Hello !
 
I think we have a lot of topics about the VRM issues and I think a topic who resume all of this would be nice !
 
1- How much cards are dead ? For me, I think it's 3, maximum 4. So nothing anormal, just unlucky customers for me... This is electronic, it can have troubles like that...
 
2- This cards was overclocked ? 
 
3- This cards had the original ventilation curve ?
 
4- This cards are in the hands of EVGA right now ?
 
5- Why there are several reviews about VRM temperatures who are 20 degres more than others ?
A ventilation curve too agressive for silence can be the reason ?
 
Please dont come here to tell that its not normal from evga, that it's a shame, in this topic I would like to have ONLY FACTS, only things that are real and we really are sure for the moment !
 
Because it's mass hysteria here but I only saw two pictures, and all the websites use the same picture again and again...

By the way, sorry for my very poor english ! :D

NucleusX
If you came to this forum looking for clarity on the situation, i doubt you will find it.
All you're gonna find is a tsunami of speculation and generic corporate responses.
Somewhere in-between all that is the truth, but right now its too early to tell.

That's what I think indeed... between the evga guys who are "no problem guys, everything is fine" and the customers who are "this is the end of the world, all evga cards are gonna blow up"... its hard to find the truth and to know what's really going on...

Maybe I made this topic too soon but if someone has real information no speculation this is the place to be ! :D

But if there isnt a truth in all of that, there are a lot of topics for just 2 or 3 dead cards so...

same questions like me  ;)

earlier or later we will know.

but i hope very soon. to go safe i ordered the pad`s too.

Well...
 

• We know those 1080 FTWs blowing up had faulty VRMs and that was corrected on the cards shipped after August 31st, according to EVGA. This issue is also limited to 1080 FTWs, EVGA has stated officially no other GTX 10 models were affected by such.
• We also know the 1070 & 1080s with ACX 3.0 coolers are suffering from quite higher VRM temps than the competition, due to an inferior cooling solution + lack of thermal pads between the mid plate and heatsink. 
• The 1060 situation on all these VRM temps is even worse, reaching temps of almost 120ºC.
• It's not yet clear if those cards suffering catastrophic failures, as reported by some people, have faulty VRMs or it's just the result of thermal runaway. This of course is not equating the endless amount of variation between builds which may or may not cause the mishap.
• We also don't know if these cards with "hot VRMs" will last as long, since hot temperatures degrade components faster.
• The thermal pads which will be provided by EVGA will in fact lower the VRMs operating temp, yet they will still run hotter than most of the competition. 

loveha
we have the heatsinks not touching the VRAM in some instances.
 

That's a 1080 only issue, correct? I haven't yet seen 1070s with those issues.

loveha

brokencross

loveha
we have the heatsinks not touching the VRAM in some instances.
 

That's a 1080 only issue, correct? I haven't yet seen 1070s with those issues.

Are they not built identical? I don't know for sure, but I thought they were.

Think so yes, but every thread that popped up complaining of such was about a 1080. Plus everyone who had a 1070 and inspected their card noticed nothing abnormal.
I haven't inspected mine because I can't be bothered to take it out.
And yes, I fully agree with you. I can't care less if mine blows up.
EU directives protect us customers in full and I will demand a refund if anything happens. And I advise every EU user on this forum to do the same.
 
btw, everyone who has a 1070 FTW should definitely check this:
https://www.reddit.com/r/...tw_under_load/d96jcxn/

I can tell you that my 1070SC is having ram issues. I get a checkerboard crash when ever i put even a 50mhz overclock on the ram. It happens in every game. That is caused by Vram over heating. I ran a 980Ti in the same case for years. It would never get above 74C on the core. My 1070 runs max 72c and it still get this crash anytime i put any overclock on it. 

brokencross

dwrecording
I can tell you that my 1070SC is having ram issues. I get a checkerboard crash when ever i put even a 50mhz overclock on the ram. It happens in every game. That is caused by Vram over heating. I ran a 980Ti in the same case for years. It would never get above 74C on the core. My 1070 runs max 72c and it still get this crash anytime i put any overclock on it. 

Have you updated the bios?

In addition to updating the VBIOS, there is absolutely, 100%, NO guarantee by EVGA for overclocking results that were not done at the factory.
 
Not even 1 MHz more. 
 
Overclocking is free performance on top of stock performance, and no graphics company that I'm aware of guarantees any overclocking results beyond what comes from the factory.
 
Yes, it sucks to get a card that can't be pushed very far. It happens, but at least the 1070 is a very powerful card to begin with - and VRAM overclocking doesn't help as much as core clock overclocking on the top end models. 

brokencross

arestavo
In addition to updating the VBIOS, there is absolutely, 100%, NO guarantee by EVGA for overclocking results that were not done at the factory.
 
Not even 1 MHz more. 
 
Overclocking is free performance on top of stock performance, and no graphics company that I'm aware of guarantees any overclocking results beyond what comes from the factory.
 
Yes, it sucks to get a card that can't be pushed very far. It happens, but at least the 1070 is a very powerful card to begin with - and VRAM overclocking doesn't help as much as core clock overclocking on the top end models. 

Well of course, no vendor guarantees you every card will do X overclock.
But either way, like you say, overclocking the VRAM brings only marginal improvement. Nothing that can be really noticeable. 

Neither does overclocking the core unless you're benching. Real world gains are roughly what 1-7fps more depending on clocks on the 1070's? Nothing really noticeable unless you're getting low fps to begin with. Guess it's free performance tho.

After reading the forums in the past few days i decided to share my opinion and impressions on the issue, i don't like vulgar, rude or any sort of heated discussions so i'll try to make this my first and last post about the issue, i'm pretty sure the mods already got their hands full with so many people fighting between themselves.
 
I'll be honest and admit i was scared when i came back from a 2 weeks business trip and found out that most PC/Hardware forums and communities were all making an uproar about cards exploding and that i was in risk of getting my cat killed if i didn't RMA my 1080 FTW right away, it all happened at the same time in such short time frame between the reported cases that i was completely confused (why did all these cards go up in flames in the same week to begin with?)
 
But after cooling my head and searching a bit it doesn't sound that much of a doomsday like people are trying to make it look. First off everytime time someone posts a picture or video it's always the same pictures and the same videos, not saying everyone is lying but the facts points that may be a dozen of cards with this issue at best, this isn't like Galaxy Note 7.
 
I got my 1080 FTW back in July and so far i never got a single issue. No black screen, no fans spinning 100%, no crashes and no instabilities. Nothing, nada and i play games daily after finishing my work. Then again i got a high-end full tower case (Corsair 780T) so airflow and cooling isn't an issue, anything better than this only if i go liquid. But i can understand the fears, no one wants the trouble of having their cards having any issue or worse, spitting flames. This could have changed my mind if i heard about this back in July, i almost went with a Strix since EVGA cards are taking too long to appear in local stores where i live.
 
Personally i'll just keep using the card as i have been always doing in the last few months and call EVGA should anything happen, i'm not a fanboy (this is my first EVGA card) but EVGA is the only brand to offer a 3 years warranty in my country, even Asus and Gigabyte offer only 1 year here. If an Asus card were to die after 1 year i was pretty much screwed up, with EVGA i can at least RMA it.
 
I'll keep following these discussions since it's something that is pretty hard to ignore, so far everything points out to be the VRM but i can't get this feeling off that something may be adding up to make that happen, maybe a combination of hot VRM with something else that people have yet to discover.
 
Take care everyone

brokencross

• We know those 1080 FTWs blowing up had faulty VRMs and that was corrected on the cards shipped after August 31st, according to EVGA. This issue is also limited to 1080 FTWs, EVGA has stated officially no other GTX 10 models were affected by such.

Where was this official statement made?  Is there a link?

Thanks!

So in fact, this picture as nothing to do with the vrm overheating ? 
 
http://icdn2.digitaltrends.com/image/nvidia-1070-vrm-damage-1280x720.jpg
 
There isnt a single case about this overheating ?

I know there is a problem but is it possible that we all overeacting about this ?