GTX 680 4GB vs GTX 580 3GB - performance test

A forum where development builds are posted for testing by the community.
Forum rules
NOTE: The software in this forum is not %100 reliable, they are development builds and are meant for testing by experienced octane users. If you are a new octane user, we recommend to use the current stable release from the 'Commercial Product News & Releases' forum.
User avatar
smicha
Licensed Customer
Posts: 3151
Joined: Wed Sep 21, 2011 4:13 pm
Location: Warsaw, Poland

Marcus,

With pleasure.
3090, Titan, Quadro, Xeon Scalable Supermicro, 768GB RAM; Sketchup Pro, Classical Architecture.
Custom alloy powder coated laser cut cases, Autodesk metal-sheet 3D modelling.
build-log http://render.otoy.com/forum/viewtopic.php?f=9&t=42540
User avatar
smicha
Licensed Customer
Posts: 3151
Joined: Wed Sep 21, 2011 4:13 pm
Location: Warsaw, Poland

Marcus,

BTW - could you change the title of the topic (just remove "on Octane 3.03")?
I'll post results for different versions.
3090, Titan, Quadro, Xeon Scalable Supermicro, 768GB RAM; Sketchup Pro, Classical Architecture.
Custom alloy powder coated laser cut cases, Autodesk metal-sheet 3D modelling.
build-log http://render.otoy.com/forum/viewtopic.php?f=9&t=42540
User avatar
smicha
Licensed Customer
Posts: 3151
Joined: Wed Sep 21, 2011 4:13 pm
Location: Warsaw, Poland

Here are latest results for new beta nvidia driver 310.33 - seems that "hallway" problem gone.
Attachments
RC1_nvidia310.33.jpg
3090, Titan, Quadro, Xeon Scalable Supermicro, 768GB RAM; Sketchup Pro, Classical Architecture.
Custom alloy powder coated laser cut cases, Autodesk metal-sheet 3D modelling.
build-log http://render.otoy.com/forum/viewtopic.php?f=9&t=42540
User avatar
abstrax
OctaneRender Team
Posts: 5509
Joined: Tue May 18, 2010 11:01 am
Location: Auckland, New Zealand

smicha wrote:Marcus,

BTW - could you change the title of the topic (just remove "on Octane 3.03")?
I'll post results for different versions.
Done. And thanks for the benchmark update.

Cheers,
Marcus
In theory there is no difference between theory and practice. In practice there is. - Yogi Berra
User avatar
smicha
Licensed Customer
Posts: 3151
Joined: Wed Sep 21, 2011 4:13 pm
Location: Warsaw, Poland

RC2 scores. Hallway problems again:

Image
3090, Titan, Quadro, Xeon Scalable Supermicro, 768GB RAM; Sketchup Pro, Classical Architecture.
Custom alloy powder coated laser cut cases, Autodesk metal-sheet 3D modelling.
build-log http://render.otoy.com/forum/viewtopic.php?f=9&t=42540
User avatar
smicha
Licensed Customer
Posts: 3151
Joined: Wed Sep 21, 2011 4:13 pm
Location: Warsaw, Poland

Image
3090, Titan, Quadro, Xeon Scalable Supermicro, 768GB RAM; Sketchup Pro, Classical Architecture.
Custom alloy powder coated laser cut cases, Autodesk metal-sheet 3D modelling.
build-log http://render.otoy.com/forum/viewtopic.php?f=9&t=42540
User avatar
abstrax
OctaneRender Team
Posts: 5509
Joined: Tue May 18, 2010 11:01 am
Location: Auckland, New Zealand

Hi, thanks for the feedback. Unfortunately I can't reproduce your results here on 1 GTX 690 GPU: The Hallway PT gives 2.03 Ms/s and the Hallway PMC 1.67 Ms/s. A GTX 690 GPU should be a little bit slower than a GTX 680 GPU. -> I think there must be something else going on. Could you try the last WHQL certified driver (306.97)?

Cheers,
Marcus
In theory there is no difference between theory and practice. In practice there is. - Yogi Berra
User avatar
smicha
Licensed Customer
Posts: 3151
Joined: Wed Sep 21, 2011 4:13 pm
Location: Warsaw, Poland

abstrax wrote:Hi, thanks for the feedback. Unfortunately I can't reproduce your results here on 1 GTX 690 GPU: The Hallway PT gives 2.03 Ms/s and the Hallway PMC 1.67 Ms/s. A GTX 690 GPU should be a little bit slower than a GTX 680 GPU. -> I think there must be something else going on. Could you try the last WHQL certified driver (306.97)?

Cheers,
Marcus
Marcus,

I did tests again and something is strange. WHL 306.97 gives the same results on RC2 as the beta driver. But I couldn't reproduce results for RC1 on the beta driver - still 1.6Ms/s for the hallway scene (I had 2.31Ms/s). I'll try some more tests and let you know if anything changes.

PS. In PT (with one gpu, my second gpu is at RMA) my computer is so slow. I need to wait seconds to pull down menu or to change any values in Octane. In DL and PMC this is not the case. Isn't it strange?
3090, Titan, Quadro, Xeon Scalable Supermicro, 768GB RAM; Sketchup Pro, Classical Architecture.
Custom alloy powder coated laser cut cases, Autodesk metal-sheet 3D modelling.
build-log http://render.otoy.com/forum/viewtopic.php?f=9&t=42540
User avatar
smicha
Licensed Customer
Posts: 3151
Joined: Wed Sep 21, 2011 4:13 pm
Location: Warsaw, Poland

Results for RC3 on EVGA GTX 680 4GB watercooled OC
Attachments
rc3.jpg
3090, Titan, Quadro, Xeon Scalable Supermicro, 768GB RAM; Sketchup Pro, Classical Architecture.
Custom alloy powder coated laser cut cases, Autodesk metal-sheet 3D modelling.
build-log http://render.otoy.com/forum/viewtopic.php?f=9&t=42540
User avatar
abstrax
OctaneRender Team
Posts: 5509
Joined: Tue May 18, 2010 11:01 am
Location: Auckland, New Zealand

The hallway numbers still seem wrong to me. With RC3 I achieve on one 690 GPU on Windows 7 the following sample rates for the hallway scene:

PT: 2.10 Ms/s
PMC: 1.73 Ms/s
DL: 18.82 Ms/s

So something is going wrong in your system. How does it behave if you don't overclock? Did you disable Windows Aero, SLI and PhysX? The latter shouldn't make any differences, but you never know...

Cheers,
Marcus
In theory there is no difference between theory and practice. In practice there is. - Yogi Berra
Post Reply

Return to “Development Build Releases”