Announcement

Collapse
No announcement yet.

AMD Will Release Mantle Programming Guide, API Reference This Month

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

  • AMD Will Release Mantle Programming Guide, API Reference This Month

    Phoronix: AMD Will Release Mantle Programming Guide, API Reference This Month

    This month AMD is planning to finally make their Mantle graphics API more "open" by releasing a 450-page programming guide and API reference for Mantle...

    Phoronix, Linux Hardware Reviews, Linux hardware benchmarks, Linux server benchmarks, Linux benchmarking, Desktop Linux, Linux performance, Open Source graphics, Linux How To, Ubuntu benchmarks, Ubuntu hardware, Phoronix Test Suite

  • #2
    RIP Mantle, useless single vendor effort. 3dnow was a failure, but at least x86-64 was widely adopted.

    Comment


    • #3
      Point 2) is the important one in the blog.

      It indeed sounds like a lot of Mantle is going into glNext/Vulcan.

      Comment


      • #4
        ROFL. Mantle now even more open!

        Comment


        • #5
          Originally posted by Kano View Post
          RIP Mantle, useless single vendor effort.
          Great ideas sometimes came from the single vendor effort.

          3dnow was a failure, but at least x86-64 was widely adopted.
          Well, part of it 3dnowprefetch is supported on new Intel CPUs too.

          Comment


          • #6
            Originally posted by Kano View Post
            RIP Mantle, useless single vendor effort. 3dnow was a failure, but at least x86-64 was widely adopted.
            accumulate, min, max, sqrt
            idk why people trash 3dnow so much, i think it was good

            Comment


            • #7
              3dnow was used by very few games like Quake to get a similar speed compared to Pentium 2. It was before the rise of the Athlon, but AMD basically never gained the speed trophy after Core 2 Quad was out. Mantle was the one and only way to let their hardware shine on their own slow CPUs. If they wanted to shine in more than a handful games they would have talked to Nvidia/Intel before. It was a marketing gag, not more.

              Comment


              • #8
                Originally posted by Kano View Post
                Mantle was the one and only way to let their hardware shine on their own slow CPUs.
                So what, AMD working on their hardware that sounds normal

                S3 again working on their hardware (Savage chip i think) first iplemented s3tc, to save bandwidth and vram usage and it is still used even today .

                Comment


                • #9
                  Nothing is against similar efforts like DX12 or GLnext (call it Vulcan if you want). But: Mantle was never designed to make Nvidia gfx faster or Intel. It was a filler between the real solution, nothing more. They needed it just more early than others...

                  Comment


                  • #10
                    Originally posted by Kano View Post
                    Mantle was never designed to make Nvidia gfx faster or Intel.
                    Is there something Nvidia designed so that directly make AMD cards go faster? Each vendor work on their hardware and if it is usefull for other vendors they will consider adopting it.

                    Part of Mantle design will go into DX12 and Vulkan, we can thank AMD for that effort or no.

                    They needed it just more early than others...
                    So, if nvidia or intel need something more early then others, will they wait for others? Most likely no, they will just implement it for their hardware first of course, just like AMD do.
                    Last edited by dungeon; 02 March 2015, 03:02 PM.

                    Comment

                    Working...
                    X