MSI Caught Overvolting GTX 660 Ti, 670 Power Edition Cards?

[Nelly]

What do you make of this... If true, does this effect the GTX 680 Lightning?

If a company changes its product in the middle of a production run after an error becomes public, you know something’s fishy. If Nvidia steps in as well, that’s a good indication there’s something serious going on. Here’s what happened at MSI.

Our colleagues over at Tom’s Hardware Germany seem to have caught MSI red handed, overvolting the GPUs on the GTX 660 Ti and GTX 670 Ti Power Edition boards (Google Translate) to achieve a higher and longer lasting GPU boost state by basically circumventing the PWM controller. In other words, MSI was cheating. Perhaps no one would ever have known if it hadn’t been for one side effect. The increased voltage can cause the system to refuse to POST.

Full Story here >>http://www.tomshardware.co.uk/MSI-GTX-660-670-overvolting-PowerEdition,news-40278.html

The full article on Tom’s Hardware Germany (Google Translate) covers the topic in more depth >> http://www.tomshardware.de/MSI-GTX660Ti-Overvolting,testberichte-241108.html

[kromosto]

lol what is the problem with overvolting? also tom like exaggerating these kind of things. imo problem in here is refuse to post situation not the overvolting.

[tajoh111]

"We are currently not aware of any other vendor providing this same level of performance and we’ve worked with NVIDIA to ensure that new production models will limit this free overclock boost you currently get. Our new production models with normal GPU Boost function will be on sale next month."

This is pretty funny actually and depressing at the same time. Tomshardware is actually provided a disservice here and MSI basically says thanks for telling on us, lets tell the consumers who to thank. If you didn't know earlier, Nvidia was limiting voltage on basically all their cards. Basically MSI is saying well because you told on us, we are limiting the amount of voltage and back tracking on the overclock ability of our videocards because Nvidia is forcing us to.

I.e Remember the gtx 680 lightning had overvolting properties that let it go beyond other boards hence it's overclocking limits were 100 mhz or higher compared to other cars. Nvidia didn't like that so the next update made it so this wasn't possible anymore..

http://www.guru3d.com/articles_pages...weaking,1.html

"And then the really bad news ...

While I was writing this article I learned that NVIDIA just issued new BIOS files to the AIC partners and is frowning upon voltage tweaking outside their limitations. As such all new batches Lightning cards will have BIOSes where their limit of 1.175V is enforced, even in the LN2 BIOS. MSI has to follow that directive or probably face the fact that they will not be able to purchase the GPUs anymore. "


Considering this made it to retail cards and not just review samples, this is hardly a cheat. Some cards might of had problems but what cards don't have overclocking problems every once in a while and as long as MSI will warranty it, I am happy they are circumventing some of Nvidia strict voltage regulation they put on to curb overclocking.

Now because of tomshardware they are cutting this off on their value cards. i.e the non-lighting cards as well. Considering this is xtreme systems, i think a well deserved you is in order for tomshardware.

[massman]

Giving higher performance for free?

THANKS MSI!

[K404]

Like it or not, nVidia can set whatever rules they want when it comes to the same and use of their parts. Especially when they're worried about RMA rates.

Just makes things easier for ATI

[CrazyNutz]

oh nooooo! Send them to the cornfield! Tom's hardware what a bunch of dopey maroon's.

Looks like sensationalism has crept it's way into our hobby. Sad day

[SubZero.it]

Tom's?I don't even read the news...

[NEOAethyr]

I was thinking about testing this theory back when I 1st got the card, but didn't get to it.

My theory was that they up'ed the voltage by 0.1v, and then decreased the voltage in the bios by that same amount.
The soft supposedly always reads off the stock voltage anyways, so the user would never know.
Therefor allowing an extra boost of voltage of 0.1v over the normal limit.

The only way to test this theory is flash a stock bios to one of the 2 bios bank thingy's and read it out on a dmm to see if it's 0.1v lower then stock.
Crash galore too, it would probably need a bit of fiddling in the clocks before flashing.

I was thinking the same could be applied to a stock card.
Hard mod it by 0.1v, down volt it by the same in the bios, driver thinks it's got 0.2v of headroom +/-, software readings say it's all stock..., win win sorta thing.

[zanzabar]

the overclocked card has higher default voltage and NV banned voltage control. oh no they found a loop hole to give people what they want.

[tajoh111]

the overclocked card has higher default voltage and NV banned voltage control. oh no they found a loop hole to give people what they want.

Yeah like seriously. I will reiterate. you tomshardware. I can understand Nvidia's reasoning, which I personally don't approve of and am mad at them for, but i understand it is for business reasons and RMA rate. But Toms reasons? To be a stick in the mud or an enemy of overclocking.

Looked for this so called issues and I couldn't find anything, atleast when looking for not working cards for the gtx 660 TI power edition. Basically the only problem was that one toms already put up and that could be something else.

http://www.newegg.com/Product/Produc...82E16814127696

Look at the reviews for the power edition. 78 or 92% 5 star rating, 6 or 7% 4 star rating and a single 1 star rating which got a card DOA.

[NEOAethyr]

Here beit da review in ? ->
http://www.tomshardware.de/MSI-GTX66...-241108-3.html

I can't read very much in german lol, and google translate has never worked in opera for me for some odd reason.
I'm probably the only person laughing at the 88% thing, 88 lol...
Whatever.

This was bound to happen someday though anyways.
Just look at amd and intel, they've put locks on there parts for along time now.
Though usually the high end rev's are unlocked.
You pay $600 for a card (lightning), one that's pimped out ready to clock, and you're not allowed to clock past the norm? :\

I'm glad now I got the lightning 680 when I did, even after the initial guru3d report of locking the tdp and voltages on the lighting it still wasn't locked.
Mine isn't.
Though no more software updates for the msi overclocking util, latest ver limits it (saw that a few weeks ago).
But then again supposedly there's a way of bypassing any limits, at least on the older ver's, by hacking the process in memory using a game hacking util of some sorts (art money for example, etc).

It's the end of the world, no more clocking of nvidia cards lol.
I'm joking... it's not that big of a deal, it's like moving back into the 90's and early 2000's, gotta have to hard mod these things now.

I've never had any post issues on my card btw.

[B-Shot]

Isn't the card MSI's problem if it fails?

[tajoh111]

They aren't particularly failing though. Tom's hardware is exaggerating like crazy. His only link for proof is a small 9 reply thread that is very vague and doesn't come up with much detail. I have tried to look for these issues but came up dry for the most part. A decent amount of gtx 670 have failed, but that was across multiple vendors which got a bad batch. Evga was most caught up in this. Look at the gtx 660 ti PE and they are pretty reliable and they overclock better to boot.

[jimba86]

i Hope nvidia will get rid of this "boost" function on their cards..boost is for sissys..this is XS..we "BOOST" To the XTREME!

[BeepBeep2]

Giving higher performance for free?

THANKS MSI!

Yes, and yet again another trash article from Tom's Hardware.

[SKYMTL]

Is this some kind of sick joke? MSI has EVERY RIGHT to circumvent NVIDIA's overly picky specifications since it is MSI that covers warranty terms and NOT NVIDIA

Cheating my ass.....

In many cases there is no controversy until one looks for it. Case and point: that article.

[gumballguy]

Is this some kind of sick joke? MSI has EVERY RIGHT to circumvent NVIDIA's overly picky specifications since it is MSI that covers warranty terms and NOT NVIDIA

Cheating my ass.....

In many cases there is no controversy until one looks for it. Case and point: that article.

Exactly what I was thinking.

[NEOAethyr]

I could see the issues with nvidia wanting everything to spec on there chips, it's there chip afterall.
It would be like an unlicensed nintendo game if it wasn't.
For example they could restrict sli access.
I dn, that's not a good way of explaining it lol.

It's bs anyways.
It's not there pcb... do whatever...
I dn.
Make a jumper... that should doit right?, that's what I would figure.
But yet they jump on evga's butt too... for there dongle non the less, which is a hardware solution too.

There should be no limits on a good card out of the box, I think it's bs.
Same goes for a motherboard.

Whatever though I guess, hard mods are gonna be the way of the future I guess.
It's not a big deal though I think, because the cards will still clock a bit without voltage adjustments anyways.
It's not so bad. ^^
But it is a step back..

[cowie]

WTF oh no msi was adding more voltage so they can have a good overboosting card,them bastards how dare them.

ok ok it was thruogh pwm and not the ic itself so what?

If you guys can help make hard mods for the 660 ti hawk when it releases that would be fine by me.

[CrazyNutz]

Off topic a bit, but somewhat pertaining to the subject: Nvidia sure is putting the hammer down hard on not allowing manufactures to push the performance of kepler. Why? Are they trying to hold GK104 back now so they can re-lable it for the next GPU series with higher clocks like we have seen in the past?

Nvidia has already jumped on MSI in the past with the 680 lightning. MSI allowed 1.21v via the LN2 bios, Nvidia caught wind of this and made them change it back to 1.15.

Anyhow if anyone is interested, I can mod your 660ti bios: http://www.xtremesystems.org/forums/...ers-Club/page2

[kiwi]

So what is the score MSI vs Nvidia?

[cowie]

3-0 msi

At least for cap-ler(sic)

1.lightning got 1.37v with ln2 bios and ab 2.2.3
2. 670 pe got alittle more volts then most
3 660 ti same as above

all this after nv was saying no voltage control.
BUT nv it looks is taking the baseball and going home so partys over for anyone new to the capler game.

Did i mention i would hope to see 660 hawk hardmods?

[naokaji]

how is higher performance cheating?

as long as it results in a genuine perfomance advantage its fine.

[SoF]

the vgpu and boost crap introduced by nvidia with 6XX series sucks...and if you think it can't be any worse...
there is nvidia with some more bs on hw-limitations...

thanks MSI for putting some light into this...

[Russian]

Here's what's really going on...

http://www.brightsideofnews.com/news...nnovation.aspx