Card Preformance Comparison

[Monody]

I was just wondering how accurate Video Card Database was versus the reviews you can find at sites like GameSpot.

http://www.gpureview.com/show_cards....=560&card2=549

http://www.gamespot.com/features/618...result;title;0

http://www.gamespot.com/features/618...result;title;0

Mostly curious because it seems there is a bit of a difference between what the ranking on Database gave the Nvidia cards versus what the performance showed from the games in the GameSpot reviews.

[theeDEATHMASTER]

tomshardware.com
charts on right side where it says charts. Numbers and stuff are different because of different setups used. Hell even setups that are exactly similar can give slightly different performances.

[MTL]

Just because its easier to read, i use the Guru3d charts
http://www.guru3d.com/index.php?page=vga_1
[screenshot]http://www.guru3d.com/vga/vga-3dmark06.png[/screenshot]
[screenshot]http://www.guru3d.com/vga/vga-fear.png[/screenshot]

oh and the gpureview site says the ATI 4000 series cards are coming out 5-31-08....is that confirmed?

[oldkawman1]

You do have to remember, that SYNTHETIC benchmarks may not be representative of a cards true performance in a particular game. Also, it does seem that NEW cards all seem to do very well in a game benchmark, but, it may not even work in the actual game due to some issue.

So, you really cannot take any benchmark as a "true and complete" representation of how well it will run the actual game.

[Slippster]

yes but in a test where you have differing cards and/or drivers ,
you have to have some constant control as a consistent test vehicle
where everything is EXACTLY THE SAME EACH TIME YOU TEST,
across all tested cards and/or drivers,
and 3D Mark accels at this.

then go ingame and see how you do and compare the "seat of the pants" feel,
acording to the 3D mark score.

and then add differing drivers then you have a constant test bed with the same video and then you can compare and contrast the results in the 3D mark scores along with your personal ingame setup in the games you play.

i see it as an essential tool when im looking at new drivers.

i frequent Xtreme G forums and when new drivers come out they all use 3D mark to test the drivers for frame rates, but also Image Quality and glitches ingame can make results not too consistent so it comes down to your personal config many times and how optimised your pc is for gaming.

ie, are you turning everything off in the background?
are you defragged?
have you moved the swap file to a 2nd hard drive?

have you overclocked your CPU and GPU?
frame rates are directly tied to front side bus speeds.

faster FSB = faster frames.
also faster ram = faster frames.

and ive said it too, "i dont play 3D Mark, i play UT3"...