarding JACK would be > > > > appreciated. It seems from looking at the code, JACK does not care > > about > > > > blocksize variable but I could be easily missing something. > > > > > > > > Best, > > > > > > > > Ico > > > > > > > > > > > > > >cheers > > > > >M > > > > > > > > > >On Tue, Oct 07, 2014 at 05:30:55PM +0000, Jonathan Wilkes via Pd-list > > wrote: > > > > >>Hi list, What does the blocksize entry box in audio dialog > > properties do? And how does that value relate (if at all) to the value > > output by sys_getblksize? > > > > >>Thanks,Jonathan > > > > >>_______________________________________________ > > > > >>[email protected] mailing list > > > > >>UNSUBSCRIBE and account-management -> > > http://lists.puredata.info/listinfo/pd-list > > > > > > > > > >_______________________________________________ > > > > >[email protected] mailing list > > > > >UNSUBSCRIBE and account-management -> > > http://lists.puredata.info/listinfo/pd-list > > > > > > > > > > > > -- > > > > Ivica Ico Bukvic, D.M.A. > > > > Associate Professor > > > > Computer Music > > > > ICAT Senior Fellow > > > > DISIS, L2Ork > > > > Virginia Tech > > > > School of Performing Arts - 0141 > > > > Blacksburg, VA 24061 > > > > (540) 231-6139 > > > > [email protected] > > > > www.performingarts.vt.edu > > > > disis.music.vt.edu > > > > l2ork.music.vt.edu > > > > > > > > > > > > _______________________________________________ > > > > [email protected] mailing list > > > > UNSUBSCRIBE and account-management -> > > http://lists.puredata.info/listinfo/pd-list > > > > > > _______________________________________________ > > > [email protected] mailing list > > > UNSUBSCRIBE and account-management -> > > http://lists.puredata.info/listinfo/pd-list > > > > > > > > > > > > > > _______________________________________________ > > > [email protected] mailing list > > > UNSUBSCRIBE and account-management -> > > http://lists.puredata.info/listinfo/pd-list > > > > _______________________________________________ [email protected] mailing list UNSUBSCRIBE and account-management -> http://lists.puredata.info/listinfo/pd-list
