On the more commodity front, we've had good success with Dell Dual Xeon
Precision workstations, with varying sets of capture and display cards,
depending on the situation.
Here's one configuration:
Dell Workstation 650:
Dual 2.8 GHz Xeon
2GB Memory
Quadro 4 NVidia 900XGL Dual-Head Video
2 - 210 Osprey Video Capture
SoundBlaster Live
Matrox g200 MMS Quad Video (PCI)
-randy
At 05:30 AM 9/10/2004, Osland, CD (Chris) wrote:
> > Behalf Of Deb Agarwal
> >
> > I have heard several of you mention that you are running a single
> > machine AG node.
> > I am planning to upgrade the hardware for my node and want to
> > try running
> > my room based on a single machine. Does anyone have specs suitable
> > for a single machine node running Windows XP or RedHat 9.0
> > that can transmit
> > at least 2 videos, operate 1 audio, drive 4 displays, and
> > keep up with
> > the video streaming
> > in to a well attended meeting in a venue? Lets say up to 30 arriving
> > video feeds.
>
>Here's a repeat of the info I sent out a few weeks ago,
>with the name of the audio card manufacturer corrected
>(I got it wrong in the original e-mail).
>
>--->
>
>Ours are:
>
> twin 2.8 GHz Xeon on Supermicro motherboard (I think)
> 1 Gb memory
> Winnov Videum 4400 VO video capture (4 streams, one card)
> Matrox Parhelia 256 Mb display card (we only needed 2 main
> displays)
> onboard 100 Mbit ethernet
>
>Spot the non-information about sound. Here's the story.
>
>I got Echo Audio cards (stereo, balanced I/O) cards but when we
>enabled them, they played back for about half a second and
>then produced a lovely blue screen. Turns out their cards
>don't work with Xeon processors - the memory access mechanism
>is subtly different from Pentiums, and the Echo Audio cards work
>with Pentiums and Athlons, but not Xeons.
>
>In a panic (days to go before VERY high profile demo - UK
>Minister for Science!) I found that the onboard AC97 audio
>chipset actually was full duplex, so added a balanced-unbalanced
>interface (we use Sonifex, but any will do) and a couple of
>stereo mini-jack to twin phono leads and we were away.
>
>I have now found that E-Mu 1212m cards do support Xeons
>(including twin Xeons) and so, as they are cheaper than the
>bal-unbal interface and save 1U rack space, will try those in my
>next nodes.
>
>On the display front, we are about to try the Matrox QID which
>has, at long last, surfaced - 4 outputs from a single PCI-X or
>AGP slot.
>
>--->
>
>We haven't done any stress tests, but from experience with
>3 single processor 1400 MHz Athlons and much slower display
>card (Matrox G200 PCI) I would be very surprised if the above
>doesn't cruise along with 20-30 incoming video feeds.
>
>Hope that helps
>
>Cheers
>
>Chris
>
>____________________________________________________________________
>Chris Osland Office tel: +44 (0) 1235 446565
>Digital Media and Access Grid Medialab tel: +44 (0) 1235 446459
>BIT Department Access Grid room tel: +44 (0) 1235 445666
>e-mail: c.d.osl...@rl.ac.uk Fax: +44 (0) 1235 445597
>
>CCLRC Rutherford Appleton Laboratory (Bldg. R18)
>Chilton, DIDCOT, Oxon OX11 0QX, UK
>
>[The contents of this email are confidential and are for the use of
>the intended recipient only. If you are not the intended recipient
>do not take any action on it or show it to anyone else,
>but return this email to the sender and delete your copy of it.]
>
>
>
>
>
> > -----Original Message-----
> > Thanks for any help,
> > Deb
> >
> > --
> > ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
> > ~~~~~~~~~
> > Deb Agarwal e-mail:daagar...@lbl.gov
> > MS50B-2239 phone :(510)486-7078
> > Lawrence Berkeley National Lab URL: http://www-itg.lbl.gov/~deba
> > Berkeley, CA 94720
> > ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
> > ~~~~~~~~~
> >
