Announcement

Collapse
No announcement yet.

Digging Deeper Into AMD's UVD Code Drop

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

  • #61
    Originally posted by bridgman View Post
    Interesting... that means FSF is effectively demanding that the open source community have the ability to buy less functional hardware (typically at a lower price) and update it to the more functional / more expensive hardware without paying the vendor, or at least saying "if we can't have it for free then you can't sell it to anyone else".
    I think most people would agree that this practice is dishonest anyway. The important thing to note here is that the hardware is NOT less functional. The hardware is the same, and only the firmware is different between the models. I think consumers are in their right to feel ripped off by intentionally gimped firmware.

    I understand the economics for doing so, but it's kind of incompatible with Free software.

    Comment


    • #62
      Originally posted by bridgman View Post
      Interesting... that means FSF is effectively demanding that the open source community have the ability to buy less functional hardware (typically at a lower price) and update it to the more functional / more expensive hardware without paying the vendor, or at least saying "if we can't have it for free then you can't sell it to anyone else".

      I understand that this is conceptually similar to the "anti-Tivo-ization" provisions of GPL v3 (at least from a distance) but seems like a much more blatant "functionality grab" ("all your extra-cost features are belong to us") than I'm used to seeing.

      That said, presumably the scenario you describe would require a microcode build that was specific to a single device (with some kind of internal unique ID) to prevent the "premium microcode" from being installed on another device and that is not what we are doing.
      It's simpler than that. The argument is you sold me hardware, it's not mine, thus I can do what I want with it including hacking he firmware to unlock new features. Of course the FSF would strongly prefer you gave them the source of the original firmware to make this job easier, but the vendor may see it differently for obvious buisness reasons.

      That said if the only differentiation between a workstation card with superior colour quality and a desktop card is a few lines of code in the firmware, I dare say the effort to gouge the workstation cards sorta sucks.

      Comment


      • #63
        Originally posted by ownagefool View Post
        It's simpler than that. The argument is you sold me hardware, it's not mine, thus I can do what I want with it including hacking he firmware to unlock new features. Of course the FSF would strongly prefer you gave them the source of the original firmware to make this job easier, but the vendor may see it differently for obvious buisness reasons.

        That said if the only differentiation between a workstation card with superior colour quality and a desktop card is a few lines of code in the firmware, I dare say the effort to gouge the workstation cards sorta sucks.
        The whole point is that microcode is part of the hardware we sold you, whether it be ROM-resident or RAM-resident.

        Are you saying that charging extra for additional functionality/performance is somehow bad ?
        Test signature

        Comment


        • #64
          Originally posted by benmoran View Post
          I think most people would agree that this practice is dishonest anyway. The important thing to note here is that the hardware is NOT less functional. The hardware is the same, and only the firmware is different between the models. I think consumers are in their right to feel ripped off by intentionally gimped firmware.
          But it's all OK if the firmware is stored in ROM rather than RAM ?
          Test signature

          Comment


          • #65
            Originally posted by bridgman View Post
            But it's all OK if the firmware is stored in ROM rather than RAM ?
            Yes. Then it's the company's loss for selling the same hardware for much less profit, with no way to make it use its full potential, not mine.

            If it was updateable, then it would be my loss if I bought a Radeon and could not turn it into a FirePro, if the only difference was the updateable firmware.

            Comment


            • #66
              Originally posted by curaga View Post
              Yes. Then it's the company's loss for selling the same hardware for much less profit, with no way to make it use its full potential, not mine.
              OK... so let's say we store the microcode in flash on the chip and program it just before we ship.

              Charging different prices for different levels of functionality would still be "dishonest", right ? Then we pop the write protect bit and it suddenly becomes fine ?

              Originally posted by curaga View Post
              If it was updateable, then it would be my loss if I bought a Radeon and could not turn it into a FirePro, if the only difference was the updateable firmware.
              ... even if we charge for the upgrade, it would somehow be "your loss" if you were unable to upgrade for free ?

              Can you explain that one to me without using "it costs you the same" because obviously it doesn't -- there's extra development and qualification work for the additional functionality which has to be worked into the selling price somehow. If we didn't offer a range of price/functionality options then everything would have to sell for a price somewhere in between consumer and workstation SKUs.

              Do you really feel "the world would be a better place" if consumer SKU prices went up 15-20% across the board but included workstation features & functionality ?
              Last edited by bridgman; 08 April 2013, 02:12 PM.
              Test signature

              Comment


              • #67
                Originally posted by bridgman View Post
                OK... so let's say we store the microcode in flash on the chip and program it just before we ship.

                Charging different prices for different levels of functionality would still be "dishonest", right ? Then we pop the write protect bit and it suddenly becomes fine ?
                It's instructive to think of this in more physical terms.

                Let's say the back of every GPU sold has a physical switch. Consumer cards have the switch defaulted to off, pro cards default to on.

                Otherwise, there is no difference, except cost.

                To use the card, you have to accept an EULA that says you aren't allowed to flip the switch.


                Now, i think a whole lot of people would simply flip that switch and ignore the EULA. And those that don't, would probably be pretty pissed off that they couldn't.

                But obviously i see the business reasons behind it.


                If there is no switch that can simply turn on the new functionality, then that's a completely different situation.
                Last edited by smitty3268; 08 April 2013, 02:30 PM.

                Comment


                • #68
                  Originally posted by bridgman View Post
                  Can you explain that one to me without using "it costs you the same" because obviously it doesn't -- there's extra development and qualification work for the additional functionality which has to be worked into the selling price somehow. If we didn't offer a range of price/functionality options then everything would have to sell for a price somewhere in between consumer and workstation SKUs.

                  Do you really feel "the world would be a better place" if consumer SKU prices went up 15-20% across the board but included workstation features & functionality ?
                  The obvious solution here is to provide special pro drivers that include the extra functionality and testing, and consumer drivers that do not.

                  And document things so that the OSS drivers can implement whatever they want. There is no extra testing that AMD has to pay for in that case, because it's 100% work done by the community.

                  Comment


                  • #69
                    Originally posted by smitty3268 View Post
                    The obvious solution here is to provide special pro drivers that include the extra functionality and testing, and consumer drivers that do not.
                    That's what we do... pro drivers run on pro hardware, and consumer drivers run on consumer hardware.
                    Test signature

                    Comment


                    • #70
                      Originally posted by bridgman View Post
                      That's what we do... pro drivers run on pro hardware, and consumer drivers run on consumer hardware.
                      You conveniently left out the last 3rd. Allowing 3rd party developers to do whatever they want.

                      Comment

                      Working...
                      X