Hey, I have a RBA from 2019. I’ve run it as a desktop for the past 3 ish years. It does NOT have a battery. When I’m not using it I leave it off and unplugged.
I use it for Photoshop and intermittent gaming. It’s about 5 years old and within those 5 years I rarely saw its GPU temp go over 74C. In fact it was running fine and two hours later (off and unplugged the whole time) I booted it up and launched Tekken 8, its GPU temp hit 95 within seconds, as soon as the opening cinematic started- and I assume it throttled to 10FPS in the menu. The GPU was only under ~30% load, drawing maybe 13-15 watts. The CPU temps were fine, as expected and consistent with those just a few hours ago.
- I immediately opened it to check for obstructions in the vents, everything was clean, I’d cleaned it a few months ago. I couldn’t see any damage anywhere.
- The next day, I replaced the thermal compound on both dies with Arctic MX-6. Dies and chips looked fine visually once cleaned. Nothing changed about the GPU temps, still 90C+ under ANY load. CPU was still normal and not overheating at all, in fact it seemed a bit cooler, I think because of the new compound.
- I ran a few antivirus scans, which turned up nothing.
- I tried to update the bios from the Razer site, said it was up to date, the GPU drivers and Windows were up to date.
- I reinstalled Windows later that day. I am still on Win 10. Again, the CPU temp was normal, and it was under 100% usage at times when I was setting the system up again.
- I tried launching DMCV after all of this: the GPU hit 90C+ within seconds even under 10% load, I killed DMCV, and to keep the GPU reporting I opened the Nvidia preview render in the Nvidia control panel, but doing this was caused it to report 99C and shut down due to the extreme heat?? In an attempt to get its fans back on and cooling I restarted the system. It reported 99C three more times on startup and shut down each time due to heat. To get the fans on, I’d start it again each time. It did not shut down again once it reported 92C but I was too afraid to test the GPU again.
Two things I find odd though:
- The sensors are saying 86-90 but the metal chassis is MUCH cooler than what I’m used to it feeling like when it said it was running cooler. (When it was around 74C, the chassis was VERY hot to the touch to me).
- While the fans sound normal as in they sound smooth and even with no new grinding or oscillating noises, they are spitting max speed at 90C and considerably slow down when it hits 70C. The other day and for the past 5 years, they would still be at a high speed until it was cooler than that, maybe 65-58C. It basically 'acts' like 70C~ is a nice cool temp now.
Also:
- There are absolutely no visual artifacts, glitches, or crashes other than those when the GPU overheated to 99C. I've never heard of a GPU dying this way, and all of this came on from one boot to the next on the same day.
I’ve disabled the GPU
Anyone know what could be causing this behavior? I'm at a loss as to why this might be happening, I’d really love to fix this thing if I can.