Overclocking is not meant for high end products!

The whole point of overclocking is to get a cheap product for which you know may be clocked A LOT more than it is... that goes for pretty much any component, "get" a product that's priced sometimes twice as much by overclocking the cheap one (obviously "get a product" means "get a performance of a product" no need to explain really). People usually miss this value of overclocking and replace it with ... oh I am gonna get that 1fps more even if I had to spend tons of money on cooling... or whatever. Silly.

My Radeon 9800 NP is by default clocked at ~ 300/300 so after flashing it to pro and overclocking it I got a product that runs 430/380 speed which is pretty much your 9800XT level.

And I got that for half the money I would have payed for the XT, and all it took is a little bit of research on released models of 9800 line of products

I bet you that we are not talking 2-5 fps gain here.

Similar story goes for my Barton 2500. It runs at 3200+ performance level for more than half the money.

Call it a steal!

It’s a shame the heatsinks were not removed to reveal the speed of the memory.
This would have shown if it were the memory or the memory subsystem. I'm guessing the ATI had 2.5ms and the Nvidia had 2ms memory. Still this should have allowed the memory speed to well exceed the (2.5ms) 800 and (2ms) 1000 barriers, especially with volt modding.

"So, we can conclude that this particular game is more dependent on the processor speed, rather than the graphics card performance."

So why test the game(s) at 1024x768 or 1600x1200 this is more a Processor or Chipset performance test then a graphics card test, start testing at 640x480 AGAIN that should do! At least some difference should come up.

fuck the world