Microsoft on DirectX 10

[Hanners]

Extreme Tech have posted up a pretty interesting interview with Microsoft's Chris Donahue (Director of Business Development in the Games for Windows Entertainment and Devices Division) and David Blythe (Software Architect with the Graphics Platforms Unit). Nothing much new is covered per se, but it's well worth a read to get up to speed with what's new in DirectX 10, what didn't make the cut, and what we might be seeing further down the line DirectX-wise.

Check out the article at this link.

 

[JHoxley]

Interesting catch - thanks for the link.

Would've been nice if they'd gone into a bit more detail on the "why its not on XP" - I know the reasoning (as I'm sure many people reading this forum will), but I still come across far too many people who dont really understand it

Cheers,
Jack

 

[Wunderchu]

I agree, thanx for the heads up

 

[LeStoffer]

Yes, thanks. A bit on the few optional features:

... the most expensive features are the ones that end up being optional. And there's a similar one for being able to texture filter 32-bit texture formats. The big deal is, doing 32-bit floating-point processing inside the texture processing unit or the blend unit is a relatively large hardware cost. So not every hardware vendor wanted to build that.

 

[JHoxley]

So not every hardware vendor wanted to build that.

Now I'm curious as to which IHV(s) that applies to.

Jack

 

[Acert93]

Optional? I thought they were doing away with that. If it is not required they might as well not even mention it. I have no issue with IHV's offering new features, but if they are not going to be supported across the board in the API that means adoption is low. Unless one of the IHV's was able to hit all the 10+1 milestones as well having a 10+1 feature in D3D10... oh well.

Thanks for the heads up and news guys

 

[KimB]

Now I'm curious as to which IHV(s) that applies to.

Jack

All of them

(I'd be willing to bet)

 

[KimB]

Anyway, now that I've read the article, I'd like to mention that the interviewee was really out of touch on two points:
1. Procedural generation. It's not happening. I'm with John Carmack on this one: procedural generation just cannot be applied as a general technique, and thus won't be done to any meaningful extent (in terms of performance) in games, because it adds far too much to the development time to be worth it. It may be useful for content generation tools to make lives easier on developers, but it most likely isn't going to be used for the content itself.
2. There's just no f'in way that there are going to be any DX10-only games any time soon, unless Microsoft specifically bankrolls those games to evangelize Vista.

 

[Hanners]

2. There's just no f'in way that there are going to be any DX10-only games any time soon, unless Microsoft specifically bankrolls those games to evangelize Vista.

*cough*Halo 2*cough*

 

[Tim Murray]

Maybe Chris Donahue will come and answer some of our comments?

 

[Demirug]

*cough*Halo 2*cough*

I don’t expect that any of the “Vista Launch Games“ will be D3D10 only. As no developer should have hardware at the moment it is nearly impossible to develop anything as large as a game.

Maybe we will see some Games that require “DirectX for Vistaâ€￾ to run well. Even if Richard Huddy don’t think so.

 

[tEd]

yeah halo2 has been confirmed d3d9 but vista only

 

[Demirug]

yeah halo2 has been confirmed d3d9 but vista only

I know. But the question is still why only for Vista? There could be technical reasons but at the moment I would not set much money on this horse.

 

[KimB]

There's just no way it's for technical reasons. It's for marketting reasons. Of course, if the game is ported straight to Vista, it may take significant work to make it run on XP, but it was still a marketting decision that caused it to be a Vista game in the first place.

 

[SugarCoat]

Who cares about Halo 2. By the time its released for Vista the title will be about 2 years from its origonal release.

Alan Wake is a more juicy subject.

 

[compres]

I really don't like the idea of vista only games. As much as I hate MacOS I really wish it takes up a lot of market share, maybe that way devs will code games for many platforms, includding linux and windows Xp.

 

[Pete]

Just stumbled across the interview and thought it made for a great read. Nice to see that Hanners has already introduced it (nicely).

 

[Ailuros]

Anyway, now that I've read the article, I'd like to mention that the interviewee was really out of touch on two points:
1. Procedural generation. It's not happening. I'm with John Carmack on this one: procedural generation just cannot be applied as a general technique, and thus won't be done to any meaningful extent (in terms of performance) in games, because it adds far too much to the development time to be worth it. It may be useful for content generation tools to make lives easier on developers, but it most likely isn't going to be used for the content itself.
2. There's just no f'in way that there are going to be any DX10-only games any time soon, unless Microsoft specifically bankrolls those games to evangelize Vista.

(2) is what worries me mostly for the time being.

As for (1) IHVs as well as Microsoft have nearly half a decade research time for it.

 

[KimB]

Well, there are a few surfaces that lend themselves well to procedural generation. But implementing those as shaders would be rather counterproductive, as it'd just take so much more developer work. Procedural generation is great on the content creation side, but there's no reason why it should be shoehorned into shaders alongside textures. For one, it makes it more difficult to handle resource limitations in a shader library that allows either a texture or a procedurally-generated texture.

Procudural-type effects will, I claim, be almost completely limited to effects to simulate extra detail, such as by applying something akin to the detail textures seen in Unreal/UT. There may also be macro textures to prevent tiled textures from looking tiled from a distance, but I suspect that other techniques like John Carmack's MegaTexture will win out instead.

 

[Parousia]

Anyway, now that I've read the article, I'd like to mention that the interviewee was really out of touch on two points:

I think all the interviewee wanted was a response, not that he's not knowledgeable.

1. Procedural generation. It's not happening. I'm with John Carmack on this one: procedural generation just cannot be applied as a general technique, and thus won't be done to any meaningful extent (in terms of performance) in games, because it adds far too much to the development time to be worth it. It may be useful for content generation tools to make lives easier on developers, but it most likely isn't going to be used for the content itself.

You mean there could be things John Carmack said that anyone could disagree with?

Common sense usually rules.

2. There's just no f'in way that there are going to be any DX10-only games any time soon, unless Microsoft specifically bankrolls those games to evangelize Vista.

Of course, you're talking about independent developers or developers with huge backing from their publishers. "any time soon" is open for discussion. Games that require DX10 would be different from games which "Microsoft specifically bankrolls those games to evangelize Vista". Timing is the keyword. I also suspect the console industry/situation/market would play a part too.