DeanoC's comments about SM3.0

[Evildeus]

Wouldn't the best approach be implement SM2.0, 3.0 and others if cost effective?

I would prefer this approach. Developers already has to code for non-shader hardware, PS 1.1, SM 2.0 and now SM 3.0. On top of this they have to spend a lot of time to get good performance on NV3x, so 2.a and 2.b would probably distract to much.

I'm a bit of a purist so I would have preferred that we never saw 2.a with NV3x (and just pure SM 2.0) like I would have preferred SM 3.0 on R420 and no 2.b. But that is just me.

That is what i think. SM1.0, 2.0, 3.0 and others if cost effective.

 

[Anonymous]

Reading between the lines of your posts, it seems that ATI is skipping SM3.0 altogether.

Not sure whether or not they will "skip" it, but I have a gut feeling that ATI is pushing very hard at moving towards SM 4.0. I doubt they will play up SM 3.0 much at all because of this (and for other obvious reasons too, given the nature of their current gen hardware). Also, I recall DaveB making a comment that SM 3.0 is more of a "stepping stone", which leads me to believe that SM 4.0/Longhorn/DirectX Next is the next major inflection point that ATI is really gunning for.

 

[jimmyjames123]

Reading between the lines of your posts, it seems that ATI is skipping SM3.0 altogether.

Not sure whether or not they will "skip" it, but I have a gut feeling that ATI is pushing very hard at moving towards SM 4.0. I doubt they will play up SM 3.0 much at all because of this (and for other obvious reasons too, given the nature of their current gen graphics hardware, and also the nature of the XBOX2 graphics hardware). Also, I recall DaveB making a comment that SM 3.0 is more of a "stepping stone", which leads me to believe that SM 4.0/Longhorn/DirectX Next is the next major inflection point that ATI is really gunning for.

 

[LeStoffer]

Reading between the lines of your posts, it seems that ATI is skipping SM3.0 altogether.

Not sure whether or not they will "skip" it, but I have a gut feeling that ATI is pushing very hard at moving towards SM 4.0. I doubt they will play up SM 3.0 much at all because of this (and for other obvious reasons too, given the nature of their current gen graphics hardware, and also the nature of the XBOX2 graphics hardware). Also, I recall DaveB making a comment that SM 3.0 is more of a "stepping stone", which leads me to believe that SM 4.0/Longhorn/DirectX Next is the next major inflection point that ATI is really gunning for.

Actually, you should play attention to the SM 3.0+

Xbox 2 will most probably be SM 3.0+ where + = some additions and fine tunning for Full Speed Features Only.

 

[Dave Baumann]

AFIAK, at the moment, ATI's SM3.0 part is not the same as the XBox and their SM4.0 part isn't either, but that will be closer to the XBox part.

 

[Farid]

AFIAK, at the moment, ATI's SM3.0 part is not the same as the XBox and their SM4.0 part isn't either, but that will be closer to the XBox part.

If i remember and understand Dave Ortons comments/hints correctly, their next part (R500/520?) (which will obviously be SM3 compliant) will be derived from the R300 architecture, then they will have a part (R600?) that is derived from the Xenon part (R400)...

 

[LeStoffer]

AFIAK, at the moment, ATI's SM3.0 part is not the same as the XBox and their SM4.0 part isn't either, but that will be closer to the XBox part.

Hmmm, interesting. I assumed that they would do something similar to nVidia's NV2A (Xbox) chip that was pretty close to the rest of their architectures at that time.

BTW: You seem to hint that Xbox won't follow PC DirectX specs since its neither SM 3.0 nor SM 4.0. That is even more interesting!