thats not the reason FOr the DX10 Vista problems of compatibility. They completly Reworked the whole API not because the want to force people to switch but becasue the way Vista handles device Drivers. Remember the Drivers sit between the Hardware and The APi(Direct X). The changes they made Broke compatibility but the advantages far outweight it in my opinion.
GFX Wise i never looked at LFS as pretty game, and theres only a hand full of things you would really need DirectX10 for that directX9 could not do.
i think LFS is a directX9 game. its just it seems the developers don't use any pretty Shaders.
Sidenote does nyone know how the damage modem is implemented, Vertex Shaders?
No, you can't say that because OpenGL is only a graphics API, but DirectX contains stuff for sound, input, networking etc. So the only thing you could say is something like SDL+OpenGL owns DirectX (although I'm not sure if SDL contains any networking stuff)...
As for a Linux version: I would really appreciate it! But I guess that there are a lot of challenges: I don't know how FFB works on Linux, afaik there is no support for it. And I don't know how well designed LFS is, by 'designed' I mean application architecture, it would help if there is a clear separation of the model (the sim) and the 'view' (all the operating system calls). In this case Scawen could just hire some guy to port the 'view' to Linux (and/or MacOS, why not??) and continue to improve the sim (the model)...
And while writing I just came across another question: What is the difference between e.g. Direct3D 8 and 9? Is it just some technical limitations like shader length? I mean you can do stuff like HDR in D3D 8, you just have to do it yourself (multiple shader passes)... And what is possible with D3D 9 that is not possible using OpenGL? How can people say, that 'game xy looks better than game uv because it is DirectX 9'? Shouldn't it be 'game xy looks better than game uv because it uses complex per-pixel lighting that just happens to be implemented in D3D 9 but could have been done in OpenGL too?
And this leads me to a final statement: Devs,consider using OpenGL! Why? because it will support everything that is needed for good graphics (like said, I think today you can do everything that you can do with D3D with OGL), it is a widely accepted standard and it is supported by a lot of platforms (Linux, old Win versions including XP, MacOS etc.). This way you will be able to write a sim with the latest graphics technology and make it available to a wide audience. I believe that not everyone that uses XP will switch to Vista (what's exactly the deal with DRM?) and when MS brings DX 10 only for Vista, you can do the same stuff with OpenGL that will be supported on XP too...
And I think you have to rewrite parts of the app for full Vista compatibility (or don't you?) and in this process you could try to make an abstract hardware/operating system layer that makes it easier to port the whole app to other operating systems...
You answered that one yourself. With OpenGL you have to do everything yourself. With DirectX you can sometimes rely on "premade" effects that make your work a bit easier.
The (small) downside is that some effects, like the DX lensflare or HDR, always looks the same in all games. But on the other hand it took you 15 minutes to implement and bug-check the feature.
OpenGL is great when you want to do stuff yourself. You have a special problem that you want to solve. Take OpenGL for the task. But when you want a random GFX engine that just throws loads of polys and textures at the user and don't want to spend a lot of time with it, use DirectX.
DirectX has it's coding-advantages. The object-orientated view makes Dx well structured and easy to handle. OpenGL can look like mess sometimes, if you don't organize your code very well.
Oh, and by the way, SDL has a great net-library called "SDLnet". I used to work with it and it's a really handy piece of code with a good documentation.
If you think the re-write of the driver system has anything to do with improvements to performance etc you have been living under a rock. Its all about DRM and making the DRM harder to bypass, nothing else.
BTW. It is now official that 32bit Vista will not allow ANY viewing of HD content at all because it might might be possible to copy it. The 32bit base does not have all the extra hardware protection that the so called "tursted computing" uses to protect content from being copied. This means it could be cracked which the RIAA amd MPA did not like. Solution, disable HD content totally.
It appears more and more people are saying XP is their last MS OS, myself included. I might even look at Linux more seriously but even so it's just too late now for a Linux port, the code base is too large.
As stated by others, there is far too much code to change from GFX, Network, Sound, Controller interfaces and any bit of code that results in user interaction. That does not even include library calls that might be present under Windows and not in Linux.
The Rewrite of the drivers is one of the reasons why compatibility is broken M$. It really has nothing to do with DRM is the case of directX.
Blame the Studios not M$ they were the ones that preasured them. No need to worry because am sure couple months after release someone would of cracked it and we will all be watchinig it in VideoLan.
Maybe it would be worth for the devs to look into Cider (http://www.transgaming.com/ind ... ;func=display&ceid=24). It seems to be an easy way to get Win software working on Intel Macs and Linux. This way they could evaluate the interest in a LFS version for other operating systems and if there's a big interest, they could start to solve the remaining problems (I think Cider won't support FFB for example...).
The only drawback might be the money involved, I don't know how expensive it would be to use Cider...
If it was easy to do i'd join the OpenGL brigade because i'm moving away from PC's to Mac with every new feature announcement of Vista. But it isn't easy, and it would slow down development of the game.
I suspect a Linux version of the server would not be too much to expect though.
Anyway I think Scawen already wrote off doing an OpenGL port, so I guess we're lumbered.
