13700k crashing most often comes from unstable bios power rules, ram xmp settings, old microcode, or heat, and a few targeted checks can stop it.
If your pc reboots mid-game, drops to desktop, or blue-screens with a WHEA error, it feels random. It rarely is. Most crashes leave a trail: a timestamp, a thermal spike, a voltage jump, or a repeatable trigger like one specific game scene.
This guide walks through the fixes that pay off first, then moves into bios, memory, and hardware checks that catch the stubborn cases. You’ll end with a stable baseline you can trust, plus a clean way to prove what changed.
What These Crashes Look Like In Real Use
Crashes tend to fall into a few buckets. Naming the pattern helps you pick the right test instead of changing ten settings at once.
- Spot the failure type — Note if it is a reboot, a freeze that needs a power button hold, an app crash, or a bsod with a stop code.
- Log the trigger — Write down what was running, the last thing you clicked, and whether it happens under load, at idle, or during sleep and wake.
- Save the timestamp — Match the crash time to Windows logs so you can see whether it is cpu, memory, gpu, storage, or a driver chain.
Event viewer clues that point to the next step
Open Event Viewer, then check Windows Logs and System. Look for WHEA-Logger entries around the crash. A WHEA_UNCORRECTABLE_ERROR (bug check 0x124) signals a hardware error captured by Windows’ WHEA layer, so it often points at instability in cpu, memory, pci express, or power delivery.
If you see repeated WHEA-Logger warnings without a full bsod, treat them as early warnings. They can show up days before a hard crash.
Quick pattern map for common symptoms
| What you see | What it often points to | First thing to try |
|---|---|---|
| bsod 0x124 or WHEA-Logger 18 | cpu core stability or voltage limits | set bios to intel default settings |
| game closes to desktop, no bsod | ram xmp edge case or gpu driver | turn off xmp, then retest |
| freeze with audio loop | memory, imc, or power spikes | reduce ram speed one step |
| crash after sleep or wake | bios firmware, usb, or driver state | update bios and chipset |
Fast Windows Checks Before You Touch The BIOS
Start in Windows so you do not lose your current settings. These checks are quick, low risk, and often reveal the real culprit.
- Check temperatures — Use a hardware monitor and watch cpu package temp, cpu power, and clocks during a run that normally crashes.
- Update chipset drivers — Install the latest motherboard chipset package, then reboot before testing again.
- Update graphics drivers — Do a clean install if you see game-only crashes or driver resets in the logs.
- Verify system files — Run sfc /scannow and a disk check if crashes started after a rough shutdown.
Read the temps like a detective
Intel lists a Processor Base Power of 125 W and a Maximum Turbo Power of 253 W for the i7-13700K, so heavy loads can push cooling hard when the board lets the chip boost freely.
If temps pin near the thermal limit during a crash, treat cooling as a real cause, not a side note. Thermal throttling is not the same as a crash, yet sustained heat can make borderline voltage settings fall over.
One clean stability pass in Windows
Pick one repeatable test and use it every time you change a setting. A short Cinebench loop, a ten-minute OCCT cpu test, or the game scene that triggers the crash all work. The rule is simple: same test, same duration, same logging each time.
Turn off Fast Startup while testing. It can blur cold boot results by mixing shutdown and hibernate states. During tests, disable any cpu overclock profiles in vendor apps, since they can reapply settings after a reboot without you noticing.
13700K Crashing Fix Checklist In BIOS
Motherboards often ship with “auto” rules that chase speed, not stability. For Intel 13th and 14th gen desktop chips, Intel has described instability tied to sustained high voltage and temperature, and board vendors have rolled out bios updates with new microcode and default profiles meant to reduce that risk.
- Update the bios — Install the newest stable release from your board vendor, then load defaults once after the flash.
- Use intel default settings — Select the vendor option that applies Intel Default Settings, then save and retest before changing anything else.
- Set sane power limits — If your bios offers manual PL1 and PL2, start with Intel-style limits for your chip and cooler.
- Turn off multi core enhancement — Disable vendor boost extras that raise voltage and current beyond Intel’s target rules.
Where to find the right bios profile
Different brands name it differently. MSI has documented “Intel Default Settings” in bios updates meant to improve stability. ASUS has published guidance tied to microcode updates such as 0x129 and adjusted factory defaults. Some Gigabyte boards show an “Intel Baseline” option. Intel has said the preferred fix is the Intel Default Settings style profile, not a baseline profile used for other tuning goals.
Power limits that stop surprise reboots
If your board has been running unlimited turbo power, your cpu may sit at high wattage for long stretches. Setting limits can calm voltage spikes and keep temps in check. Start with a conservative setup, test stability, then decide if you want to raise limits later.
- Set PL1 and PL2 — Try 125 W for PL1 and 253 W for PL2 as a baseline, then test your worst workload.
- Set turbo time — Use the board default or a moderate window so the chip can boost, then return to PL1.
- Set load line calibration — Pick a middle level that avoids big droop and avoids big overshoot.
Offset undervolts and auto voltage rules
If you’ve been tuning, undo it for your baseline. A negative core voltage offset can pass light loads and then fail under heavy boosts.
- Clear manual offsets — Set core voltage mode back to auto, remove negative offsets, and retest your crash trigger.
- Keep current limits sane — If you raised ICCmax or similar limits, return them to default values in the intel profile.
- Leave CEP on — If your board has current excursion protection, keep it enabled while you chase stability.
Memory And XMP Tuning That Stops Silent Errors
Ram instability can look like a cpu problem, since the cpu’s memory controller is the one talking to the sticks. XMP can be stable on one board and flaky on another, even with the same kit.
If you suspect ram, aim for a clean A/B test. Turn off XMP, boot at stock JEDEC, then rerun your crash trigger. If the crashes vanish, you found a path forward.
- Disable XMP — Boot once at stock, then test the same workload that used to crash.
- Lower one step — If you want XMP speed, drop memory frequency one notch and keep timings on auto.
- Test with memtest — Run a full memory test pass, not just a quick scan, then retest in Windows.
- Watch system agent voltage — Keep SA and IMC related voltages on auto unless you know your board’s safe ranges.
Signs your crash is ram-leaning
Look for crashes that happen faster with more apps open, crashes that shift when you change ram slots, or WHEA warnings tied to memory or pci express. A game that runs for hours one day and dies in minutes the next can also point at silent memory errors stacking up.
Stable first, fast second
Once stock is stable, climb back toward your target speed in small steps. Change one thing at a time, then run the same repeatable test. If you change speed, keep timings on auto at first. If you change timings, keep speed fixed.
Cooling, Power, And Mounting Checks That Matter
When a 13700k sits near its thermal limit, tiny voltage shifts can push it over the edge. Cooling issues also hide in plain sight: a pump that ramps late, a fan curve that is too quiet, or a cooler mount that is slightly uneven.
- Reseat the cooler — Clean off old paste, apply fresh paste, then tighten evenly in a cross pattern.
- Check pump and fan curves — Make sure the pump runs at a steady high duty and fans ramp before temps spike.
- Verify cpu contact — Check for bent pins, socket debris, and uneven pressure that can cause hot spots.
- Inspect psu cabling — Reseat the 24-pin and cpu EPS cables, and avoid loose adapters on high draw systems.
Why power delivery can look like a cpu fault
A weak or aging psu can sag under transient load. The pc may reboot with no bsod and no neat error message. If crashes show up right as a game loads a new scene, or right as a stress test starts, power transients are on the suspect list.
Prove The Fix And Keep It Stable Over Time
After you make a change, prove it with a plan that hits cpu, memory, and real games. A single pass is not enough when you are chasing random crashes.
- Run a cpu stress — Do a sustained cpu test that heats the chip and holds clocks steady.
- Run a memory test — Use a full pass, then follow with a mixed cpu plus ram load in Windows.
- Play your crash game — Run the same map, same settings, same time window that used to fail.
- Check event viewer again — Confirm WHEA warnings are gone, not just hidden.
- Save a bios profile — Store the stable config in a named bios slot so you can return to it after updates.
When it is time to swap parts
If crashes continue at stock settings, with Intel Default Settings applied, with XMP off, and with temps under control, a hardware fault becomes more likely. At that point, isolate by swapping one part: ram, psu, then gpu. If you can reproduce crashes on another board, that points toward the cpu.
Intel has also described a Vmin shift instability issue tied to reliability aging under elevated voltage and temperature on some 13th and 14th gen desktop parts. If your chip has been run hard for long periods on aggressive auto rules, and it no longer holds stability even at sane defaults, an RMA request may be the cleanest path.
Once you land on a stable setup, resist the urge to stack tweaks. If you want to tune again, change one setting, retest, then move on. That rhythm keeps your pc fast and predictable, and it keeps 13700k crashing from creeping back.