The new PC freezes immediately during startup for demanding games.
The new PC freezes immediately during startup for demanding games.
Your computer appears to experience instability during certain demanding games. Recent crashes have occurred while trying to play Fortnite and launching the MW2 Open Beta. When you observe the crash, internal lights remain on briefly before you're asked to restart. Reinstalling Windows hasn't resolved the problem. Error messages appear in the Event Viewer, but this doesn’t seem to assist. Diagnostic tests on components show no overheating issues—CPU ran normally, GPU handled tests well, memory passed checks. Here are your system details:
PC Parts List: [link]
CPU: AMD Ryzen 7 5800X 3.8 GHz 8-core
Cooler: Cooler Master MASTERLIQUID ML240L RGB V2
Motherboard: Gigabyte X570 AORUS ELITE ATX AM4
RAM: Corsair Vengeance RGB Pro 16 GB (2 x 8 GB) DDR4-3200 CL16
Storage: Silicon Power A60 1 TB M.2-2280 NVME
HDD: Seagate Barracuda Compute 2 TB 3.5" 7200RPM
GPU: Asus GeForce RTX 3070 Ti 8 GB TUF GAMING
Case: be quiet! Pure Base 500DX ATX Mid Tower
Power Supply: EVGA SuperNOVA GA 750 W (80+ Gold)
If you have more details, feel free to share. This situation is hindering your ability to use the PC effectively.
Good setup. I aim to create something similar soon. Still unsure about choosing Ryzen 3, 4 or 12th gen; I’m not optimistic for 13th. I don’t use many eco-friendly options. The first steps with everything electric are checking if it’s powered on and then verifying the power cycle. In your case, the video card power cable is what matters. Make sure drivers and bios are updated. There’s an older test that spots minor RAM issues missed by memtest86. Run it on a single memory stick for a short time and see if it still works; repeat with the other. If one functions while the other doesn’t, the faulty stick likely has a problem—possibly in the motherboard, which might fix itself with a bit of cleaning inside the slots. If both fail, it’s less likely to be RAM, but it could point to a heat issue. Then check the PSU and finally the graphics card. Edited September 25, 2022 by Bombastinator
Thanks for your input! I'll address your questions directly. I've been using this PC for roughly a month and a half, and it initially functioned well. Recently, crashes have started occurring within the last week. I checked that all cables were securely connected and confirmed my drivers were up to date. I also performed memtest86 on one drive at a time; if needed, I might run another test. Before reinstalling, I cleaned the memory modules. During stress tests, neither the CPU nor GPU hit 80°C. You might want to explore more demanding benchmarks like Cinebench or Unigine Heaven. These are widely used and reliable. I haven’t seen them perform exceptionally well recently, though they worked fine a few weeks back without any issues. I also attempted to cap GPU power via MSI Afterburner, but the crashes persisted. The most consistent way to trigger a crash is by launching the MW2 Beta—when I double-clicked it, my screen went black, all peripherals shut down, and I was prompted to log back in. Interestingly, the internal lights stayed on during that process. I’ve tried running games at minimal settings, but crashes still occur. The only game I can reliably play is Valorant, yet I’m unsure why it works while others fail.
The issue with the older tests is quite frustrating. One attempt might take weeks or even minutes for a single stick, based on the symptoms. That’s why tools like memtest86 were created. They’re built on the idea that two memory sticks rarely both get memory issues. It’s a bit of a trial-and-error test, though—it can spot very minor problems.
When it comes to bench tests, you never really know if they’ll work for your specific case, but they do generate heat. If that’s the issue, it could point to something else besides the CPU. The ones I’m familiar with are Prime95 and FurMark. They’re basic power tools that crank up the temperature to test performance.
If the problem lies in a component other than the CPU, these can help identify overheating elsewhere. Personally, I use a powerful 4300 cfm fan to improve airflow. I remove the side panel, direct the fan’s exhaust, and check if the issue resolves. That usually reveals whether it’s a heat-related problem.
Regarding power supplies, I understand the concern with new CPUs and GPUs from brands like Ampere, AMD Zen 3, RX6xxx, and 12th gen. They sometimes face transient spikes that trigger safety measures. There’s an updated ATX standard called atx3.0 that addresses this. For older PSUs, a solid 100W rating is usually enough, but a 650W minimum is recommended for better performance.
Also, larger PSUs often come with more power cables, which can be useful in certain setups. A card with multiple ports means both cables can reach it directly.
The unusual part for ME is being asked to log back in. Did your whole system restart completely or were you simply removed from Windows? As others recommended, try warming up your parts (I’d check your CPU) and observe if that triggers problems. If not, I’d focus on checking updates. *PS: Beta games aren’t ideal for diagnosing issues. I recall being permitted to test EVE online beta, which would cause my GPU to behave erratically and potentially crash my PC. In that case, it was the software under development that caused the trouble.*
That idea about beta games is pretty accurate. They often ask for reports on oddities like floating rocks or terrain cracks, but those aren't the only issues. I used "login" to mean a complete system restart. This game seems to have memory problems or frequent updates across it. If a BSOD appeared, it would make things worse for me. Sometimes crashes happen so quickly that the machine can't recover in time to show a crash screen.
I only changed my BIOS and it seemed to work perfectly. No issues reported! Thanks for all the ideas!