• Welcome to Overclockers Forums! Join us to reply in threads, receive reduced ads, and to customize your site experience!

Win 8 vs Win 7: >4 thread gaming performance

Overclockers is supported by our readers. When you click a link to make a purchase, we may earn a commission. Learn More.

Stryfe

Member
Joined
Mar 4, 2002
Location
Winnipeg, Manitoba, Canada
In a post on a forum for Battlefield 4, a user was talking about how the game is designed for the next gen consoles and their 8 core CPUs. He also made the case that as a result of the use of >4 threads: 1) the game performs better on hyperthreaded Intel quad cores 2) On non-hyperthreaded quad cores the game performs better on Windows 8 than Windows 7 as Windows 8 is better at thread management.

The whole post is here:
http://forum.symthic.com/battlefield-4-technical-discussion/5511-update/index5.html#post188156

I've been looking around for more information on this but so far haven't found much. Does anyone have any links or information they could share on the subject? (Game performance with >4 threads in Win 8 vs Win 7, which game doesn't matter)
 
Last edited:
This write-up shows very little difference i5 to i7. They don't compare W7 to W8, though.
Those results are interesting. If you compare the minimum FPS of the stock i5 to the stock i7 you see a 13% increase for the i7. Granted the i7 does have a 5% higher turbo frequency. But even if you compare the i7 to the i5 @ 4.2ghz it still has a slight advantage despite a frequency deficit. These results are contrary to most other games where having hyperthreading enabled on the i7 usually decreases performance.

So this does seem to confirm that there is performance to be gained in Battlefield 4 by being able to run >4 threads. Now I'd just love to see a comparison of an i5 in Win7 to the same i5 in Win8.
 
I wonder if this is the same test that was done at 1280x720 that I have seen posted here (and many other sites) before... (link blocked from the office). I will have to check this out later, but if it is, that low resolution can really skew results... I would like to see that testing done at a non CPU bound res and see if the performance difference between W7 and W8 still hold true...
 
I think it was a 1080P test, ED. Here's the chart...

1.png

Some other interesting benchmarks for the BF4 beta (I'm still taking them all with a grain of salt, since it's beta, but I would imagine results will be somewhat similar at launch)...

GPU benchmarks
2.png

More GPU (no AA & AA)
3.jpg

4.jpg

vram usage
5.jpg

micro stutter
6.jpg

AMD CPU usage
7.jpg

Intel CPU usage
8.jpg

CPU-limited
9.jpg

RAM usage (though, it contradicts this review, which states 8GB+ was being used!!)
10.jpg
 
Last edited:
Yep, that first graph is the one I was referring to. As they mentioned in the article you can see the top three results on the chart are actually all GPU bound. It makes for a good test to show the difference hyperthreading makes in terms of minimum FPS in BF4. I just wished they'd used the same generation for the Intel CPUs, the i7 probably has an unknown advantage just due to optimizations in the newer architecture
 
I think I need more caffeine... does that first graph ninja so kindly posted show the effect of HT?

I see a 2500K (no HT, quad core) a 3980X (w HT, hex core) and an i3 3220 (w HT, dual core).

There are clockspeed differences and physical core count differences all involved there, no? I would have like to have seen a 2600K at say 4Ghz, with and w/o HT. Then perhaps the Hex at 4Ghz, with and w/o HT. And last that i3 3220 at 4Ghz as well, w and w/o HT.

That would be more conclusive, no? Am I missing something or just being too granular, LOL?!
 
I would have liked to see an HT vs non-HT comparison as well. There are just too many variables in there as-is.

I sorta took that first graph as 4 vs. 6 cores (plus HT) = very small difference, so even less of a difference between 4 core (no HT) vs. 4 core (with HT).

If I only had beta access... (though, I sold my i7, so I couldn't really do a comparison anyhow ;))


EDIT: I had that 2nd-from-the-bottom chart mislabeled (I think the units are FPS?). It sort shows HT vs non-HT (2500k vs 2600k). Looks like a fairly sizeable difference (10FPS), but I'm not sure I trust those results. Do we already have a thread with people posting BF4 beta benches?
 
Last edited:
Oops, I forgot about the existence of hex core Intel CPUs and was thinking the 3960X was a quad core w/HT. Yeah, that does make the comparison much less valuable.

Once the game is released I may have to test this for myself on my 2600k with and without hyperthreading and then, if I'm feeling ambitious, install Win8.1 and run the tests again.
 
That [H] review is weird with the memory...(vram) At 2560x1440, a lesser resolution, with default Ultra settings, I was sitting at 2.4GB vRAM use.

As far as a thread about BF4 beta benchmarks. I created one in the Games section a few weeks ago, yep. :)
 
windows 7 and window 8 have a big difference. I Have used it both and windows 8 is way better not only graphic but also its performances.
 
Please, share some links supporting those points as most I read, there is no difference in graphics (looks). I haven't seen testing done on BF4 since its release (W7 vs W8).
 
windows 7 and window 8 have a big difference. I Have used it both and windows 8 is way better not only graphic but also its performances.

Agreed.

Also, want to know if the builds were both equally fresh installs. not fair to compare a 2 year old Win7 install to a fresh Win8 install.
 
I have W7 and W8 PCs running side-by-side (1st one in sig is 8, 2nd is 7). The Wiindows 8 machine boots 2-3 seconds faster, but I would say that has more to do with the 4th gen vs. 2nd gen i5 and the SSD difference (EVO vs. M4) than the OS. Both operating systems were installed within one day of each other. Gaming performance while swapping the same GPU back and forth (7950 and 7970), yielded almost identical results. Any difference could, likewise, be explained by the other subtle hardware differences.
 
Back