So, apparently MOSTLY everyone needs to return their RTX 3080 / 3090 because of "cheap" components. I'm def returning my EVGA 3090.

SixFootDuo

Supreme [H]ardness
Joined
Oct 5, 2004
Messages
5,825
So, apparently MOSTLY everyone needs to return their RTX 3080 / 3090 because of "cheap" components. I'm def returning my EVGA 3090.

No way IN HELL am I going to accept this on a $1800 with tax video card.

But, then again, my card is very stable. I've not had any crashing but, will my components degrade over time?

Wait, yes, I do have a strange crash. Timespy crashes right off the jump. Let me go turn off my overclocking tool within Geforce Experience.

Trust me, watch this video.

 
Go right ahead - more for others!

BTW, EVGA tested this, fixed it, and isn't shipping cards with this issue:

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

I have read that at least 1 person is having issues with game ready drivers and the EVGA 3090 FTW3 Ultra, but the studio drivers work.


Thanks for this! I have the RTX 3090 XC3 Ultra with 5 + 1. I am going to try and find more pictures of the RTX 3090 XC3 Ultra's and see whats up.
 
Hmm, I guess if it was a 3090 XC3 it could get close. Either way its in spec and works.

Evga says it's ok but is it? Just going to take them at their word? I wouldn't if I was spending that much, I agree with the OP
 
So, apparently MOSTLY everyone needs to return their RTX 3080 / 3090 because of "cheap" components. I'm def returning my EVGA 3090.

No way IN HELL am I going to accept this on a $1800 with tax video card.

But, then again, my card is very stable. I've not had any crashing but, will my components degrade over time?

Wait, yes, I do have a strange crash. Timespy crashes right off the jump. Let me go turn off my overclocking tool within Geforce Experience.

Trust me, watch this video.



EVGA *shouldn't be* affected, they caught the problem before launch. Makes me wonder about the response by nvidia if this was a known issue before launch though. There is still a lot of research into the issue going on, but if you're still having the issue it's not due to the capacitors as EVGA pulled their stock and fixed it before launch.
 
My problem with this is, why are AIBs charging us more for lower quality than the FE cards? Remember the days when OC cards were a few bucks more and real custom cards were ranging around ~50$...
 
  • Like
Reactions: Vader
like this
Yeah EVGA already said their cards are not affected. I know my 3090 has the same layout as the 3090 FE. So, we are good bro!
 
Thanks for this! I have the RTX 3090 XC3 Ultra with 5 + 1. I am going to try and find more pictures of the RTX 3090 XC3 Ultra's and see whats up.
Found another person that has issues with the game ready drivers, this time with the GPU not downclocking at desktop with nothing open with his 3090 FTW3. The Studio drivers fixed it too. The original dude was limited to about 15% GPU usage in all games, and geforce experience caused all his games to minimize - he's the one who originally found out that the Studio drivers fix those issues (he has a 3090 FTW3).

Hopefully Nvidia pulls their heads out quickly as they marketed the 3090 as a gaming card :LOL::LOL::LOL::LOL:
 
Last edited:
So, apparently MOSTLY everyone needs to return their RTX 3080 / 3090 because of "cheap" components. I'm def returning my EVGA 3090.

No way IN HELL am I going to accept this on a $1800 with tax video card.

But, then again, my card is very stable. I've not had any crashing but, will my components degrade over time?

Wait, yes, I do have a strange crash. Timespy crashes right off the jump. Let me go turn off my overclocking tool within Geforce Experience.

Trust me, watch this video.


Its not down to component quality, its down to the size (and number) of capacitors used on each section.
If yours is working fine then you dont have the problem.
Its easy to check anyway by looking at the back of your card.
 
From that EVGA post, yours is the fixed one since it has 5+1. EVGA maintains no bad ones were shipped to customers, period.
I’m sorry but there’s no such thing as perfect manufacturing, no matter the product. There will always be at least some products that have issues, even after a supposed fix. And eventually something will go wrong with the ones that work 100% right now, which is why warranties exist. There’s no such thing as perfect production. Some just have the worst luck in general, like my cousin that, no matter what she buys, she has to return something 3 out 4 times before she wins the manufacturing lottery. Some companies have lower failure rates, but not one is immune to issues.
 
I’m sorry but there’s no such thing as perfect manufacturing, no matter the product. There will always be at least some products that have issues, even after a supposed fix. And eventually something will go wrong with the ones that work 100% right now, which is why warranties exist. There’s no such thing as perfect production. Some just have the worst luck in general, like my cousin that, no matter what she buys, she has to return something 3 out 4 times before she wins the manufacturing lottery. Some companies have lower failure rates, but not one is immune to issues.
I believe he was specifically addressing the current MLCC / POSCAP debacle - not the manufacturing defects per million.
 
I’m sorry but there’s no such thing as perfect manufacturing, no matter the product. There will always be at least some products that have issues, even after a supposed fix. And eventually something will go wrong with the ones that work 100% right now, which is why warranties exist. There’s no such thing as perfect production. Some just have the worst luck in general, like my cousin that, no matter what she buys, she has to return something 3 out 4 times before she wins the manufacturing lottery. Some companies have lower failure rates, but not one is immune to issues.

Yeah, not sure what I am going to do. I think in 30 days, before my window is up, hoping that 3090 come back in stock and stay in stock so I can get something else. I'm very uncomfortable with 5 + 1
 
Yeah, not sure what I am going to do. I think in 30 days, before my window is up, hoping that 3090 come back in stock and stay in stock so I can get something else. I'm very uncomfortable with 5 + 1

Huh interesting. I have a 3090 XC3 Ultra as well, but my layout is 4+2.....odd how yours is 5+1
 
Huh interesting. I have a 3090 XC3 Ultra as well, but my layout is 4+2.....odd how yours is 5+1
I read that the XC3 got a silent revision though I thought it was the 3080 XC3 that was mentioned. If it got a revision I definitely wouldn't want the earlier version.
 
Man what a cluster.... Doesn't nvidia have to apove aib cards? If so this is 100% on them.... I wonder if it's because they tried to push the core harder then they thought they where going to have to or something? Wonder if nvidia/aib are going to push out a sw fix that caps the boost clocks and say it's fixed for day one buyers or rma all the cards?

This doesn't bold well for the 300x life span if any cap degradation causes it to fail
 
Man what a cluster.... Doesn't nvidia have to apove aib cards? If so this is 100% on them.... I wonder if it's because they tried to push the core harder then they thought they where going to have to or something? Wonder if nvidia/aib are going to push out a sw fix that caps the boost clocks and say it's fixed for day one buyers or rma all the cards?

This doesn't bold well for the 300x life span if any cap degradation causes it to fail
That’s why u make sure to get a good warranty from an AIB. That’s why I always go with EVGA.
 
The 3080/3090 release seems to be a paper launch, but even if AMD\Nvidia could release say 1 million cards vs 10k or 100k cards, this release seems to be, would they? Nvidia\AMD live on cutting edge with videocard releases, that it limits release allocation to allow for risks like this (where recall is in order) to release earlier with less testing time. Compared to CPUs and Consoles releases which have to assume are tested with ample time to find any major issues in house since releases are much larger and much more costly to recall.
 
EVGA *shouldn't be* affected, they caught the problem before launch. Makes me wonder about the response by nvidia if this was a known issue before launch though. There is still a lot of research into the issue going on, but if you're still having the issue it's not due to the capacitors as EVGA pulled their stock and fixed it before launch.
This was only a rampant problem with AIB cards, the founders edition users have not been reporting these issues nearly as much (if at all).
This technically shouldn't have anything to do with nVidia from the information I've gathered, they sent out the minimum specs needed for the AIB manufacturers.
Its up to them how they want to design and build out their board power delivery and filtering which is the alleged cause of the issues.

For anyone interested theres a on going list of the different poscap/mlcc designs from all the manufacturers here:
https://www.reddit.com/r/hardware/comments/izmi1k/ampere_poscapmlcc_counts/
 
Last edited:
This was only a rampant problem with AIB cards, the founders edition users have not been reporting these issues nearly as much (if at all).
This technically shouldn't have anything to do with nVidia from the information I've gathered, they sent out the minimum specs needed for the AIB manufacturers.
Its up to them how they want to design and build out their board power delivery and filtering which is the alleged cause of the issues.

For anyone interested theres a on going list of the different poscap/mlcc designs from all the manufacturers here:
https://www.reddit.com/r/hardware/comments/izmi1k/ampere_poscapmlcc_counts/
Or what if nvidia sent then bad specs/cores? The fact this has never happened before on reference design cards and does now says someone/something was given out as bad specs, or something is more sensitive then it should be....when does anyone remember any card crashing when it's not over clocked/to hot? I really doubt 90% of the aib said hey let's make cards under nvidia spec...had it be one company cards I could see that happening

Fo all we know it doesn't happen to fe cards because they bin the chips or something
 
This was only a rampant problem with AIB cards, the founders edition users have not been reporting these issues nearly as much (if at all).
This technically shouldn't have anything to do with nVidia from the information I've gathered, they sent out the minimum specs needed for the AIB manufacturers.
Its up to them how they want to design and build out their board power delivery and filtering which is the alleged cause of the issues.

For anyone interested theres a on going list of the different poscap/mlcc designs from all the manufacturers here:
https://www.reddit.com/r/hardware/comments/izmi1k/ampere_poscapmlcc_counts/

Yes, nvidia DID not follow their own reference design in this cycle, which was known from the beginning. That's why their boards look so different than AIB cards. Also, NVIDIA gave minimum specs to AIB's which they followed. Now cards are showing instability issues. Now we know why NVidia didn't want to follow their own reference design documentation ;). The manufactuers *should* have found this before shipping, which EVGA did and is why they had a recall prior to launch, so good on them for finding this. One of the issues was AIB's couldn't test cards out due to missing drivers which they didn't get until right before launch, when the initial batch of manufacturing, etc was done. If someone like zotac took a few out of packages to then test and they all worked, they had no reason to think the others may have issues. NVidia knowing what EVGA found should have been more forward with other AIB's about possible issues with using their lower spec reference design and the possibilities of crashing (wether they did this or not, we don't have confirmation of, so who knows what or when AIB's knew anything). This was a rushed launch from the get go and nvidia dropped the ball pretty hard in more ways than one. I'm wondering if they are going to have recalls of affected cards, or just push out a bios "fix" that reduces boost clocks. The ball is in nvidia's court, but they've been very silent on the issue. Probably stuck between pissing off their fans/customrs and pissing off their AIB's and trying to figure out how to save some face with both.
 
Or what if nvidia sent then bad specs/cores? The fact this has never happened before on reference design cards and does now says someone/something was given out as bad specs, or something is more sensitive then it should be....when does anyone remember any card crashing when it's not over clocked/to hot? I really doubt 90% of the aib said hey let's make cards under nvidia spec...had it be one company cards I could see that happening

Fo all we know it doesn't happen to fe cards because they bin the chips or something
The manufacturers followed the minimum spec. The FE edition did not and was built somewhere in the mid-level spec. The mfg's didn't get together and decide to make under-specced cards, they made cheap cards to the low end spec to sell near MSRP. Also, you are correct that higher binned parts probably would do better even with slightly crappier power, and there were rumors (not sure if they were substantiated) that nvidia was keeping good binned parts for themselves (not sure how many or w/e, but they weren't keeping the lowest spec dies).
 
My EVGA 3090 FTW has been a little crashy (comparatively), but what's to say that isn't driver-related? Supposedly that model follows Nvidia's FE and isn't affected. I'm not returning anything until some driver updates arrive and I'm certain this isn't just simply software. Performance on this thing is out of this world and literally exactly what I wanted, so I'm not about to return something after 2 days that might be fixable via drivers or firmware.
 
My EVGA 3090 FTW has been a little crashy (comparatively), but what's to say that isn't driver-related? Supposedly that model follows Nvidia's FE and isn't affected. I'm not returning anything until some driver updates arrive and I'm certain this isn't just simply software. Performance on this thing is out of this world and literally exactly what I wanted, so I'm not about to return something after 2 days that might be fixable via drivers or firmware.

Lowering the boost clocks below a certain point seems to fix the crashing for people with this issue. You could always check if that fixes your issues too and if it does I would think that most likely points towards hardware not drivers though it could be PSU related or something.
 
The manufacturers followed the minimum spec. The FE edition did not and was built somewhere in the mid-level spec. The mfg's didn't get together and decide to make under-specced cards, they made cheap cards to the low end spec to sell near MSRP. Also, you are correct that higher binned parts probably would do better even with slightly crappier power, and there were rumors (not sure if they were substantiated) that nvidia was keeping good binned parts for themselves (not sure how many or w/e, but they weren't keeping the lowest spec dies).
Thing is the minimum spec should be 100% stable or it's not really a minimum spec is it? Also there's no way all the aibs even evga would make a card that wouldn't work my bet is nvidia test cores they sent out worked 100% or where caped in the bost clocks for testing then when they got the for production cores something changed/ the uped the clocks or something and the fact it's not all cards but only some suggest to me something has a defect and making it more sensitive to what caps are used.

But for all we know it could be people just uesing lame psu/cables or drivers
 
Lowering the boost clocks below a certain point seems to fix the crashing for people with this issue. You could always check if that fixes your issues too and if it does I would think that most likely points towards hardware not drivers though it could be PSU related or something.

I have a 1000w Corsair without a ton of devices plugged into it, so I'd like to think my PSU is fine. At least one of my crashes was with Horizon Zero Dawn, which is kinda notorious to begin with. I only have 2 weird crashes. One was with Precision X's curve test and the other was with Borderlands 3 and a whole lot of Mayhem 10 chaos. Borderlands tends to be pretty stable, so that's and odd one. Only had it crash in similar circumstances on my 2080Ti once.
 
Back
Top