Forceflow

Out of the blue, following behaviour: PC boots out of cold state When I put it to sleep, and try to wake it later, half of the time, I hear a click, the sleep light goes out permanently and it doesn't react to power button presses anymore. I have to fully shut off (with the I/O button) the PSU before I can start my PC again. Faulty / failing PSU? (A 3 year old Corsair RMX850) The fact that I can sometimes get it to boot fully and click around in Windows leads me to believe that the actual components are fine, and it's just the power required to wake from sleep that is failing in the PSU?

Not comfortable opening up my card. It was just this one game doing a very specific workload doing this. I've let it sit on Cyberpunk busy street with path tracing right now, which never crosses 71 degrees at 99% usage.

Launching the Superposition Benchmark sees it climb from 30C idling to 73 degrees over about 5 seconds

Hey there, Been happily running a Asus TUF 4070 TI in my system. Today, I started a new game for the first time (the rumoured crappy-but-so-crappy-it's-good "Wanted: Dead"), and something occured that I've never seen before: my GPU, for just a few seconds, instantly shot to 83 degrees, fans on full blast, instantly thermal throttling. It's the first time I've ever seen this behavior, and something that i cannot replicate again on subsequent launches. I thought my card might be broken, so I ran the following tests: OCCT 3D standard test: 78-79 degrees, 30 minutes, no errors Furmark: 75 degrees Superposition benchmarks: 3 or 4 runs, nothing out of the ordinary Cyberpunk Path Tracing: Let it sit for 30 minutes on a busy intersection, 99% GPU usage: 70-72 degrees Alan Wake Path Tracing: Let it sit in the forest section of the game for 15 minutes: 99% GPU usage, 70-72 degrees What can be the cause of this weird spike? Am I being too concerned?

Sorry, I meant to apologize to previous poster!

This was GPU-Z output -> Excel. Thought that would be easier than giving the full timeseries, my apologies. Oh and yes, this is full stock fan curve.

Gotta note that this is a worst case (benchmark/stress test) scenario. Under gaming scenarios (Cyberpunk with RT), the overall temps are 5 degrees lower, deltas the same.

Started worrying about my 4070Ti Asus Tuf OC hotspot temp. Ran a Superposition bench, and plotted the GPU temp, hotspot temp and the delta between them. Does this look normal? Thanks!

So I've got nothing to worry about and this behaviour is normal with my cooling setup?

All stock, no PBO.

Hey, was testing my 5900X under an NH-D15, and with CB R23 or Aida64 I can't push it past 72C on an hour of the hardest multicore tests, which is good. But there is a game (Midnight Suns), which during saving, briefly (2 or 3 seconds) pushes it to 84. So I tested with prime95 AVX (small FFT's) on a few cores (1-3) only and doing this pushes my CPU to 90, pegged. Doing the small FFTs test on a few cores without AVX puts them in the 75C range. Doing the small FFTs test on all cores moves it back down to 70C range. TL;DR my question: can single core heavy AVX loads push the Tctl temp of a 5900X well beyond what a good cooler can handle?

Might be where my problem is ... i went for a HUGE case, and I think my airflow is sub par. Triple Noctua 140mm in the front, 120 mm bequiet in the back. It's the Fractal Design XL with the closed front. I prefer quietness

Hey there, Loading my new GPU at 100% pushes the temps to 79-80 degrees. Is this normal for this card? All reviews mention lower temperatures, but that's probably in open benches. thanks!

Here are some more pics https://imgur.com/a/qlcgMwu

https://imgur.com/a/QAL9u4d Scratched my Tomahawk X570 a bit when I tried to undo a PCIe latch between my beefy Noctua cooler and my GPU with a long screwdriver. PC powers on, works fine, ran through OCCT power test.