Nostyke
from reddit :
 

So, as an Electronics Engineer and PCB Designer I feel I have to react here.xHowever, concluding from this that a POSCAP = bad and MLCC = good is waaay to harsh and a conclusion you cannot make. Both xx prone to cracking- prone to piezo effect-x

..... I happen to have a EE background and the piezo effect is my fav physics thingy.... so... 2 things.... if you strike it does it produce electricity and worse, when electricity is applied to it does it vibrate?  if so that's a horrible flaw considering they also say it's prone to cracking! 

https://forums.evga.com/Message-about-EVGA-GeForce-RTX-3080-POSCAPs-m3095238.aspx
 

kevinc313

Mattler
Do we know if EVGA is gunna fix that? I was hoping for the FTW card. But if it has that I'm going to get the ASUS Strix because it has all 6.

I wouldn't sweat it too much.  The two in the center seem to be the important ones.  2080 Ti have two similar arrays with option for a big cap, some 2070's and 2080's actually use the big caps.

Well that's good.

guitarwar241
This issue was fixed after their initial testing. Nothing to worry about now. 

Good.

Think its been addressed, believe its 2MLCC

Thehalfdemon

HeavyHemi

javelina1

HeavyHemi

INGREDCOLD

HeavyHemi
This is a technically inept and absurd thread. For most applications POSCAPs are superior.
Wow...the internet 'experts'. Why is everyone claiming POSCAPs are cheaper and lower quality?

Do think any one was saying that. mlcc is better at filtering out electronic noise 

No, they are not. In fact they are cheaper and generally inferior. Holy cow...utterly bassackwards.  

What's your perspective to the issue?  would appreciate any thoughts and insight you have.  Anecdotal evidence thus far pointing towards POSCAPs vs MLCC?  I also work at a large Semiconductor outfit, but not as a EE but rather IT.  I'd ask some of my designer friends regarding this, but I'm having to WFH for a bit and no chance to stroll up via the water fountain chats...

My perspective on this is that it would not be the use of POSCAPs which are superior to cheap MLCCs being the issue. My perspective is, if there is an instability, is design not component choice in this specific example. Other than that,  without actual schematics I would just be guessing.

I'm just here to see if HeavyHemi eats crow... Jumps on forums, flashes credentials and talks down to folks having a discussion and turns out he was flat wrong. 

Some people belong on QC teams, others on forums. Looks like everyone was cast correctly.

HeavyHemi

NexusPhase

HeavyHemi
This is a technically inept and absurd thread. For most applications POSCAPs are superior.
Wow...the internet 'experts'. Why is everyone claiming POSCAPs are cheaper and lower quality?

For most sure. Are you 100% sure this fits in to "most"? That's the reason for this thread. So we can talk about it

As a MSEE, a member of IEEE and JEDEC, manager of the ATE division for one of the largest semi companies in the world (now retired). Yes in fact I am. Just what do you believe anyone here who is not a professional is going to be doing anything but expressing their opinions? Mostly based on what some amateur on the internet posted.

I find it intersting that someone claiming to be such an expert in this field has FAILED to notice that the caps in queston ARE NOT POSCAPS (A Panasonic Product, not a generic term).  POSCAPS have distinctive markings on the components as shown here:
 
https://industrial.panaso...ymer-capacitors/poscap
 
The black rectangular caps being discussed here and used in the Bypass section behind the GPU are actually called SP-Caps and are made by a variety of manufacturers.  (Any "real expert" would know that)
 
I have also noticed that the actual value of these caps seem to be all over the map from one GPU brand to another. (220, 330 to 470 microfarad)  This tells me that NVIDIA has been playing fast and loose with the AIBs in this regard simply because NVIDIA SIGNS OFF on ALL AIB designs, and as a direct result, they approved the use of 6 SP-Caps over any MLCCs as well as SP-Caps with values as low as 220.
 
Lastly... The MLCC route is actually MORE expensive to produce on the card than the use of SP-Caps.  Only ASUS saw fit to invest in a 100% MLCC solution for their Bypass designs on both the TUF and Strix models. Using 100% MLCCs on the back of the TUF GPU is especially admirable considering the MSRP.
 

I just want there to be a full deep dive on this so we can get this settled once and for all.  I just want to know what's the safest bet as someone who plans to watercool and OC to high, but not excessive levels.  I don't want to buy something that's handicapped from the gate if it meant the AIB saved a few dollars on capacitors...

KrotosTheTank
I just want there to be a full deep dive on this so we can get this settled once and for all.  I just want to know what's the safest bet as someone who plans to watercool and OC to high, but not excessive levels.  I don't want to buy something that's handicapped from the gate if it meant the AIB saved a few dollars on capacitors...

I'd say the FTW3 and most likely the STRIX are safe bets in terms of AIB's. But I'd wait while they figure the problem out right now.

ty_ger07

ShaconBacon

KrotosTheTank
I just want there to be a full deep dive on this so we can get this settled once and for all.  I just want to know what's the safest bet as someone who plans to watercool and OC to high, but not excessive levels.  I don't want to buy something that's handicapped from the gate if it meant the AIB saved a few dollars on capacitors...

I'd say the FTW3 and most likely the STRIX are safe bets in terms of AIB's. But I'd wait while they figure the problem out right now.

What problem? There is no problem right now.



Some cards will be unstable from the factory and will have crash to desktop problems. There have always been some percent which are defective, and there will always be some percent that are defective. That is the imperfect world we live in. That doesn't mean that all of the video cards have a problem or that something needs to be fixed.

There are tons of people on Reddit and Twitter posting videos of their new graphics cards having a lot of defects. The cards with the most issues are the cards that used 6 POSCAPS instead of 4 POSCAPS + 2 MLCC. Watch the video by JayzTwoCents titled "The RTX 3080 Launch can't get any worse... Right? Wrong..." cause he explains it more in depth. I am not saying there won't be issues wit these cards, but there seems to be enough issues to look into it.

ty_ger07

ShaconBacon

ty_ger07

ShaconBacon

KrotosTheTank
I just want there to be a full deep dive on this so we can get this settled once and for all.  I just want to know what's the safest bet as someone who plans to watercool and OC to high, but not excessive levels.  I don't want to buy something that's handicapped from the gate if it meant the AIB saved a few dollars on capacitors...

I'd say the FTW3 and most likely the STRIX are safe bets in terms of AIB's. But I'd wait while they figure the problem out right now.

What problem? There is no problem right now.



Some cards will be unstable from the factory and will have crash to desktop problems. There have always been some percent which are defective, and there will always be some percent that are defective. That is the imperfect world we live in. That doesn't mean that all of the video cards have a problem or that something needs to be fixed.

There are tons of people on Reddit and Twitter posting videos of their new graphics cards having a lot of defects. The cards with the most issues are the cards that used 6 POSCAPS instead of 4 POSCAPS + 2 MLCC. Watch the video by JayzTwoCents titled "The RTX 3080 Launch can't get any worse... Right? Wrong..." cause he explains it more in depth. I am not saying there won't be issues wit these cards, but there seems to be enough issues to look into it.

None of EVGA's video cards have 6 SP-CAPs. So, what you are talking about is irrelevant. Again, see the link I posted above. Nothing to fix here. Move on. Carry on the discussion in some non-EVGA thread or non-EVGA forum if you want to have a thread relevant to what you are talking about.

Bro chill. I was trying to help a guy who said he didn't want to buy a broken card from an AIB. I never said anything about EVGA having broken cards, I even recommended EVGA cards to him. I am no longer replying in this thread. You need to find a better attitude rather than being rude to people just trying to help others.

This was only for the review boards i think

mjdoja
This was only for the review boards i think

That's correct.  It was the pre-production models only.  All the final retail versions are fine.

GG007

HeavyHemi

NexusPhase

HeavyHemi
This is a technically inept and absurd thread. For most applications POSCAPs are superior.
Wow...the internet 'experts'. Why is everyone claiming POSCAPs are cheaper and lower quality?

For most sure. Are you 100% sure this fits in to "most"? That's the reason for this thread. So we can talk about it

As a MSEE, a member of IEEE and JEDEC, manager of the ATE division for one of the largest semi companies in the world (now retired). Yes in fact I am. Just what do you believe anyone here who is not a professional is going to be doing anything but expressing their opinions? Mostly based on what some amateur on the internet posted.

I find it intersting that someone claiming to be such an expert in this field has FAILED to notice that the caps in queston ARE NOT POSCAPS (A Panasonic Product, not a generic term).  POSCAPS have distinctive markings on the components as shown here:
 
https://industrial.panaso...ymer-capacitors/poscap
 
The black rectangular caps being discussed here and used in the Bypass section behind the GPU are actually called SP-Caps and are made by a variety of manufacturers.  (Any "real expert" would know that)
 
I have also noticed that the actual value of these caps seem to be all over the map from one GPU brand to another. (220, 330 to 470 microfarad)  This tells me that NVIDIA has been playing fast and loose with the AIBs in this regard simply because NVIDIA SIGNS OFF on ALL AIB designs, and as a direct result, they approved the use of 6 SP-Caps over any MLCCs as well as SP-Caps with values as low as 220.
 
Lastly... The MLCC route is actually MORE expensive to produce on the card than the use of SP-Caps.  Only ASUS saw fit to invest in a 100% MLCC solution for their Bypass designs on both the TUF and Strix models. Using 100% MLCCs on the back of the TUF GPU is especially admirable considering the MSRP.
 

Why would you find it interesting I would stick with the terminology used by the OP? I guess you missed my point this entire thread is technically inept. MLCC could be more or less expensive depending upon other factors, such as the stability of the PWM section.  We don't have the design drawings in hand...thus back to point one: this thread is technically inept.

Urieldagda
I think another issue here is that there were some issues, yet most people neglect the fact that not all issues are going to be this issue. Even though some brands may now use all MLCC or some moved to a mixture, there will still be issues with OC stability and stability in general on some setups.  I'm willing to bet some will have issues OCing to 2Ghz just because they're silicon lottery losers. Some will have problems with poor ventilation. I'm hearing about GPU scheduling issues that Microsoft has to fix. The drivers have to mature.
 
The problem right now that I'm seeing all over the internet with this is so many people thinking all of the crashing has a single hardware cause due to capacitors.

Exactly. 
 
The exaggeration is running rampant and the usual OC instability norms, thermals, etc. will be blamed elsewhere.