XFX 9600GSO 384mb vs Sapphire HD4670 512mb

Test Setup 

For this review, we will be using a test setup consisting of parts normally found in a mid- to high-end PC. A common mistake made when benchmarking graphics cards is that the rest of the PC system isn’t sufficient enough to test the GPU to its limits. This results in a bottleneck situation, where the system can only run at the speed of its slowest component. For this reason, the test configuration chosen below has been specially selected to give each of the graphics card on test the headroom they require in order to produce the best results. To ensure we had no CPU bottlenecks we also overclocked the Q6600 to 3.6ghz for all of the tests. A fresh Vista installation along with Service Pack 1 updates were installed for each video card, along with the latest video card drivers available at the time of the review.
 
Test Setup
  
 
Overclocking
 
With the 9600GSO weighing in at an already impressive 580mhz Core/1400mhz Memory (up on the stock 550mhz employed by other manufacturers despite this not being any sort of special edition card), I was intrigued to see how much further it could be pushed, especially considering that this is not the range topping XXX edition which can clock to 680mhz on the core.
 
I was also interested in how much further the HD4670 could be pushed given that it is based on a 55nm fabrication process. Cooling however may become an issue for this card, given that the stock cooler is half the size of the 9600GSO’s.
 
 
Impressive is not the word. Taking the 9600 GSO up to 700mhz/2000mhz certainly made a noticeable difference to 3DMark06 adding an extra 2000 points. The HD4670 actually managed a 100mhz overclock too but the overclock didn’t have such a drastic effect on the 3DMark score. We managed to pin another 50mhz on the memory but any more resulted in a BSOD. Let’s see how the synthetic benchmarks run with the cards set back to their factory settings…