gamegpuCall of Duty Black Ops III Beta CPU Benchmark

Page 2 - Seeking answers? Join the AnandTech community: where nearly half-a-million members share solutions and discuss the latest tech.

YBS1

Golden Member
May 14, 2000
1,945
129
106
I find it funny you use the term destroy when you're already above 150fps minumums. Who gives a flying turd at that point?

No one does, but we aren't discussing playable framerate, we're discussing games taking advantage of more cores.
 

cmdrdredd

Lifer
Dec 12, 2001
27,052
357
126
No one does, but we aren't discussing playable framerate, we're discussing games taking advantage of more cores.


So? When you are above 150fps I don't care if it uses more cores. There is nothing there to justify the extra cores in my view. There is nothing getting destroyed. It is just your choice of wording that doesn't make sense when you look at the performance of both. I don't deny that one is faster but I think it's a little overzealous to point to this as an example of why cores matter. If you get what I'm saying.

Show me a game that goes from borderline unplayable to 60fps just based on the CPU and you are on to something.
 
Last edited:

LTC8K6

Lifer
Mar 10, 2004
28,520
1,576
126
Nice showing for the AMD CPU's.

I'm not sure an 8 thread cpu at 4Ghz being beaten/tied by a 4 thread CPU at 3.4Ghz is a nice showing.

Or the 6 thread CPU at 3.5 being beaten by the 4 thread CPU at 3.5.
 

YBS1

Golden Member
May 14, 2000
1,945
129
106
So? When you are above 150fps I don't care if it uses more cores. There is nothing there to justify the extra cores in my view. There is nothing getting destroyed. It is just your choice of wording that doesn't make sense when you look at the performance of both. I don't deny that one is faster but I think it's a little overzealous to point to this as an example of why cores matter. If you get what I'm saying.

Show me a game that goes from borderline unplayable to 60fps just based on the CPU and you are on to something.

The particular numbers are irrelevant, in the next game it may be an even more ridiculous figure say 240 vs. 310, or it may be very relevant to the user experience say 58 vs. 86. We don't even know in this particular case if this was the lightest map in the game or a worst case scenario. The point is games are most likely to increasingly take advantage of more cores.

About the "destroyed" wording...Say AMD next week came out with Zen (or whatever) and it bested Intel by 20% in h264/265 encoding, we'd all mostly be in agreement it "destroyed" the Intel in that comparison. Someone enters the discussion and states "I've viewed both output files and they are identical. Thus the user experience is the same!" We'd all do a collective face palm. I was discussing the performance difference, despite the fact the 5960 is at a significant clock speed disadvantage. The fact you can't see it in this particular example doesn't mean the performance difference isn't there, you may very well be able to see it next time.
 

SPBHM

Diamond Member
Sep 12, 2012
5,076
440
126
I'm not sure an 8 thread cpu at 4Ghz being beaten/tied by a 4 thread CPU at 3.4Ghz is a nice showing.

Or the 6 thread CPU at 3.5 being beaten by the 4 thread CPU at 3.5.

if they performed like this on most games it would be pretty good for AMD, but in that case they would also probably cost more, but for this game specifically it's quite good for AMD, I think performance per $ is more important here than performance per/clock/thread...
 

LTC8K6

Lifer
Mar 10, 2004
28,520
1,576
126
if they performed like this on most games it would be pretty good for AMD, but in that case they would also probably cost more, but for this game specifically it's quite good for AMD, I think performance per $ is more important here than performance per/clock/thread...

I make up the price difference in the electric bill. :biggrin:
 

Blitzvogel

Platinum Member
Oct 17, 2010
2,012
23
81
Nice showing for the AMD CPU's.

FX-8xxxs are technically octo-cores, so I guess that makes it easy for porting from octo-core x86 consoles. Though, I think both the PS4 and Xbone use one or two cores just for the OS :hmm:

It is funny how dual-module APUs and i3s are quite a bit more capable than the consoles in terms of straight x86 performance. Not at all hard to build a console killer these days, except in the case of bad ports.......
 

cmdrdredd

Lifer
Dec 12, 2001
27,052
357
126
The particular numbers are irrelevant, in the next game it may be an even more ridiculous figure say 240 vs. 310, or it may be very relevant to the user experience say 58 vs. 86. We don't even know in this particular case if this was the lightest map in the game or a worst case scenario. The point is games are most likely to increasingly take advantage of more cores.



About the "destroyed" wording...Say AMD next week came out with Zen (or whatever) and it bested Intel by 20% in h264/265 encoding, we'd all mostly be in agreement it "destroyed" the Intel in that comparison. Someone enters the discussion and states "I've viewed both output files and they are identical. Thus the user experience is the same!" We'd all do a collective face palm. I was discussing the performance difference, despite the fact the 5960 is at a significant clock speed disadvantage. The fact you can't see it in this particular example doesn't mean the performance difference isn't there, you may very well be able to see it next time.

To me it seems like you're looking for any excuse to try to justify to everyone that more than 4 cores is worth the extra cost. No amount of reasonable arguments will sway that view.

The facts remain that you said one CPU got destroyed by another when neither was below 150fps minimums that is not destroyed.
 
Last edited:

YBS1

Golden Member
May 14, 2000
1,945
129
106
To me it seems like you're looking for any excuse to try to justify to everyone that more than 4 cores is worth the extra cost. No amount of reasonable arguments will sway that view.

The facts remain that you said one CPU got destroyed by another when neither was below 150fps minimums that is not destroyed.

I personally don't give a crap if the lot of you are running Pentium 4s. I'm not even trying to justify a price difference, I couldn't care less if someone wants to save a buck. I do find it a tad odd though that the very same people that comment "but games", "but games" while championing their 4 cores chips will discount every single instance of the 6/8 cores out performing them. Bad port, game sucks anyway, framerate too high to matter, etc., etc.

It strikes me much the same way as all of the comments I saw to the effect of "Sandy Bridge to Skylake = Meh..Sidegrade" immediately after the 6700K reviews. Now, is it worth the money to go from SB to SL?? That's up to the individual, but it most definitely is not a sidegrade.
 
Last edited:

tential

Diamond Member
May 13, 2008
7,348
642
121
I personally don't give a crap if the lot of you are running Pentium 4s. I'm not even trying to justify a price difference, I couldn't care less if someone wants to save a buck. I do find it a tad odd though that the very same people that comment "but games", "but games" while championing their 4 cores chips will discount every single instance of the 6/8 cores out performing them. Bad port, game sucks anyway, framerate too high to matter, etc., etc.

It's a benchmark shown to show what core scaling is.... you don't actually say the framerate is too high so the bench isn't relevant. Not everyone has 980Ti.
A person with a single GPU will will see some benefit and increased minimums compared to others with weaker CPUs.

Do we seriously not understand how to interpret this benchmark now?
 

YBS1

Golden Member
May 14, 2000
1,945
129
106
It's a benchmark shown to show what core scaling is.... you don't actually say the framerate is too high so the bench isn't relevant. Not everyone has 980Ti.
A person with a single GPU will will see some benefit and increased minimums compared to others with weaker CPUs.

Do we seriously not understand how to interpret this benchmark now?
Umm...I think you might have directed that at the wrong person. I'm not the one saying the benchmark isn't relevant.
 

guskline

Diamond Member
Apr 17, 2006
5,338
476
126
My 5960x with a single 980TI shows a min of 3fps, max of 139 and an average of 89.749 fps measured by Fraps.

My 4790K with CF 290s shows min of 2 fps, max of 119 and average 71.877. I doubt CF is enabled in the game.
 
Last edited:
Aug 11, 2008
10,451
642
126
I personally don't give a crap if the lot of you are running Pentium 4s. I'm not even trying to justify a price difference, I couldn't care less if someone wants to save a buck. I do find it a tad odd though that the very same people that comment "but games", "but games" while championing their 4 cores chips will discount every single instance of the 6/8 cores out performing them. Bad port, game sucks anyway, framerate too high to matter, etc., etc.

It strikes me much the same way as all of the comments I saw to the effect of "Sandy Bridge to Skylake = Meh..Sidegrade" immediately after the 6700K reviews. Now, is it worth the money to go from SB to SL?? That's up to the individual, but it most definitely is not a sidegrade.

Well, lets face it, a high end 4 thread quad intel is still faster than anything from AMD in the vast majority of games, and that is stock. At best, for AMD, 8350 and 4690k are pretty much equal, and the intel probably has slightly more overclocking headroom. Not to mention that if you max overclock, the 8350 is going to use a lot of power.

So I think it is still quite reasonable to buy something like a 4690k for gaming. You dont really have to discount the advantages of hex cores or hyperthreading to still be satisfied with a non-HT quad. I dont really think anyone is trying to say hyperthreading or hex cores (intel) dont have an advantage in some games, but whether or not the advantage is worth the extra cost is up to the purchaser. My personal opinion is that in a less than 1K rig the cost savings of an i5 are probably enough to justify it, but if you start getting in the 1K plus cost range, then a hex core or i7 are probably the way to go.
 

SimianR

Senior member
Mar 10, 2011
609
16
81
I'm not sure an 8 thread cpu at 4Ghz being beaten/tied by a 4 thread CPU at 3.4Ghz is a nice showing.

Or the 6 thread CPU at 3.5 being beaten by the 4 thread CPU at 3.5.

I was talking more from a perf/$ point of view. Gonna take you awhile to make up that difference on your electric bill. I'm running an i5 4670k myself, but if the 8350's performance was this consistent in all games I probably would have gone AMD.
 

cmdrdredd

Lifer
Dec 12, 2001
27,052
357
126
Umm...I think you might have directed that at the wrong person. I'm not the one saying the benchmark isn't relevant.

I'm not either, what this is showing is which CPUs can keep up good performance when you move to a CPU limited situation. It doesn't seem too demanding unless you are on an AMD platform.
 

SlowSpyder

Lifer
Jan 12, 2005
17,305
1,002
126
I'm not sure an 8 thread cpu at 4Ghz being beaten/tied by a 4 thread CPU at 3.4Ghz is a nice showing.

Or the 6 thread CPU at 3.5 being beaten by the 4 thread CPU at 3.5.



Ultimately what matters is how it plays on a system, if the hardware can deliver a good experience or not. All the other stuff is meaningless to someone playing a game. It'd be like saying Intel did poorly here given their R&D budget is more than 12x AMD's. The truth is CPU's that cost barely $100 can drive a 120Hz monitor if this bench is representative of actual performance. I don't see how it's a bad showing. I hope more games use this engine, more better threaded games would be great.
 

TeknoBug

Platinum Member
Oct 2, 2013
2,084
31
91
Hmm so optimizations from the consoles are being carried over. This is a good sign...now I'm waiting to be surprised by dx12 support.
Not really sure if that's a good thing, 90% of console ports runs like crap on PC, BO3 beta on PC has mouse input lag and some textures look like crap, and doesn't run well higher than 1080p (sluggish on 1440p).

This game running ~120-150fps even on a moderate PC while barely 60fps on a PS4 (there's a few laggy spots on some maps)...
 

Erenhardt

Diamond Member
Dec 1, 2012
3,251
105
101
Not really sure if that's a good thing, 90% of console ports runs like crap on PC, BO3 beta on PC has mouse input lag and some textures look like crap, and doesn't run well higher than 1080p (sluggish on 1440p).

This game running ~120-150fps even on a moderate PC while barely 60fps on a PS4 (there's a few laggy spots on some maps)...

I will help you correct yourself, here:
medish.jpg


If Top GPU SLI system is your moderate PC, than we have nothing to talk about here, and anywhere else. </alternate_reality>
 
Aug 11, 2008
10,451
642
126
I was talking more from a perf/$ point of view. Gonna take you awhile to make up that difference on your electric bill. I'm running an i5 4670k myself, but if the 8350's performance was this consistent in all games I probably would have gone AMD.

Yea, that is the problem with FX. Only equal to intel in the best case scenarios, and that only to an i5, while using more power. And much worse in some games, even newer ones. Just doesnt seem worth it to save a hundred bucks or less in a rig that will last several years.

I will grant that the FX is more attractive than previously, since more games are performing well on it, the price has come down, and they are somewhat more efficient. But intel is still the way to go in a mid to high end rig.