[hardware.info]Radeon HD 7970 and GeForce GTX 680 tested with 10 CPUs

[Final8ty]

Two graphics cards, ten processors, ten benchmarks

If you like playing games in Full HD resolution with all fancy graphical effects enabled, you obviously need to invest in a high-end graphics card. If you're looking at buying one of the current high-end cards, such as the AMD Radeon HD 7970 GHz Edition or the Nvidia GeForce GTX 680, it's worth taking a closer look at the rest of your computer. Will your PC be able to squeeze the most out of these expensive powerhouse video cards, or do you have a bottleneck hiding somewhere? With a speedy Core i7 processor you obviously need not worry, but what if you have something like an Intel Core i3 or AMD A6? Hardware.Info examined just that scenario, so read on to find out.

http://uk.hardware.info/reviews/3714/radeon-hd-7970-and-geforce-gtx-680-tested-with-10-cpus

 

[Fx1]

did they split the AMD and Intel charts to make it harder for people to see how far behind AMD are currently.

1080p and 3x1080p bit of a jump.

 

[chimaxi83]

did they split the AMD and Intel charts to make it harder for people to see how far behind AMD are currently.

1080p and 3x1080p bit of a jump.

What does this even mean? Is every post of yours going to be retarded A vs B trolling?

 

[Lepton87]

GTX680 gaining over 12% of performance by switching from 3770k to 3960X seems strange in BF3. It almost catches up to 7970 with that CPU and people are complaining about AMD's drivers. Overall it seems more sensitive to CPU performance. 3770K to 3820k and performance dips while AMD's card performs the same.

 

[Grooveriding]

I didn't pay much attention to the AMD CPU results, they are pretty much all behind Intel results, but the Battlefield 3 results are pretty interesting.

The GTX 680 is a fair bit behind the 7970GE across all resolutions with any CPU, but, once they use the SB-E chip it gets much closer to the 7970GE suddenly compared to using the Ivy and Sandy chips... interesting. Wonder why.

 

[Lepton87]

BF3 results are very strange, why 7970 is faster with pentium then core i3? Also how on earth 3770k is more than 2x faster than 3570k at 3x1080p?

 

[nismotigerwvu]

Across the board the results seem just, off. I think we might be seeing a case where the number of variables might not have been kept in check (ram speed/timing, not using a fresh install, bios setting, ect..) or maybe even a small sample size. No accusations of misconduct or anything, I think they might have just been trying to tackle a bigger project than they had resources on hand.

 

[Lonyo]

I didn't pay much attention to the AMD CPU results, they are pretty much all behind Intel results, but the Battlefield 3 results are pretty interesting.

The GTX 680 is a fair bit behind the 7970GE across all resolutions with any CPU, but, once they use the SB-E chip it gets much closer to the 7970GE suddenly compared to using the Ivy and Sandy chips... interesting. Wonder why.

4 core game and multithreaded drivers?
it's not the SB-E that gets faster, it's 6 core SB-E.

 

[Lepton87]

Across the board the results seem just, off. I think we might be seeing a case where the number of variables might not have been kept in check (ram speed/timing, not using a fresh install, bios setting, ect..) or maybe even a small sample size. No accusations of misconduct or anything, I think they might have just been trying to tackle a bigger project than they had resources on hand.

About that sample size, they didn't say anything about running the tests multiple times and getting rid of outliers etc.

4 core game and multithreaded drivers?
it's not the SB-E that gets faster, it's 6 core SB-E.

I don't think he meant 3820 only EE
IMHO from the results it is clear that they didn't properly take care of run to run variations.

 

[BD231]

So the trend of more cpu intensive drivers from nvidia continues, nothing new. They've always had higher overhead. Having spent 300 hours using an nvidia card on BF3 its clear to me nvidia uses more filtering methods to smooth out gameplay and add more of a gloss to their 3d image quality.

The difference is plain as day but i would imagine there's a penalty for it.

 

[Grooveriding]

4 core game and multithreaded drivers?
it's not the SB-E that gets faster, it's 6 core SB-E.

Yeah, I should of been more specific. They did have a 3820 there, but I meant the 3960X. I know BF3 performs better on SB-E, but I would of expected the gains to be equal using either video card. The 680 is slower than the 7970GE by a pretty consistent margin until they use the 3960X, at which point it gets closer to the 7970GE's score than it does with any other CPU.

 

[dragantoe]

poop! i thought buying a g860 would give me better performance than the a6

 

[The Alias]

I didn't pay much attention to the AMD CPU results, they are pretty much all behind Intel results, but the Battlefield 3 results are pretty interesting.

It's also interesting to note that AMD's low-end processors score better than Intel's cheap processors in combination with high-end graphics cards. That means you're better off with an AMD A6-5400K (about £56) or AMD A10-5800K (about £99), than an Intel Pentium G860 (about £52) or an Intel Core i3 (about £91). Particularly the Pentium and the Core i3 have a negative impact on the performance of the graphics cards.

intel bias or troll ? There are more cpu besides just the high end

 

[Lepton87]

Yeah, I should have been more specific. They did have a 3820 there, but I meant the 3960X. I know BF3 performs better on SB-E, but I would have expected the gains to be equal using either video card. The 680 is slower than the 7970GE by a pretty consistent margin until they use the 3960X, at which point it gets closer to the 7970GE's score than it does with any other CPU.

53 59 64 65 64 73
71 68 74 74 74 76

I wouldn't call that consistent. The lead is only pretty consistent across 3 CPUs. What's also strange is that nV gets 69fps on 3 of AMD's cpus from A10 to FX8350. More than on 3770k.

 

[frozentundra123456]

Across the board the results seem just, off. I think we might be seeing a case where the number of variables might not have been kept in check (ram speed/timing, not using a fresh install, bios setting, ect..) or maybe even a small sample size. No accusations of misconduct or anything, I think they might have just been trying to tackle a bigger project than they had resources on hand.

Yea, their results seem rather odd. I also find the way they presented to data very difficult to interpret. I would have preferred to see a comparison of AMD vs Intel at various price points instead of the data presented separately for each manufacturer. The dual core Intel results seem not correct somehow. I have a hard time believing that the i3 is consistently slower than the A10, and that there is so little difference between the i3 and the Pentium. According to Anand's bench, the i3 is equal or faster than the A10 in every game tested, whether at low or high resolution. Also strange that the i3 tested is a Sandy Bridge rather than Ivy, but that should not really make much difference in the results.