Shaders 1.4 in 3dmark03? Fair or unfair?

[ET]

Much more useful than having the benchmark deciding what to use.

Why is that? It is simply reflective of reality. Based on my understanding, Doom 3, for example, will automatically choose which rendering path to use (although I do expect that we will be able to tweak some config files to force it into a different code path).

But Doom3 isn't a benchmark. A synthetic benchmark is something to help you measure performance of specific features, in this case of the graphics subsystem. It's not a means, as I see it, to give your machine a "score", and it's not supposed to be representative of games. Games are representative of games. A collection of games can be considered a benchmark. Synthetic benchmarks, OTOH, are tools for finding bottlenecks in your system, for determining which features are better implemented on a specific system, that kind of thing.

Even if you want 3DMark to be representative of games, there's no reason not to give the ability to drop to 1.1, to give it both the benefit of representing games (which is doubtful anyway) and of allowing you to do specific measurements. It could answer such (IMO interesting) questions as how much the Radeon 8500 suffers when using 1.1 shaders vs. 1.4 shaders.

 

[Glonk]

You should definitely take time to review the specs! :?

Can someone please enlighten me, rather than tell me to go read the specs?

The specs don't do much for me...

 

[Dio]

Can someone please enlighten me, rather than tell me to go read the specs?

I did an extensive post in this thread comparing the DX8 PS versions, and was then of course corrected and clarified in several ways.

http://www.beyond3d.com/forum/viewtopic.php?t=4292

The brief summary: 1.4 is about twice as capable as 1.3 (which is why it typically halves the number of passes required), while 1.3 offers little of significance over 1.1.

 

[ET]

Can someone please enlighten me, rather than tell me to go read the specs?

The specs don't do much for me...

Ask not what the specs can do for you, but what you can do for the spec. Erm.

You might want to check the "Is PS 1.4 better than PS 1.3?" thread.

 

[mboeller]

IMHO;

I don't think that the inclusion of PS1.4 is unfair, because as Carmack said, the GF-Ti series is as fast as the R8500 in Doom3 despite the fact that the R8500 uses (mostly) single-pass rendering and the GF-Ti series uses multi-pass rendering ( see an old .plan from him ). So IMHO the use of PS1.4 in the new 3DMark03 levels the field more or less instead of introducing more bumps.