Most games push GPU usage to 100% and crash.

[mRecords]

First of all, english is not my native language, so I apologize if I explain something innaccurately.

I bought a new GPU (specs at the end) and now most games crash somewhere between a couple seconds and a few minutes. They either freeze or just push me back to desktop.
What I did so far:

- I checked the GPU is plugged into the correct socket (by my tech guy). It is.
- I checked temperature with two games opened simultaneously at ultra settings (before they crash). Temp is no higher than 70°C.
- I checked drivers. All up to date. I tried older drivers. Nothing changed.
- I checked BIOS was up to date. It wasn't. Now it is. Nothing changed.
- I checked the event viewer (I hope I translated that correctly) from Windows. It doesn't register any entry.
- Checked that power is enough to feed the GPU. It should be.
- I had W11. Did a full clean up and started from scratch with W10. Still the same.
- Config the GPU from the NVIDIA panel to max performance. Nothing.
- Set the games to the lowest possible settings. They still push the GPU to 100% even on the menu.
- Tried underclocking.

Now what's odd is that it happens with MOST games, but not all. State of Decay 2, DOOM Eternal, both Ori games, Back 4 Blood (and I can keep on going) crash but it doesn't happen with League of Legends (although I suspect that has to do with the fact that it has very low sytem reqs) and, here's the weird thing for me, Grounded to its max settings which stays between 14-27% GPU usage.

Any help would be appreciated.

Specs (everything but the GPU has less than a year old):

Thermaltake Smart 500W power supply.
120GB SSD + 1TB hard drive.
2x 8GB RAM
Gigabyte B450M DS3H V2 Motherboard.
AMD Ryzen 5 5600 processor
NVIDIA gtx 1660 super ventus GPU.

[mRecords]

Update: Grounded crashes. It just takes longer.

[cSg|mc-Hotsauce]

Only Steam games crash?

[Rumpelcrutchskin]

Most likely PSU. Cheap Thermaltake PSUs are rubbish quality and 500W is kinda borderline for this build, especially when PSU is rubbish and wont put out the advertised power.

[_I_]

Originally posted by mRecords:

Thermaltake Smart 500W power supply

^ this is the problem

replace it asap before it tries to keep running the system when its drawing more power than it can safely deliver within spec
thats what causes other hardware to fail

[Bad 💀 Motha]

Thermaltake JUNK 500W is your problem. Say no more... toss it.
Do not even power that PC on one more time; order a decent quality PSU before it up and kills your stuff.

Look at BeQuiet 600W Gold Certified or better.

While you might not need like 650-850 Watts now, you might if you ever get a better GPU.

[kiwikev]

From memory the DS3H had a poor/weak vrm setup as well for both b450 and b550 versions.

Check what the vrm temperature is too

And as suggested above get a better branded PSU

[_I_]

its solid meh
4+2+1
4 with cooling can give the cpu cores 100w
+1+2 to the igpu and imc, they dont need alot, since the cpu has no igpu, those are basically unused

it is weak, boards do not tell you the mosfet temps
only way to check that is if its throttling before the cpu is above 90c

but the problem is the thermaltake psu
the 'smart' lineup is complete trash

[Bad 💀 Motha]

Reset Cmos does NOT ensure there is no OC. And the defaults on many of these Motherboards is utter rubbish. You need to double check that the board maker does not have the default as having some branded CPU OC. As this is quote common nowa and for the last couple generations of Motherboards.

You're also using would can easily be considered a terrible MB for a good CPU that most likely in order to run correctly and stable, may require a better board.

Folks with B350, B450 and any A series Motherboard really have no business installing CPUs above Ryzen 7 3xxx series. You may end up in a scenario where you can't unleash the CPU fully, or you actually prematurely kill the Ryzen 5xxx series CPU because despite BIOS update, such cheap boards "mostly" can't handle these properly in real world.

X370, X470, B550, X570 boards all should be fine however for Ryzen 5xxx series CPUs

[Bad 💀 Motha]

And how is the Monitor connected?

I've seen plenty of instances of faulty HDMI cable where the signal was lost and then restored and this causing the screen res to change as a result.

[GOD RAYS ON ULTRA™]

Those PSUs don't offer 500 watts to the 12 volt rail. You're only getting 450. That's 37.5 amps, really weak. On top of that they have poor protection. These smart series can only protect your PC one time and that's it. The next time your PC will die do to the poor design.

Originally posted by _I_:

Originally posted by mRecords:

Thermaltake Smart 500W power supply

^ this is the problem

replace it asap before it tries to keep running the system when its drawing more power than it can safely deliver within spec
thats what causes other hardware to fail

I would listen to this comment ^^

[Bad 💀 Motha]

If the system NEEDS 350-450 watts then you want at least a Bronze 600-650 watt PSU

[mRecords]

Originally posted by cSg|mc-Hotsauce:

Only Steam games crash?

:winterbunny2023:

Nope.

Originally posted by kitty:

What is the resolution of your monitor? That is important and you did not tell us.

1366x768.

Originally posted by Bad 💀 Motha:

And how is the Monitor connected?

VGA/HDMI

Ok, I'm just trying to understand here why would the PSU be the problem. I understand what you guys say, what I don't get is why this is happening now. I've been using it for a year with an rx 6600 xt. According to the manufacturer's web sites, the rx 6600 asks for 500W and the gtx 1660 asks for 450W.
I get that my PSU is garbage. I don't get why it's a problem the exact same day I put a new GPU that, theoretically, needs less power.

[Bad 💀 Motha]

So VGA to HDMI?

Are you bloody serious

[mRecords]

Originally posted by Bad 💀 Motha:

So VGA to HDMI?

Are you bloody serious

I never joke. Well I do, sometimes. I don't see the point of your comment. You have something to add?

[Tonepoet]

The Thermaltake Smart is a 500 watt 80+ certified P.S.U., and the 1650[www.nvidia.com] isn't that demanding. Only seems to want a minimum system power of 300 watts, and it also looks like it's a 75 watt card (despite having a 6 pin connector). Now granted maybe that figure increases pending on the processor, but the Ryzen 5500 is a fairly average processor in terms of T.D.P., so I kind of doubt that's the problem.

I'm thinking there might be a driver conflict. If I recall correctly it's usually recommended to uninstall old G.P.U. drivers before installing the new ones, and you haven't exactly checked that.

I don't know how to uninstall G.P.U. drivers, but you can probably test this hypothesis pretty easily by making a linux thumb drive and booting off of that to see if it still crashes.

If it's a hardware problem as with the P.S.U., then I imagine the card should crash regardless since it shouldn't have enough power draw for it to operate under load either way. If it's something software based as with the drivers, then it likely shouldn't be exhibited on a fresh O.S. installation.

I'd tell you to install Windows on an external drive to eliminate other variables, but unfortunately Windows doesn't let you install it on an external drive. I also don't think you want to wipe data on your internal drive just for trouble shooting purposes.

You can make an external bootable linux thumb drive pretty easily using unetbootin[unetbootin.github.io].