• We’re currently investigating an issue related to the forum theme and styling that is impacting page layout and visual formatting. The problem has been identified, and we are actively working on a resolution. There is no impact to user data or functionality, this is strictly a front-end display issue. We’ll post an update once the fix has been deployed. Thanks for your patience while we get this sorted.

5960x vs 5820k same clock gaming performance?

hunkeelin

Senior member
There's review out there claiming 5820k have more fps compare to 5960x and 5930k. I'm curious if anyone can confirm that. Given if all three chips is clocked at 4.4ghz why does the 5820k performs better? I don't understand.


Thanks
 
Are they all at stock clocks in the review? If so, 5820's stock + turbo advantage vs. the 5960 easily explains it. Also, how large are the differences?
 
Yes, please link this review. I have never heard of results like this. At the same clocks,5820 and 5930 should be very close. If anything, the 5930 should be slightly faster, because i believe it has more cache. And in pretty much every test I have seen on game.gpu, the 5960x, even at the stock 3.0ghz is somehow faster than either of the other two at stock.
 
Are they all at stock clocks in the review? If so, 5820's stock + turbo advantage vs. the 5960 easily explains it. Also, how large are the differences?

This.

At stock speeds, the 5820k will most likely turbo more aggressively with a smaller number of cores, but not a huge amount of difference. If a game is coded to really benefit from more cores, you should see small bump in performance on the 5960. If both CPUs are behaving similarly (same speed, same # cores active) the 5960 will have a small (1-3% advantage at most) due to more cache.
 
5820K max turbo = 3.6GHz

5930K max turbo = 3.7GHz (which I run at).

5960X max turbo = 3.5GHz

Turbo speeds there would be a small difference, running all three @ 3.0GHz I doubt it. Only the X has 5MB more cache.
 
Yes, please link this review. I have never heard of results like this. At the same clocks,5820 and 5930 should be very close. If anything, the 5930 should be slightly faster, because i believe it has more cache. And in pretty much every test I have seen on game.gpu, the 5960x, even at the stock 3.0ghz is somehow faster than either of the other two at stock.

You are wrong. The only difference aside from the stock clocks is the number of PCI-E lanes 5930K has 40 while 5820K has 28 lanes. That makes 5820K good for only up to 3 graphics cards. I think I can live with that. Both CPUs feature 15MB of L3 cache. The tests you saw are garbage, 5930K is the fastest CPU in games. Also I believe you are mistaken and have confused some CPUs I think there is only one HW-E in the gamegpu charts and that is the 5960X.
http--www.gamegpu.ru-images-stories-Test_GPU-Action-Watch_Dogs_Bad_Blood_-test-wd_proz.jpg

I was right, they don't test any HW-E CPU other than the 5960X. I don't like their selection of CPUs, it can give you the wrong idea that 8 cores might be very useful in games. There are a few CPUs that are faster than the 5960X in games, that is 5930K, 4790K, 4960X and in most cases the 5820K but they omitted all of them and that makes it seem like the INTEL flagship is the fastest gaming CPU. You fell for that and what about tech illiterate people?
 
Last edited:
^ i didn't get my 5960x for games. My plan right now is get a customer water loop and oc the shit out of it. Then after 4 years when I change pc i'll go ln2 and dump it.


Bad language is not allowed in the technical forums.
Markfw900
 
Last edited by a moderator:
You are wrong. The only difference aside from the stock clocks is the number of PCI-E lanes 5930K has 40 while 5820K has 28 lanes. That makes 5820K good for only up to 3 graphics cards. I think I can live with that. Both CPUs feature 15MB of L3 cache. The tests you saw are garbage, 5930K is the fastest CPU in games. Also I believe you are mistaken and have confused some CPUs I think there is only one HW-E in the gamegpu charts and that is the 5960X.
http--www.gamegpu.ru-images-stories-Test_GPU-Action-Watch_Dogs_Bad_Blood_-test-wd_proz.jpg

I was right, they don't test any HW-E CPU other than the 5960X. I don't like their selection of CPUs, it can give you the wrong idea that 8 cores might be very useful in games. There are a few CPUs that are faster than the 5960X in games, that is 5930K, 4790K, 4960X and in most cases the 5820K but they omitted all of them and that makes it seem like the INTEL flagship is the fastest gaming CPU. You fell for that and what about tech illiterate people?

You are correct. But the 5960x is 10% faster than the 3970x. I seriously doubt a Haswell E six core is 10% faster than a 3970x, although it is possible. I agree the results seem strange, but the 5960x is consistently at the top of the charts, vs haswell quad and ivy hex core. And that is at stock clocks, in which the 5960x has a considerable deficit. The OP was asking about performance with all cpus overclocked to the same speed, and I think it is impossible that a 5820k would be faster under those conditions, unless the 5960x is throttling due to overheating.
 
Just wanted to comment on this part.

Isn't the IPC difference from SB-E to HW-E about 10%?

More, because SB-E to IB-E is already 7%.

You are correct. But the 5960x is 10% faster than the 3970x. I seriously doubt a Haswell E six core is 10% faster than a 3970x, although it is possible. I agree the results seem strange, but the 5960x is consistently at the top of the charts, vs haswell quad and ivy hex core. And that is at stock clocks, in which the 5960x has a considerable deficit. The OP was asking about performance with all cpus overclocked to the same speed, and I think it is impossible that a 5820k would be faster under those conditions, unless the 5960x is throttling due to overheating.

At the same frequency the 5820k should be a tiny bit slower than the 5960X simply due to the L3 cache size. HW-E is easily more than 10% faster than the 3970X, IB-E is already 7% faster at the same clocks and HW-E is in turn about 7% faster than IB-E.
 
Back
Top