Tweet
Hello good folks,
I'm looking for some hands on impressions of users using NVLink with either Geforce or Quadro GPUs. Given the situation with the Ampere launch shortages I'm thinking of potentially NVLinking my 2x 2080tis in my workstation. The rest can potentially go out of core but I'd be really great to have at least a pair NVLinked.
I guess at this point I'm just wondering how NVLink in V-Ray is looking like. Is it stable? Does it enable at least somewhat close to 2x VRAM buffer? Do you need to split meshes so each chunk can fit onto a GPU (f.e if you have a single 14GB mesh, does it get split over onto the other GPU?)? Any and all other impressions
Any and all input would be kindly appreciated
I'm looking for some hands on impressions of users using NVLink with either Geforce or Quadro GPUs. Given the situation with the Ampere launch shortages I'm thinking of potentially NVLinking my 2x 2080tis in my workstation. The rest can potentially go out of core but I'd be really great to have at least a pair NVLinked.
I guess at this point I'm just wondering how NVLink in V-Ray is looking like. Is it stable? Does it enable at least somewhat close to 2x VRAM buffer? Do you need to split meshes so each chunk can fit onto a GPU (f.e if you have a single 14GB mesh, does it get split over onto the other GPU?)? Any and all other impressions
Any and all input would be kindly appreciated
