Return-Path: Received: from mail-qc0-f169.google.com ([209.85.216.169] verified) by media-motion.tv (CommuniGate Pro SMTP 4.2.10) with ESMTP-TLS id 4924343 for AE-List@media-motion.tv; Thu, 13 Dec 2012 17:00:43 +0100 Received: by mail-qc0-f169.google.com with SMTP id t2so1775330qcq.28 for ; Thu, 13 Dec 2012 08:04:40 -0800 (PST) X-Google-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=google.com; s=20120113; h=mime-version:in-reply-to:references:from:date:message-id:subject:to :content-type:x-gm-message-state; bh=+65X+q9871zFr5OjgbQQGYnoC3r6LHKyZJZO9TxL59E=; b=n8kMclnO7gjrUgAYlg33IiweNvwLS327Qr2jsF4f4a4SfHI6aMSGwpFhXw/zUF9W7l UvmF0CelTO+kHYc6IkYAgJQsXlgCx810wfNoyMMH3DoJbS3c7ZDSR6BlGFmhU+E/gdSS /Zt8wtMor6Dtlv47eN0PlrK9G7FRQXp9EUzv7sLcs+aHGUCQRQ117PlvXiDEzwTu+zXm p+St5iW/1NsPB5KM+lyoYhG6RKnM4XzJrgYAjsU7srZaCGYiaYX06atd8F+DlmPB2Ylw lOR7WBLhHFaf0U2AZGLSd6auUuiKoTYy9QawxuzfpRBNuzS4wkknQAlC+eHY21VxcW/1 UAXQ== Received: by 10.224.146.74 with SMTP id g10mr1191195qav.93.1355414680213; Thu, 13 Dec 2012 08:04:40 -0800 (PST) MIME-Version: 1.0 Received: by 10.49.60.161 with HTTP; Thu, 13 Dec 2012 08:04:20 -0800 (PST) In-Reply-To: References: From: Robert Kjettrup Date: Thu, 13 Dec 2012 17:04:20 +0100 Message-ID: Subject: Re: [AE] 32core benchmarks To: After Effects Mail List Content-Type: multipart/alternative; boundary=089e0111ce9e899a5804d0be127f X-Gm-Message-State: ALoCoQmaedVHANR+uwKjNeD0t+rNLSyNK1V7a1lSPbNeIkpKp8St2PwzjQ73vTyq+6vkg/UMY0kX --089e0111ce9e899a5804d0be127f Content-Type: text/plain; charset=ISO-8859-1 Yes, my K5000 took ~22 min and according to Danny the GTX680 is about ~15min, so that does tell that the K5000 is a underclocked 680. at least the openGL performance in Cinebench (and Softimage that i use) is good :-D - Robert 2012/12/13 rendernyc > Seems the CUDA performance of the K5000 is much lower than the GTX680 so > far on Roberts test. > Anyone else have a K5000 that can spare 15-20 minutes to run the test? > > > > > On Thu, Dec 13, 2012 at 10:10 AM, Dave Bittner wrote: > >> On the issue of lower OpenGL performance from the GTX series of cards, I >> did some research, and the conclusion seems to be that NVidia did some >> internal bandwidth bottlenecking on the GTX cards to keep them from >> competing with the much more expensive Quadro series. As gaming cards, the >> GTX series is designed to get images rendered to the screen quickly, but >> for OpenGL it is limited in getting those rendered images back to the CPU, >> which is what we need for graphics work. The CUDA pipes are wide open, from >> what I could find. >> >> This stuff came up as part of my blog post exploring where the >> bottlenecks actually are in GPU accelerated rendering. >> >> >> http://www.pixelworkshop.com/2012/05/02/slot-card-racing-mac-pro-gpu-speed-tests/ >> >> Dave Bittner - Pixel Workshop Inc. >> www.pixelworkshop.com >> 410.381.8555 >> Twitter @bittner >> >> On Dec 12, 2012, at 4:44PM, Robert Kjettrup wrote: >> >> > Hi >> > >> > Just for fun i tried running my new 16core (32 threads) Z820 through >> some of these tests, and comparing it there also seems to be some odd >> numbers for your GPU in cineBench. I have the new Quadro K5000 that as far >> as i know is a lower clocked version of the GTX680 you have in your >> machine, but my OpenGL fps number are much higher. I only got the K5000 >> because i have had some bad experience with my former GTX card that just >> crashed the OpenGL driver when it was under load for longer periods. Would >> have gone with the GTX680 if i trusted the gamer cards more, but "known" >> stability was prioritised. >> > >> > CineBench scores: >> > OpenGL 81.45 fps >> > CPU 25.07 pts >> > >> > And running GeekBench (the free 32bit version) i can see that Teddy has >> some impressive memory numbers, higher than i get. Is the memory linked to >> the CPU clock frequency also? 4.5Ghz is a bit higher than my 3.1Ghz :-D >> > >> > Integer Processor integer performance 32414 >> > Floating Point Processor floating point performance 42123 >> > Memory Memory performance 7071 >> > Stream Memory bandwidth performance 8585 >> > Geekbench score: 28360 >> > >> > not to show off, but i thought some of you guys would like to see the >> numbers of the current top of the line intel xeon >> > well, but i also guess that this machine has cost ALOT more than what >> you spent on yours. Intel/HP sure know how to price their current highend >> stuff :-) >> > >> > - Robert >> > >> >> >> >> >> >> +---End of message---+ >> To unsubscribe send any message to >> > > > > -- > danny princz > > exposedideas.com > --089e0111ce9e899a5804d0be127f Content-Type: text/html; charset=ISO-8859-1 Content-Transfer-Encoding: quoted-printable Yes, my K5000 took ~22 min and according to Danny the GTX680 is about ~15mi= n, so that does tell that the K5000 is a underclocked 680. at least the ope= nGL performance in Cinebench (and Softimage that i use) is good :-D

- Robert


2012/12/13 rendernyc &= lt;rendernyc@gmail= .com>
Seems the CUDA performance of the K5000 is much lower than the GTX680 so fa= r on Roberts test.
Anyone else have a K5000 that can spare 15-20 minute= s to run the test?

=A0


On Thu, Dec 13, 2012 at 10:10 AM, Dave Bittn= er <dave@pixelworkshop.com> wrote:
On the issue of lower OpenGL performance from the GTX series of cards, I di= d some research, and the conclusion seems to be that NVidia did some intern= al bandwidth bottlenecking on the GTX cards to keep them from competing wit= h the much more expensive Quadro series. As gaming cards, the GTX series is= designed to get images rendered to the screen quickly, but for OpenGL it i= s limited in getting those rendered images back to the CPU, which is what w= e need for graphics work. The CUDA pipes are wide open, from what I could f= ind.

This stuff came up as part of my blog post exploring where the bottlenecks = actually are in GPU accelerated rendering.

http://www.pixelworkshop.com/2012/05/0= 2/slot-card-racing-mac-pro-gpu-speed-tests/

Dave Bittner - Pixel Workshop Inc.
www.pixelworksho= p.com
410.3= 81.8555
Twitter @bittner

On Dec 12, 2012, at 4:44PM, Robert Kjettrup <robert@stvmayday.dk> wrote:

> Hi
>
> Just for fun i tried running my new 16core (32 threads) Z820 through s= ome of these tests, and comparing it there also seems to be some odd number= s for your GPU in cineBench. I have the new Quadro K5000 that as far as i k= now is a lower clocked version of the GTX680 you have in your machine, but = my OpenGL fps number are much higher. I only got the K5000 because i have h= ad some bad experience with my former GTX card that just crashed the OpenGL= driver when it was under load for longer periods. Would have gone with the= GTX680 if i trusted the gamer cards more, but "known" stability = was prioritised.
>
> CineBench scores:
> OpenGL 81.45 fps
> CPU 25.07 pts
>
> And running GeekBench (the free 32bit version) i can see that Teddy ha= s some impressive memory numbers, higher than i get. Is the memory linked t= o the CPU clock frequency also? 4.5Ghz is a bit higher than my 3.1Ghz :-D >
> Integer =A0 =A0 =A0 Processor integer performance =A0 32414
> Floating Point =A0 =A0 =A0 =A0Processor floating point performance =A0= =A042123
> Memory =A0 =A0 =A0 =A0Memory performance =A0 =A0 =A07071
> Stream =A0 =A0 =A0 =A0Memory bandwidth performance =A0 =A08585
> Geekbench score: 28360
>
> not to show off, but i thought some of you guys would like to see the = numbers of the current top of the line intel xeon
> well, but i also guess that this machine has cost ALOT more than what = you spent on yours. Intel/HP sure know how to price their current highend s= tuff :-)
>
> - Robert
>





+---End of message---+
To unsubscribe send any message to <ae-list-off@media-motion.tv>



--
danny princz

exposedideas.com

--089e0111ce9e899a5804d0be127f--