There are rumblings of Microsoft favoring the idea of requiring unified shaders in the future. With Ati’s support for this and Nvidia’s reluctance, will Nvidias’s involvement with the Ps3 jeopardize its influence with Microsoft? Furthermore what tangible consequences could this have?
Nvidia losing influence due to Ps3 involvement?
- Thread starter nelg
- Start date
To me it seems that both Microsoft and Sony want to be at the center of the "Digital Home". Nintendo, IMHO, is more focused on just game hardware and software.Ostsol said:
In terms of the future of Direct X. Do you think that Microsoft would be more inclined to map closer to ATI's architecture, even when there may be no technical merit?
the difference between nintendo and sony, from microsofts perspective, is that nintendo is content being basically a video game only company. sure, they have recently broken into animation, but i think that is more a natural progression (their characters have been featured in numerous animated shows and movies). sony, on the other hand, very much wants to be a leader in the "does everything set top box" arena (games, dvr, movies, music jukebox, sell ratings info to nielson, ect...). a market that microsoft is also fighting for. windowsXP MCE isn't an accident, it's a glimpse of the future.
so nintendo isn't really competing with microsoft, they are competing with sony for video game supremacy. and microsoft wants sony out of the living room.
so nintendo isn't really competing with microsoft, they are competing with sony for video game supremacy. and microsoft wants sony out of the living room.
Maybe influence is the wrong word. Looking at the example that I gave, unified shaders. If equal technical arguments could be made wether or not future DX implementations should require this, do you think that MS may be persuaded against Nvidias position, in part, due to their involvement with Sony?Reverend said:
I thought we already discussed the idea of unified pipelines being required by the spec.
http://www.beyond3d.com/forum/viewtopic.php?t=18740
Basically, unless the API itself demands some form of control over the process of allocating ALUs, there's no reason to require it. I think that at most, the shader specification will require the same capabilities in both vertex shader and pixel shader ALUs. Unified shader pipelines is more of an architectural design decision, rather than a feature.
What MS could do is more of what they are already accused of doing: listening to a single IHV on individual rendering feature specifications, and ignoring the competitors. In the past it has reportedly been pixel shader precisions, texture formats, and such. It's much more specific. They might even go so far as to specify how exactly an implementation of a feature such as texture filtering must work (that is, a required equation to produce a specific output). General architecture should always remain up to the IHV, though.
http://www.beyond3d.com/forum/viewtopic.php?t=18740
Basically, unless the API itself demands some form of control over the process of allocating ALUs, there's no reason to require it. I think that at most, the shader specification will require the same capabilities in both vertex shader and pixel shader ALUs. Unified shader pipelines is more of an architectural design decision, rather than a feature.
What MS could do is more of what they are already accused of doing: listening to a single IHV on individual rendering feature specifications, and ignoring the competitors. In the past it has reportedly been pixel shader precisions, texture formats, and such. It's much more specific. They might even go so far as to specify how exactly an implementation of a feature such as texture filtering must work (that is, a required equation to produce a specific output). General architecture should always remain up to the IHV, though.
What I agree on is that there are things that Microsoft could do that are detrimental specifically to NVidia or ATI. Whether they would do this or not is the real issue, though. Of course, some people think that they already have done so by making the minimum requirement for full precision FP24 under PS2.0. We can go back a few years to the original Radeon and it's "pixel shaders". Merely conspiracy theories? Perhaps so, perhaps not.
Personally, I don't think that there would be such a huge technological advantage for Microsoft even if NVidia weren't helping Sony. As such, what NVidia is doing is not going to be particularily detrimental to Microsoft's success. After all, a large part of their profits from the XBox 2 (and Sony's profits from the PS3) will be due to past reputation and to the game developers that are already working with them. It is true that developers may shift their concentration to or away from a specific console as it becomes more apparent as to which suits their needs better, but for the most part things are likely to remain relatively stable.
Personally, I don't think that there would be such a huge technological advantage for Microsoft even if NVidia weren't helping Sony. As such, what NVidia is doing is not going to be particularily detrimental to Microsoft's success. After all, a large part of their profits from the XBox 2 (and Sony's profits from the PS3) will be due to past reputation and to the game developers that are already working with them. It is true that developers may shift their concentration to or away from a specific console as it becomes more apparent as to which suits their needs better, but for the most part things are likely to remain relatively stable.
DukenukemX
Newcomer
nelg said:
I'm pretty sure that Nvidia is on Microsofts sh*t list after Xbox. This is probably the reason why ATI is working with Microsoft on Xbox2.
Nvidia has tried to influence Microsoft before and it didn't end well. This is probably how Nvidia never got DX9 the way they wanted it.
Now I wonder since ATI is working with Microsoft and Nintendo don't you think that will jeopardize ATI's influence with Microsoft?
Drak said:Now that's it's only a two pony show (Sorry XGI), what if NVIDIA decided to drop out of WGF/DX10 whatever? That will make WGF/DX10 a vendor specific API like GLIDE. OpenGL will become the true multi-vendor industry-wide API
I assume you're kidding here as I don't think it's likely NV will commit corporate suicide.
No more than Sony making Windows-running Vaios, or Toshiba making PC drives with Windows drivers, or Ati in Nintendo, or IBM being in the Xbox2, PS3 AND the next N console.
It's all very incestuous you know...
It's all very incestuous you know...
Mariner said:Drak said:Now that's it's only a two pony show (Sorry XGI), what if NVIDIA decided to drop out of WGF/DX10 whatever? That will make WGF/DX10 a vendor specific API like GLIDE
I assume you're kidding here as I don't think it's likely NV will commit corporate suicide.
It was very clever of Microsoft to tie WGF/DXNext into the GUI of longhorn. Have the SDKs for DXNext been sent to the developpers yet? Otherwise, NVIDIA could just as well have supported only DX9.x and moved entirely to OpenGL since it's one of it's strengths. Or they could have used the threat to strongarm Microsoft.
Bouncing Zabaglione Bros.
Legend
Can you imaging Nvidia putting "Not Compatible With Windows" on all their boxes? I doubt it, because it would hurt Nvidia not just for aftermarket sales, but all the OEMs would dump Nvidia. You're not going to see Dell shipping PCs with anything other than MS Windows and a compatible graphics card.
Nvidia has to follow MS, as the NV30/NV35 debacle showed.
Nvidia has to follow MS, as the NV30/NV35 debacle showed.
radar1200gs
Regular
I could certainly imagine nVidia putting "Not just for Windows" on the boxes, and given their history of strong alternative OS support it would be no idle boast either.
