CUDA errors

[samsue]

Hi samsue,
first: using Nvlink you have tu set on Out of core.
second: keep in mind Nvidia removed 2 x 2 Nvlink support (geforce and quadro) so it doesn't work on your slave.
In your case, you would unplug the second Nvlink, then set Out-of-core during slave setup.
You will be able to use Nvlink only using 2 GPUs master + 2 GPUs slave; you cannot use Nvlink using all of 6 GPUs.
If you really want to use Nvlink, instead of out-of-core only, you shoud have a secondary 2x GPU slave.
I hope it helps.
Rgards
Paride

I was reviewing some topics, I decided to test in my Workstation, SLI to have vRam pool with nvlink ( 2x 2080 Ti 11 gb so 22gb vram )

1. Is the vraym pool working?

2. You said that if I'm using nvlink, I have to activate the out of core, but so Octane uses cpu ram

[paride4331]

Hi samsue,
first: using Nvlink you have tu set on Out of core.
second: keep in mind Nvidia removed 2 x 2 Nvlink support (geforce and quadro) so it doesn't work on your slave.
In your case, you would unplug the second Nvlink, then set Out-of-core during slave setup.
You will be able to use Nvlink only using 2 GPUs master + 2 GPUs slave; you cannot use Nvlink using all of 6 GPUs.
If you really want to use Nvlink, instead of out-of-core only, you shoud have a secondary 2x GPU slave.
I hope it helps.
Rgards
Paride

I was reviewing some topics, I decided to test in my Workstation, SLI to have vRam pool with nvlink ( 2x 2080 Ti 11 gb so 22gb vram )

1. Is the vraym pool working?

2. You said that if I'm using nvlink, I have to activate the out of core, but so Octane uses cpu ram

3dsmax_SEkR343oUI.png

Hello samsue,
don't forget Nvlink works with geometryes only, the textures will go to Ram.
Regards
Paride

[samsue]

any idea? on my workstation works fine, but the server, crashs always

[samsue]

can these problems be related to a bad video card?

-----

bad video card