280x Unigine Heaven 4.0 score, lower than card benchmark in review

Jay76

Distinguished
Mar 9, 2011
63
0
18,640
I have a Gigabyte Windforce 280x OC rev 2. I ran Unigine Heaven 4.0 a few times and can't come close to the 49.6 fps, from this reviews Unigine Heaven 4 benchmark.
http://www.fudzilla.com/home/item/32990-gigabyte-r9-280x-oc-windforce-reviewed?tmpl=component&print=1
I am at 41.8fps. Same settings, 1920x1080p, tessellation-Extreme, AA=Off.
GPU, 1155mhz, and memory 1600mhz.(6600.)
Software overclocked my cpu to 4.2ghz also using the Asus AI thing.
This is about the best I could get. FPS: 41.8,
Score:1053, Min FPS:7.2,Max FPS:97.4
Are there other settings that I need to tweek? Or did I just get bad card? Or is it normal for the review card to benchmark better?

i7 4770K
Asus Z87 A
Gigabyte Windforce 280 OC rev 2
Crucial Ballistix 8GB 1600
Samsung Evo 250 GB
Pioneer BR/B
Corsair AX 750 PSU


 
Solution
This is also due to the silicon lottery, your card simple just won't overclock as high as the reviewer got it to...also, This isn't a great review to be basing your performance off of. A good review will tell you what version of drivers they are using, and what other hardware they are using. You likely are getting the realistic fps for your card, while the review is an amd biased review that is fibbing slightly about their numbers
your cpu would be the thing software overclocked (though software overclocking cpu's is only a good idea on really high end boards).
Other than that, updating your drivers would be the first solution. The reason i say this is that you are still coming within 8 fps of someone else. This is a gap that can be closed with proper drivers

Also, no two graphics cards will always perform the same way, it is how the 'silicon lottery' works. His card may overclock better than yours, or vice versa.

It IS normal for a reviewcard to score maybe 2-3% higher than the consumer cards, but not much difference than that.
 
Also from what i read in that review, they do not tell you what they are using as a test bench (listed no other parts besides the gpu). For all you know they could have used a 3960k at like 4.8 ghz which would definately account for some of the fps difference you are seeing
 

Jay76

Distinguished
Mar 9, 2011
63
0
18,640
Yes, I meant CPU. I downloaded and installed the latest amd drivers prior to running it. So I should overclock the CPU in the Bios instead?

Also while I am able to run Unigine at the same GPU overclocks as in the review, it isn't stable at those overclocks when I game. The game crashes and gives me some type of amd driver failure notice.
 
This is also due to the silicon lottery, your card simple just won't overclock as high as the reviewer got it to...also, This isn't a great review to be basing your performance off of. A good review will tell you what version of drivers they are using, and what other hardware they are using. You likely are getting the realistic fps for your card, while the review is an amd biased review that is fibbing slightly about their numbers
 
Solution

zielritter

Honorable
Nov 4, 2013
80
0
10,660
I originally had a Gigabyte 270x that would not overclock past 1130mhz core without crashing during BF4. I wanted to give Crossfire a try and picked up an XFX that zooms along at 1225mhz no matter what I throw at it. I get 40fps (39.9) on the extreme preset for Unigine Heaven with the XFX. Ended up returning Gigabyte because Crossfire was sh*t on BF4.

I've read that GB pick lower "asic" silicon and therefore may not overclock or perform as well as an MSI or an ASUS that is comparable on paper. Don't quote me on that though.
 

TRENDING THREADS