• Some users have recently had their accounts hijacked. It seems that the now defunct EVGA forums might have compromised your password there and seems many are using the same PW here. We would suggest you UPDATE YOUR PASSWORD and TURN ON 2FA for your account here to further secure it. None of the compromised accounts had 2FA turned on.
    Once you have enabled 2FA, your account will be updated soon to show a badge, letting other members know that you use 2FA to protect your account. This should be beneficial for everyone that uses FSFT.

6950 Unlock Shaders - NICE!

P

polive

[H]ard|Gawd
Joined
Sep 28, 2005
Messages
1,283
I have to give the credit to here:
http://www.overclock.net/ati/923129-hd-6950-810-1250-mhz-shader.html

It takes the core to 880 and 1250 mem, which I think the mem is the same, but the core went up quite a bit with a +12 % in voltage in CCC no artifacting using 3 loops so far in Unigine Heaven.

It was an awesome result, not a 6970 flash, but just unlocking the shaders and going exactly to the first post there, I got awesome results, my windows experience rating increased from 7.8 to 7.9 and Unigine Heaven went up quite a bit in fps with normal tess on at 1920x1200. I didn't have the previous result but now it's almost 1500 score, with a visible difference.

If you have a 6950 and haven't unlocked your shaders DO IT!

BTW, No; Benchmarking is not my favorite game, but I can't imagine that my real games won't see a difference. I don't have time to play much at the moment but if there's any issues I'll post for sure on this thead!
 
I originally had my card flashed to a 6970, but yesterday I flashed it to an unlock 6950 BIOS and my overclock so far has gone up 50mhz on core and 50mhz on the memory. I am guessing it is because of higher voltage to core and looser memory timings.
Maybe I am out of the loop too, but I am sure this might help someone.
 
I was just trying to help for people that maybe hadn't seen it or were on the fence about doing it. I have been outta the loop a bit, and heard some stuff about dangers of going the 6970 route. ;)
 
It's always exciting to do a bios flash :) And always good to hear success stories at the end!!

Just glad you are happy with the results.
 
The actual overclock gives far more of a performance boost than the shader unlock itself, though neither hurts obviously.
 
Keep on trucking. Here is mine, card feels pretty much max'd out.

1000/1400 @ 1.3v

To add some perspective, the additional 50mhz core from 950 @ 1.175 to 1000 @ 1.3 seems to be good for an additional 3% frame rate in Unigine. Not really worth the additional heat at this point.



Unigine1000_1400.jpg
 
Last edited:
People who complain about radeons not overclocking very well, I don't get it, 25% is an amazing overclock really.
 
Are you running any aftermarket cooling on the card, or stock hsf?

I'm going to bench my 6850 unlocked tonight at these settings and see how it compares with stock clocks. I've been trying to compare my 6950 results to the ones in reviews online (of stock 6950 and 6970) but it's hard because often times when I bench my card with Heaven 2.1 with not enough explanation of the settings they used in the review, my card either comes way over their mark or way under their mark. Now I can see your settings, clocks, and can hopefully do an apples to apples comparison. Also, my hardware is 2 generations old compared to what most people benchmarking the cards are running (775 quad core) so I wonder how much that plays into my abnormal scores.

I'm also running the unlocked numbers with powertune at -20% because I'm cheap and don't want to pay for electricity and air conditioning if I don't have to, so FWIW, the card runs fine at -20%, including in BFBC2 and Test Drive Unlimited 2.
 
I ran Heaven 2.1, but at 1920x1080 (display limited), same settings as you had above, and got 44.9fps. Not great, even for stock settings, but I've got older hardware, and run it at powertune -20%. It's actually not that bad, only 2.1fps under the techgage review, which maybe means that the unlock somewhat makes up for my slower system speed? Who knows. I'm very happy that the card stayed under 70c the whole benchmark though and the heat-up period previous to actually running the benchmark. Not bad for a 25c room.
 
Dr. Gumby, that's not really all that bad; the screenshot posted here completely owned my single card score which is overclocked mildly. Also +1 rep for -20% powertune ;)

Kind of OT but I downloaded the heaven unigine benchmark after reading this thread and benched with single card and double card CF and noticed that I got a lower min fps with CF than with single card. Also got me wondering if it would have been much difference if I owned a nf200 based motherboard so that I can get 16x 16x CF.
 
Run the benchmark through two or three times. the first benchmark run sometimes has very bad min frame rates.

I got higher in everything going from single to crossfire and I have two x8 lanes.
 
http://www.hardocp.com/article/2010/08/23/gtx_480_sli_pcie_bandwidth_perf_x16x16_vs_x8x8/

Great review about 16x vs. 8x for PCI-E lanes with modern video cards :D

Bottom line:

"If you are running on a 30" display at 2560x1600 or below, an x8/x8 SLI or CFX configuration will perform the same as a x16/x8 or x16/x16 configuration. The only time that you should even be slightly concerned about running at x8/x8 is when you move up to a multiple display setup. When we pushed the GTX 480 SLI at 5760x1200 we saw up to a 7% difference in performance between x8/x8 and x16/x16, in favor of x16/x16, but that was in one game only."

Definitely check out the article though!
 
You must log in or register to reply here.
Back
Top