Announcement

Collapse
No announcement yet.

AMD To Exploit OpenCL Potential In LibreOffice

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

  • #51
    Originally posted by Nille View Post
    I never see an Statement that they don't want do it.



    And again. The Blender Guys blame AMD, but for me its not clear if its really an fault in AMDs OpenCL Implementation or that Nvidias and Intels Implementation allow an not Specified Scenario.

    And last as my PoV as an User. Then the Blender Guys have to implement an Workaround that it will work and stop crying like little pussys that AMD stole there candys.
    This. 2+ years of BS. Seriously, stop it.

    There is legal open source code. In svn. There are pletny of open discussions on mail lists, blenderartists.org, etc. And someone step up and again call us "crying like little pussys".

    Prove it. Show me 1 statement in Blender Cycles that is broking any OpenCL standard in Blender or SHUT UP. You can find all info in blender.org, in main svn trunk.

    AMD, FIX YOUR SHITTY COMPILER AT LAST AND KEEP YOUR PAYD FORUM MONKEYS ON SHORT LEASH.

    Comment


    • #52
      Originally posted by storm_st View Post
      AMD, FIX YOUR SHITTY COMPILER AT LAST AND KEEP YOUR PAYD FORUM MONKEYS ON SHORT LEASH.
      And you Cry again ... i have already say thats its not clear for me on which side the fault is. Btw in this case blender has to Prove that AMDs Implementation is wrong....

      Under the Line i don"t care who is wrong! it has to work! So if its really AMDs fault try to find a workaround so that it work and try to get in contact with AMD. Everything that i currently found about this are the blender people(? or only fanboys? ) that blame AMD.
      Last edited by Nille; 06 July 2013, 08:45 AM.

      Comment


      • #53
        Originally posted by curaga View Post
        Very much situation dependant. I'd be inclined to think it increases power use for the spreadsheet scenario, having to power up more shader cores, move things to VRAM.
        HSA are APUs. So there is the same memory for GPU and CPU. E.g. the Current x264 OpenCL work is on an APU faster as with an extra Card.

        Comment


        • #54
          Originally posted by Nille View Post
          i have already say thats its not clear for me on which side the fault is.
          So, you dont have a clue? Great brave position of real man. BTW, it is not clear to me and maybe other readers, do you eating shit ot killing kittens, I must say that is bad, dont do that!

          Comment


          • #55
            You don't has an Point and start flaming?

            Comment


            • #56
              Originally posted by Nille View Post
              And you Cry again ... i have already say thats its not clear for me on which side the fault is. Btw in this case blender has to Prove that AMDs Implementation is wrong....

              Under the Line i don"t care who is wrong! it has to work! So if its really AMDs fault try to find a workaround so that it work and try to get in contact with AMD. Everything that i currently found about this are the blender people(? or only fanboys? ) that blame AMD.
              Go to devgurus.amd.com and you will find more examples of the problem.

              The code has no problems with Opencl standard so it's amd's fault.

              Vray is a good example of a render that cannot use opencl cause this bug, yes there's a vray for amd version... but it lacks a lot of features that you can achieve with cpu, or nvidia cards.

              The way vray has solved the problem is reducing the size just including less features to the render, blender can compile too if you go and deactivate almost everything, another workaround is just chop all cycles in little pieces and pray to monesvol that the amd compiler didn't choke to dead while compiling (a lot of work taking into account that nobody cant ensure that it'll work at the end).

              Comment


              • #57
                Originally posted by elanthis View Post
                Oh my lord don't argue about stupid stuff. Apparently you missed out on the rising popularity of hyperbole. He isn't trying to literally say only 3 people use Blender, only pointing out the vast difference in numbers of users between a general-purpose software suite and a specialized tool.
                Of course I had understood that, and as I had explained in a other message, is not a valid argument.
                Considering users installing the full suite of LibreOffice, a minority of them actually need enhanced with OpenCL. Instead, GPU rendering is always good for Cycles. So if you compare a minority who need OpenCL for LibreOffice with the large majority who would benefit in Blender with OpenCL, I really do not think that the difference is so big.

                Comment


                • #58
                  Originally posted by curaga View Post
                  Just like autovectorization, once cpus have fully integrated gpu units with unified address space and no penalty for using the gpu, it's the job of the compiler to make use of it with existing code.

                  It's irresponsible to require random app foo developers to learn a domain-specific language for a minority use case, when the compiler should be able to do a good enough job for them.
                  OK, where's the compiler that can or will be able to do this on a 32bit system? There are already cases where projects can't be compiled on S390 (31-bit) due to low ram, and a couple years ago turning on PGO in Firefox was going to force cross-compiling the x86 binaries from amd64.

                  Comment


                  • #59
                    Originally posted by Ibidem View Post
                    OK, where's the compiler that can or will be able to do this on a 32bit system? There are already cases where projects can't be compiled on S390 (31-bit) due to low ram, and a couple years ago turning on PGO in Firefox was going to force cross-compiling the x86 binaries from amd64.
                    Huh, what does 32-bitness have to do with the issue?

                    Comment


                    • #60
                      Bettter replace VCL with something like Qt.

                      Comment

                      Working...
                      X