I was super excited when i heard the new cards had nvlink, and Blago (i think?) even suggested that 2 cards =22gb gpu ram.
However today i see a review on guru3d which casually states that on this version of nvlink, memory sharing is disabled. Extremely dissapointing if so. Ahh well back to my trusty old cheap 12 gb titan x’es.
I am pretty sure guru3d are wrong.
Best,
Blago.
Well i expect of anyone has a clear idea of final spec i imagine it would be you guys, so thats reassuring 
i fear they are right.
They are not.
We will have a blog post very soon to end up the speculations.
Best,
Blago.
Thanks for clearing that up Blago.
The Puget guys are wrong. We have tested NVLink here and it does work and it allows you to increase the memory available for GPU rendering with V-Ray GPU. Of course, the software that you use must be able to take advantage of this specifically - it doesn’t happen automatically for everything. For the moment we are not aware of any software other than V-Ray GPU that can use it, so no wonder the Puget guys couldn’t figure out what to do with it. I’m sure this will change with time.
Best regards,
Vlado
Awesome news!!!
Also can’t wait for the blog post Blago mentioned. Looking forward to that
Really cool!
Cheers!
I’m going to buy a new custom build workstation soon and may have the money to install two RTX 2080 ti. If I understand you guys right, it’s best to connect them via NVLink to share memory - right? And what about active shade/ipr - is it possible to split the 2 cards just for the previews? Or is it not necessary?
Cheers!
Still don’t know if nvlink will share memory for the consumer level cards chaos has only tested pre released quadros.
We have tested 2080 and 2080ti, NVLINK works okay. There are few setup things that one should consider, blogpost about the topic will be released soon.
Best,
Blago.
I have these cards in my cart just waiting for your blog 
Long story short - as long as you can enable SLI, the GeForce NVLINK RTX will work on Windows https://www.chaosgroup.com/blog/prof…idia-rtx-cards
Best,
Blago.
Great news, thanks for the info Blago!
awesome… what sort of ETA can you give us on RT core support in a proper release of vray?
As there is no even public NVIDIA SDK that can support RT Cores with CUDA, we can’t really give any ETA yet.
Best,
Blago.
I have a setup right not with 4x 2070 rtx cards . I do run out of memory in some of my scenes and I would love to hear your suggestions on the best solution.
Could i for example add one quadro rtx 8000 with 48 gb ram to fix these issues? Or would that leave me only able to render with one card in those heavy scenes and the others idle?
I haven’t really understood how the video card ram works and would love for someone to explain it to me
If you want to use the maximum vram and only one card has that amount, you can only render with this one card. Only nvlink doubles your vram. And it does so only once. Can’t connect 4 cards together to get four times the amount.
Thank you Manuel!