kennethbored
This happened to me about a week ago. I couldn't find any post and thought maybe it was just an isolated incident. I actually just received my new RMA'd EVGA 1070 SC edition and now I'm considering getting a refund for it. 
 
 

It is certainly an option they will do for you.  I decided, since the guy I talked to on the phone was so open and honest that I would give the newly manufactured one another try.  After seeing the people on this forum though it really makes me want to go far far away from EVGA however.

ZCTxCHAOSx

HeavyHemi

ZCTxCHAOSx

HeavyHemi

ZCTxCHAOSx

Sajin

ZCTxCHAOSx

GloR1ouS_
So basically you have no proof it's the vrm. The 1070sc is a reference pcb. Pull apart a founders edition card and compare it to the 1070sc and tell me there's a vrm issue. Cards fail, just like every part in life can fail. Automobile parts fail way more often than computer parts.

It is a matter of oversight in regards to the VRM cooling.  EVGA has already admitted to this oversight hence the new Thermal Pad Modifications being offered and/or full replacement of cards for those that want to send them in. As well as newly produced ones being corrected.  Please don't be the typical know it all that seem to roam these parts.

The article also states that the pads are optional and the cards run within spec without the pads. Looks like evga will now be adding pads to all newer cards because the internet has blown up because of missing pads. The pads don't even help reduce temps... http://forums.evga.com/FindPost/2572799

Man this Sajin guy is just begging to get me banned. Unfortunately it is not going to work.  First of all the only real way to fully make this right is to build the VRM into the heat sink.  Second these new pads DO reduce the temperature  from over 100c to ~75c.  I swear someone needs to put this guy in a deep dark Silence of the Lambs type hole where he belongs.   At least there you will have company with your same intelligence, dirt and worms.

The offer is still open for you to show this 'massive' VRM failures you're claiming. I've no doubt you'll ignore the request again. The only person responsible for your 'know-it-all' insulting posts are you.

Fine, I'll give into your limited mental capacity needs and google for you... https://www.google.com/search?q=EVGA+vrm+problems&oq=EVGA+vrm+problems&aqs=chrome..69i57.6691j0j4&sourceid=chrome&ie=UTF-8   They are even talking about it over on ASUS   Get a life please!

In other words, nothing other than the echo machine of the same thing repeated over and over.  I guess you're getting the attention you needed.

Of course its the same thing echoed over and over.   You want it all to be completely different topics?  What in the actual heck is you problem.  This is why I didn't respond to you the first time. You have zero mental capacity. ZERO!

So you agree, there are two separate issues. The VRM/memory issue running hotter than other models, and the isolated failures of VRM components. I'm glad we finally settled your confusion. You didn't answer because you had nothing to back your claim of 'massive' hardware failures on which you based your call for everyone to boycott an EVGA product. Such a serious claim as yours requires evidence. Offer is still open.

HeavyHemi

ZCTxCHAOSx

HeavyHemi

ZCTxCHAOSx

HeavyHemi

ZCTxCHAOSx

Sajin

ZCTxCHAOSx

GloR1ouS_
So basically you have no proof it's the vrm. The 1070sc is a reference pcb. Pull apart a founders edition card and compare it to the 1070sc and tell me there's a vrm issue. Cards fail, just like every part in life can fail. Automobile parts fail way more often than computer parts.

It is a matter of oversight in regards to the VRM cooling.  EVGA has already admitted to this oversight hence the new Thermal Pad Modifications being offered and/or full replacement of cards for those that want to send them in. As well as newly produced ones being corrected.  Please don't be the typical know it all that seem to roam these parts.

The article also states that the pads are optional and the cards run within spec without the pads. Looks like evga will now be adding pads to all newer cards because the internet has blown up because of missing pads. The pads don't even help reduce temps... http://forums.evga.com/FindPost/2572799

Man this Sajin guy is just begging to get me banned. Unfortunately it is not going to work.  First of all the only real way to fully make this right is to build the VRM into the heat sink.  Second these new pads DO reduce the temperature  from over 100c to ~75c.  I swear someone needs to put this guy in a deep dark Silence of the Lambs type hole where he belongs.   At least there you will have company with your same intelligence, dirt and worms.

The offer is still open for you to show this 'massive' VRM failures you're claiming. I've no doubt you'll ignore the request again. The only person responsible for your 'know-it-all' insulting posts are you.

Fine, I'll give into your limited mental capacity needs and google for you... https://www.google.com/search?q=EVGA+vrm+problems&oq=EVGA+vrm+problems&aqs=chrome..69i57.6691j0j4&sourceid=chrome&ie=UTF-8   They are even talking about it over on ASUS   Get a life please!

In other words, nothing other than the echo machine of the same thing repeated over and over.  I guess you're getting the attention you needed.

Of course its the same thing echoed over and over.   You want it all to be completely different topics?  What in the actual heck is you problem.  This is why I didn't respond to you the first time. You have zero mental capacity. ZERO!

ZCTxCHAOSx

HeavyHemi

ZCTxCHAOSx

HeavyHemi

ZCTxCHAOSx

Sajin

ZCTxCHAOSx

GloR1ouS_
So basically you have no proof it's the vrm. The 1070sc is a reference pcb. Pull apart a founders edition card and compare it to the 1070sc and tell me there's a vrm issue. Cards fail, just like every part in life can fail. Automobile parts fail way more often than computer parts.

It is a matter of oversight in regards to the VRM cooling.  EVGA has already admitted to this oversight hence the new Thermal Pad Modifications being offered and/or full replacement of cards for those that want to send them in. As well as newly produced ones being corrected.  Please don't be the typical know it all that seem to roam these parts.

The article also states that the pads are optional and the cards run within spec without the pads. Looks like evga will now be adding pads to all newer cards because the internet has blown up because of missing pads. The pads don't even help reduce temps... http://forums.evga.com/FindPost/2572799

Man this Sajin guy is just begging to get me banned. Unfortunately it is not going to work.  First of all the only real way to fully make this right is to build the VRM into the heat sink.  Second these new pads DO reduce the temperature  from over 100c to ~75c.  I swear someone needs to put this guy in a deep dark Silence of the Lambs type hole where he belongs.   At least there you will have company with your same intelligence, dirt and worms.

The offer is still open for you to show this 'massive' VRM failures you're claiming. I've no doubt you'll ignore the request again. The only person responsible for your 'know-it-all' insulting posts are you.

Fine, I'll give into your limited mental capacity needs and google for you... https://www.google.com/search?q=EVGA+vrm+problems&oq=EVGA+vrm+problems&aqs=chrome..69i57.6691j0j4&sourceid=chrome&ie=UTF-8   They are even talking about it over on ASUS   Get a life please!

In other words, nothing other than the echo machine of the same thing repeated over and over.  I guess you're getting the attention you needed.

Of course its the same thing echoed over and over.   You want it all to be completely different topics?  What in the actual heck is you problem.  This is why I didn't respond to you the first time. You have zero mental capacity. ZERO!

So you agree, there are two separate issues. The VRM/memory issue running hotter than other models, and the isolated failures of VRM components. I'm glad we finally settled your confusion. You didn't answer because you had nothing to back your claim of 'massive' hardware failures on which you based your call for everyone to boycott an EVGA product. Such a serious claim as yours requires evidence. Offer is still open.

Never said there was "massive failures"  There is however a massive amount of defective hardware in the community right now just waiting to fail.  The cards have only been out a few months now so over time more and more actual failure reports will surface. Or may not even surface as a lot of people like me may just think its a general hardware failure, send in for RMA and never know what happened.  I wouldn't have even known, or thought to look if I would not have been linked on twitter.  We are just now getting to the sweet spot where the over stress is starting to take effect.  
 
Now I am here defending the general consumer, and opening thier eyes to what EVGA is not being very open or upfront about.
 
Game Set Match. Leave me alone now.

ZCTxCHAOSx

GloR1ouS_
So basically you have no proof it's the vrm. The 1070sc is a reference pcb. Pull apart a founders edition card and compare it to the 1070sc and tell me there's a vrm issue. Cards fail, just like every part in life can fail. Automobile parts fail way more often than computer parts.

It is a matter of oversight in regards to the VRM cooling.  EVGA has already admitted to this oversight hence the new Thermal Pad Modifications being offered and/or full replacement of cards for those that want to send them in. As well as newly produced ones being corrected.  Please don't be the typical know it all that seem to roam these parts.

You realize these pads are strictly for the custom pcb models and not the reference? The SC edition is literally a founders edition card with the acx 3.0 cooler slapped on it.

Just wondering why the ..70 series cards have all the problems. They are always requiring bios updates also. Most other cards are fine. Really not trying to be a jerk, just curious. Are they made in a different factory? Lesser components?

GloR1ouS_

ZCTxCHAOSx

GloR1ouS_
So basically you have no proof it's the vrm. The 1070sc is a reference pcb. Pull apart a founders edition card and compare it to the 1070sc and tell me there's a vrm issue. Cards fail, just like every part in life can fail. Automobile parts fail way more often than computer parts.

It is a matter of oversight in regards to the VRM cooling.  EVGA has already admitted to this oversight hence the new Thermal Pad Modifications being offered and/or full replacement of cards for those that want to send them in. As well as newly produced ones being corrected.  Please don't be the typical know it all that seem to roam these parts.

You realize these pads are strictly for the custom pcb models and not the reference? The SC edition is literally a founders edition card with the acx 3.0 cooler slapped on it.

Yes slapped on it without proper thermal protection applied over the VRM area.  The PCB makes no difference.  Once again, your point is really no point at all. Keep chiming in only to look foolish.

Shockjockey
Just wondering why the ..70 series cards have all the problems. They are always requiring bios updates also. Most other cards are fine. Really not trying to be a jerk, just curious. Are they made in a different factory? Lesser components?

A lot of that depends on manufacture date and if Samsung vs Micron memory was used.  The overheating issue affects the back plated ones prior to a certain date.

LotsOfPosters
An EVGA tech has said, on this forum, only the 1080 FTWs had a 3-4% faulty VRM rate.

Several people have posted the phrase quoted above, but I don't think it is an accurate interpretation of what EVGA stated.

The quote from EVGA_Jacob is at http://forums.evga.com/m/..aspx?m=2530081&fp=3 and says:

EVGA_Jacob
.... This previous black screen issue was caused by On Semiconductor (VRM IC Manufacturer) VRM IC’s being out of spec, triggering the OCP (Over Current Protection) and will not cause any damage to your system. The percentage of the IC’s that were out of spec has been confirmed by On Semiconductor and is approximately 3% to 4%. ...

Note that the 3-4% estimate is NOT for the "percentage of cards", but rather for the "percentage of IC's" affected. I believe there are 12 of those IC's on each card?

If it is 3-4% of ICs, and there are 12 of them on each 1080FTW, then (for cards shipped before the supply fix on Sept-1)... There is a 36-48% chance of having at least (1) out-of-spec VRM (although we don't know whether (1) bad VRM is enough to cause the problem). Again, this applies only to 1080FTW's shipped from EVGA before Sept-1, and the symptom is "black screen and fans go to 100%".

Bottom line, I don't think we know what the "percentage of affected cards" is. Based on 3-4% of ICs being bad, the minimum would be 3% (assuming all 3% rate and all 12 have to be "bad" to cause failure) and maximum would be 48% (assuming 4% and only one bad IC can cause the failure).

How does everyone else interpret that?

Yes, you have interpreted the quote correctly.
 
Until someone does a thorough comparative analysis of the hardware components used in cards any talk of design flaw is pointless.
The design may be sufficient for properly working components.
If any of the IC's are substandard or are generally of lesser quality that would lend them to perform to a lesser degree of acceptability as compared to competitors cards then any attributions as to cause of the higher temps are frought with confounds.
There still has been no acceptably valid or reliable science reported. 

Scott_W

Note that the 3-4% estimate is NOT for the "percentage of cards", but rather for the "percentage of IC's" affected. I believe there are 12 of those IC's on each card?

If it is 3-4% of ICs, and there are 12 of them on each 1080FTW, then (for cards shipped before the supply fix on Sept-1)... There is a 36-48% chance of having at least (1) out-of-spec VRM (although we don't know whether (1) bad VRM is enough to cause the problem). Again, this applies only to 1080FTW's shipped from EVGA before Sept-1, and the symptom is "black screen and fans go to 100%".

Bottom line, I don't think we know what the "percentage of affected cards" is. Based on 3-4% of ICs being bad, the minimum would be 3% (assuming all 3% rate and all 12 have to be "bad" to cause failure) and maximum would be 48% (assuming 4% and only one bad IC can cause the failure).

How does everyone else interpret that?

Wow, I didnt notice that.
I wish it is 3% of gpu and not 48 %   (!!)   
 
I just bought  a 1080 ftw  second hand and I am very worried
Although it has no major problem, I'm not sure to keep it because if it has weakened components, it does not live old, and for that price ... this is not what I ' hope !

 
I would see when I receive it.
 
What about EVGA costumer service with second hand?