Re: the new iBook...

[Andrew Pask <andrew@kaleidacousticon.com>]

Hey Kevin! Bet you wondered why I sent you that.


Cheers

A




> On 5/4/01 at 10:14 AM, kevin@unitcircle.com (Kevin Goldsmith) wrote:
> 
> > Can you talk about your experiences with the 828?  I'm having a hard 
>time
> > finding reviews of it, but I'm thinking seriously about picking one up.
> > Anyone using one of these with a PC?
> 
> 
> Quoted From the Max/MSP list, So not applicable to PC, sorry.
> 
> Hi all,
> Well, much to my surprise my MOTU 828 multichannel firewire box arrived
> today. If you don't know what I am talking about, go here:
> http://www.motu.com/english/motuaudio/828/body.html
> 
> What I've got: g3/500 powerbook. also testing on G4/450. OS 9.1. Most
> recent Max/MSP. For all these tests I used a signal vector size of 16
> samples. 
> 
> Here is some of what I have found thus far:
> 
> 1. I've been using a MOTU 2408 mkII on my desktop, so here are some
> differences for those who are more familiar with this card (mainly 
>PCI-324
> card). There is no longer an app to change settings on the card. instead
> changes are made via a control panel or control strip module. also
> installed are 2 extensions which are needed in order for the mac to talk 
>to
> the card. changing card settings is a breeze!
> 
> 2. asio latency. for me this is critical since most of the stuff I do
> involves me live with my clarinet...so latency is just something I don't
> want. not overwhelmingly great news in this category. the minimum sample
> buffer size, which to a large extent determines input/output latency, is
> 128. PCI-324 ASIO driver is 64, FYI. Using Les Stuck's latency tester,
> this equates to about a 6.893 ms in Max/MSP. But...there is a problem.
> Testing some of my slightly more involved patches (i.e. more involved 
>than a
> single sfplay~ or groove~) I've noticed that with this buffer size, my
> g3/500 pb just can't keep up. lots of hiccups and at times DSP just 
>decides
> to stop working and needs a reboot. This is the case with both groove~ 
>and
> sfplay~, just to eliminate possibilities of a slow hard-drive. Setting 
>the
> buffer size to the next level (192) cures all. no stutters, hiccups, 
>stops.
> all works fine. 192 = 9.796 ms latency in Max/MSP using same tester as
> before. On my G4/450, none of these problems exist. A buffer of 128
> samples works great with none of the problems found on my powerbook. I
> contacted MOTU tech support via email and got a reply today(!) regarding
> this...basically they say there is enough of a difference in CPU's 
>between
> the 2 to account for this.
> 
> As an FYI, I also tried this using Peak and MOTU's Audio Desk. Both could
> handle playback fine with the 128 sample buffer, but as soon as you added
> any effects I got the same problems found in my Max tests.
> 
> 3. cpu usage. looks like you take a bit of a hit for the using the ASIO
> driver...but I also saw this with the driver for the PCI-324 card. what
> takes up 14% using the sound manager takes up about 22% using the ASIO 
>828
> driver. The good news is that there is very little difference in CPU 
>usage
> between 2 channels and 8 channels of output. So...perhaps worth the
> tradeoff. I personally think so.
> 
> Despite the fact that I can't get the sample buffer size below 192, I 
>still
> really like this card. it's just too cool to have multichannel output
> coming from my powerbook. Not only that, I can plug it into any firewire
> mac, install a few drivers, and I've got multichannel output wherever I
> go...and I only need one card for my g4 and powerbook. Great for the 
>David
> Beaudry Road Show. Other perks of the card are 2 mic preamps for 
>channels 1
> and 2 (that is if you aren't too picky about your mic pre-amps), s/pdif
> in/out, an in and out ADAT light pipe, and ADAT sync.