Announcement

Collapse
No announcement yet.

Linux gamer base

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

  • Linux gamer base

    Since it was asked to move this into a new topic from the wishlist thread, and I had a few things I wanted to respond to..

    Now, having said this, for all the people claiming it's 'great', it's not all a bed of roses by any stretch of the imagination- most of the space is bronze or silver rated (using the WINE rating system ratings...) at best, even WITH Cedega and Crossover Games.
    I don't think many people would say Wine/Cedega/CrossOver are great solutions. If anything, it's more of a last-ditch effort, which just goes to show how much Linux users really want to try playing these games. It may not be a very good solution for actually playing, but it's rather telling about the potential user base for publishers.

    So, if people are willing to spend $20 on a poor substitute, why bother making a native title?
    Because it's a poor substitute. A native version would (read: should) be of better quality than the Windows version run through Wine/etc. A better quality product brings in more users, and makes for happier customers. Happy customers are repeat customers..

  • #2
    I'll just drop a thought I had...

    I do wonder about the prospect of porting 100% Wine compatible games. If publishers/dev studios can be coaxed into having a Linux release built against Winelib, assuming the game is 100% Wine supported.

    Comment


    • #3
      Originally posted by Chris View Post
      Because it's a poor substitute. A native version would (read: should) be of better quality than the Windows version run through Wine/etc. A better quality product brings in more users, and makes for happier customers. Happy customers are repeat customers..
      Considering that everyone's buying that instead of "crappy", "old", and "overpriced" native titles- why risk ANY cash on it when we don't have to do anything other than make the Windows version </devil's advocate>

      Seriously. That is a goodly portion of what's going through quite a few busninessmen's heads right at the moment- if they give it even a first thought. Using WINE to promote "sales" figures doesn't help- especially with the poor sales figures and HIGH piracy figures we're seeing right at the moment.

      It is all about money. If they expend any making a Linux port, it's a risk compared to the Windows version or even a MacOS version- because those are business KNOWNS. Linux, as far as they're concerned, is an unknown- if they do it, they risk losing a lot of cash doing it with no assurances (Loki's failure, the piracy of LGP content, etc. REALLY helps there...NOT!) of seeing ANY of it back. If they don't do it and license it out, they want to see their pound of flesh that's due NOW, instead of later- because they could have IP leakage or deals that end up with the Linux game publisher owing them LOTS of money (Loki owed Id something between 250-500k in royalties that never got paid. Sure, it's money they'd probably have never seen- but a businessman does NOT see things that way in many cases. It was logged on the books and never collected- it's a loss.). In light of that, it's better to just do Windows titles and ignore us- after all, we've got that WINE thingy, right?

      Comment


      • #4
        Originally posted by niniendowarrior View Post
        I'll just drop a thought I had...

        I do wonder about the prospect of porting 100% Wine compatible games. If publishers/dev studios can be coaxed into having a Linux release built against Winelib, assuming the game is 100% Wine supported.
        There'd be a small performance gain- but all the issues you find with WINE use, including performance loss in at least some of the cases, would be still present with it. All you're doing with that is taking the DLL loader out of the picture.

        Comment


        • #5
          Originally posted by Svartalf View Post
          There'd be a small performance gain- but all the issues you find with WINE use, including performance loss in at least some of the cases, would be still present with it. All you're doing with that is taking the DLL loader out of the picture.
          I was just thinking if it would be easier to coax them into releasing the Linux port of a product A if the Windows game worked very well on Wine. I understand that the performance hit would still remain if there was any for the said product.

          Comment


          • #6
            Originally posted by niniendowarrior View Post
            I was just thinking if it would be easier to coax them into releasing the Linux port of a product A if the Windows game worked very well on Wine. I understand that the performance hit would still remain if there was any for the said product.
            Why go to the bother of that when it runs in WINE? There's extra steps involved (learning how to compile with GCC instead of VC++, learning how to make binaries that work nicely across a range of Linux versions (Something apparently Runesoft's not figured out yet... ), and so forth)- WINE itself will run it, why spend money doing that stuff? Just to please what seems to be only a couple thousand? Riight. That's more money in our pockets or more money for the next title.

            Comment


            • #7
              Originally posted by Svartalf View Post
              Why go to the bother of that when it runs in WINE? There's extra steps involved (learning how to compile with GCC instead of VC++, learning how to make binaries that work nicely across a range of Linux versions (Something apparently Runesoft's not figured out yet... ), and so forth)- WINE itself will run it, why spend money doing that stuff? Just to please what seems to be only a couple thousand? Riight. That's more money in our pockets or more money for the next title.
              I understand that. But, you cannot deny that Transgaming is getting more companies jumping on Cider. They just snagged Ubisoft and they have EA under their belt. There's something in it, that's missing in Linux, I think.

              Comment


              • #8
                Originally posted by niniendowarrior View Post
                I understand that. But, you cannot deny that Transgaming is getting more companies jumping on Cider. They just snagged Ubisoft and they have EA under their belt. There's something in it, that's missing in Linux, I think.

                I completely agree, if it takes a bandaid like cider to show the potential market in the beginning, then that is what should be done. After they get some real numbers coming from current titles by people that buy games developed with a product like cider, they will at least know that the demand is there.

                Comment


                • #9
                  Originally posted by niniendowarrior View Post
                  There's something in it, that's missing in Linux, I think.
                  Copy protection.
                  That's the whole reason you pay for Transgaming Cedega/Cider (so I'm told).

                  Commercial games need commercial copy protection, it's all proprietary.
                  Forget that there's probably a surefire and easily implementable way to do it on linux (hash the UUID, some other CD/DVD signature, or some cryptoloop).

                  Comment


                  • #10
                    Originally posted by Svartalf View Post
                    There'd be a small performance gain- but all the issues you find with WINE use, including performance loss in at least some of the cases, would be still present with it. All you're doing with that is taking the DLL loader out of the picture.
                    Hmmm... Reading this discussion again suddenly lit up a bulb in my head... Not only would there still be overhead, but it could be really, really decreased if studios be using OpenGL rendering for their Winelib compatible games, leaving DirectX for stuff like Networking and Input (I'd rather not have them use DirectX for Network as Windows networked games on Linux are a PITA to setup... especially if the game can't get your local IP address and requires some /etc/hosts magic to do so). But still taking out the biggest chunk of overhead (the renderer) for wine-compatible apps, might actually be a viable first-step to show (800lbs) publishers and studios how viable a Linux market really is, based on real figures, not guess-timates.

                    Comment


                    • #11
                      Originally posted by Thetargos View Post
                      Hmmm... Reading this discussion again suddenly lit up a bulb in my head... Not only would there still be overhead, but it could be really, really decreased if studios be using OpenGL rendering for their Winelib compatible games, leaving DirectX for stuff like Networking and Input (I'd rather not have them use DirectX for Network as Windows networked games on Linux are a PITA to setup... especially if the game can't get your local IP address and requires some /etc/hosts magic to do so). But still taking out the biggest chunk of overhead (the renderer) for wine-compatible apps, might actually be a viable first-step to show (800lbs) publishers and studios how viable a Linux market really is, based on real figures, not guess-timates.
                      Unfortunately, when one takes on anything other than D3D from DirectX, it's a tarbaby of titanic proportions. Moreover, we don't have wireline protocol info on DirectPlay (Uh, we wouldn't have used Grapple for Ballistics if we could have done that one- that may have changed in recent times, but I don't think WINE's got that one down any better than it already has. You get DirectX support (incl. D3D...) by way of loading MS' DLL's and making them think that WINE's a 3D accelerator driver, etc. We don't HAVE DirectX past pieces of D3D/DirectDraw for 8.x. We ought to be utilizing SHARK and HLSL2GLSL, coupled with a modernized wrapper that RealtechVR developed a long while back for this and ditching that whole mess, but we're not there.

                      It's all a hack, Thetargos. Honest.

                      Again...all you're doing is telling them we're interested in second rate solutions against Windows binaries when you do ANYTHING with WINE to make the game happen. Whether it's the loader or Winelib.

                      Comment


                      • #12
                        I see. I understand that supporting Wine (paradoxically) is supporting Windows. And we've pretty much beaten that horse to death. However, the idea of using it as a marketing to try and attract developer attention was what I thought might be worth.

                        Comment


                        • #13
                          Not only would there still be overhead, but it could be really, really decreased if studios be using OpenGL rendering for their Winelib compatible games
                          Just using OpenGL for rendering in Windows has its own benefits as well. You can get D3D10 level capabilities in OpenGL without having to require Vista. It'll still require a capable card, of course, but the capabilities are there and degrade gracefully. Eg. just because you may not have a D3D10 card doesn't mean you can't get any features above D3D9. These things work just fine in OpenGL under XP. That fact that Wine can just pass these right through to the Linux drivers for nearly no effort is icing on the cake here.

                          The same could be said about OpenAL.. though there aren't really any hardware drivers for that under Linux (but you can get 3D surround sound with it in Linux, which is more than the Windows software driver offers).

                          Comment


                          • #14
                            Originally posted by Thetargos View Post
                            I see. I understand that supporting Wine (paradoxically) is supporting Windows. And we've pretty much beaten that horse to death. However, the idea of using it as a marketing to try and attract developer attention was what I thought might be worth.
                            Unfortunately, it doesn't work that way, friend- you're still thinking "common sense", which doesn't quite apply here. When you show them something that allows them to write for Windows and allows them to run, even if it's unstable, that code on Linux- they hear the "Windows" part that they DO understand, but don't hear any other parts of the conversation. Seriously. WINE doesn't help things in that regard.

                            Comment


                            • #15
                              Originally posted by Chris View Post
                              Just using OpenGL for rendering in Windows has its own benefits as well. You can get D3D10 level capabilities in OpenGL without having to require Vista. It'll still require a capable card, of course, but the capabilities are there and degrade gracefully. Eg. just because you may not have a D3D10 card doesn't mean you can't get any features above D3D9. These things work just fine in OpenGL under XP. That fact that Wine can just pass these right through to the Linux drivers for nearly no effort is icing on the cake here.

                              The same could be said about OpenAL.. though there aren't really any hardware drivers for that under Linux (but you can get 3D surround sound with it in Linux, which is more than the Windows software driver offers).
                              1. D3D and OpenGL/OpenAL are both Hardware Abstraction Layers, they exist because when these 3D games first appeared, you had accelerated versions for specific cards (s3Virge Edition, 3DFX Edition, ATi Rage Edition).
                              2. The HALs were created to give companies a common interface. MS created D3D, and the OpenGL open source community does OpenGL. Can't remember who does OpenAL.
                              3. The games have to be programmed for these HALs in order to reach as wide an audience as possible.
                              4. Have you tried running a DX8/DX9 application with a DX7 card? Go ahead, try it. World of Warcraft on a Radeon 7000/7500 in D3D. Let's see if you get past the opening cinematic without it crashing. Try OpenGL if you dare. (FYI, I run a IGP 320M, which is the equivalent of a 7000 on shared memory.)
                              5. Games are programmed with some backward compatibility, they will run older versions, but only so much. Guild Wars is an example of a DX8 fallback with DX9 features.
                              6. Wine will only pass OpenGL. If a game was programmed with D3D only, you're SOL if it doesn't work. Wine will translate some D3D, but you're now going from one HAL to another HAL.
                              Sorry I had to rip, that D3D9/10 comment irked me.

                              EDIT:
                              Okay, I re-read your post, I do agree somewhat. OpenGL gets those features by using some CPU power I suspect. Long time ago I remember running Need for Speed 3 with OpenGL in software mode, and it just was not possible to play at all. OpenGL, like MS, goes through upgrades. If an OpenGL card isn't 2.0 capable, it'll still run at 1.4, 1.3, though you will not get the GPU features of the higher level.
                              Last edited by me262; 07-10-2008, 07:40 PM. Reason: Now I understand...

                              Comment

                              Working...
                              X