there is significant dis-information re the provenance of shaders in D3D.
no, we are not simply implementing what nVidia says.
earlier this year there was a long thread on the OGL mailing list about just this. I
am sure the archive of that mailing list will be informative, various folk were eating
popcorn whilst the debate raged :-).
wrt to shader support:
D3D
Radeon8500 supports ps 1.1, as does gF3; from DX 8.0.
Radeon8500 also supports ps 1.2, 1.3, and 1.4; that were added in DX 8.1.
then you have a choice to implement shaders ( 1.4 ) that would let you get
more out of the R8500, but accept 2 code paths.
or stay to 1.1 and get maximal compatibility with 1 shader code path.
OGL
gF3 register combiners do not work on Radeon8500.
Radeon8500 shaders do not work on gF3. there is no compatibility at all.
I dont see what the pain is there with D3D since the "sticking with ps 1.1" path gives
a clean cross-hardware implementation. So yes, its significantly better on the D3D
side here.
And since ps 1.1 does provide gF3 level functionality, its hardly limiting yourself in
a real sense to stick to it. Its more a matter of code-paths vs feature exposure. To
get to some of the extra-cool bits of an R8500 you need to use 1.4 and accept multiple
code paths. But you can certainly forego that and appear to be "just a gF3" with ps
1.1.
yes, DX 9.0 does add significant new features. a high-level language for shaders is
just one.
interestingly, OGL 2.0 closely resembles the DX 9.0 plan. DX 9.0 features were
discussed at Meltdown in July. 3DLabs presented its 2.0 ideas to the ARB in September.
I think that speaks for itself.
this is moving off-topic. if anyone wants to continue this, private email will be best.
> -----Original Message-----
> From: Joachim Diepstraten
> [mailto:[EMAIL PROTECTED]]
> Sent: Monday, November 19, 2001 1:48 PM
> To: [EMAIL PROTECTED]
> Subject: Re: [JAVA3D] Okay I have to ask
>
>
> Ehlo Philip
>
> > here is one that is not currently a reason, but will likely
> be one in the future.
> >
> > in the future as J3D adds shader support, that is going to
> be easier on D3D since D3D shader versions for pixel shaders
> work across IHV hardware offerings; as in ps 1.1 will work on
> both gF3 and R8500.
> >
> > Compare that to the current state of OGL pixel shader
> extensions which are incompatible across IHV hardware offerings.
> No offence here (I know you're from Microsoft), your statement is of
> course right at the moment, but isn't it so that DX is just
> implementing
> the things Nvidia wants in their next version of Geforce-line
> instead setting a real overall vendor-independent standard?
> (If you look
> at DX8 shader and compare them to nvidia shader-extension in OGL the
> difference isn't very big)
> Shader-Programming in DX and in OGL with vendor extensions is
> a real pain and I don't see a significant change in DX8.1 or
> 9 for this.
> (I hope I'm wrong here). So OGL 2.0 really is to set new grounds.
>
> And as Java3D 1.4 lies at least now 2 years in the future.
>
> EOF,
> J.D.
>
> --
> Realtime Raytracer in JAVA
> http://www.antiflash.net/raytrace
>
> ==============================================================
> =============
> To unsubscribe, send email to [EMAIL PROTECTED] and
> include in the body
> of the message "signoff JAVA3D-INTEREST". For general help,
> send email to
> [EMAIL PROTECTED] and include in the body of the message "help".
>
==========================================================================To
unsubscribe, send email to [EMAIL PROTECTED] and include in the body
of the message "signoff JAVA3D-INTEREST". For general help, send email to
[EMAIL PROTECTED] and include in the body of the message "help".
