Hi,
I have 1xEVGA 980ti hybrid and 1x Asus 1080 founders edition.
If I compare the render times between the 2 cards, the results are very similar...The 1080 is just 10-15 seconds faster on an 8 min scene render.
I have checked the temps and clock speeds and everything is working at full speed.
Also tried with Octane 3.05 and 3.06 that uses CUDA 8 and the results are very similar.
Do you guys have the same issue, or it's just me? Maybe Octane is not optimized for Pascal yet?
980ti and 1080 almost the same performance in Octane?
Forum rules
Please add your OS and Hardware Configuration in your signature, it makes it easier for us to help you analyze problems. Example: Win 7 64 | Geforce GTX680 | i7 3770 | 16GB
Please add your OS and Hardware Configuration in your signature, it makes it easier for us to help you analyze problems. Example: Win 7 64 | Geforce GTX680 | i7 3770 | 16GB
Win 10x64, AMD 1950x 16C/32T, RAM 32Gb, 4x EVGA GTX 1080Ti, Octane-for-C4D 4.x.x, Nvidia 398.82
Hi Yan,
quoting from v3.06 release note:
quoting from v3.06 release note:
ciao beppeComplete switch to CUDA 8
We completely switched the kernel compilation to CUDA 8. We had to make quite some changes to work around kernel crashes caused by CUDA 8 and to avoid slowdowns caused by the new toolkit. After all these changes were made rendering became a bit faster in most scenes. The difference is not tremendous (2-4%), but there are some cases where is made a larger difference (5-10%).
Please let us know if there are more or less issues with kernel crashes and if you experience any major slowdowns with specific scenes.
I don't think that should surprise & the reason is pretty simple: You're are running 980Ti in ideal conditions (cooled up properly) so it is able to boos accordingly. now 1080 in Your case is air-cooled & thus it's probably not boosting under full load. 980Ti should run between 110 to 140 depending on cooling - what are You getting? Then 1080 should score somewhere close to 140 on hacked bench, again, what are You getting?
All in all You're comparing top end card from Maxwell range to mid range for Pascal - their performance is very similar, so do not hope for too much of difference. It's the same story as we had with 580 vs 680 or 780 vs 980. bottom line do not compare cards side by side - just because it's new it doesn't have to be faster & feel free to blame nvidia for their naming schemes. however doing a bit of research before buying would help.
All in all You're comparing top end card from Maxwell range to mid range for Pascal - their performance is very similar, so do not hope for too much of difference. It's the same story as we had with 580 vs 680 or 780 vs 980. bottom line do not compare cards side by side - just because it's new it doesn't have to be faster & feel free to blame nvidia for their naming schemes. however doing a bit of research before buying would help.
Thank you for your replies.
The 1080 scores 150 in hacked benchmark and the 980ti 140. They are both a bit oberclocked.
I knew at first that 1080 is close to 980ti, but that was some time ago when Octane was starting to work with Pascal.
I was hoping for some optimization especially on CUDA 8 with the new release, but from my tests the speed is the same.
It seems that the video card itself is close to the 980ti in any application.
I'll stick on buying just ti version cards from now.
The 1080 scores 150 in hacked benchmark and the 980ti 140. They are both a bit oberclocked.
I knew at first that 1080 is close to 980ti, but that was some time ago when Octane was starting to work with Pascal.
I was hoping for some optimization especially on CUDA 8 with the new release, but from my tests the speed is the same.
It seems that the video card itself is close to the 980ti in any application.
I'll stick on buying just ti version cards from now.
Win 10x64, AMD 1950x 16C/32T, RAM 32Gb, 4x EVGA GTX 1080Ti, Octane-for-C4D 4.x.x, Nvidia 398.82