Nothing overclocked at current:
Intel(R) Core(TM) i7-5820K CPU @ 3.30GHz
GTX 980 (single card)
16GB DDR4 RAM
SSD

Nothing overclocked at current:
Intel(R) Core(TM) i7-5820K CPU @ 3.30GHz
GTX 980 (single card)
16GB DDR4 RAM
SSD
In response to the CPU usage of the game : all along the tests i did, the benchmark only ever used 1 CPU core, and frequency wasn't even @ max. (Big ass i7 5960X 8C@4.4GHz/4C@4.5GHz/2C@4.6GHz here, but it doesn't affect how the game use CPU, it's only ever one core, frequency will not be a good comparison between AMD / INTEL, but the cpu usage is more bandwidth than actual operation on the cpu from the game/benchmark)
But your CPU can be a bottleneck for your GPU.
Regarding 2.0 Bench vs 3.0, did you do both tests today with the same settings (DX9 Maximum for a fair comparison) ?
You can observe a small performance drop with DX11, for a small quality upgrade.
Last edited by Saifaa; 04-27-2015 at 11:30 PM. Reason: Add of the quote

I have not run the 3.0 bench under dx9 yet, only 11 maxIn response to the CPU usage of the game : all along the tests i did, the benchmark only ever used 1 CPU core, and frequency wasn't even @ max. (Big ass i7 5960X 8C@4.4GHz/4C@4.5GHz/2C@4.6GHz here, but it doesn't affect how the game use CPU, it's only ever one core, frequency will not be a good comparison between AMD / INTEL, but the cpu usage is more bandwidth than actual operation on the cpu from the game/benchmark)
But your CPU can be a bottleneck for your GPU.
Regarding 2.0 Bench vs 3.0, did you do both tests today with the same settings (DX9 Maximum for a fair comparison) ?
You can observe a small performance drop with DX11, for a small quality upgrade.
Player
Hey there, i've been posting this earlier in a lodestone blog post.
Although not that big a difference, display mode affects the score, so i suggest OP to add another recommendation : fullscreen (somewhat easier to play around with, you can use different resolution, and games perform better in fullscreen, better for benchmarking purposes)
Details in the blog post.
1920x1080 Fullscreen : 21578
2560x1440 Fullscreen : 14877
3840x2160 Fullscreen : 7603
Ask if you want other test settings

Pretty much what I was expecting, i'm planning on replacing this old machine anyhow and I do have another 7950 to crossfire in future when I change my motherboard, I'm actually surprised I was still able to hit "very high"
![]()
Last edited by Firepower; 04-27-2015 at 11:24 PM.



I am happy with my results. Not sure if cross-fire was working though, as even during the benchmark my AMD 7870 reached mid 70s temp.
Even though I have a 7870 it registers it as 7800, which is what my second card connected to the cross-fire is.
![]()

It probably didn't do crossfire because its in "borderless windowed mode" set it to full screen and try again, that score is very low for a crossfire set of 7870's or even 7850's.
I posted my benchmark already for DX11 so I thought i'd do some more
1920x1080 Dx11 Max Full screen Score: 5710, Average Frame Rate: 44.300
1920x1080 DX-9 Max Full screen Score: 8936, Average Frame Rate: 73.175
1920x1080 DX11 Max DX9 settings FS : 8419, Average Frame Rate: 66.084
In the DX11 client when using DX9 settings it showed noticeable drop off in the battle scenes (first scene, last scene) and similar performance across all the others, I'm not sure if the graph is delayed but when referenced next to the dx9 score the peaks and troughs of the graph were appearing earlier.
I will likely do this again when I switch my cards over after getting my R9 280 back from RMA.
Last edited by Firepower; 04-28-2015 at 12:21 AM.



Hrm, I never thought it would make that big a difference, but apparently it does. I tried it out, Full Screen my score was 7080.
Only problem seems to be that my main card still got really hot by the end of the benchmark, it was at at least 87 degrees.
When I went into the character creator it actually went up to 91 at one point when I checked after seeing obvious signs of my card being overheated (flickering in some parts of the graphics) and I had to close it down quickly to cool it off.
I am a bit saddened that at least my one card still runs hot. Haven't figured out how to see both simultaniously since the CCC only seems to let me view performance on one at a time as far as I can tell.
|
|
![]() |
![]() |
![]() |
|
|
Cookie Policy
This website uses cookies. If you do not wish us to set cookies on your device, please do not use the website. Please read the Square Enix cookies policy for more information. Your use of the website is also subject to the terms in the Square Enix website terms of use and privacy policy and by using the website you are accepting those terms. The Square Enix terms of use, privacy policy and cookies policy can also be found through links at the bottom of the page.



Reply With Quote










