NVIDIA Kepler speculation thread

It's pretty old news that the voltages are locked and controlled on the Kepler cards. It's the reason why for example you need an add on hardware to raise the voltages on the EVGA Classified card and why MSI stopped shipping Lighting 680s with adjustable voltages.

What exactly are you blaming nVidia here?
 
There appears to be some misunderstanding regarding overvolting on NVIDIA GPU's. The reality is that any add in board partner (whether it is EVGA, MSI, Zotak, etc) can provide overvolting, but if they push beyond a certain voltage threshold limit, then there will be no warranty from NVIDIA on that part. That is a reasonable and fair stipulation on NVIDIA's part. So the final decision is left to the AIB partners whether or not they want to have their parts warrantied by NVIDIA, and the AIB partners have to decide whether or not the higher heat, louder acoustics, and lower reliability are a good tradeoff for overvolting and no warranty on products sold to end users.

Here is the official response from Senior PR Manager Bryan Del Rizzo @ NVIDIA:

Bryan Del Rizzo @ NVIDIA said:
Green Light was created to help ensure that all of the GTX boards in the market all have great acoustics, temperatures, and mechanicals. This helps to ensure our GTX customers get the highest quality product that runs quiet, cool, and fits in their PC. GTX is a measureable brand, and Green Light is a promise to ensure that the brand remains as strong as possible by making sure the products brought to market meet our highest quality requirements.

Reducing RMAs has never been a focus of Green Light.

We support overvoltaging up to a limit on our products, but have a maximum reliability spec that is intended to protect the life of the product. We don’t want to see customers disappointed when their card dies in a year or two because the voltage was raised too high.

Regarding overvoltaging above our max spec, we offer AICs two choices:

· Ensure the GPU stays within our operating specs and have a full warranty from NVIDIA.

· Allow the GPU to be manually operated outside specs in which case NVIDIA provides no warranty.

We prefer AICs ensure the GPU stays within spec and encourage this through warranty support, but it’s ultimately up to the AIC what they want to do. Their choice does not affect allocation. And this has no bearing on the end user warranty provided by the AIC. It is simply a warranty between NVIDIA and the AIC.

With Green Light, we don’t really go out of the way to look for ways that AICs enable manual OV. As I stated, this isn’t the core purpose of the program. Yes, you’ve seen some cases of boards getting out into the market with OV features only to have them disabled later. This is due to the fact that AICs decided later that they would prefer to have a warranty. This is simply a choice the AICs each need to make for themselves. How, or when they make this decision, is entirely up to them.

With regards to your MSI comment below, we gave MSI the same choice I referenced above -- change their SW to disable OV above our reliability limit or not obtain a warranty. They simply chose to change their software in lieu of the warranty. Their choice. It is not ours to make, and we don’t influence them one way or the other.

In short, Green Light is an especially important program for a major, new product introduction like Kepler, where our AICs don’t have a lot of experience building and working with our new technologies, but also extends the flexibility to AICs who provide a design that can operate outside of the reliability limits of the board. And, if you look at the products in the market today, there is obviously evidence of differentiation. You only need to look at the large assortment of high quality Kepler boards available today, including standard and overclocked editions.
 
Last edited by a moderator:
Slight annoyance on the bolded part: having specific warranty conditions is influence. It's just false to say it isn't. That kind of political bullshit annoys me sometimes.

That said, it does seem to me a reasonable restriction.
 
That's rubbish from NV and that team in Germany. WTH people, what's wrong with you?
I have the same reaction about your statement. In fact, I'd love to understand what's wrong with the Nvidia policy?

There's a reason why chip companies overvolt during reliability tests: it's the easiest way to dramatically accelerate aging processes and to check how a chip will behave x years from now without actually waiting x years.

What's so unreasonable to deny warranties to products that chose to operate outside the reliability limits of the product? Please explain.

Ad absurdum: should partners allowed to overvolt to 10V, reducing the lifetime to seconds, and still be given warranty? No? Do you agree then there should be some limit?
 
I have the same reaction about your statement. In fact, I'd love to understand what's wrong with the Nvidia policy?

There's a reason why chip companies overvolt during reliability tests: it's the easiest way to dramatically accelerate aging processes and to check how a chip will behave x years from now without actually waiting x years.

What's so unreasonable to deny warranties to products that chose to operate outside the reliability limits of the product? Please explain.

Ad absurdum: should partners allowed to overvolt to 10V, reducing the lifetime to seconds, and still be given warranty? No? Do you agree then there should be some limit?

What happens to those who bought 680 Classifies cards with EVBot ;)
http://www.evga.com/forums/tm.aspx?high=&m=1755509&mpage=1#1755596
Take a look at the reaction in the thread.
 
EVGA's customers will be fine, because EVGA will warranty these products. Presumably EVGA will have replacement cards with EVBot available for some time for those who need an RMA on the original card, and if they run out of that, then EVGA will replace the original card with a newer Geforce GPU with similar (or better) performance.
 
EVGA's customers will be fine, because EVGA will warranty these products. Presumably EVGA will have replacement cards with EVBot available for some time for those who need an RMA on the original card, and if they run out of that, then EVGA will replace the original card with a newer Geforce GPU with similar (or better) performance.

Getting a new gpu isn't the point here with the classified and EVbot. They are paying for a feature on the card, not just the gpu. And what official information do you have that they will replace an evbot card for an evbot card? Got a link?
 
What happens to those who bought 680 Classifies cards with EVBot ;)
http://www.evga.com/forums/tm.aspx?high=&m=1755509&mpage=1#1755596
Take a look at the reaction in the thread.
The board vendor is customer in this case. I didn't realize I has to spell that out. How the board vendor deals with its customers is their problem once they decide to exceed reasonable operating conditions. That's the whole point of the PR guy.

Frankly, it would have been smarter to not allow out-of-spec overclocking at all. This way, you prevent this kind of tempest-in-a-teapot (aka Big Scandal) controversy completely.
 
The board vendor is customer in this case. I didn't realize I has to spell that out. How the board vendor deals with its customers is their problem once they decide to exceed reasonable operating conditions. That's the whole point of the PR guy.

Frankly, it would have been smarter to not allow out-of-spec overclocking at all. This way, you prevent this kind of tempest-in-a-teapot (aka Big Scandal) controversy completely.

But that's not really the way I see it.
Now that the evbot feature isn't there what will be the price of the classified 680 now?
Can someone with a evbot be able to sell their card for the same price used at the new classified w/o? Talking about demand here.
And most importantly will the older card be on par with it's next gen replacement?

Those are the questions I'm seeing. We can't answer all of them with certainty now (is the new classified in retailer's hand's yet?) but I'm sure it will be once next gen cards are released. :smile:

But are you honestly blaming the customer for buying a product they thought would be on the market until the next gen product came out (PR comment)?
 
Last edited by a moderator:
Getting a new gpu isn't the point here with the classified and EVbot. They are paying for a feature on the card, not just the gpu. And what official information do you have that they will replace an evbot card for an evbot card? Got a link?

Naturally there is no guarantee that one will receive the exact same card, but EVGA expects to have EVBot RMA replacement cards for the next year (or more?). If an EVGA customer needs a replacement card after that timeframe, then the replacement card will be of equal or greater performance, and that could actually be quite fortuitous for the customer (irrespective of overvolting feature) because the replacement card could use true next generation GPU technology. Anyway, the fact that EVGA is willing to provide a full warranty on EVBot products with overvolting capability should be plenty of peace of mind for the end user.

See EVGA's response here: http://www.evga.com/forums/fb.ashx?m=1756648
 
Last edited by a moderator:
Naturally there is no guarantee that one will receive the exact same card, but EVGA expects to have EVBot RMA replacement cards for the next year or two. If an EVGA customer needs a replacement card after that timeframe, then the replacement card will be of equal or greater performance, and that could actually be quite fortuitous for the customer (irrespective of overvolting feature) because the replacement could use a true next generation GPU. Anyway, the fact that EVGA is willing to provide a full warranty on EVBot products with overvolting capability should be plenty of peace of mind for the end user.

See EVGA's response here: http://www.evga.com/forums/fb.ashx?m=1756647
I think we all understand how their warranty works. But what's being pointed out are the feature(s) not the GPU. It's good that they at least try to address it but they aren't making any guarantees.
 
But are you honestly blaming the customer for buying a product they thought would be on the market until the next gen product came out (PR comment)?
Yes, I'm blaming the customer, EVGA, for using the chips outside the specs given by their supplier.

The end consumer has nothing to do with this: if EVGA made promises to them, they should honor those promises. The RMA dealings between Nvidia and EVGA are not something an end consumer should be exposed to. EVGA has higher margins by running the chips out of spec. This comes at the risk of having more failures in the field. No big deal: just send a new card, paid out of EVGA's pocket, but don't expect Nvidia to refund the failed chip.

I see no controversy here?
 
I think we all understand how their warranty works. But what's being pointed out are the feature(s) not the GPU. It's good that they at least try to address it but they aren't making any guarantees.

Sure, I know that some people were looking forward to making use of the overvolting feature, but I'm not sure how practical a feature it is on the EVGA GTX 680 Classified. When comparing Overclocking vs. Overclocking+Overvolting, the power consumption, GPU temperatures, and noise levels increase by ~ 10% on average ( http://www.anandtech.com/show/6096/evga-geforce-gtx-680-classified-review/8 ), while gaming performance increases by only ~ 3% on average ( http://www.anandtech.com/show/6096/evga-geforce-gtx-680-classified-review/9 ) . That is a poor tradeoff to make in my opinion, even without factoring in potentially reduced long-term reliability when overvolting.

Here is what Anandtech said about overvolting this card:

Anandtech said:
As for the impact of overvolting, unfortunately as it turns out overvolting didn’t greatly improve our results. With a core voltage of 1.275v (0.1v over stock) we were able to hit a base clock of 1286MHz, which is just 75MHz higher than what we hit on the stock voltage of 1.175v. In fact both in absolute and relative terms we gained more from our initial stock voltage overclock than we did from the overvolting overclock. Which is not to say that overvolting doesn’t help – clearly it does – but in our limited experience overvolting isn’t unlocking any kind of amazing clockspeed. It’s merely pushing GK104 a bit higher than it can go on stock voltage.

Memory overvolting doesn’t have a particularly great impact either. With a memory voltage of 1.65v we were able to squeeze out another 100MHz with the GTX 680 Classified’s 4GB of GDDR5. Much like the core overclock this is an improvement, but not greatly so.

Ultimately overvolting the GTX 680 Classified most certainly improves its overclockability, but at least on air you won’t be hitting any obscene clocks. Even then we don’t seem to be particularly temperature limited, so it’s not clear whether better (but not extreme) cooling would improve our overclocking results.
 
From what I understand, MSI's behavior here is a bit questionable (since it's kind of a bait & switch) but NVIDIA isn't doing anything I'd object to.
 
If you believe that gf104 wasn't intended for the mobile sector, then you must think AMD engineers are incompetent. That being, Nvidia somehow designed a better mobile GPU than AMD by designing a purely focused desktop and somehow it overcame AMD ability to design a mobile part.

Eh? Better mobile GPU than AMD? Pitcairn is smaller and just as fast in mobile. I'd say the win goes to AMD for the better "mobile" GPU in that case.

Regards,
SB
 
Yes, I'm blaming the customer, EVGA, for using the chips outside the specs given by their supplier.

The end consumer has nothing to do with this: if EVGA made promises to them, they should honor those promises. The RMA dealings between Nvidia and EVGA are not something an end consumer should be exposed to. EVGA has higher margins by running the chips out of spec. This comes at the risk of having more failures in the field. No big deal: just send a new card, paid out of EVGA's pocket, but don't expect Nvidia to refund the failed chip.

I see no controversy here?
Is there any information out there that suggests that the previous models in consumers hands now are defective? Like I said before, in what way will price be effected on such cards that no longer offer this ability? Resale value and performance vs next gen parts will also be something of interest. It's not a controversy but a discussion.

Sure, I know that some people were looking forward to making use of the overvolting feature, but I'm not sure how practical a feature it is on the EVGA GTX 680 Classified. When comparing Overclocking vs. Overclocking+Overvolting, the power consumption, GPU temperatures, and noise levels increase by ~ 10% on average ( http://www.anandtech.com/show/6096/evga-geforce-gtx-680-classified-review/8 ), while gaming performance increases by only ~ 3% on average ( http://www.anandtech.com/show/6096/evga-geforce-gtx-680-classified-review/9 ) . That is a poor tradeoff to make in my opinion, even without factoring in potentially reduced long-term reliability when overvolting.

Here is what Anandtech said about overvolting this card:
They admit that overvolting does help. Gains on return of the over volting may vary but people were willing to take the risk. That's really the gist of it. From my understanding the only way to get higher then 1.21V, for oc, is with evbot.

Edit:
I wonder what the results would have been on OC/OV had they tested more physx based games besides batman: AA... (Borderlands 2, Mafia II, etc).
 
Last edited by a moderator:
Eh? Better mobile GPU than AMD? Pitcairn is smaller and just as fast in mobile. I'd say the win goes to AMD for the better "mobile" GPU in that case.

Regards,
SB

I hate going on who have the best. .. But when you read review you understand quickly the problem, it is specially viewable on an Anandtech one when they review the 680M vs 7970M. the problem of performance was coming from the Enduro... with notebook who allow to disable it.. the performance of the 7970M is really higher.

On the other aspect, Optimus is yet better if you think to the battery life, not if you play lol .. my last concern is about 3dgfx and other benchmark i have see, we was know the GK104 is not good at computing, but seeing it come so much ashamed in the mobile space with all the gfx tests ( Catia, Maya etc ) vs what is a lower clocked speed of the 7870 Pitcain...

12.9 bring allready some fix with Enduro ( going with large fps gain ), and the second fix is coming in some days. ( let say this month. )
 
Last edited by a moderator:
Back
Top