Aurora R11, RTX 3090 repeat crashes, bad replacements

It certainly seems that way, however ... I'm not sure how many machines/cards Dell is selling compared to the reported failures here. 10 or so here seems like a lot, but might be a small-fraction of total Dell-OEM Nvidia RTX cards sold. 

IIRC, the 3xxx cards (or was it 2xxx cards) got off to a rough start (bad memory was it ?). Anyway, I didn't even like seeing that , but that should be in the past. I heard 4xxx cards were coming-out soon.

My GTX-1070 cards are rock-solid, but with all these reports ... I'm considering switching back to AMD after all these years.

Are you on any other forums or groups? I was wondering if everyone elses MSI cards are fine? Also, retail-MSI compared to eVGA or Zotac for example.

Thanks for the reply. It could certainly be caused by sampling error. However, what worries me is that Dell only offers a 1 year warranty versus Nvidia (or most other manufacturers of GPU) offers 3 years. 

My system is at the end of its warranty period after the next replacement. I would have no reassurances how long that would last. Given that the GPU accounts for the majority of the computer's value, it makes me very worried and I think others need to be aware of this when making a purchase as well. 

Yeah, I don't Over-Clock anything either. I'll take stability, reliable, and long-lasting any-day (over 5% faster).

The RTX-3090 is a nice card and I personally think you deserve a 100% working one. You can always extend your warranty. 

I remember when eVGA offered life-time warranties on Retail cards. I RMAed one once. My retail XFX-AMD cards were Lifetime (just got a new one a while back ... sent me a newer one).  Looks like my retail-MSI is only 3-years.

With your work-flow, I was hoping you had other data-points from co-workers or something.

Personally, I think the 3090 is overkill for the average user. I have a 3080 and I game at a power limit of 65% of the card, giving me a maximum of around 200 Watts of power draw and it performs well enough for the latest games at 2K resolution with all eye candy turned on.

Granted I only game at 60 Hz, but I also do not reach anything over 55 Celsius on that card when I game.

My ears and power bill thank me for that. At 100% power limit that card will draw around 320 Watts and generate a lot of extra heat. Not really performs any better FPS, just draws way more power.

I just gotta mention that there was one person that was having issues with crashing and her Alienware machine was placed by a wall covering the left side of the machine which is the air intake and exhaust port on that left panel. 

I have no issues with my R11 and the 3090. The Dell 3090 has great reviews so your the first I’ve have heard of having issues. Could it be other issues like Ram or a bad bios update? Both system and gpu vbios. Did they test you psu? 
I’ve seen many tech videos that crashes are mostly a bad Ram stick. Could it even be a bad solder on the mobo ? I think it’s worth extending the warranty and getting your R11 fixed by the right tech. But for your question about the Dell 3090 I have not have or heard any issues with this gpu. I got the R11/3090 because of the 24gb vram and for VR simulation game “DCS”  the gpu is popular for this kind of game play. I hope you can get the right answer and get it fixed by the right person.

Also, are you using MSI Afterburner for the gpu fan speed to keep the temps down? It’s a must have program. Are also using the R11 at High Performance Mode while gaming and working?

just a thought 

I wanted to keep this post technical. But I felt like I should share my support experience for this as well for others who are considering purchasing a computer from Dell for both work and personal use as it has been less than satisfactory for me.

First replacement: Dell couldn't diagnose over chat so I sent it into depot. Process was fine but returned with GPU that failed soon after

Subsequently once that failed, Dell rearranged on-site service as part of warranty. However, 2 appointments were made and confirmed both times but ended up being no shows (notified AFTER the appointment time, the first time Dell support over WhatsApp actually messaged directly for me to wait for the whole time-window, which was an unreasonable 9-6pm). I understand this was due to a part shipping delay, but when I was making the appointment I was reassured both times that the part should arrive. So after having to take time off work twice and losing more than a week of use and having to use backup computers for work, this is an issue that is still not resolved. 

I should add that I requested the alternative of having the replacement be sent to me directly if possible and I can ship back the defective one. However, I was told only on-site service was possible. 

Thanks for your input. I have not modified the fan curve. As for the cooling point, I understand and have observed poor temperatures and throttling during operation and have resorted to running it with the side panel open with no obstructions on the sides or above.

I thought it was the GPU (and apparently Dell did as well for the first replacement) given the fact that the PC is stable under CPU/Memory tests (cinebench and memtest) but crashes upon doing GPU benchmarks or gaming (see a sample image below, right before it crashed which occured the first time I really got a chance to use my replacement GPU within a week). 

Upon further searching, there is another post that had similar initial symptoms and quite a few people replied about similar experiences. (https://www.dell.com/community/Alienware-Desktops/Aurora-R11-randomly-crashes-to-black-screen/td-p/7748242/page/7).