hi,
just to mention that i have installed a fresh copy of os x 10.7.4 on a secondary hd, with no toolkit and cuda drivers 4.2.10 and the results with 3.03/3.02x1/3.02x2 are exactly the same as with 10.7.5 and cuda 5.0.36.
ciao beppe
OctaneRender™ Standalone v1.0 beta3.02x [experimental]
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.
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.
Here we go, hope it helps
- Attachments
-
Win 7 64bits / Intel i5 750 @ 2.67Ghz / Geforce GTX 470 / 8GB Ram / 3DS Max 2012 64bits
http://proupinworks.blogspot.com/
http://proupinworks.blogspot.com/
I posted scores from the test scenes but on 3.03 for GTX 680 4GB here:
http://render.otoy.com/forum/viewtopic.php?f=33&t=24319
http://render.otoy.com/forum/viewtopic.php?f=33&t=24319
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
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
A big thanks to all who contributed benchmark results.
I think (/hope) we have found the issue for Linux and Mac OS. I rebuilt the two beta 3.02x version for Linux and Mac OS. All Linux/Mac users who are interested in the performance of Octane, could you please re-test with the new binaries?
Thanks a lot,
Marcus
I think (/hope) we have found the issue for Linux and Mac OS. I rebuilt the two beta 3.02x version for Linux and Mac OS. All Linux/Mac users who are interested in the performance of Octane, could you please re-test with the new binaries?
Thanks a lot,
Marcus
In theory there is no difference between theory and practice. In practice there is. - Yogi Berra
hi,
yeah, at a first look seems that the issue is gone
a bit better with 3.02x 1 then 2 but i will post all the results during the day
thanks a lot
ciao beppe
yeah, at a first look seems that the issue is gone
a bit better with 3.02x 1 then 2 but i will post all the results during the day
thanks a lot
ciao beppe
the results
ciao beppe
ciao beppe
- Attachments
-
v1rc1
chess-images
PT 4.56
PMC 3.14
DL 8.89
chess-proc
2.54
1.92
6.56
hallway
1.12
0.99
9.63
benchmark
1.78
1.44
6.15
screws-
2.75
1.32
4.90
roughly 2.5-3.5% slower than 3.03
chess-images
PT 4.56
PMC 3.14
DL 8.89
chess-proc
2.54
1.92
6.56
hallway
1.12
0.99
9.63
benchmark
1.78
1.44
6.15
screws-
2.75
1.32
4.90
roughly 2.5-3.5% slower than 3.03
i7-3820 @4.3Ghz | 24gb | Win7pro-64
GTS 250 display + 2 x GTX 780 cuda| driver 331.65
Octane v1.55
GTS 250 display + 2 x GTX 780 cuda| driver 331.65
Octane v1.55
Yes, this was a test release to see what the changes in render speed are due to the rewritten CUDA framework (which eventually allowed us to properly support Fermi and Kepler GPUs). To exclude the performance changes due to different shading due to the new BRDF, this version still used the old BRDF model.tmpdev wrote:Forgive my ignorance, but versions x1 and x2 both say on the splash screen (old BRDF). Is this right? Was there only a 3.03 version for the mac?
The latest release is 1.0 RC1, which you can find in the release candidates testing forum.
Cheers,
Marcus
In theory there is no difference between theory and practice. In practice there is. - Yogi Berra
v1rc3
chess-images
PT 4.92
PMC 3.44
DL 9.71
chess-proc
2.54
1.98
6.86
hallway
1.24
1.06
10.66
benchmark
1.79
1.51
6.37
screws-
2.90
1.43
5.11
pretty good speed. getting back near 2.59 levels and in some cases better.
here and there we have picked up better than 10 % from the lowest performing builds we had before thanks to our hardworking coders
the only obvious odd one out still is screws PMC which remains relatively slow I think.
chess-images
PT 4.92
PMC 3.44
DL 9.71
chess-proc
2.54
1.98
6.86
hallway
1.24
1.06
10.66
benchmark
1.79
1.51
6.37
screws-
2.90
1.43
5.11
pretty good speed. getting back near 2.59 levels and in some cases better.
here and there we have picked up better than 10 % from the lowest performing builds we had before thanks to our hardworking coders
the only obvious odd one out still is screws PMC which remains relatively slow I think.
i7-3820 @4.3Ghz | 24gb | Win7pro-64
GTS 250 display + 2 x GTX 780 cuda| driver 331.65
Octane v1.55
GTS 250 display + 2 x GTX 780 cuda| driver 331.65
Octane v1.55