[PlanetCCRMA] JACK Optimization

[Nexxus Six]

Fernando,
Thank you for the fast reply... I was also trying some unusual combinations last night... like 32 for buffer, and 16 (or more) for period. Is there any trade offs for using low buffer / high period? 2 Periods or less definitely does not work (tried that last night)... more buffering works, but the latency then gets large (341mS or bigger).

I was able to find this:
http://www.alsa-project.org/main/index.php/Matrix:Module-hda-intel

Would any of this be good when using Jack?

Thanks,
NexxuSix



----- Original Message ----
From: Fernando Lopez-Lezcano <nando at ccrma.Stanford.EDU>
To: Nexxus Six <nexxusix at yahoo.com>
Cc: planetccrma at ccrma.Stanford.EDU
Sent: Wednesday, April 30, 2008 2:23:29 PM
Subject: Re: [PlanetCCRMA] JACK Optimization

On Wed, 2008-04-30 at 12:23 -0700, Nexxus Six wrote:
> Fenando,
> Do you have information on what is the best way to fully optimize JACK
> for my onboard laptop audio? I have Intel HDA, and Realtek ALC268. I
> was playing around with the combination of Frames, Buffers, and got
> some good results, but still occasionally I would have some drops. I
> tried lower sample rates (41k samples per second) and "Output Only" 2
> channel maximum.
> 
> Is there any recommendations or general rules that apply here?

The one rule for hda intel hardware is to use 3 periods or more. 2 
does not work very well (if at all) so you are almost certainly already 
doing that. Hda intel is by design quite crappy....

There are some parameters you can pass to the kernel module when it
loads that apparently makes it work when it doesn't, I don't know if
those would help with an already working configuration. 

-- Fernando


      ____________________________________________________________________________________
Be a better friend, newshound, and 
know-it-all with Yahoo! Mobile.  Try it now.  http://mobile.yahoo.com/;_ylt=Ahu06i62sR8HDtDypao8Wcj9tAcJ
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://ccrma-mail.stanford.edu/pipermail/planetccrma/attachments/20080430/1550ea73/attachment.html