Thanks for the pointers Istvan, I edited rtalsa.c to remove the limits 
on -b and -B ( only on my system, not the CVS! ). I am able to get 
realtime audio working as low as -b4 -B16 though it is iffy, -b8 -B32 
seems rock solid. My kernel and IRQ setup is not as aggressively tuned 
as it could be though, I only ran the one patch, so I suspect that -b4 
-B16 is doable with better IRQ sharing and the latest patch. ( it just 
barely cracks out once in a while ).

As to the printout issue, it appears that the messages from Istvan's 
alsa module get printed only if -m1 or greater ( with the correct 
results ). At the same time, the incorrect messages from somewhere else 
also get printed ( showing whatever was typed in not what is being used 
) and these get printed even at -m0. Considering that anything above -m0 
doesn't work for low latency real time output, this is quite misleading. 
Below is the output before eliminating limits at the two different 
settings ( called with -b2 -B4 for examples sake ).

( at -m1 )
orch now loaded
audio buffered in 1 sample-frame blocks
ALSA output: total buffer size: 64, period size: 16
writing 4-byte blks of shorts to da

( at -m0 )
orch now loaded
audio buffered in 1 sample-frame blocks
writing 4-byte blks of shorts to dac

It's a minor issue, but sure caused me some confusion. I would also like 
to know how does one calculate the total latency in samples from the 
above? For real time fx processing I would like to know the minimum 
delay I can get, as anything above 4-5 ms becomes quite noticeable under 
certain circumstances.

Thanks
Iain


-------------------------------------------------------
This SF.Net email is sponsored by the JBoss Inc.  Get Certified Today
Register for a JBoss Training Course.  Free Certification Exam
for All Training Attendees Through End of 2005. For more info visit:
http://ads.osdn.com/?ad_id=7628&alloc_id=16845&op=click
_______________________________________________
Csound-devel mailing list
Csound-devel@lists.sourceforge.net

Iain Duncan wrote:

> realtime audio working as low as -b4 -B16 though it is iffy, -b8 -B32 
> seems rock solid. My kernel and IRQ setup is not as aggressively tuned 

Note that the limits are now changed to -b8 -B16, but maybe I will
remove that. There are other checks elsewhere, though, that ensure
that the buffer sizes are non-zero, and that -B is greater than and
an integer multiple of -b.

> It's a minor issue, but sure caused me some confusion. I would also like 
> to know how does one calculate the total latency in samples from the 
> above? For real time fx processing I would like to know the minimum 
> delay I can get, as anything above 4-5 ms becomes quite noticeable under 
> certain circumstances.

Theoretically it is something like -B / sr, but you can only
know for sure with actual tests.


-------------------------------------------------------
This SF.Net email is sponsored by the JBoss Inc.  Get Certified Today
Register for a JBoss Training Course.  Free Certification Exam
for All Training Attendees Through End of 2005. For more info visit:
http://ads.osdn.com/?ad_id=7628&alloc_id=16845&op=click
_______________________________________________
Csound-devel mailing list
Csound-devel@lists.sourceforge.net

> Note that the limits are now changed to -b8 -B16, but maybe I will
> remove that. There are other checks elsewhere, though, that ensure
> that the buffer sizes are non-zero, and that -B is greater than and
> an integer multiple of -b.

oops, I'm an idiot. I forgot --sched. with --sched I can do -b4 -B16 
with no crack up!

> Theoretically it is something like -B / sr, but you can only
> know for sure with actual tests.

I am interested in this. How would one test it? I can see how one could 
test round trip latency, which I intend to do later, but what about just 
synthesis?

At anyrate, that gives a theoretical latency of .36 milliseconds!

Thanks
Iain


-------------------------------------------------------
This SF.Net email is sponsored by the JBoss Inc.  Get Certified Today
Register for a JBoss Training Course.  Free Certification Exam
for All Training Attendees Through End of 2005. For more info visit:
http://ads.osdn.com/?ad_id=7628&alloc_id=16845&op=click
_______________________________________________
Csound-devel mailing list
Csound-devel@lists.sourceforge.net