-
×InformationNeed Windows 11 help?Check documents on compatibility, FAQs, upgrade information and available fixes.
Windows 11 Support Center. -
-
×InformationNeed Windows 11 help?Check documents on compatibility, FAQs, upgrade information and available fixes.
Windows 11 Support Center. -
- HP Community
- Gaming
- Gaming Desktops
- Re: Every cpu/gpu intensive game crashes after couple of min...

Create an account on the HP Community to personalize your profile and ask a question
02-08-2025 01:42 PM - edited 02-08-2025 03:03 PM
Hello. I got this new HP OMEN by HP GT15-2300ng gaming desktop pc in a german online shop only to discover that every game crashes no matter what I do. The longest play session I got was about 40 minutes and then the game randomly stops working, mostly without error messages. I changed every setting there is to reduce the power usage, but still, in heavy cpu/gpu game moments it seems the pc is giving up completely.
What I tried already:
- Increased the case fan speeds so they cool down the pc significantly with OMEN Gaming Hub (agressive fan curve)
- Overclocked the ram to 5200 with OMEN Gaming Hub
- Changed to XMP 1 profile in BIOS
- Updated to the latest BIOS firmware from 2024
- Updated most of the Hp support drivers that were listed for my pc
- Changed and adjusted settings in the Nvidia control panel
- Changed and adjusted the gpu settings with MSI afterburner (agressive fan curve/no overclocking at all)
- Installed Windows Update 24H2
- Changed to Balanced power saving mode in Windows
- Reduced the processor power condition to 99% under the processor energy management in Windows
- Changed and reduced in-game settings
- Used HWInfo64 in combination with HeavenBenchmark 4.0 for a 10 minute benchmark without over the top thermals
So far, nothing helps. If I wouldn't game, the pc runs fine most of the time. It's just the fact that I bought it to replace my already old pc.
My best guess is that the 600 watt psu is too low for my the actual power usage. I heard about people online suggesting to undervolt the cpu and that would fix many issues. Is that true?
Solved! Go to Solution.
Accepted Solutions
02-21-2025 08:06 AM
02-08-2025 04:03 PM - edited 02-08-2025 04:09 PM
Greetings @Beltano
Excellent troubleshooting.
Heaven Benchmark did not crash. That's not expected from my perspective.
A network problem could be indicated if the PC only crashes when doing online gaming.
Yes, a failing PSU can cause weird stuff.
I don't think your PC's BIOS does voltage adjustments.
You have an Intel CPU. Check the status of the CPU by following steps at this Site.
Check the CPU, RAM, and graphics using OCCT free.
Regards
02-09-2025 02:38 AM - edited 02-09-2025 06:19 AM
Hello Bill_To
Thank you for the reply.
Heaven Benchmark did run without any issues, although it was only the lowest preset. Around 10 minutes later there were some temperatures in Hwinfo64 higher than normal but still under the thermal threshold. I think around 84 degrees Celsius was the highest number somewhere, but no crash.
What I didn't mention is that I use wifi where I live. Since months our router is doing weird things like some devices cannot establish a connection with ipv4. It's been this way wihout any explanations.
Right now I'm using the Dual Force feature in the Omen Gaming Hub, which means ethernet connection from my own repeater plus using a wifi that is more or less stable. I've noticed that the crashes happen randomly, but you are right it's mostly online games. That's one reason.
The custom BIOS from HP is missing every important setting there is. I didn't know beforehand that you basically have no real choice there. Everything else is in the Omen Gaming Hub.
I'm thinking about upgrading to a 850 watt psu just to be safe and switching out all case fans, because they are too loud. Additionally more case fans to cool down everything to the maximum. Maybe even upgrading to a better cpu cooler.
I will check what you did recommend me.
Are there any screenshots that I could provide for more troubleshooting?
Most importantly, do you think that the random crashes can be fixed completely without having to send in the pc and getting a replacement?
02-09-2025 08:03 AM - edited 02-09-2025 08:21 AM
Greetings @Beltano
My pleasure.
I would return the PC to the seller if you are not happy. But this has to be done within the return/exchange time period.
It is best to run the PC in the stock configuration (no overclocking) to check stability.
84 C on the CPU is acceptable, but toasty. 84 C on the graphics card is not good.
Only do Ethernet to check online performance.
Run the tests per my first response to check the PC's components.
Let the Forum know how the tests go.
I would guess you are not happy if you are thinking about component upgrades for a brand new PC.
Regards
02-09-2025 09:14 AM - edited 02-09-2025 09:35 AM
Hello Bill_To
84 C was in the tab named thermal gpu max limit. So far, the Intel PCH (no idea what exactly this is) is sitting at around 55 C in idle, which might indicate that it's too hot in the pc case.
In fact, the stock HP fan settings are only suggesting about 800 rpm, but that's too low and caused my cpu in the first hours to overheat to over 80 C in idle. Also, I noticed the case is lacking in air flow in general, so adding more fans is needed and helping when gaming.
I thought about not buying this pc because of bad air flow, but took the risk. All other pc's within the same price range were sold out and I wanted to be quick. Bad mistake. I should've waited much longer to decide which other pc out there in the online shop is more stable. Overall, I would say that I'm mixed about it. My first pc build I put together years ago went poorly and maybe I damaged some parts early on, so buying a pre-built one was better in my opinion.
Also, I tested the game Grim Dawn today with max settings in 1080p and locked to 177 fps. The gpu was under 60 C all the time, no crash this time. The latest installed Nvidia gpu driver on my pc was unstable for many other people, so I reverted back to the december driver.
Here is the Intel Diagnostic result:
This was a good result I guess
The other test results will come in the next days.
If I really wanted to undervolt the cpu, is there even a possibility doing that outside of BIOS?
02-09-2025 11:36 AM - edited 02-09-2025 03:18 PM
Greetings @Beltano
84 C on the graphics card is not good. Most graphics cards will thermal throttle at about 82 C.
55 C on the PCH is fine.
80 C, under a heavy load, on newer Intel CPUs is okay.
The Intel CPU diagnostic test looks good.
I am not aware of any way to modify Intel CPU settings on locked HP MBs. IXTU used to do this if an Intel "K" CPU was present. But I think doing this is no longer possible.
Plus, you have a locked 14700F CPU.
Try IXTU to see what can be done. I'm pretty sure it's not going to work.
Regards
02-11-2025 08:02 AM - edited 02-11-2025 10:22 AM
Hello Bill_To
Oh, I didn't know that. I'll keep the cpu/gpu temperatures in check while gaming. So far, only during the Benchmark the one number was too high.
As of today the nvidia web helper was the reason that the game No Man's Sky crashed out of nowhere. I checked the event page and it said that this service stopped working. Other critical events were pointing into the direction of a hardware failure. It must be a defect gpu driver installation.
Also Microsoft WMA Host and GraphicsPerfSvc stopped working. The crashes in the last couple of days were eye opening though.
02-11-2025 02:51 PM - edited 02-11-2025 02:53 PM
Hello Bill_To
After hours of playing different games it's safe to say that my pc is stable. Reinstalling the gpu driver was the solution!
In the OCCT app, should I use the standard settings and do all the benchmarks there is? And for how long is it good?
02-11-2025 03:40 PM
Greetings @Beltano
Good to see the PC is stable.
I use OCCT standard settings.
Do the OCCT: CPU and Memory Stability Test.
I usually run this for about 20 minutes.
You can also do the 3D test and VRAM tests.
Or you can do the Combined test to test everything at once.
Regards
02-12-2025 09:34 AM - edited 02-12-2025 09:58 AM
Hello Bill_To
More or less stable right now.
The 20 minute combined OCCT Stability Test went perfect with no errors. Forgot to make a screenshot from the main monitor.
Here are the temperatures right after the Stability Test. Notice the max temps in the cpu tab. These max temps were from the game sessions today while playing Planetside 2. Multiple crashes + bluescreen under heavy load.
Here are the archived windows reports. Notice the critical hardware failure, from LiveKernelEvent Code 141. What is going on?
Planetside 2 is extremely heavy on the cpu, but is too old for newer optimization methods. Under Nvidia control panel I used a combination of Normal & High perfomance mode, single & multithreaded optimization. In-game settings are set to high, but not ultra at all, with 177 fps capped. Maybe this game doesn't like the new pc hardware at all.