Announcement

Collapse
No announcement yet.

Mesa Now Supports A Bit More Of OpenGL 3.0

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

  • #11
    Originally posted by elanthis View Post
    OpenGL isn't some isolated graphics standard. It's a specification for allowing developers to access the features found in actual, real hardware. That's it. The fact that some of those features are patented is highly unfortunate, but we all expect OpenGL to fully expose the hardware capabilities, because otherwise OpenGL is immediately useless and the few holdouts remaining will all just jump to the Direct3D bandwagon.

    The whole reason people are complaining about these patents is because we actually fucking need those features, not because we want to have Mesa change a single version number in its interface. A version number which is almost meaningless because you can implement the entirety of OpenGL 4.1 in an OpenGL 2.0 driver thanks to the extensions feature in OpenGL. There's no reason that Mesa can't have every single non-patented OpenGL feature in its core today other than a lack of manpower. The version number is a convenience at absolute most.
    If this is the case, then when people buy the "actual, real hardware" then they have an implied license to use any technology in that hardware.


    "Implied licenses often arise where the licensee has purchased a physical embodiment of some intellectual property belonging to the licensor."

    Given that users, and even FOSS developers, who have purchased a graphics card, which is a physical embodiment of some intellectual property, have an implied license to use that selfsame intellectual property, then there should be no issue in writng support for accessing that selfsame intellectual property in layers of FOSS software which sit above the hardware. The intellectual property is implemented in the hardware, so IP owners asking for royalties from software projects for layers above the hardware would be double-dipping.

    Comment


    • #12
      (software) patents suck!

      Originally posted by pingufunkybeat View Post
      Aren't software patents great? It's an open standard which you are not allowed to implement.
      Indeed. Fortunately they don't apply here in Europe and if I may, I humbly suggest any European residents present on this forum to sign this petition to help making sure it stays that way. Poland even completely rejected such a concept which makes me seriously consider moving here if those incompetent morons in our government sanction it (and I gotta say they've proven beyond any doubt they're much more than enough corrupt and stupid to do so).
      I hope Adarion doesn't mind if I quote his "Stop TCPA, stupid software patents and corrupt politicians!"

      Comment


      • #13
        Very frustrating

        Originally posted by hal2k1 View Post
        If this is the case, then when people buy the "actual, real hardware" then they have an implied license to use any technology in that hardware.


        "Implied licenses often arise where the licensee has purchased a physical embodiment of some intellectual property belonging to the licensor."

        Given that users, and even FOSS developers, who have purchased a graphics card, which is a physical embodiment of some intellectual property, have an implied license to use that selfsame intellectual property, then there should be no issue in writng support for accessing that selfsame intellectual property in layers of FOSS software which sit above the hardware. The intellectual property is implemented in the hardware, so IP owners asking for royalties from software projects for layers above the hardware would be double-dipping.
        Anything that get's implemented in Mesa will be available through the software driver (softpipe or llvmpipe) which means there is no real hardware behind it.

        However, in this case, a court has already ruled in ATI's favor that implementing this support in their driver did not infringe upon the patent, so it's extremely irritating that Mesa won't add it to theirs. Their stance seems to be, even though we know this isn't patented, we'll probably be sued anyway so we won't bother with it.

        At this point, i think we only have 2 hopes.

        1. Intel realizes this is important as their hardware gets faster and makes it a priority.

        2. Some dev realizes what a disaster this is and hosts their own version of Mesa somewhere with all these changes, and publishes it around so that people are aware of where to get it. I don't really mean fork mesa, but just keep a patch set up to date with the latest Mesa that people can easily use.

        Comment


        • #14
          Originally posted by pingufunkybeat View Post
          Personally, I'm not worried about the patent situation in terms of usability. If the functionality is available in a separate branch, we can still fetch that (in the free world) and compile it, just like we do with the freetype stuff and LAME.

          What is worrying me is that some of this patented stuff will end up not being developed at all, because nobody wants to risk it.
          That is soo funny.
          You just don't understand how monopolization works. You could OUTLAW by pain of death the implimentation of closed standards and people would create more of them than you can shake a stick at.
          This is how it works in the real world. Develop closed standard pass it out. Let people get used to it and once it's sufficiently dominant start rapidly and heavily enforcing patent on it. That's the software side. The hardware side is bit more difficult. They impliment closed standard, supplement it to make it more attractive then jack up license and chip costs. Intel is CONSTANTLY doing this to memory. They did it with RDRAM and they continue doing it with all their little goofy sever drams. Goal complete control of ram market, goal complete control of competitive advantage, our systems cost -15 bucks your systems cost +15 bucks. It never works so new tact. You should be hearing about it before too long. Intel has been a very naughty boy.

          All laws are based on force and enforceability. If 50 percent of the people will not willingly follow it is a bad law. If 66 percent of people willingly accept an inferior position through laws you can start developing a "normalcy bias" in peoples thinkings. If you can get 83 percent to accept it you can develope a very strong normalcy bias and completely screw peoples heads off into lala land making unacceptable things so deeply entrenched that people will not fight back against it. The use of force and the basis of authority and monopolization of punishment or violence into one groups hands has enjoyed a 50 percent monopoly for most of human history. A legtimatized authoritarian regime. They are shooting for 83 percent but have lost huge market share.

          Comment


          • #15
            Originally posted by elanthis View Post
            Nope. Open means that the standard is available to all, and that the standard itself is not proprietary. The fact that something in that standard is patented is an unfortunate coincidence, not intentional.
            I find that extremelly hard to believe. Also, I don't grant standardization bodies a monopoly on the use of the word "open". Actually, the doublespeak definition of "open standard" that you refer to is by no means what institutions such as the EU and other countries across the globe understand by that term.

            Comment


            • #16
              Originally posted by yotambien View Post
              I find that extremelly hard to believe. Also, I don't grant standardization bodies a monopoly on the use of the word "open". Actually, the doublespeak definition of "open standard" that you refer to is by no means what institutions such as the EU and other countries across the globe understand by that term.
              Indeed. Microsoft's definition of "open standard" is interesting:

              Vijay Kapoor, national technology officer, Microsoft, defines what open standards are as follows: "Let's look at what an open standard means: 'open' refers to it being royalty-free, while 'standard' means a technology approved by formalised committees that are open to participation by all interested parties and operate on a consensus basis."

              Comment


              • #17
                Originally posted by hal2k1 View Post
                Indeed. Microsoft's definition of "open standard" is interesting:

                Vijay Kapoor, national technology officer, Microsoft, defines what open standards are as follows: "Let's look at what an open standard means: 'open' refers to it being royalty-free, while 'standard' means a technology approved by formalised committees that are open to participation by all interested parties and operate on a consensus basis."
                Translation: You are free to do what we tell you to do and you are not free to do what we don't tell you to do or tell you not to do.

                Comment


                • #18
                  Originally posted by droidhacker View Post
                  Translation: You are free to do what we tell you to do and you are not free to do what we don't tell you to do or tell you not to do.
                  That's quite the way Microshaft does things. They don't want their customers to have free choice

                  Comment


                  • #19
                    Originally posted by elanthis View Post
                    About 13 more years for the floating-point patent.
                    Yay for math being patented.

                    Everyone needs to help spread the word to fight patents, and I really wish companies would join together to do so as well. And no, the Open Invention Network doesn't count, because it's just a patent pool, and that simply reaffirms patents. They need to be fraught against, not reaffirmed.

                    Comment


                    • #20
                      The risk with these features being in separate branches is bitrot. Even if they, say the float one, apply now, will they in Mesa 7.12?

                      Would be nice if someone outside of these patents' influence would keep these features alive. Perhaps it would even become the de-facto Mesa source for everyone who doesn't give a shit about the sw patents.

                      Comment

                      Working...
                      X