The computer turns off when i play halo infinite

[IamPufik]

The computer turns off when i play halo infinite. Many hours of testing in OCCT passes without problems, the situation does not repeat itself in any other game or task.

[MrGergoth]

Попробуй Свойства системы - дополнительно - Загрузка и восстановление - Параметры - Выполнить автоматическую перезагрузку (отключить)
Открой Просмотр журналов событий, попробуй найти ошибку в момент отключения.
Попробуй изменить графический драйвер на более старую версию, удалив его через DDU

[Solid Gamer]

Check your RAM with MemTest (separatly)

[Defective Dopamine Pez Dispenser]

What does Event Viewer say after these crashes in Custom Views > Administrative Events or Windows Logs > Application? Likewise Reliability Monitor? Double click any logs generated when the crashes happened (there will always be some) and see what they say. You'll want to include that in any crash report submitted as well.

In the mean time, since system shutdowns shouldn't be happening when a game crashes, do some troubleshooting to ensure you aren't having some instability somewhere that the game is revealing. OCCT is a decent test suite, but not as intensive for certain issues as I would like for narrowing down problems.

• In Windows Explorer right click each of your drives, select Properties > Tools > Check and run the disk check utility. Do this for all volumes.
  
• Open a cmd prompt (Windows Key, type "cmd" without quotes) as Administrator (right click the command prompt that appeared when you typed cmd, and again select Run As Administrator,) then type or paste sfc /scannow and hit enter. Allow it to finish.
  
• In the same cmd prompt, type or paste DISM /Online /Cleanup-Image /CheckHealth and hit enter. Allow it to finish. If errors that could not be fixed were found, run both of those commands three times sequentially or until no errors are found.
  
• Follow any (or all) of the options here to test your memory stability (easier to paste this than a step by step): https://www.tomshardware.com/how-to/how-to-test-ram
  
• Run an application like Crystal Disk or your storage vendor's proprietary S.M.A.R.T. utility to check your drives' health.
  
• Download CPUID's HWMonitor from a reliable source or from the author, and run HWMonitor. Ensure your voltages, temps, etc. are within known safe tolerances for your hardware components under high load conditions (gaming, benchmarks, etc.) Run a few games or benchmarks that you know have run stably on your system before now and monitor your resources via HWMonitor while they run.
  
• Download the free trial of AIDA64 Extreme from a reliable source or from the author website. Once it's installed, under Tools select System stability test, and under the available options leave everything default but tick the "stress GPU" box. You will get a pop-up asking you for permission to change your TdrDelay most likely. Click Yes. Close everything else you're running other than AIDA64 and HWMonitor, then click "Start" at the bottom of the window. Allow this to run for an hour.
  
• While it's running, use the information graphs provided to monitor your voltages, temperatures, and your fan speeds. Ensure your fan speeds are responding as they should and at their proper RPMs, that your voltages are in the correct ballpark for their respective rails (minor divergence from the listed voltages is okay, but you're looking for overvoltages or big spikes or drops that could contribute to system instability.) On the voltage graph, you want those rails to be straight lines over the course of the hour long stress test. A stable PSU should provide rock solid stable voltages under load. If you do see significant drops or spikes that shouldn't be there, especially if they coincide with system instability, consider changing your PSU.
  
• While running this test, use HWMonitor or another utility to also monitor your GPU temperature independently.
  
• When it's been an hour, click Stop. When you do, make a note of how quickly your temps fall. Your fan RPMs should drop almost immediately, and your temps should as well, for all components. Make a note of any lingering high temperatures, as that should not be the case.
  
• Disclaimer: Although in a thermally and otherwise stable build, you should be perfectly safe throwing this at your components, if you do suffer from a thermal or other instability in your system, the following two stress tests have at least some potential to reveal those issues by damaging your hardware. Proceed at your own risk and only if you are confident you know what you are doing. You should be afforded some protection if you performed the above steps without issue, but there is still the possibility of thermal instability not yet revealed that this test could exacerbate. You have been duly warned.
  
• Download Prime95 and install. Run it, and under Test, select Small FFTs and click OK. Monitor your CPU temperature via HWMonitor. If you aren't sure which sensor is your CPU, don't worry - you'll know shortly, because it's going to get warmer. Assuming you don't see any immediate problems with your temps or voltages, continue to run this test for an hour. After an hour click "Test" in the upper left of the window and click "stop." Click Exit to close Prime95. Small FFTs simulates a very real world like workload but does so in a way that should utilize 100% of your cores and drive them to temps you will almost certainly never see in the real world. If your CPU can handle this within safe temperature tolerances, you can pretty confidently say you do not have thermal issues with your CPU.
  
• Download FurMark from a reliable source or its author. Install it, run it, close everything else you're running, and click GPU Stress Test. (Take the warning message seriously and proceed at your own risk. Again, in a thermally stable GPU with a properly seated HSF and no hardware or other defects, no unstable overclocks, etc. you should be fine. For the record, I have never in all my years using it across myriad builds had a restart, crash, or issue with FurMark. But if your GPU does have thermal or other defects, this test will likely reveal them, and if it does, could potentially cause a crash, or damage your GPU. FurMark won't cause these issues. It will only reveal them.)
  
• Monitor your GPU temperature via the on-screen graph in FurMark. If you see it beginning to approach temps that you know to be unsafe for your particular GPU, hit Esc to end the test and click Quit. However, if temps are safe and stable, allow the test to run for an hour and check back periodically for temp spikes or instabilities.

Assuming all of that checks out, you almost certainly have a rock solid build, and your hardware is likely not the issue. You can run these tests for longer in the off chance a voltage instability is happening only intermittently or what have you (some people run them for 24 hours) but that's not strictly necessary and especially in the case of FurMark is probably unwise in terms of the life span of your part.

Generate a DxDiag report and attach to a crash report here: https://support.halowaypoint.com/hc/en-us/requests/new and be sure to include details of all the steps taken above and a copy of any event log information found in Event Viewer about the crash.

[long chin man]

short answer: power supply issue

[MrGergoth]

Nvidia drivers can cause pc shotdown\reboot, there was reports in New World and BF2042 (driver crash was causing the system to crash).

[Chubbazwyn Griffith]

It's funny how these graphical issues only occur to people playing on toasters and old gear.

Makes ya think.

[FeilDOW]

Full specs with make and model of PSU would sure help.

[n#n#m#]

My PC was tripping and randomly entered hibernation state while I was sitting in Infinite's main menu. Updated drivers for my GPU and everything is ok ever since.