markedpk

deathlessdd

gsrcrxsi

deathlessdd
jayztwocents did show the difference between a MSI 3090 and a evga 3090 the fact is evga doesnt stay near the power set limit no matter what game it plays. Which is very odd what is the point of having a power limit set if it goes 22% over the power limit? Think evga has some issues with this for sure. 

because the power % isnt showing what he thinks it's showing.
 
watch the video and many different times you'll see values ranging from 300W-400W all showing the same ~100%.
 
so 300W is 100% and 400W is also 100%? right.
 
it's most likely getting that % value from looking at the pcie slot power vs pcie slot power limit.

Clearly this isnt how the power limit works if you set it to 102% it shouldnt goto 120% period. I have a msi 3090 and a evga and I can confirm they do not act the same when setting a power limit it stays at the power limit it doesnt go over it. If the bios has a limit of 102% instead of 100% it shouldnt go over that.

I wonder how accurate the read outs even are though

charlesborner

8thWonder
 
Well the crashes aren't normal. The fans go at 100% and the PC stays on so god knows what heat damage has been done. Just because I managed to play around and get mine to boot it doesn't mean there's no long term heat damage or a flaw in the design. My PC wouldn't boot after I finally got it to boot by selecting the alt bios on my motherboard. When it loaded in to windows the screen was flickering like crazy, I managed to fix by re installing the normal BIOS (was on the 450w one previously) so I'm effectively running it on reduced load. 
 
Pretty sure I can kill the card if I go back to the 450w bios and play new world uncapped if that's what you're suggesting?

Okay, when something is going on and the card is about to die, crashing IS normal.
This is the card protecting itself.
Is it desirable behavior?
If you're trying to game?  No.
If you're trying to preserve your card?  YES!

The problem is, these sort of protections have a limited shelf life.
You can only over-current so many times before bad stuff REALLY starts to happen to the device.

So, do you prefer a card that crashes?
Or a card that dies?

Duppex77
isn't Jayztwocents on EVGA payroll??

He always seems to get their GPU's before other YouTubers.

Xavier Zepherious
by looking at the pic above - im assuming an induction coil blew
it's that or a liquid capacitor
 
if thats the case it may be due to dirty power at high loads
coil whine is indications of dirty power trying to smooth out the incoming power
caps inline pass AC thru them - coils block or try to
 
for a coil to go - would also mean overpower
 
too early to speculate 
definitely with that pic it's not the problem from before
 

neteng101
 
 
One of the 20A fuses blew but the damage around it must have taken some serious loads to do.  Its one of the 3 PCIe power connector inputs, I'd guess #2 since that runs hot on all FTW3 cards.

rogueSkib

gsrcrxsi

nezff

the same crappy article posted again.
 
EVGA says they are RMA-ing cards. nothing more.
 
Jay's comments about them cross-shipping is hearsay and being blown out of proportion. there is no evidence that they are doing this. and I haven't seen a single person confirm who RMA'd their card that they are.

I was bricked by New World on Tuesday night. On Wednesday, EVGA accepted my RMA. They called me back later on and said they would be shipping me a new card immediately (cross-shipped!).
I got the new card on Friday, and couldn't get it to work (No Signal on my monitor). I posted a thread on these forums, and someone helped me figure out that I needed to reset my CMOS on the motherboard (see my post about Troubleshooting on the front page).

Outofstock4ever
i would also go for a refund, at least until they acknowledge the problem and go for a full redesign of the PCB, or you'll just enter the RMA loop until the warranty expires