Unigine Valley 1.0 Benchmark Thread

Page 11 - Seeking answers? Join the Tom's Hardware community: where nearly two million members share solutions and discuss the latest tech.
Status
Not open for further replies.
Tman450- FX-8320@ 4.0 Ghz- XFX DD 7850 @ 1000/1235- 36.3 FPS- 1518

7f21461610abd4ca529c3d66ed8352ba.3b3185e9d399849a3eaee44b58958f3c.jpeg


KEEP IN MIND, my Monitor is 1600x900 max not TOO far away from 1080 though.. i do have a second monitor hooked up, so that may have had a very small perfmorance impact- i keep a CPU/GPU/RAM monitor widget on it.
 


acutally, 1600x900 is pretty far from 1920x1080

1600x900 is 1440000 pixels, 1920x1080 is 2073600 pixels, which equates that 1600x900 is 69.4% of the pixel difference, or roughly 70%. 30% pixel increase is pretty significant.
 
Still doing some tweaking, these clocks could be a little better voltage wise (1250Mv). 1267/1520+ On mem

EDIT: I am on air. I wish I cooled have a heavy block EK on my PCB. But my series of cards have higher ASIC quality which means that it's better off on air instead of ln2 or water. Now if I had a MATRIX 7970, then I'd be better off going under water or ln2 since the ASIC quality is typically lower.
 


Why would ASIC be on average, higher on the DC2 rather than the Matrix? Nothing spectacular about the DC2 models other than the cooler. The TOP editions are the flagship of Asus.
 


Because...
http://postimg.org/image/n97qf2su1/0456973e/
The Matrix is really best for ln2/water cooling.
 

The tops are just the pre-oced version... lol, I still have the save board and cooler. Triple slot cooler + unlocked without hotwire + 7970 = awesomeness. The TOP editions actually had really bad reviews since people didnt know how to tame them correctly.
 
It may be pointless to put certain cards under water. But the 7970, no matter what make or model, is not one of those cards. They get hot, especially the VRMs, and the cooler they are the better they overclock.
 


Your right on that note, but all GPU's will get hot if you run furmark
 

I can... I hacked into his rig, then looked where he said it was. Simple as cake. I can upload his if you'd like!
 


To put up a screenshot, you have to first upload the picture to a hosting site like www.photobucket.com or any free hosting site. Once it is uploaded, you can get a link which can be posted here,
 

I'm sure I could, but I have no need to. Every game I play right now is as smooth and fluid as synthetic motor oil. I only overclock when I need to. Overclocking for the sake of it makes no sense to me. Why would I push my rather expensive cards when they're running perfectly fine as is? One day, years from now, when they can no longer run games well, I will OC them to get a few years more out of them. Also keep in mind, as the OP stated, this benchmark tends to favour nVidia cards so I'm taking it with a grain of salt. It is a beautiful benchmark to watch though.
 
 
HD7870 pitcairn

1000/1200 with latest driver and increased room temperature %0 tdp setting

of3fib.png


1250/1525 with latest driver and increased room temperature %15 tdp setting

35hpgs8.png



1100/1200 with very old catalyst driver %5 tdp setting

351dc49.jpg


With mild oc

1270/1525 with very old catalyst driver %20 tdp setting

j9wmtt.jpg



Each catalyst setting is "application-controlled" and standard.
 
Yeah they dont talk about that in most benchmark posts and reviews.
Very easy to change settings in GPU control panel and tweak extra FPS while still using the settings on the benchmarking program.
I could get over 40 fps on this bench with same clocks and looked like it was at bench settings
so these threads are kind of bogus in some ways
 


Reminds me of that Gold Plated GTX 480 that used to float around. :lol:

Nice scores on the 7870 BTW.

 
Yeah that 7870 is matching up to my OCd 660Ti
nice numbers

yeah bench threads are interesting
really need people with a good honor system.
When I just did this recent bench my control panel settings werent at default but also I know that from prior benches where the settings were at default I did have similar score.
I should run it with Nvidia control panel reset to see the difference
 
Status
Not open for further replies.