• Competitor rules

    Please remember that any mention of competitors, hinting at competitors or offering to provide details of competitors will result in an account suspension. The full rules can be found under the 'Terms and Rules' link in the bottom right corner of your screen. Just don't mention competitors in any way, shape or form and you'll be OK.

My 3Dmark11 scores E8500 vs i7 2600K cpu's

chaparral

chaparral

chaparral

Caporegime
Joined
27 Nov 2005
Posts
25,124
Location
Guernsey
My 3Dmark11 score with E8500 @3800mhz cpu / SLI 580 gtx
Untitled-42.png




My 3Dmark11 score with i7 2600k @4400 or 4600mhz (can't remember which speed :o) / SLI 580 gtx
P11522.png
 
straxusii said:
Classic example of how irrelevant CPU is compared to GPU in most games.
Click to expand...
Well, it really depends on how you look at it. I definiely agree that a fast dual-core techinically fast enough with high-end graphic, but the biggest problem is with some damn console ports not really run well on a dual-core CPU, and I'm afraid there's nothing much we can do about the poor portings that's been going on for like...forever.
 
The problem is there are really very few bad console ports, often cited as the worst case is GTA 4, which IS NOT A BAD PORT. Its just shocking coding, it doesn't run slower on a PC than a console, it runs like turd on both platforms, it uses all the general console tricks to hide its slowness and fails, its got shadows from 15 years ago, get so freaking slow going over a bridge at night while doing streetlight effects it makes my eyes want to bleed.

Theres few if any other console games that required a fast quad core to run a PC version of.

AS for the tests, despite almost doubling the cpu score, four times the physics score, 3 times the combined score, the biggest jump in actual graphics test performance is 14% for test 1, around 11 % for test 4, and the two middle tests are closer to a 5% boost.

So yes, CPU's are largely irrelevant, CPU reviews run a top end gpu at half the resolution you should with reduced settings to show the difference.


Sure at a rubbish resolution and settings a dual core athlon something or other can only get you 80fps in some game, while a quad core Sandy can give you 762fps, but when you up the resolution and settings and your graphics card can't churn out more than 70fps at that point, the extra power of the Sandy core is going to waste.

Doesn't mean they are bad or pointless, there are more cpu limited games, they are more power efficient, you'll be able to have other things going on in the background, or video on a second screen, or youtube with gpu acceleration etc.

But in terms of gamings, a pretty expensive upgrade to something several times faster offers less than a 10% boost on average, and that difference will show up in real games.

If you go from a 4850 to a 5850 you'd see more than a 10% boost........ gpu's are where your upgrade money should go for IF your primary concern is gaming.

For instance if you could only afford X amount that allowed you a cheap E8500 and a 580gtx, or a full on Sandybridge setup and a 460gtx which would be faster?
 
drunkenmaster said:
For instance if you could only afford X amount that allowed you a cheap E8500 and a 580gtx, or a full on Sandybridge setup and a 460gtx which would be faster?
Click to expand...

It depends who you're asking. Competive gamers with 120hz displays would choose the latter every time. You can run at lower settings on the Sandybridge+460gtx to achieve 200fps average, whereas the E8500+580gtx may struggle to get above 60fps in the same scenario, regardless of settings.
 
phatboy said:
good lord! that is one hell of a jump mate!
Click to expand...
But it one hell a jump down for me :(

As my i7 motherboard packed up and am back to using my old E8500 cpu & board till i get my i7 motherboard fixed/replaced

Skellig said:
Why only 3.8 Ghz for the E8500? Mine's at 4 Ghz due to the fact that I don't need or want it any higher. But there are 24/7 E8500's at 4.3+
Click to expand...
It only at 3800mhz because am only using the crappy intel stock cooler due to the reason above
(I Used to run this E8500 at 4200mhz with watercooling ;) )
 
Last edited:
Generally speaking and to a certain point you will get better min and average FPS with a faster CPU. So benchmarks are not the whole story.

Unless you have money to spare its better to throw cash at the GPU if you play games.
 
does 3dmark 11 still use nvidias GPU for physics? because it looked like you had physics on the cpu for the first one but the gpu doing them for the second.

in the older 3dmark this would give you a huge score increase
 
LiamJ said:
The last test should be used to compare performance in games. The GPU and Physics tests use the full power of the GPU/CPU regardless of other hardware.

Look at the graph I made for my setup in this thread: http://forums.overclockers.co.uk/showpost.php?p=18195464&postcount=1

You can clearly see how my CPU is bottlenecking, and that is why the OP's score improved massively with a CPU upgrade.
Click to expand...

The Graphics FPS have not "massively" increased though. Depends if you play games or spend all your time with benchmark programs I suppose.
 
That's the point. The GPUs work at 100% in those tests regardless, so the score is the same. Of course though, games do not work this way, they work combined, as in the final test.
 
You must log in or register to reply here.
Back
Top Bottom