GTX 970 Bios Modded [Extreme OC] 2Ghz,1.1v

[Tristerin]

@characteress its being called into question simply because what you are saying doesn't make sense, nor are you providing what the Tech World REQUIRES as proof.

 

BTW I can boot up @ 4.0 ghz on stock voltages on my R7 1700 - I can even get it to game but its not stable lol.  So you have to give the community, who you are trying to show you have accomplished something insane, proof.

 

Please, please, prove it.  That means that there is hope for the used GPU market.  When the 1060 miners were cracked...they brought proof not just claims.  Your taking an old, outdated card and making it super relevant.  Do so with the proof required or be called a liar until you do.

[unsungheroist]

8 minutes ago, TheSLSAMG said:

Run Fire Strike and send a screenshot or link to the result.

I second this. WHERES THE LINK!

[TheSLSAMG]

3 hours ago, characteress said:

So this Superposition run is "overclocked", correct?

[characteress]

2 minutes ago, TheSLSAMG said:

So this Superposition run is "overclocked", correct?

yup. I will run Firestrike soon(i am downloading it -- 1mb/s internet)

[TheSLSAMG]

11 minutes ago, characteress said:

yup. I will run Firestrike soon

Okay chief. The reason no one believes you (including me) is because from what you have posted so far, running at 2.1GHz makes no sense. Under liquid nitrogen, the highest clocked GTX 970s are sitting around 1700-1980MHz. In fact, the fastest Superposition 1080p Extreme result on HWBot is 2973 at 1644MHz on the core.

 

And before you pull the "my CPU is less powerful than that CPU and that's why the scores are different", behold. A stock GTX 970 Strix on air cooling at 1178MHz is pulling 2103 in Superposition 1080p Extreme, 43 points higher than your supposed 2GHz result.

 

3 hours ago, characteress said:

While I am downloading 3DMark(I did stock vs overclocked on 1080p medium and it was ~4500(stock) vs ~7200(overclocked)

Another thing. This is a 60% increase in score for a 92% increase in clocks. I don't expect clock scaling and score scaling to be perfectly linear, but I'd expect them to be closer than that.

[Tristerin]

Believe me OP we wish you were able to prove it.  Id go buy a 970 off Ebay RIGHT NOW to mess with if this is actually real.

[flibberdipper]

15 minutes ago, TheSLSAMG said:

And before you pull the "my CPU is less powerful than that CPU and that's why the scores are different", behold. A stock GTX 970 Strix on air cooling at 1178MHz is pulling 2103 in Superposition 1080p Extreme, 43 points higher than your supposed 2GHz result.

Let's also not forget that in Superposition your CPU will have minimal impact on score unless it's roughly as fast as a constipated shit.

[characteress]

https://www.3dmark.com/3dm/36579140

strangely benchmarks scores didn't improved using overclock

[characteress]

https://www.3dmark.com/3dm/36579406 HardwareMonitoring.hml

as you can see on this html file(from msi afterburner) it was running at 2164mhz(only downclocking when changing benchmarks, cuz it isn't rly needed for GPU to max out clock if there isn't any load)

[TheDankKoosh]

37 minutes ago, characteress said:

https://www.3dmark.com/3dm/36579140

strangely benchmarks scores didn't improved using overclock

 

20 minutes ago, characteress said:

https://www.3dmark.com/3dm/36579406 HardwareMonitoring.hml

as you can see on this html file(from msi afterburner) it was running at 2164mhz(only downclocking when changing benchmarks, cuz it isn't rly needed for GPU to max out clock if there isn't any load)

The bios is just fooling the card into believing that it is running those clock speeds, it's fake, just manually overclock on a bios with a high power limit.

[TheSLSAMG]

22 minutes ago, characteress said:

https://www.3dmark.com/3dm/36579406 HardwareMonitoring.hml

as you can see on this html file(from msi afterburner) it was running at 2164mhz(only downclocking when changing benchmarks, cuz it isn't rly needed for GPU to max out clock if there isn't any load)

It's clear it isn't actually running at 2164MHz based upon the benchmark scores, no matter how the software is misreading it (due to the spoofed VBIOS)

[Tristerin]

As suspected, and not because we didn't WANT you to be right, but because its just not in the numbers or the silicone lottery for that chip (or likely you wouldn't be the first to do this..no offense)

[characteress]

28 minutes ago, Mr. horse said:

What? Really an html file that can be simply edited? No screen shot of the log? Like asked of more then once?

 

sory buddy but everything screams fake at this point.

u asked for logs, that file is directly from msi afterburner. Screenshot? I already took millions... Still think it's fake? Download the bios I uploaded on 1st msg and test it by yourself. Still wanna think it is fake? Just ignore this thread then lol

[Tristerin]

1 minute ago, characteress said:

u asked for logs, that file is directly from msi afterburner. Screenshot? I already took millions... Still think it's fake? Download the bios I uploaded on 1st msg and test it by yourself. Still wanna think it is fake? Just ignore this thread then lol

No one believes you because you aren't proving it, correctly.

 

I to, can create the SAME exact "proof" you did.  Can you give us a Firestrike score link?  That helps prove it because you cant edit that.  That would be step 1 in the right direction.

 

Currently under your post I would NEVER purchase a 970 to play with this BIOS.  If you proved it...well Im already on EBAY Id snag one up the moment it was believable...why?  Because it would probably smash my Fury (which I would in turn sell for more than the price of the used 970) in benchmarks if it was true.

[Tristerin]

Hell please prove it, Ill buy a 970 and a waterblock right now lol just to push the voltages to the extreme to see what this "bios" can really do.  But right now this 1 guy giving very vague and hastily put together "proof" isn't cutting it.  Are you so excited (if this is true) that you are actually getting chemical dumps and unable to focus to get this proved out?  You'd be an internet legend.  Do it bro, do it.

[Tristerin]

Here I will go first, this at least proves my 3.8ghz OC no question but doesn't PROVE its on stock voltages (it is) (now its your turn)

 

https://www.3dmark.com/3dm/36583905?

 

But this helps:

(this is at 1600x900 throttling my card lol...my other Ryzen rig is on a 2k 144hz panel...Fury sings)

 

 

[xAcid9]

10 hours ago, characteress said:

https://www.3dmark.com/3dm/36579140

strangely benchmarks scores didn't improved using overclock

IMHO, this is the case of software reading error because you messed up the power table. Your card did not clocked at 2.1Ghz thats why the score is similar or regressed. You probably get better result with default BIOS and manually overclock it with MSI Afterburner.

 

You set power limit to 570w but your card only use 30%(~170w)

[unsungheroist]

14 hours ago, characteress said:

u asked for logs, that file is directly from msi afterburner. Screenshot? I already took millions... Still think it's fake? Download the bios I uploaded on 1st msg and test it by yourself. Still wanna think it is fake? Just ignore this thread then lol

[bebeto_maya]

Ok, I decided to test these BIOS, because a i have this graphic card. It works and is real. But it doesn't improve performance because the reduction of GPU voltage reduces clock, too. I have a lot of madness in my mind, putting my hardware in real damage risk, but a tried, because its for a few minutes and overheat isnt problem in short time (since you dont goes to beyond of 90°C).

Results: When I ran the board to 2164MHz and memories at 4000Mhz, I had a series of screen artifacts. I then downgraded to the BIOS (3800Mhz in memory). I got stability, but the performance was terrible, I lost 10% with overclocking. When I returned to the old settings, my board running at 1500MHz/GPU and 3800MHz/memory (already overclocking), my performance was 10 to 15% better according to Passmark Performance Test 8.0 (200 FPS in directX 11).

That is, this superoverclock is a scam bored. Doesnt work. You will not gain performance.