Announcement

Collapse
No announcement yet.

oneAPI-Focused UXL Foundation Now Collaborating With The Khronos Group

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

  • #11
    Originally posted by Jumbotron View Post
    First of all yes, AMD should immediately cancel and drop ROCm even though they have a well earned reputation of developing initiatives and frameworks that do not garner industry wide support even though initiatives and frameworks showed technical merit at first. Dropping ROCm will not do much further damage to their reputation for that reputation has been earned over the last 25 years. oneAPI now has the big development momentum in the marketplace and has industry wide support through foundations and groups like UXL and Kronos. They as well as Intel have a credible plan and are excuting on that plan in a way that AMD never could do nor can they now. Hardware is nothing without software. The industry knows and accepts AMD as the best x86 CPU manufacturer and Xilinx now owned by AMD is the industry leading FPGA solution over Intel’s Altera. Also the industry has accepted that AMD’s Infinity Architecture is the heterogenous fabric of choice of tying together GPUs particularly across pods. But that’s where the industry support stops for AMD. When it comes to GPUs and software stacks to run those GPU workloads it’s Nvidia and CUDA….full stop. Not so much AMD and certainly not Intel at least with GPU tech as of now. That’s changing. Not so much with Intel suddenly becoming a GPU powerhouse even with buying out half of AMD’s GPU team a few years back. But as they make their REAL GPU play this year and going forward in the HPC and AI segment Intel will have a leg up in penetrating that hardware market with great compilers and software frameworks like oneAPI. Not to mention that oneAPI was designed not just for GPUs but all accelerators. oneAPI is the only credible competitor to CUDA….full stop. So what that El Capitan and Frontier are using AMD software. Those systems were designed YEARS ago and the only halfway credible solution at the time was AMD because this was pre-Pat Gelsinger Intel which was rapidly going into the toilet and perpetually stuck at 14nm+++++++++. By the time Capitan and Frontier are decommissioned oneAPI will be only non Nvidia solution left and even then it can run on any and all AMD CPUs and GPUs and FPGAs.
    As far as you poo-pooing foundations. Agreed…until there are only those left who have universal support by all industry players with the biggest of all being Intel. And Intel heads them all up. From interconnects to APIs to even endorsing AMD’s Infinity Architecture now called UALink which doesn’t care if ROCm is flowing through that fabric or oneAPI. Everybody in the industry that’s not named Nvidia are all in these groups…::
    CXL
    UXL
    oneAPI
    UAlink
    Ultra Ethernet
    And Intel heads three of the five with the exception of UAlink and Ultra Ethernet but Intel endorsed both so now the industry is comfortable supporting it and not just AMD and Cisco.
    This is the stack that is going to compete against Nvidia, CUDA, NVLink and Infiniband. With the above in mind….tell me with a straight face ROCm stands a chance of ever seeing widespread adoption in the industry. ROCm is the 3DNow! of compute.
    i do not share your vision because C++ was a language is death.

    i think in the long run Rust writen Vulkan Compute will take over.

    also keep in mind that CUDA is well established and ROCm/HIP is just a legal trick to allow ZLUDA​ to happen.

    to me its pretty clear that oneAPI/intel will lose big here because C++ is death.

    CUDA+ROCm/HIP+ZLUDA​ in the short time and Rust+Vulkan Compute in the long run will eat it alive.
    Phantom circuit Sequence Reducer Dyslexia

    Comment


    • #12
      Originally posted by qarium View Post

      i do not share your vision because C++ was a language is death.

      i think in the long run Rust writen Vulkan Compute will take over.

      also keep in mind that CUDA is well established and ROCm/HIP is just a legal trick to allow ZLUDA​ to happen.

      to me its pretty clear that oneAPI/intel will lose big here because C++ is death.

      CUDA+ROCm/HIP+ZLUDA​ in the short time and Rust+Vulkan Compute in the long run will eat it alive.
      As a fantasy I would agree. But the literally trillions of lines of C/C++ code in place right now not to mention being put into production going forward means your vision won’t ever have a hope of even halfway becoming a reality until well into the 2040s at the earliest.
      Last edited by Jumbotron; 14 June 2024, 02:09 PM.

      Comment


      • #13
        Originally posted by Jumbotron View Post
        As a fantasy I would agree. But the literally trillions of lines of C/C++ code in place right now not to mention being put into production going forward means your vision won’t ever have a hope of even halfway becoming a reality until well into the 2040s at the earliest.
        "literally trillions of lines of C/C++ code in place"

        you know that this is called code rot software decay



        they did not solve this problem yet. more code and a trillion lines of code does not make the situation better.
        Phantom circuit Sequence Reducer Dyslexia

        Comment


        • #14
          Originally posted by qarium View Post

          "literally trillions of lines of C/C++ code in place"

          you know that this is called code rot software decay



          they did not solve this problem yet. more code and a trillion lines of code does not make the situation better.
          What I meant was is that if all C/C++ projects stopped this instant and we went over to Rust or something else you would still have a huge viable market for C/C++ coders for decades simply for maintenance, analogous to COBOL programmers coming back for Y2K and beyond. But for the next few decades there is no real alternative to C++ in the realm of HPC and AI. Sad…but true because C++ is truly a hideously bad engineered language.

          Comment


          • #15
            Originally posted by Jumbotron View Post
            What I meant was is that if all C/C++ projects stopped this instant and we went over to Rust or something else you would still have a huge viable market for C/C++ coders for decades simply for maintenance, analogous to COBOL programmers coming back for Y2K and beyond. But for the next few decades there is no real alternative to C++ in the realm of HPC and AI. Sad…but true because C++ is truly a hideously bad engineered language.
            to this COBOL argument i can fully agree its "Sad" but true. but tell me is the basis of the COBOL argument really what you want to promote ?

            you literally say Intel products is for legacy only LOL. well Legacy is a big market of course.
            Phantom circuit Sequence Reducer Dyslexia

            Comment


            • #16
              Originally posted by Jumbotron View Post

              Remember 3DNow! as an alternative to Intel’s MMX ?
              3DNow! is an alternative to Intel’s SSE (SSE1 - SSE3). And 3DNow! was useful for some time.
              Now industry uses AMD64 with Intel's SSE & SSE2 plus Prefetch command from 3DNow! instruction set.
              Today Windows requires Prefetch command (from 3DNow! ).​


              Originally posted by Jumbotron View Post

              What I meant was is that if all C/C++ projects stopped this instant and we went over to Rust or something else you would still have a huge viable market for C/C++ coders for decades simply for maintenance, analogous to COBOL programmers coming back for Y2K and beyond. But for the next few decades there is no real alternative to C++ in the realm of HPC and AI. Sad…but true because C++ is truly a hideously bad engineered language.
              There are 3 languages for HPC computing: C, C++, Fortran (and all three are supported by Nvidia). And C/C++ for HPC use a lot of additions for HPC.
              Rust is good, but not for HPC.
              Fortran is better for HPC because now it becomes DSL for HPC, and participants can add needed HPC extensions in language standard because HPC is exclusive client.
              Last edited by Svyatko; 24 June 2024, 04:26 PM.

              Comment

              Working...
              X