Announcement

Collapse
No announcement yet.

NVIDIA Publishes Reference Documentation For Volta's Display Hardware

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

  • #21
    Originally posted by kpedersen View Post
    Though avoiding sh*t like Gnome 3 and Wayland which are over-reliant on an accelerated GPU also helps haha
    what idiot would buy volta if he does not need accelerated gpu?
    Last edited by pal666; 13 April 2018, 09:52 AM.

    Comment


    • #22
      Originally posted by starshipeleven View Post
      6.000 lines with registers for the Volta display hardware still better than Zero documentation, how NVidia usually does by default.

      Comment


      • #23
        Originally posted by starshipeleven View Post
        "Clickbait"
        No. The headline is not misleading, innacurate, or irrelevant to readers of this site.

        I think you're just butthurt from getting your hopes up after a cursory reading of the headline, or maybe you were just itching to use that fancy image you found.

        Save your criticism for something that deserves it. It's not exactly like there aren't plenty of more justifiable targets.

        Comment


        • #24
          Originally posted by Michael View Post
          I guess you can consider NVIDIA's own email clickbait then as it's titled "NVIDIA display hardware documentation update"
          Yeah, don't listen to him.

          Comment


          • #25
            Too bad you need nvidia for deep learning. Wonder when AMD will catch up with that?

            Comment


            • #26
              Originally posted by PeeJay View Post
              Too bad you need nvidia for deep learning. Wonder when AMD will catch up with that?
              They're supposed to have Vega fabbed on a 7 nm node, later this year (for machine learning; with mass market applications in 2019).

              https://www.anandtech.com/show/12233...nm-vega-on-7nm

              Vega slightly surpassed Pascal on training, and narrowed the gap on inferencing, thanks to its 2x fp16 and 2x int16 operations. But, of course, we all know about Volta...

              As for software, they have OpenMI. For that, you'll need the ROCm stack, for which the kernel support (on Vega) is still absent from mainline as of 4.17. Hopefully 4.18...
              Last edited by coder; 16 April 2018, 06:14 PM.

              Comment

              Working...
              X