[ardour-users] [Fwd: Re: Is 64 bit dual core supported and stable yet?]

[Fernando Lopez-Lezcano]

On Fri, 2005-12-02 at 22:17 -0500, George Chan wrote:
> I was just about to go out and buy a dual core AMD processor &
> motherboard, but after reading this email (see below), I'm going to go
> with a dual CPU single core Opteron with an ASUS K8N-DL motherboard.
> Tyan motherboards are nice, buy way too expensive for me.
> 
> You may find this link helpful:
> http://www.tomshardware.com/cpu/20051107/index.html
> 
> If future dual core AMD CPUs remain in a 940 pin configuration, I can
> always upgrade to a Dual CPU Dual Core system.  That may not happen
> though.  I recently read that the next generation of Dual Core AMD CPUs
> will have a different pin configuration.
> 
> If there was no design fault in AMD's dual core offerings, I would like
> to have gone with a Dual CPU Dual Core setup.... buying a dual cpu
> motherboard with one dual core cpu to start with, then adding a second
> cpu when I had some extra cash.
> 
> Hope this is helpful.

Interesting link....

There are other considerations for audio machines that may be important
(or not), like for example cooling needs and the associated fans. 

Obviously going with the latest generation of processors is not the best
cost/performance tradeoff :-) I don't have a link right now but I read
an interesting article not long ago on precisely that and I think
something like an Athlon 3000+ had one of the best cost/performance
ratios. 

-- Fernando

> -------- Forwarded Message --------
> From: Paul Davis <paul at linuxaudiosystems.com>
> Reply-To: paul at linuxaudiosystems.com
> To: Joel White <cv223 at comcast.net>
> Cc: ardour-users at lists.ardour.org
> Subject: Re: [ardour-users] Is 64 bit dual core supported and stable
> yet?
> Date: Tue, 29 Nov 2005 09:34:28 -0500
> 
> On Mon, 2005-11-28 at 23:15 -0500, Joel White wrote:
> > Even thought the results are good so far, I still get occasional xruns like:
> > 
> > **** alsa_pcm: xrun of at least 118.867 msecs
> > (this particular xrun interrupted a recording with my band, which is 
> > _very_ bad! I was even running capture only and with a longer period, 
> > just to be sure xruns wouldn't happen!)
> > 
> > and the odd:
> > 
> > delay of 1131462.000 usecs exceeds estimated spare time of 5606.000; 
> > restart ...
> > (this type of error comes mostly when audio connections change, but also 
> > at other times, although the delay isn't usually quite so long)
> > 
> >  From my understanding, both of these time periods indicate something 
> > very wrong (an xrun of 118+ msecs?? a delay of 1.13+ _sec_?!?!), but I 
> > haven't been able to track them down, yet.  I've run Ardour recording 8 
> 
> this is caused by a design fault in AMD's dual core offerings. they have
> already made it clear that future designs will rectify the problem.
> there is a slowly evolving discussion on jackit-devel to try and decide
> what to do about it.