The games
The last two days the game began to crash to desktop without any errors. In Windows journal, it also had no records. Protest the CPU, the graphics card and RAM (Linpack, OCCT, furmark, MemTest) no problems were found.Windows 10 Pro 1809 with the latest updates. Config of a computer in profile. Games that were tested: Far cry new dawn, Battlefield 4, and 5. What could be the problem?
Ar4er2016
Can the Internet something picked up, try to fix via a restore point.
Programs dxdiag and msinfo create reports and see if there's error messages. There should be a separate section of the report with the errors fixed Windows. The error code you can try to Google something. If no errors there, then you have to remember not did the tweaking Windows did not turn off all the excess. If disabled, you have to turn it back on.
About picked up something on the Internet... cannot be excluded, but for these purposes, there are boot disk images with antiviruses. You can swing, to boot from it to check.
vitalik76
Reset the system just in case to rule out the software part, the problem remained. However now the game sometimes go to a black screen without shutting down, but often also just closed as before. Like signs care card in defense. But overheating seems not. Raised the maximum speed on the cooler, but it didn't help. While playing Battlefield 5, the maximum temperature was 72 degrees. Mostly less than 70.
Ar4er2016
Maybe not enough video memory, RAM, die die, how they are loaded?
Memory on graphics card 8gb, RAM 12 GB at the same running battle 5, the memory remains with the stock. The RAM ran Memtest, no errors.
Ar4er2016
On overah recently there was a case that because of the heat the user overheated the graphics card and then in games he started regular flights (though it and on the desktop have become artifacts will occur). And since he's due to overheating corny degraded chip to correct anything already it is impossible but in rare cases, helps as you can greatly reduce frequency of the chip.
Try to reduce the frequency of the chip as possible and check whether the crashes, if you stop the chip degraded but not critical.
Ar4er2016
Look, it's summer. Possible temperature VRM (MOSFETs) high. We need to see the condition of the thermal pads on the VRM + the presence of the dust tube on the radiator of the graphics card. If you have VRM temperature sensors, as some of the Radika, you do not need to climb.
1. The experience similar to the instability of RAM. Memory can the tests pass, and the game crachit! On overah full of assertions of this thesis.
2. Video card.
You can check putting the RAM with the percent in the runoff. It will be clear.
If glitches remain, significantly lower frequencies of videocards. If it helps, then gradually test VC after service.
But if it is in warranty and 100% the reason in it, can best be attributed to STS.
It is possible that the same card could come from the factory with non-standard frequencies. You can try to reduce them a bit. If the bug appears consistently, it is better to repair.
Ryazancev
Now run Furmark for half an hour, the temperature is above 80 does not rise, judging by the VRM sensor in AIDA warmed to 78 degrees. No crashes, no black screens. So the graphics card can be deleted or not?
dilat
The power supply is new, you can tell Deepcool dq750st, changed last fall. Just for the reason that with the old unit the computer went to reboot in games and synthetic benchmarks.
Ar4er2016
Throw overclocking in the beginning with percent and memory. On the motherboard can overheat VRM. Maybe memory is not enough pitulova.
If the bug will remain, lower the frequency of the Graphics card.
look SMART drives.
Maybe even BP gives the drawdown at a continuous load on the 12 Volt line.
Maybe a New version of Windows Vyezhivaetsya. Can't hurt to check a known-working OS.
By process of elimination, step by step, it is possible to find the cause. That's just going to take some amount of time on a consistent diagnosis.
Sometimes even a PSU with platinum certificates is buggy, but rarely.
In short, be patient or get a competent friend (can we call him to drink the beer to the guests). In an extreme case, bring it to STS yet the situation has not worsened.
What graphics АИДЫ64 and Afterburner seem, in the time of departure?
Ryazancev
The lower the frequency, the RAM seems to have helped, at least during the 40 minutes played in the battle 4 and 5 without crashes. Will have other games to check.
Ryazancev
Now after all the tests it's safe to say that the problem was due to overheating RAM. Are two sets from different manufacturers, one with radiators, the other without them. Most likely the problem is in them, but too lazy to figure it out. Thanks for the advice. Honestly RAM is not thought.