Game Performance: Quad vs Duo

One of the benefits of keeping awake the whole night and sleeping after Sehr is you get about 4 extra hours with nothing to do. Here's how I utilized about 8 of them. ;)

I wanted to see how four cores perform against two in games, how much of a difference over clocking makes as well as how well Crossfire scales to add extra FPS. Almost all of the websites use very expensive and latest motherboards/processors which are not available in Pakistan and if they are they are extremely unaffordable. Both the processors and graphic cards I benchmarked offer the best 'bang for the buck' right now.

Components used and explained.

Intel Q6600 = Quad ( four cores )

Intel E5200 = Dual Core (NOT Core2Duo)

HIS 4850 = Single ATI 4850

HIS 4850x2 = Two 4850 running in Crossfire mode

Motherboard = MSI P35 Platinum

Ram = Corsair 2 GB DHX

All tests were run at a resolution of 1920 x 1080 with all settings on High/Max or Ultra.

Note: My Crossfire tests were a bit crippled by the fact that the 2nd PCIx slot was X4 not X16 which resulted in the 2nd card performing at 86 % instead of 100 % at x16.

On with the results !

First we have the GPU and CPU intensive World in Conflict.

Next Company of Hero's

And last the Fantastic Devil May Cry 4 (which is Quad optimized)

There you have it, comments appreciated !

wow! very good work indeed. i like the layout of graphs. Quad doesn't provide much difference after O'Cing. I bet in multitasking quad will beat the hell out of duo. again very good work. you should put this thread in your signatures.

two thumbs up for you man! good effort!

Thanks and I think I should add a word on what FPS are and what difference it makes to game play.

Fps is frames per second, which means how fast your computer can redraw a single picture every second. The faster it can redraw the smoother your picture will be. You television for example is displayed at 25 frames per second, DVD movies run at 30 fps.

In games, 30 fps is the minimum (below that the game will be jerky), 40-45 fps per sec are really good, 60 fps are optimal and 72 fps is the maximum a human eye can perceive. Anything over that IMO is not noticeable.

So looking at the results above, I would not notice much of a difference in Devil May Cry 4 while playing on a E5200 at stock speed with a single 4850 vs Q6600 over clocked to 3.2 with 2x4850's.

World in conflict would roughly play 18 % better & Company of Hero's would play 14 % better.

^abbas did you monitor the temperatures while doing all this. what was the cards temperature on crossfire in over locking mode (in load). somehow i feel e5200 would be hotter than the Quad...

edit::::one more thing::: r u using stock cooling or external cooling technique.???

Well done Abbas! :)

I ran a test on Crysis with max settings on the following C2D system:

C2D e8400 running at stock 3Ghz speed

9600GT with 512mb DDR2 RAM


Min: 41

Max: 85

Avg: 56.554

Is it OK performance?

@Kratos24 I did not monitor the temprature but the Q6600 did not go over 70C and the Duo did not go over 60 C. The duo actually runs cooler than the quad. Four cores generate more heat than two. The gpu temp is around 55 C in idle (4850's run hot). I am using a Zalman 8700 (HTPC air cooler) for the Q6600 and the Q6600 stock heatsink for the E5200.

@Specter. Thanks. The performance is great at 56 fps average enjoy !

I made another chart at a friend's request for 1280x1024. Will post that here as well.

please run the same test but quadcore 4 coreQ9950 3.23Ghz VS core 2duo E8400 3core 3.01Ghz and i bet E8400 will beat Q9950

gud job man...nice benchmarks

Impressive work Bro!

gud work

DMC4 and COH do not support quad core, and if people think why the E5200 is so close to the Q6600, then its simple:

1. It says Dual core but it is really core2duo (Damn marketing Dept)

2. Its a 45nm CPU.

3. The 45nm CPU 800MHz is equal or more to a 65nm CPU 1066Mhz

4. The reason why Q6600 is still ahead cuz the L2 cache is 8Mb total, which means 4Mb per each dual-core die. Where as E5200 has 2MB in its dual-core die.

5. go to this link if you don't believe me:

Test same hardware using games:

Unreal Tournament 3

Supreme Commander