Blue Screen Only While Gaming Check Heat VRAM and the Graphics Driver
A PC that is stable all day but blue-screens twenty minutes into a game is telling you something precise: the failure needs load, heat, or full VRAM to appear. This guide separates those three suspects with tests, not guesses.
Watch the full method in the video - identifying a blue screen cause, step-by-step
Gaming-only crashes are actually good news diagnostically - the trigger is known. Under game load, three things change versus desktop use: the GPU gets hot, its VRAM fills up, and the graphics driver works at full stretch. A fourth guest sometimes joins: the power supply struggling at peak draw. Work through them in the order below - temperature first because it is free to check, then the driver because it is the most common, then VRAM, then power. And keep the honesty rule in mind: one crash proves little; a repeatable trigger plus a dump pattern is evidence.
GPU monitoring during a game (example values)
GPU Temp93 C
GPU Load99 %
VRAM Used7.0 / 8 GB
Example readings only - 93 C under full load is too hot for most cards and points at cooling first.
Example monitoring values. Watch your own numbers while the game runs - the pattern at the moment of the crash is the clue.
1
Check Temperatures First - It Costs Nothing
Heat failures grow with session length: crash at 30 minutes, then 20, then 10 as the case warms up.
Install a monitor like HWMonitor or use the GPU maker's overlay
Play the crashing game with temperatures visible
GPU above roughly 85-90 C, or CPU near 100 C, at crash time means cooling is suspect number one
Open the case: dust-clogged fans and heatsinks are the usual story, especially on older PCs and laptops
Clean dust, confirm every fan spins, and retest before buying anything
2
Read the Dump - Gaming Crashes Usually Name the Driver
Open the newest file from C:\Windows\Minidump in BlueScreenView after a game crash.
Names like nvlddmkm.sys (NVIDIA) or amdkmdag.sys / atikmpag.sys (AMD) point straight at the graphics driver
A stop code of VIDEO_TDR_FAILURE (0x116) means Windows tried to reset the GPU and it did not answer - driver, heat, or a struggling card
If the dump instead blames random different files each time, skip ahead - that is the hardware pattern, covered below
3
Fix the Graphics Driver Properly - Clean, Not Layered
Years of driver-on-driver installs corrupt easily. Do one clean pass.
Download the current driver from NVIDIA or AMD directly - not a driver site
Boot into Safe Mode and run Display Driver Uninstaller (DDU) to remove the old driver fully
Restart and install the fresh driver
If the crashes began right after the newest driver release, install the previous version instead - newest is not always stable
Retest with the same game and the same settings
4
Test VRAM and Power - the Two Silent Suspects
If heat and driver pass, the remaining gaming-only causes are the card's memory and the PSU.
VRAM: lower texture quality in the game (which cuts VRAM use) and play - if crashes vanish, run a dedicated VRAM stress test like the one in OCCT; errors mean the card's memory is failing
Underclock test: reduce GPU core/memory clocks slightly with the maker's tool - a card that becomes stable underclocked is a card that is ageing
PSU: crashes at the exact moment of peak action, sometimes a hard restart with no dump file at all, point to power delivery - test with a known-good PSU if you can borrow one
Be honest about the outcome: a GPU that fails VRAM tests at stock clocks is a replacement conversation, not a settings conversation
Reinstalling the game or Windows almost never fixes gaming-only blue screens - the game is just the load generator. The fault is in heat, driver, VRAM, or power; the game merely reveals it.
Watch the Complete Video Guide
The full blue screen identification method - dump files, stop codes and the fix - shown live, step-by-step.
The PC restarts during games with no blue screen and no dump file. Why?
A missing dump after a sudden restart usually means the crash never reached Windows - classic signs of power supply overload, severe overheating protection, or a motherboard-level fault. Check temperatures and suspect the PSU before software.
Will capping FPS help diagnose it?
Yes, it is a useful test. Capping FPS or enabling V-Sync lowers GPU load, heat and power draw. If crashes stop with a cap, your problem lives in the load family - heat, power, or an ageing card - not in the game files.
My GPU temperature is 82 C in games. Is that the problem?
82 C is warm but normal for many cards under full load. Heat becomes the prime suspect nearer 90 C and above, or when crashes clearly get sooner as the session warms up. Numbers plus pattern together - neither alone.
This guide is based on the Technical Anshu blue screen video - stuck on any step, ask in the video comments and you will get an answer.