Futuremark's First Monthly Media Newsletter Is Here!

[Tim Murray]

Using a game engine would be pointless, why not simply use a real game? How would 3DMark be any different from any game benchmark you could run if it used a game engine?

The very fact that it doesn't use one is what makes the thing useful!

3DMark's game tests are not synthetic benchmarks, nor are they game benchmarks. They're in a third category, along with the X2 Rolling Demo, Halo's built-in timedemo, UT2003 flybys, and a few other things, that use game engines or game-like engines without actual gameplay. They're not synthetic, because they test the performance of a variety of features in a game-like situation, but they're not games because, well, they don't use gameplay.

Everybody keeps thinking of benchmarks in two categories--synthetic and game--and it's not really true anymore.

 

[digitalwanderer]

The very fact that it doesn't use one is what makes the thing useful!

That and the fact that FutureMark's benchmark is supposed to be testing for NEXT generation stuff, most of which game engines haven't been developed yet....that's why we all knew the FX wasn't going to be so hot in dx9 stuff well before nVidia admitted it. (If they have, I keep forgetting. )

 

[StealthHawk]

1. No demo's on the rails... take levels using real game engines and allow the user to play through them whilst 04 records performance. You could have synthetic versions which are on rails but are their for info only/dont contribute to the final score.

What's wrong with demos on rails? I don't see people complaining about all the canned demos being used in reviews that are on rails that come from games.

Besides, rails or not, any application used as a benchmark is succeptible to specific optimization/lowering of quality/changing of output. Look at how NVIDIA was specifically lowering the rendering quality of UT2003 for so long- which would automatically inflate all scores from botmatches, flybys, custom demos, or FRAPS recordings.

One of the criticisms of 03 was that it was too theoretical and didnt show in the best possible way how your system would perform in games. Using real game engines would resolve this issue.

The problem with this is that there were no game engines that used PS2.0 available at the time 3dmark03 was out. One of the big selling points of 3dmark has always been that it pushes the software envelop and uses features of current new generation cards that come up - something that precedes games for several months if not years. Correct me if I'm wrong, but the first shipping DX9 game was Tomb Raider AoD, and that came out some 6 months after 3dmark03...and a good engine it was not.

3DMark 2001 was riddled with many questionable optimisations, why did it not receive the same scrutiny that 3DM03 received?

Good question. AFAIK though, there were never any benchmark guidelines in place for 3dmark2001. We didn't see such obvious and blatant cheating(sky changing color). Also, I don't think websites in the beta program were given special developer versions of 2001, through which things like NVIDIA's clip planes in 03 were discovered.

 

[nelg]

I think FutureMark should let you shoot a couple of things during the test. That way they could call it a shity game with a great benchmark. Hey some people may be appeased by it.

 

[hkultala]

I think FutureMark should let you shoot a couple of things during the test. That way they could call it a shity game with a great benchmark. Hey some people may be appeased by it.

3dmark2001 had that feature; on Pro version you could drive the car of GT1

 

[ZoinKs!]

I think FutureMark should let you shoot a couple of things during the test. That way they could call it a shity game with a great benchmark. Hey some people may be appeased by it.

Some might but not everyone. There're some morons who say Tombraider: AoD benches don't count because they don't like the game...

 

[Colourless]

3DMark2001 was also made using a real game engine, MaxFX aka the Max Payne engine.

 

[StealthHawk]

3DMark2001 was also made using a real game engine, MaxFX aka the Max Payne engine.

Did all the game tests of 3dmark2001 use the MaxFX engine, or only the Max Payne test?

 

[Mendel]

By the way: There is an easter egg in 3dmark 2001 that allows playing the car game even with non registered version! Happy hunting!

Another btw: Don't you think the newsletter was a bit conveniently timed so it didn't have to tell, if newest nvidia driver was tested?

 

[digitalwanderer]

By the way: There is an easter egg in 3dmark 2001 that allows playing the car game even with non registered version! Happy hunting!

TELL!!!! ( I LOVE 3dm2k1!!!! )

 

[Hanners]

Did all the game tests of 3dmark2001 use the MaxFX engine, or only the Max Payne test?

Just the one test as far as I remember.

I think it would be great to use a real-world game engine if there was one that covered all the next generation features, but the fact is that there almost invariably isn't whenever a new version of 3DMark rolls around.

 

[Mendel]

By the way: There is an easter egg in 3dmark 2001 that allows playing the car game even with non registered version! Happy hunting!

TELL!!!! ( I LOVE 3dm2k1!!!! )

Ok, I'll spoil your hunting then.

You have to edit your project name (defaults to "My Benchmark"). so press the edit button and type to project name field this:

Holy Cow!

It is case sensitive and includes the exclamation mark

you can ignore the comments field.

Then just press ok and then press the game demo button.

More instructions in the game demo loading screen

Enjoy!

On another note: AFAIK the 3dmark2001 does NOT use Max FX engine. Not even in the Max Payne test!

 

[Neeyik]

3DMark2001 uses the MaxFX engine for all the tests but Max Payne the game is quite a different version of it.