Aurora R11, playing games, shuts down and does not restart

Are you using a displayPort connection,?  I've experienced problems when I leave 2 cables connected with DP being default, the machine acts like windows has crashed, but all the time the cable connection was confusing the machine. Nice that it works after a reseat. 

I'm using 2 monitors with display port. When the PSU shutdown et won't restart, I suspect a thermal problem or over current.

@jygarneau​  what are your CPU temps at idle and under load?? AIO liquid or air cooled CPU?

Two monitors on Display-Port is fine. When gaming, you are only using one anyway.

Install HWiNFO64 and find something overheating, if you think something is.

Be sure nothing is Over-Clocked.

Look in Reliability History Report for possible issues. There should be no (red) critical errors.

RTX-3080 is very powerful. I run MSI After-Burner for mine. Create this auto-loading Profile #1:

Core-voltage +0 mV, Core +0mhz & Memory +0mkz (so basically, Stock-Clocks),
90% Power-Limit, 83c Temp-Limit, Auto-Fans with Sync

Air cooled processor. 1000 Watts PSU. HWMonitor display max temp at 100°C but it's just a spike. My Dell G5 do the same thing. I'm using FurMark 2 with a AMD R9 200 on the Alienware to observe if the problem come from CPU. The AMD does not shutdown the PC. The CPU temp with the AMD and Cinebench R23 is around 95°C-96°C.

@jygarneau​ ,

1. Air cooled processor.

2. 1000 Watts PSU.

3. HWMonitor display max temp at 100°C but it's just a spike. 

1. How did that happen? Are all Aurora's not Liquid-Cooled any more?

2. Good, fine.

3. What is hitting 100c ? CPU or GPU. On a Alienware Desktop, nothing should be hitting 100c.

Please only talk about this one PC and it's components.

For the moment the AMD R9 200 is on the Alienware PC because I want to see if the CPU is the problem.

The Cinebench multi-cores test is used and temp checked with HWMonitor.

The 100°C is the max field on CPU global cores display in red.

But this peak seem to be only when Cinebench test start, then throttling low after (85°C and up) and after 5 minutes I see fix 95°C. The PC does not shutdown.

I want to determine if the thermal paste is the problem because I never change that paste on any of my computers.

@jygarneau​ ,

1. Cinebench test start, then throttling low after (85°C and up) and after 5 minutes I see fix 95°C. The PC does not shutdown.

2. I want to determine if the thermal paste is the problem because I never change that paste on any of my computers.

1. I suppose. I don't use it. Instead, I run Power-Supply Test (and some others) inside OCCT. Re-Encoding a long-ish video will also peg all the cores at 90-100% (but that doesn't concurrently run GPU up-high like OCCT does).

1. I use Noctua NT-H1. Not too little or too-much (watch some YT videos). Finally, primary heat-sink (or liquid-cooler block) must be down perfectly flat on chip.

The Alienware R11 works fine for 5 years and the last year i'm unable to play resource-intensive games. The FurMark 2 software have also a "CPU Burner" and run at near 100%. With this test the PC does not Shutdown. With these tests, I supposed the problem is not the CPU. I must replace the AMD for the RTX and restart these tests and see results. 

I can't test the RTX on another computer because need a 1000 Watts PSU.

Does anyone have changed the thermal paste on the RTX? Don't want change paste if is not the problem.

@jygarneau​

@jygarneau​

Does anyone have changed the thermal paste on the RTX? Don't want change paste if is not the problem.

Please Do NOT take your RTX-3080 apart. Anyway, you would not do that unless HWiNFO64 showed it over-heating.

At most it might need a (careful) 40psi air-compressor and small-brush on fan-fins cleaning.

Then use MSI AfterBurner and that Profile I posted above.

I execute the test on the Dell support web site but no errors detected.

I don't use overclocking.

I will try HWInfo64 and OCCT and replace the RTX on the motherboard.

I don't know if I can put the RTX on the last slot, physically.

Testing and return in a couple of days.

Thank you for your support.

Download Intel Extreme Tuning Utility and run the 5-minute stress test. In the bottom right corner of the screen you will be able to see what, if anything, is throttling your CPU. With the tiny pancake air cooler, throttling is most likely. Consider upgrading to the AIO liquid cooling.

XTU

I will try XTU. Thank You.

I supposed throthling the CPU never shutdown the computer?