Announcement

Collapse
No announcement yet.

NVIDIA CUDA 11.0 Released With Ampere Support, New Programming Features

Collapse
X
 
  • Filter
  • Time
  • Show
Clear All
new posts

  • NVIDIA CUDA 11.0 Released With Ampere Support, New Programming Features

    Phoronix: NVIDIA CUDA 11.0 Released With Ampere Support, New Programming Features

    NVIDIA appears to have quietly promoted CUDA 11.0 to its stable channel...

    http://www.phoronix.com/scan.php?pag...DA-11-Released

  • #2
    Poor Volta.

    Comment


    • #3
      Originally posted by eydee View Post
      Poor Volta.
      Poor Kepler (dropped) and Maxwell (deprecated).

      Comment


      • #4
        Still no GCC 10 / LLVM 10 support. What a shame.
        So you still need GCC 9 / LLVM 9 to make use of CUDA (or your distro has to provide 9 and 10 at same time).

        Crappy proprietary Nvidia sh**. Best thing would be, if the distros are brave enought to drop Nvidia CUDA (pakages).
        Last edited by Pranos; 07-08-2020, 05:12 AM.

        Comment


        • #5
          Originally posted by Setif View Post

          Poor Kepler (dropped) and Maxwell (deprecated).
          Do HPC setups still use those? (I honestly don't know.)
          Since they both lack tensor cores, I'm guessing they couldn't use CUDA 11's additions anyway.

          Comment


          • #6
            still hoping for one day nvidia opens up their drivers... at least the main kernel driver and let the community use mesa with it. similar to how amd did things. i understand the reasons nvidia has given in regards to why they haven't. its just hard to continue to accept them when we see what amd was able to accomplish.

            Comment


            • #7
              Originally posted by pieman View Post
              still hoping for one day nvidia opens up their drivers... at least the main kernel driver and let the community use mesa with it. similar to how amd did things. i understand the reasons nvidia has given in regards to why they haven't. its just hard to continue to accept them when we see what amd was able to accomplish.
              That's because AMD couldn't write their own for shit, did you ever use fglrx ?
              Last edited by Slartifartblast; 07-08-2020, 07:44 AM.

              Comment


              • #8
                Originally posted by Slartifartblast View Post

                That's because AMD couldn't write their own for shit, did you ever use fglrx ?
                fglrx was ATI. As i know, AMD had to rewrite most parts or the complete driver (even for Windows) because they didnt got the code/most of the code from ATI after they have bought it and ATI had no or worse documentations about their GPUs.
                So its not the fault from AMD.
                Last edited by Pranos; 07-08-2020, 09:57 AM.

                Comment


                • #9
                  It's actually now cuda_11.0.2 (with 450.51.05 as driver).

                  Before it was cuda_11.0.1 (with 450.36.06 as driver).

                  Comment


                  • #10
                    Originally posted by Setif View Post
                    Poor ... Maxwell (deprecated).
                    AFAICT, only sm_50 is deprecated, which is limited to Maxwell Gen1 (GTX 750 and equivalent Tesla/Quadro GPU's).

                    https://developer.nvidia.com/cuda-gpus#collapse4
                    http://arnon.dk/matching-sm-architec...-nvidia-cards/

                    Comment

                    Working...
                    X