ty_ger07
I am "claimed".
 
"It is dumb.  It is stupid." Sure, whatever.  Actually, the MOSFETs tended to run under 100c according to Nexus's report with the default BIOS, default thermal pads, and highest possible load at stock settings.  And, as we know, 100c isn't really a big issue for these MOSFETs in the first place. But, whatever.  Just ignore those facts and dive off the deep end on some fanciful rant.
 
Everything EVGA did as a service to its customers by releasing a newer BIOS and by providing extra thermal pads was not really necessary in the first place and has actually damaged EVGA's reputation by making people think that EVGA is guilty of something, or has done something wrong, or made an error in its design.  It's really quite nasty the way people talk around here and demand things.  Somebody released a scare article which passed around the internet, EVGA reassured people that it really wasn't an issue and that their testing conclusively proved more than once that it wasn't an issue, but said that they would release free improvements to those who were still concerned.  Maybe EVGA would have been better off by not providing free extra thermal pads or an additional BIOS option.  At least that way, people wouldn't assume that EVGA has a guilty conscience.
 
As far as anyone who has thoroughly researched the issue can tell, there is/was nothing wrong with the design of the card; and the extremely small percentage of people who have had this "burning up" or "blowing up" issue, were affected by a manufacturing defect which is no more common in this product series than in any product series previously.  It has all become massively overblown.  If you do some searches online, you will find many card models from many card manufacturers with the exact same sort of burns for years or decades in the past.  It is "normal" in the sense of this imperfect world we live in surrounded by imperfect humans.
 
Move along.

 
+1 to everything, nice post here ! ;)
 
I think too that was a little mistake to make the pads solution... everybody thinks that's a fault from evga now... 

When I see all the people saying 'where's my pad, where's my pad" but we dont need it ! 



When I see too all the tests on those cards, I'm new sure at 100%, there isnt a real issue with our cards... we can just say they are not the best cards for temperatures in this generation, nothing more !!!

ty_ger07

acxcoolerssuck

ty_ger07

acxcoolerssuck
the vrm may go over 110C under highest possible loading [when] some may try asus t4 bios or flash higher power limit bios or shorting shunts in order to get rid of power throttling, considered ftw is a enthusiastic card for enthusiasts, if ten phases are not marketing gimmick.

If you buy a FTW card to break world records (why would you use a FTW for that anyway?) and are voiding your warranty with BIOS mods and soldered physical modifications to the card, I am sure that you will also be cooling the card better.  Voiding your warranty and physically modifying the card to push the limits of the VRM isn't really the subject of this thread.

there is nothing to do with breaking world record. for breaking record ppl should go for classified and use the classified voltage tool to raise the voltage beyond the bios limitation. some just dislike the stupid gpu 3.0 boost and want to get the max performance from what they have bought. especially when evga ftw 1070 has only 226w max power limit on the stock bios resulting in power throttling. 
that is what dual bios does. flashing bios is safe, and once you can restore to original bios before rma then it is fine. i personally dont recommend shorting shunts by using liquid ultra (no hard mod is needed) and shorting shunts by using liquid ultra in general can be restored too. 
 
my subject is clear, it is all about vrm cooling, 110c is fine and not failure but the current output will be lower very much. and i am only asking evga for keeping the vrm temp to be under 100c for getting 35a per mosfet. if acx3.0 cant do it, then it is rubbish to me. if raising the fan speed to 2000rpm can do it, then it is fine, but acx3.0 is not strong enough as what the marketing promotion claimed, which is fine for me, and acceptable.

I can't reason with you.  Ok, fine, you are right.  Lol.  If you modify your video card BIOS, solder jumpers across the video card's shunt resistors "just for fun" (not for overclocking reasons at all), and glue on a 2000 watt electric heater to the back of the card, then yes it might catch fire.  There, you are right.  Are you happy?  Somehow, yes, that is EVGA's fault.
 
Please do the math and tell my why you would ever need 35A per MOSFET in the first place.  Do you think you can get 350 watts of load out of the card?  Oh, sorry, I forgot that you are right and that all the wires you are soldering to the card "for fun" will require more power.

pathetic...
 
fyi
https://www.youtube.com/watch?v=v1Qi8fxIi_Y
http://www.overclock.net/...many-other-nvidia-gpus
 
 
 
btw, you should ask evga why ftw 1070 has only 226w power max with such high grade mosfet used on the pcb (in total 350w @ 100c), lol. why do general end users need this, lol. not to mention the dt cards lol.

ty_ger07

acxcoolerssuck
btw, you should ask evga why ftw 1070 has only 226w power max ...

Because they didn't want the card to burn up.  They decided on a limit.  They designed all the components around that limit.  That is what engineering is.
 
As we (mostly) all agree, the card has limits and is operating properly within those limits.  You are coming along and purposely trying to ruin the card in your analogies and then trying to find a reason to say it would hypothetically be EVGA's fault.  

keep endorsing evga in you way.
like what samsung did and blamed to the note 7 users.

It's like trying to show facts to a liberal. Most will just keep on insisting that their feelings trump (no pun intended) any amount of hard facts.

Nothing that you can do besides state the facts, link the proof, and move on.

tabrez7
so, if its really the bad capacitors issue.. may i know form which batch of shipped cards  they fixed it??

It's random.  If you own the video card and it does or does not fail, then I guess you will know the answer.  A very small percentage of video cards will fail unexpectedly early in their life.
 
These things have been burning up for decades.  I don't know why it is all the sudden on people's radar now.

Can someone tell me, what is culprit at the end.
Poor design of VRM as chokes, capacitors, controllers, regulators or lack of thermal pads or contact surface.
I think only pads can't do nothing if they couldn't help to heat transfer on other surface. 
If poor design is culprit why people wait thermal pads and why EVGA spend money to send them to customers.
Than only solution is to people wait and see what will happen and EVGA will replace damaged products.
RMA rate will not be bigger than 1%.
 
But one thing is not clear for me, next year will show up some other company with exactly same problem or very similar and that's weirdest things.
We talk in previous years how much ASUS suffer because they installed cooler design for GeForce Kepler on R9-290X and lack of heatsinks over VRM cause highest temps on market and no one didn't learn lesson from that and no one will learn lesson now after this. Maybe that will not happen to EVGA but with some other company 100% in next 2-3 years, exactly same. 

loveha

Vlada011
Can someone tell me, what is culprit at the end.
Poor design of VRM as chokes, capacitors, controllers, regulators or lack of thermal pads or contact surface.
I think only pads can't do nothing if they couldn't help to heat transfer on other surface. 
If poor design is culprit why people wait thermal pads and why EVGA spend money to send them to customers.
Than only solution is to people wait and see what will happen and EVGA will replace damaged products.
RMA rate will not be bigger than 1%.

There is no culprit issue with the cards. The culprit, is us, as an over enthusiastic community who demands to much. The failure rate is normal compared to previous cards according the information in the article.
 
I will say, to fail on cooling the VRM, is stupid.Who ever engineered the cooler I hope has learned their lesson.
Rules on making a cooler. Does the component on the PCB generate heat? If yes, cool it.

Well said...
 

If failure is due to bad batch of capacitors and is standard , and the not thermal pad issue was an oversight that probably isn't the cause of the cards combusting, then going with EVGA seems to be the best solution here as the aftercare is pretty good.
 
That said, I was also the victim of the black screen 100% fan issue; at least 2 people at the same store had that issue , so that I think was what really kicked things off for people.
Bad luck with a bad batch of VRM, followed by the announcement of high temps + BIOS update and optional pads,  amplified reports of cards going up in flames.
 
It's a bad luck triple threat for EVGA this year.  At least  1 of those things they had control over however, so they are not completely blameless.

loveha
There is no culprit issue with the cards. The culprit, is us, as an over enthusiastic community who demands to much. The failure rate is normal compared to previous cards according the information in the article.
 
I will say, to fail on cooling the VRM, is stupid.Who ever engineered the cooler I hope has learned their lesson.
Rules on making a cooler. Does the component on the PCB generate heat? If yes, cool it.

I do believe that the fact that these cards have been failing a little more spectacularly/visibly has resulted in it getting more attention that what otherwise normal failure rates would, and the community running with it, and while hopefully this report will quell that and remove some of the anxiety associated with it, still can't help but feel disappointed...
 
I'm disappointed that its taken a third party to address this. I'm disappointed that the narrative has jumped between "there is no issue", to "their testing methodology is wrong", to "its a stress test, it doesn't represent real world results", to optional BIOS updates and free thermal pads that kind of, maybe, should have been there to begin with, to "no wait, its something else".  I can't help but feel that EVGA could have taken ownership of that narrative and put this fire out (sorry, no pun intended) earlier by officially stating the cause themselves or admitting to an over oversight. I know, I know, good luck finding any business that would rush out and do that but EVGA seemingly, at least to me, has been somewhat willing to do so with third parties like Gamers Nexus and JayzTwoCents going by their comments in videos posted on this topic while its played out and can't help but think if the same had been done earlier with customers/the forum directly it would have killed some of the speculating and anxiety. I don't know, perhaps maybe I just missed it.
 
Willing to admittedly that I'm making more of this than I should, and that disappointment is exaggerated by EVGA being held in high regard... if someone had asked me to guess which vendor was having cooling or component issues, I wouldn't have guessed EVGA.

PHOENIXZERO

arestavo
It's like trying to show facts to a liberal. Most will just keep on insisting that their feelings trump (no pun intended) any amount of hard facts.

Nothing that you can do besides state the facts, link the proof, and move on.

Really? Using bad analogies and political generalizations? As if "conservatives" aren't just as guilty of the same thing.
 
It probably being an issue with bad capacitors doesn't absolve EVGA of the shoddy oversights workmanship of the cooling solution and it running hotter compared to the competition because of it.

Again - "Nothing that you can do besides state the facts, link the proof, and move on."