I5-8600k Overclock Opinions
I5-8600k Overclock Opinions
I successfully ran my 8600k at 5ghz with 1.33v using Prim95 26.6 for 10 to 15 minutes without any issues, maintaining performance in the mid to high 60s. Some believe it requires a full day of testing before stability is confirmed—do I really need to push the CPU that intensely for such extended periods?
For future reference, achieving 5.1 at 1.36v and 5.2 at 1.38/1.39v seems possible. Should I aim for those settings, or is 5ghz sufficient? My gaming setup includes 1440p on a 144hz display, primarily playing League, StarCraft 2, Gears 4, Civ 6, and SimCity.
The temperatures and voltages are quite satisfactory, and you might be able to push further if you really tried. Regarding the 24-hour test, it's not necessary—your approach is adequate. For a while, I would just monitor the temperatures and record any BSODs or errors that occur.
Well, if you're checking the CPU stability, an overnight test is a wise move. You'll be fortunate to reach 5.2 at those levels. For each rise in clock speed or multiplier, you'll need a larger voltage increase, so at 5.2 you might exceed 1.4 volts. Stick with 5GHz. That's the best option. Beyond that, the benefits of higher MHz are minimal and not worth the effort.
The conditions are quite favorable, and you might be able to push further if you really tried. For the 24-hour test, it's not necessary—your current approach is enough. For a while I would just monitor temperatures and record any BSODs or errors if they occur.
Those temperatures and voltages are quite favorable, and you might push further if you really wished. For the 24-hour test, it's not essential; what you did is enough. For a while I would just monitor the temperatures and record any BSODs or errors if they occur. Why do we need a longer test? The reasoning behind that? Testing for 10-15 minutes gives confidence your CPU stays stable? You could extend the duration a bit more for temperatures or stability, but I’d suggest no more than an hour. Those stress tests impose an unrealistic strain on the CPU, something you rarely encounter in gaming. From my previous tests with Prime95 26.6 Small FFTs, the temperatures at 10 minutes were almost the same as those after an hour, indicating it’s unnecessary to go further. Also, Prime95 focuses more on temperature than stability, whereas Asus Realbench would be better for stability checks, though I don’t use it much so I’m unsure how long you’d run it with.
I would agree that the load isn't typical gaming or everyday usage, but focusing on special scenarios like Prime/IBT or similar apps is important to achieve maximum voltage and temperature. For me, 10 minutes or an hour isn't enough. My personal preference is stability at 99.9%. In most situations, temperatures should stabilize after about 10 minutes, though they might rise slowly initially. At 10 minutes it could reach around 75°C (just for discussion), but it would likely increase gradually over the next hour and possibly reach 76/77/78°C. That's my experience with Prime. I wouldn't run it continuously for 24 hours; I usually do it overnight, from around 11 pm to 7 am. If everything works well and feels stable, I'm satisfied. I don't use Realbench, never have used it. Prime fits my needs perfectly.
Sure, I’ll run it for a longer time, but not for 24 hours. Thanks for the feedback.
Probably you’re all on the same page, just like me.
You’re checking for stability while you go, which is smart.
Keep in mind that at 5ghz the chip is a great gaming processor—it’ll boost FPS noticeably.
The main difference between 5ghz and 5.1/5.2 might not be much beyond just a slight FPS increase from the GPU. At that speed, pushing the CPU further isn’t necessary. It won’t help much.
Unless you’re aiming for synthetic benchmarks or just a few extra frames, increasing clocks or voltage won’t give you a big gain. You’ll see no improvement.
Regardless of what game you play, and even if you upgrade your monitor or GPU, a 5ghz CPU is more than enough to handle any high-end card, including a 1080ti or Titan XP, plus any SLI configuration.
Enjoy your gaming, and let me know how it goes with those impressive FPS numbers!
I tested Prime95 26.6 last night for roughly an hour, experiencing a BSOD with the system running at 5ghz and 1.33v. I increased the voltage to 1.34v, which caused temperatures to rise from about 67° to around 70° ± 2° depending on the core, with only one core reaching 74°. It was usually near 72°.
In my second attempt, the PC shut down at the one-hour mark, but then went to sleep instead of crashing. After noticing the issue, I ran it for another three hours without problems, keeping the temperatures stable.
These readings are still within acceptable ranges? Should I continue testing up to six hours at 1.34v, or is the three-hour session sufficient based on how it performed at 1.33v before the crash? Thanks for supporting further stability checks regardless of time.
Absolutely, those temperatures are acceptable. I'm comfortable running my CPU at 80°C. However, the Prime Stressing temperature is normal for regular gaming sessions—you shouldn't exceed around 65°C unless the game load is high. Three hours is sufficient. It really depends on personal preference. Some players prefer higher speeds and avoid pushing the CPU beyond 80°C to maintain stability, while others prioritize maximum performance regardless of voltage, accepting the risk of crashes or overheating. It's all about balancing speed with reliability. Different situations call for different approaches.