3dmark opinion thread

[saturnotaku]

I want to gauge how everyone feels about 3dmark as a program, benchmarking tool, etc. Is the program and its results important to you? Think about when you read video card reviews and run tests on your own system when you answer this question.

Vote here.

As a 3D demo, it looks nifty. Cool way to show off your new computer to people when there are no games that support your Uber Video Cards yet.

As a stress testing program, it's iffy. I think a looped UT2003 bench at max AA/AF would be more effective at causing artifacts on an OC'd card.

As a benchmarking tool? HAH. Useless.

[digitalwanderer]

I still love 3dm2k1se for system stability testing and to compare me system pre-upgrade to post-upgrade, it's about the best all-around system tester I know for D3D.

I don't like 3dm2k3 so much since it is just so graphic card intensive that it pretty much ignores the rest of the system, I wish it would be just a bit more holistic in it's system measurements.

3dm2k is still the nut-buster for me to see if an OC is good, if it can make it thru Helicopter high-detail it can make it thru just about anything....it's also great for testing older systems out with. (I build a lot of cheap web-cruisers/kid gamer systems for families that don't have PCs, my way of giving back I guess and keeping too many old parts from clogging up my basement/garage...plus it might be a cheap old PC put together from spare bits but to a 3-5 year old it is a top-o-the-line wonderbox! )

So I guess I like 'em, but not 3dm2k3 so much 'cept for the fact that it's brought the whole "optimization" issue to light....I owe it a nod for that if nothing else.

[Hanners]

Depends what I'm using it for.

To gauge the ability of a video card to perform a particular feature or function, it's okay. To compare video cards alone (pretty much forgetting the rest of the system) in a completely synthetic scenario, then it's great. To work out how fast your machine will run UT2003/Doom III etc, it's a load of rubbish.

Assuming you mean 3DMark 2003 that is, although the same rules more or less apply to 3DMark2001.

[saturnotaku]

I'm primarily thinking about 3dmark 2001 here. The only time I ever use it is to test OC stability on my video cards. Other than that, the program is of no importance to me whatsoever and is never a deciding factor when I'm purchasing a video card.

The GTs of 3dmark are rather useless except to ascertain the effectiveness of tweaking, or perform system stability via loops.

The synthetic tests can be useful.

[b00bie]

When I look at the reviews of the cards, I look more at the benchmarks from the games rather than 3dmark2k1 or 2k3. If I'm looking for a new card, I want it to play my games the way I want them (maxed out ). I know that these scores for some people are just everything. If you don't have a good score on 3dmark2k1, your card sux0rz
So to answer your question, I don't give a damn about the 3dmark scores at all other than to see the oc results.