2x 580GTX better on 3.0 or 3.2 ?
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.
Well I just got 2x 580GTX and with the new release I wondering is it better to run the Cuda 3.2 or the Cuda 3.0 release ?
you should stay with 3.0 so you can use both of them and wait for 1.25 for the 20% extra
ciao beppe

ciao beppe
this i really sad, and i hope this would be resolved soon, am reluctant to upgradebepeg4d wrote:you should stay with 3.0 so you can use both of them and wait for 1.25 for the 20% extra![]()
ciao beppe

gigabyte mobo, core2quad, 8gb ram, x2 pny gtx 580,
hp pavilion hdx series(desktop replacement) 4gb ram, gt 8800
win7 x64bit, octane 2.3 v5
hp pavilion hdx series(desktop replacement) 4gb ram, gt 8800
win7 x64bit, octane 2.3 v5
Well hm...tobad, well atleast I can update my signature ! 

We have explained the situation in the release notes, but why are you reluctant to upgrade. Beta 2.42 CUDA 3.0 is not slower than beta 2.3 v5.anomi wrote:this i really sad, and i hope this would be resolved soon, am reluctant to upgrade
Cheers,
Marcus
In theory there is no difference between theory and practice. In practice there is. - Yogi Berra
Sorry if this is answered somewhere already but I didnt find it, do I need the Cuda 3.0 drivers installed for the Octane Cuda 3.0 version to work with 2 Cards ?
Because I have 266.58 (released Jan 18) installed and with that came Cuda 3.2.1, and starting both Octane Cuda ver 3.0 and 3.2 says both cards are active.
Again, sorry if it already was answered somewhere.
Because I have 266.58 (released Jan 18) installed and with that came Cuda 3.2.1, and starting both Octane Cuda ver 3.0 and 3.2 says both cards are active.
Again, sorry if it already was answered somewhere.
Every program uses the versions which are involved with it.
You can have a DX11 card, but if the game uses DX9, you have DX9.
You want MultiGPU? Then use Octane for version cuda3.0.
In that version, the cudart.dll v3.0 is in it and Octane use that.
face
You can have a DX11 card, but if the game uses DX9, you have DX9.
You want MultiGPU? Then use Octane for version cuda3.0.
In that version, the cudart.dll v3.0 is in it and Octane use that.
face
Win10 Pro, Driver 378.78, Softimage 2015SP2 & Octane 3.05 RC1,
64GB Ram, i7-6950X, GTX1080TI 11GB
http://vimeo.com/user2509578
64GB Ram, i7-6950X, GTX1080TI 11GB
http://vimeo.com/user2509578
Well yes I know I was just perplexed since both Octane with Cuda 3.0 and Cuda 3.2 reports both my 580GTX being active. And getting the exact same render speeds. I have SLI disabled etc. But hey it's still faster than previous versions so..face wrote:Every program uses the versions which are involved with it.
You can have a DX11 card, but if the game uses DX9, you have DX9.
You want MultiGPU? Then use Octane with cuda3.0...
face

Ok, thats strange, because 2x 2/3 cores should be faster than 1x 3/3 coresMagnusL3D wrote:Well yes I know I was just perplexed since both Octane with Cuda 3.0 and Cuda 3.2 reports both my 580GTX being active. And getting the exact same render speeds. I have SLI disabled etc. But hey it's still faster than previous versions so..face wrote:Every program uses the versions which are involved with it.
You can have a DX11 card, but if the game uses DX9, you have DX9.
You want MultiGPU? Then use Octane with cuda3.0...
face

face
Win10 Pro, Driver 378.78, Softimage 2015SP2 & Octane 3.05 RC1,
64GB Ram, i7-6950X, GTX1080TI 11GB
http://vimeo.com/user2509578
64GB Ram, i7-6950X, GTX1080TI 11GB
http://vimeo.com/user2509578
Yup that's why I'm wondering because I must have certainly done something wrong
Also GPU-Z reports Card 1 at 90% load and Card 2 at around 15-20% load with both the Cuda 3.0 and 3.2 version. Maybe I should just restart the computer between installing new drivers and disabling SLI ? 

