I have a problem, in fact I have already sent messages to evga support, but I also leave it here, a couple of weeks ago, 2 or 3 I bought an evga 2080ti ultra xc, the first days it worked excellent, without this problem
One day I returned home, and he asked me for a flash of mcu in evga precision x1
When updating, everything is correct and the next day I started with this problem of having 1 fan at 100% and in hwmonitor it indicates 0rpm
In a Cold Start it happens more often, I always enter windows, after a while I open for example "Quake 2 RTX", to test, and when the card is set to 100 performance, temperatures rise and the fan is placed in madness mode
10 minutes pass, sometimes 5 minutes, and it normalizes, and I can play without a turbine inside my cabinet
the card at its maximum temperature in 3dMARK is 73 degrees Celsius
to play it works excellent, but I have this strange problem
 
The things I have tried:
 
-desintalar msi afterburner
-Flash MCU:
-DDU + update nvidia drivers
-Update bios motheboard z390
-drain the power of system

Even so, I cannot solve this problem.
The first few days I did not behave that way, so it seems strange to me, it seems that the mcu flash caused me these problems since that day, it has this behavior.
The card is connected to 2 PCI-E 8 Pin cables
the only thing I have not tried is: make a custom fan curve, which does not fall in idle the fan
and update bios of the gpu.

Sorry for my English. my main language is spanish
I will leave some screenshot.
 
one of the screenshot is at the moment where the temperature is low (44 degrees) and the fan is at 100 &, indicating 0 rpm
 
the second screenshot is in 55 degrees, already normalized, fan 1 and fan 2 showing correct rpm.
 
Normally after this happens, I can play with the fans normally all night.
 

Hello, I followed all the necessary steps now, the PC took a little longer to start and show the post of the bios, fans 0 rpm in bios, when entering windows 0 rpm, then I started quake 2 rtx and apparently I set everything normal rpm in the fans, I send you screenshot, anyway, this does not assure me that it is solved :( tomorrow I should try it anyway in a Start Cold, by chance, will there be any bios newer than the one I have?
 
when I refer to a fan curve, I mean the ones that are predetermined in PX1, in the STEALTH preset
 
my bios is: 90.02.17.00.7D

They may not understand me, my English is regular,
 
 
When I refer to drain the power system it is exactly this:
"Turn off the PC, unplug it from the wall, unplug everything (monitor, printer, USB hub with power ...) Press and hold the" start button "20 seconds to discharge all the electricity to the PC, plug everything back in. Start up. the PC .................. "
 
Did you try to set the fans to a fixed% RPM? ..... Say 70% - 80% and check temperatures while playing
  
A: Yes, and it works, mark the 2 FAN at the correct rpm, temperatures with frost (68 to 71 degrees at maximum use (100% use of the GPU)
 
When it happens, fan 1 is 100% and it indicates 0 rpm, (fan 2 always indicates normal rpm)
It is normally when the card is being used in gaming, or in something, type video, or heavy loads (in fact it seems as if it were lack of energy from the psu to the gpu), but it only happens 10 or 5 minutes, and it returns to normal, and it happens only with the first game that I open, (in the cold start) for example in quake 2 rtx
when playing dead by daylight everything seems normal, (sometimes it happens again in between playing, and sometimes not, it seems random moments.)
I will leave a video, so that they observe the situation.
 

 
After the 10-minute thing happens with the 100% fan, it normalizes, until the other cold start.
 
I want to highlight here: the gpu is connected with 2 PCI-E 8 PIN, and this happened to me, a while ago I switched to the other 2 PCI-E 8 PIN, (my psu, corsair TX850) has 4 Molex PCI-E, and I have to test in a cold start.
 
My Cabinet is a Cooler Master Cosmos SE, it has a correct ventilation flow, and I have a fan, that they enter cold air and take out hot air (4 fans, 2 industrial Noctua 3000 rpm, 2 fan cooler master that came in the cabinet)
 
all temperatures, both gpu and cpu, are very low, at tj max.
the system is frozen, in fact the ambient temperature where the cabinet is almost 38c, and the temperatures of the components are also low.
 
that fan 1 is 100%, it gives me to think that the useful life of the fans will be shorter, that's why I'm worried.
 
 

All the monitoring software, named below, marked me 0 rpm, at the beginning (in fan 1, because in fan 2 I always dialed normally), and then dialed normally.
today, it did not happen, in the cold start, because yesterday I switched to the other pci-e cables, but I can not take it for granted, since I have to continue testing, but it was with everyone, the bad thing is that I could not get him screenshot, just take screenshot of hwinfo64 and hwmonitor
 
 
Hw monitor, gpu-z, hwinfo64, Msi afterburner, Presicion X1
 
I noticed the image with hwinfo64, fan 2 with correct rpm, fan 1 0 rpm, and had normal temperatures, (I was playing dead by daylight)

 
 
later it normalized
 

 
As I said before, yesterday I changed the cables of the 8-pin pci-e to others I had, and the problem has not happened (for now) but I also have to try with more time.
 

And, does anyone have an idea what might be happening ?, With manual fan curve it is not solved.