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

Optimal BIOS for 680 GTX overclock?

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

yoadknux

Member
Joined
May 6, 2016
Hello everyone,

Long version:

My overclocking games continue. I own a Zotac 680 GTX 2GB Stock version. By using the MSI Afterburner, by increasing the power limit to +132%, I was able to overclock the Core to +125 and the Memory to +600 (Or in total, Core: 1215, Memory: 3602) . this is actually funny, because +125/+400 causes stuttering after few minutes of Valley (benchmark), while +125/+600 can run for a few hours until I stop it. So far, it hasn't crashed in any game for me.
I've read that by increasing the applied voltage to the GPU, one is able to achieve better overclocking, and that one such way to do it is by modding the BIOS.
So, I did the following: Extracted the BIOS from GPU-Z, then edited the Power limit (132%->150%) and Voltage (1.1875V) using KGB, then flashed the new BIOS using nvflash.
I rebooted the computer, tested the new bios (with no overclock - I've reset it, just in case something goes bad with the bios) and... it crashes instantly at Valley. With or without overclock. Good thing I've backed up the old BIOS, managed to flash it back, and now I'm back into +125/+600. However, I don't understand what happened here, and I want to resolve it because I want to squeeze everything I can out of the 680 GTX! :cool:

Short version:

Modded a 680 GTX bios using KGB, in which I've increased the voltage and the power limit, flashed it, and it causes crashes during GPU usage. Flashed the old BIOS back and it's back to normal.

Any tips? Also, any "custom" ways to increase the clock speeds/get better temperatures?

Thanks!
 
Last edited:

Tír na nÓg

Member
Joined
Sep 19, 2015
Try using KBoost in PrecisionX to lock frequency and voltage.

My 780 ti is crashing around 1280/8000/1.1v without KBoost, where it holds 1380/8500/1.15v with KBoost activated.
 
OP
Y

yoadknux

Member
Joined
May 6, 2016
Try using KBoost in PrecisionX to lock frequency and voltage.

My 780 ti is crashing around 1280/8000/1.1v without KBoost, where it holds 1380/8500/1.15v with KBoost activated.
Hello and thanks for the comment! I am actually using MSI Afterburner rather than the EVGA tool. Is there a noticeable difference between those tools? Maybe different functions?

Thanks!
 

Tír na nÓg

Member
Joined
Sep 19, 2015
Nope, you can use both: start Precision X and activate Kboost. Close it and use MSI AB.

You have to do it only once (or after driver/vBios change/update).
 
OP
Y

yoadknux

Member
Joined
May 6, 2016
Off topic: I'm afraid my +125/+600 possibly causes a BSOD when playing Rocket League for a while :( This is weird though, because usually games just crash when the GPU is overclocked too far, rather than crash the whole system. I've also tested it in Valley and it was stable for over 6 hours before I stopped the application. Maybe it's unrelated?
Nope, you can use both: start Precision X and activate Kboost. Close it and use MSI AB.

You have to do it only once (or after driver/vBios change/update).
Gave it a shot. Didn't help. Exact same results.
Find already modded BIOSs is my suggestion...

Here is an informative thread for you: http://www.overclock.net/t/1289489/gtx-600-series-unlocked-voltage-bios-downloads-and-tools

:)
People recommend flashing the EVGA SC Bios, but I'm afraid it may brick my card...
 
Last edited:
OP
Y

yoadknux

Member
Joined
May 6, 2016
Hello everyone, sorry for the double post. After successfully modding the BIOS, I've managed to improve the OC by a bit - now it's 1241 / 3680. However, one thing I started noticing, is that now the GPU core fluctuates while playing Witcher 3. Is that normal?
I'm talking about fluctuations that go as low as 1241 -> 1128. The fluctuations and fast (last a second or so), and vary in "steps": 1241, 1228, 1202, 1188, 1165, 1128. GPU temperature is fine, and stays around the low 70s. The memory clock remains constant.
It looks more or less like this (1st graph is Core clock):
BYxnNxl.png


However, this doesn't happen during Valley benchmark (Or happens to a much lesser extent...), where the GPU clock can remain at 1241 for a long period of time (from minutes to hours).

Anyone ran into this problem? What can I do to fix it, and is it even worth fixing?
 
Last edited: