Question about 3DMark comparison between two cards

[Salvador]

I just purchased a EVGA 7600GS from a friend of mine and was comparing it to my PNY 6600GT. Both cards are AGP and I was using the same Forceware drivers to compare them in 3DMark2001 and 3DMark05.

The 7600GS scored about 500 pts less than the 6600GT in 3DMark2001 and only scored 300 pts better than the 6600GT in 3DMark05. I can understand the slight margin over the 6600GT in 3DMark05 because I wasn't expecting much of an increase, but what gives with the 500 pts less on the 3DMark2001 benchmark? If anything, the newer Forceware drivers would favor the 7600GS over the 6600GT, so I'm completely confused by the pathetic showing in 3DMark2001 for the 7600GS.

Any ideas?

 

[Cookie Monster]

Firstly, 3dmark really isnt the right 3d app to show which card is better. A 7600GS is 10~20% slower than the 6800ultra according to ATs benchmarks.

Secondly, im led to believe that your friend could have a more powerful CPU that could make his 6600GT score 500 more points.

 

[Salvador]

Ok.. What benchmark should I be using?

The cards were both benchmarked in my system with the same drivers. I benchmarked one card with 3DMark2001 and 05, then swapped cards and then benchmarked it.

 

[Piuc2020]

Those it really ****** matter how much marks you get?

Use a game for christs sake.

 

[Cookie Monster]

Originally posted by: Piuc2020
Those it really ****** matter how much marks you get?

Use a game for christs sake.

Agreed. You should be comparing game performance i.e real world performance, not synthetic ones.

 

[Salvador]

Then, why do review sites like Anandtech bother doing benchmark tests? I'll go ahead and take everyone's word for it that it's a better card than the 6600GT.

I just wanted to know if something was up with the card if it was scoring lower than the 6600GT on a benchmark test.