Announcement

Collapse
No announcement yet.

new video card advice

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

  • new video card advice

    Hello.
    I'm looking for a new video card. The main demands I will have for this card is HDTV/Bluray playback (vdpau support?), and also I want to do some experimenting with CUDA/OpenCL programming. I also want the card to be quiet, preferably fanless, and since my favorite games are a few years old by now I can probably go with a less expensive card from the current generation.

    I've been looking in the $70-$150 price range, maybe something like the nvidia 220 or 9800GT ? I don't know if I should consider ATI cards, all my current cards are nvidia so I'm more familiar with configuring them. Do ATI cards support vdpau and cuda programs, or something equivalent?

  • #2
    ATI cards don't support CUDA as far as i'm aware, since it's nVidia's 'thing'. This should be a good choice:

    Buy Galaxy GeForce GT 240 1GB DDR3 PCI Express 2.0 x16 Video Card 24GGS8HX2PUX with fast shipping and top-rated customer service. Once you know, you Newegg!


    If you really want it to be fanless:

    Buy ARCTIC COOLING Accelero S1 Rev.2 VGA Cooler with fast shipping and top-rated customer service. Once you know, you Newegg!

    Comment


    • #3
      Originally posted by Superewza View Post
      ATI cards don't support CUDA as far as i'm aware, since it's nVidia's 'thing'. This should be a good choice:

      Buy Galaxy GeForce GT 240 1GB DDR3 PCI Express 2.0 x16 Video Card 24GGS8HX2PUX with fast shipping and top-rated customer service. Once you know, you Newegg!


      If you really want it to be fanless:

      http://www.newegg.com/Product/Produc...82E16835186016
      AMD cards *DO* support OpenCL, which is the OPEN version of nvidia's CLOSED CUDA.

      Comment


      • #4
        Originally posted by droidhacker View Post
        AMD cards *DO* support OpenCL, which is the OPEN version of nvidia's CLOSED CUDA.
        Not entirely accurate. openCL is a GPU computing standard yes but to say it's a "open version" isn't really true. They are two separate products that try to achieve the same thing. You wouldn't call GCC for example a "open version of Visual Studio"

        Comment


        • #5
          If accelerated HD decoding is important to you, nvidia is the better choice now. XvBA can do it using ATi proprietary drivers, but the guy developing the VA-API interface gave up on the whole thing.

          The CUDA stuff can be done using OpenCL on ATi hardware, so that's not a problem.

          Comment


          • #6
            Originally posted by deanjo View Post
            Not entirely accurate. openCL is a GPU computing standard yes but to say it's a "open version" isn't really true. They are two separate products that try to achieve the same thing. You wouldn't call GCC for example a "open version of Visual Studio"
            OPEN as in OPEN FOR ALL TO USE, whereas nvidia's implementation is, as typical for nvidia, nvidia-only and blob-only, which makes the whole thing worthless.

            Comment


            • #7
              Originally posted by droidhacker View Post
              OPEN as in OPEN FOR ALL TO USE, whereas nvidia's implementation is, as typical for nvidia, nvidia-only and blob-only, which makes the whole thing worthless.
              Hardly worthless considering the number of Cuda enabled apps out there dwarf openCL/ Stream / DirectCompute apps all combined. Cuda also carries features that openCL does not have plus the documentation and resources available for Cuda exceeds all others.

              Comment


              • #8
                My beef with your comment was the word "version" as openCL is not a "version" of Cuda.

                Comment


                • #9
                  Had you said:

                  AMD cards *DO* support OpenCL, which is an open, royalty free cross platform standard for parallel computing
                  Then that would have been accurate.

                  Comment


                  • #10
                    You're picking on words without spending the time to understand the meaning/context/purpose.

                    OpenCL and CUDA are both subsets (i.e. implementations/versions/etc.) of "USE GPU AS GENERAL-ish PURPOSE PROCESSOR".

                    Is it that hard to understand?

                    Or you can look at it from a different perspective.... CUDA is a version of OpenCL. That might make more sense in the context since CUDA is a restricted proprietary implementation of what should naturally be open. Using the word "version" not to mean "subset", but to mean "interpretation" or "take on".

                    You know how tiring it is dealing with you language nazis? Spend more time worrying about the intention. We're not writing contracts here.

                    And FYI: the number of "CUDA APPLICATIONS" is irrelevant if most machines CAN'T RUN THEM!!! It is strictly MORONIC to write software that is restricted to a particular vendor's hardware! And more than that, to a particular set of drivers for that hardware!

                    Looking into a possible future, openCL has a G3D state tracker, which might hopefully one day mean openCL on intel, AMD, **AND** nvidia OPEN SOURCE drivers, as well as the current AMD blob implementation. That pretty much covers EVERYBODY EXCEPT nvidia blob users. Much more freeing, eh?

                    Comment

                    Working...
                    X