DirectX Next - Longhorn only !?

[tb]

http://www.ati.com/developer/gdc/Save_The_Nanosecond.ppt

- Will be for the next OS only
- Expect this circa early 2006
- Faster small batch performance is a very high priority, which implies a new driver model
- Will last for two or more years (DX9 lasts from Q4 2002 until the next OS)
- What gets built into Longhorn will be the basis for their next games. That is games - plural. DX Next will last at least as long as DX9 (almost 4 years)

So, will this be the most popular reason to drop all other Windows versions(95/98/Me/2k/XP) in the consumer market?

Thomas

 

[MfA]

That and Palladium.

Marco

PS. the comments for the presenter are cute

Steer people away from flow control in ps3.0 because we expect it to hurt badly. (Also it’s the main extra feature on NV40 vs R420 so let’s discourage people from using it until R5xx shows up with decent performance...)

Mention that you expect to see a 512MB board in the consumer space this year. Don’t say that it’ll be R420 based!

 

[Waltar]

This is why Directx needs to be abandoned.

OpenGL 2.0 4life. 8)

 

[Neeyik]

Ooh, did people notice slide 31?

The 3.0 shader model

Requires 32 bit floats throughout the pipeline
But that’s not necessarily full IEEE 754...
With it’s -0.0s, NANs and INFINITYs etc

It’s generally agreed that SM3.0 requires a move to 32 bit floats (some still think it’s not)
But - The spec requires it...

Steer people away from flow control in ps3.0 because we expect it to hurt badly. [Also it’s the main extra feature on NV40 vs R420 so let’s discourage people from using it until R5xx shows up with decent performance...]

Tee hee... 8)

Edit: Oops, only just noticed MfA's post.

 

[hoom]

Heh, ATI commenting on unannounced hardware

 

[ram]

"Sorry, there is no ATI.com web page matching your request."

8)


human failure sometimes is quite amusing. Very entertaining.

*prints presentation including comments*
*gets popcorn*

8)

 

[MfA]

Ohhh ... now I gotta start loving ATI too I guess.

If the geometry shader is “done rightâ€￾ it’s part of the vector processor pool – and therefore can have a vertex shader before it – and after it! Which is useful...

EX-FUCKING-ACTLY.

Just putting the "geometry shader" in front of the vertex pipeline as has been suggested a lot would be dumb fuck stupid. At least ATI isnt going that way, lets hope sanity prevails in this case.

 

[Geeforcer]

Ooh, did people notice slide 31?

The 3.0 shader model

Requires 32 bit floats throughout the pipeline
But that’s not necessarily full IEEE 754...
With it’s -0.0s, NANs and INFINITYs etc

It’s generally agreed that SM3.0 requires a move to 32 bit floats (some still think it’s not)
But - The spec requires it...

Steer people away from flow control in ps3.0 because we expect it to hurt badly. [Also it’s the main extra feature on NV40 vs R420 so let’s discourage people from using it until R5xx shows up with decent performance...]

Tee hee... 8)

Edit: Oops, only just noticed MfA's post.

LOL, that was fast.

 

[oddfellow]

Did people notice that the link no longer works?!

I think ATI just realised they dropped a boo boo

EDIT: Sorry, ram already did

 

[Trooper]

• Blend etc will be on NV40 and on our own SM3.0 parts
• Steer people away from flow control in ps3.0 because we expect it to hurt badly. [Also it’s the main extra feature on NV40 vs R420 so let’s discourage people from using it until R5xx shows up with decent performance...
• Mention that you expect to see a 512MB board in the consumer space this year. Don’t say that it’ll be R420 based!
• Note that many developers treat all DX9 NV hardware as “low end DX9†hardware because that’s the best way.
• Hence treat all NV DX9 hardware as DX8... (See previous slide)
• But be careful not to trash NVIDIA too hard here – your audience will help you out later...
• Be careful not to mention Xbox 2 – just let everyone know that‘s what you mean...
• IMpress upon the audience that targetting high end gamers is EQUIVALENT to targetting ATI hardware

This is fantstic, anything like this for NV?

 

[MfA]

BTW, this needs to be on the front page

 

[oddfellow]

Oh I love the intarweb!!

 

[Psikotiko]

Mirror here

 

[oddfellow]

Mirror here

You are THE MAN!! 8)

 

[digitalwanderer]

Mirror here

Thanks! This is sounding juicy and I can't wait to pour over this!!!!

(Hmmm....does this count as leaked slides if I publish it? :| )

 

[oddfellow]

NVIDIA’s hardware has 2-3 times the shader perf available in FF mode. But then that kills alls the useful features.
Hence treat all NV DX9 hardware as DX8... (See previous slide)

From slide 27

Can someone elaborate? (is he talking about NV3x or NV4x?)

 

[digitalwanderer]

I'm not sure, but I did just drop Mr.Huddy an e-mail requesting permission to reprint some stuff from it. (Hey, I learn! )

 

[Neeyik]

NVIDIA’s hardware has 2-3 times the shader perf available in FF mode. But then that kills alls the useful features.
Hence treat all NV DX9 hardware as DX8... (See previous slide)

From slide 27

Can someone elaborate? (is he talking about NV3x or NV4x?)

The NV3x (and possibly the NV40). The current FX cards have a far greater vertex throughput when in FF mode compared to shaders, unlike the R3xx where the throughput is the pretty much the same.

 

[Citrous]

The 3.0 shader model

Requires 32 bit floats throughout the pipeline
But that’s not necessarily full IEEE 754...
With it’s -0.0s, NANs and INFINITYs etc

It’s generally agreed that SM3.0 requires a move to 32 bit floats (some still think it’s not)
But - The spec requires it...

Steer people away from flow control in ps3.0 because we expect it to hurt badly. [Also it’s the main extra feature on NV40 vs R420 so let’s discourage people from using it until R5xx shows up with decent performance...]

I dont see this on the slide?

 

[Tahir2]

Looks like an intentional boo boo... lol.