>Simon W. Fielding wrote:
>>
>> OK - I now have a working basic FC10 system - relevant details are :-
>>
>> Hardware: Toshiba Tecra M3
>> Kernel: 2.6.26.8-1.rt15.1.fc10.ccrma.i686.rt
>> Jackd: jackdmp 1.9.1
>> ffado: 2.0-0.6.rc1.fc10.ccrma (libffado 1.999.40)
>>
>>
>> Initial cursory testing shows (no recording load, just an idle mixer
>> with nothing connected) :-
>>
>> jackd -R -dfirewire -r96000 -p16 -n3 ... jackd won't even start - uses
>> too much CPU
>> jackd -R -dfirewire -r96000 -p32 -n3 ... jackd won't even start - uses
>> too much CPU
>
>these are insanely low settings.
>i have never *ever* heard of anybody running successfully at -p16 -n3,
>and you are trying it at 96k. for the rest of us, that would be -p8!
I'm aware of that. I have no particular intention to even use this piece of kit,
let alone at those settings but I did volunteer to take a look at it so I may as
well try to be reasonably comprehensive where possible. If the info is remotely
useful to somebody, I don't feel I've wasted my time.
I'm using stupidly excessive settings to see what breaks and where, given that
this is all running on an underpowered spare laptop I've got lying round the office
which is not really up to the job.
Sorry - all that sounded a bit more pompous than I intended. Please try not to read
anything in to that.
Cheers,
Simon
imnsho, -p128 should be perfectly fine for 96k. don't waste your time
"tuning" your system for latencies below that.
there are good reasons to perform audio computations in chunks. i don't
know why people seem to get warm fuzzy feelings from invalidating their
cache lines for every new sample :)
_______________________________________________
Linux-audio-user mailing list
Linux-audio-user@email-addr-hidden
http://lists.linuxaudio.org/mailman/listinfo/linux-audio-user
Received on Tue Feb 10 16:15:01 2009
This archive was generated by hypermail 2.1.8 : Tue Feb 10 2009 - 16:15:01 EET