The processor continues to reduce its speed once the system is turned off.
The processor continues to reduce its speed once the system is turned off.
Hello everyone, I'm facing an issue with my overclock setup. Whenever I power off the PC, the processor's clock speed drops to 4.1Ghz. It was previously stable at 4.4Ghz in the BIOS, but after shutting down, it decreases again until I restart the system and then increases back up without any changes. This only occurs during a full shutdown, not just a restart. Any advice or explanation would be greatly appreciated as this is quite frustrating. Thanks.
Actually, depending on the bios version I think, that board should include auto, standard, high and extreme options. Use the + and - keys to adjust values if you're entering by name, with capitalization (required for some settings) on the first letter of the word. I used standard for my 4.5OC and it was fine. However, I switched to a Hero VIII because of that setting and the fact that DRAM voltage was only adjustable in .020v steps instead of the usual .005v steps most makers offer. Like Gigabyte did. What are you using to monitor? My suggestion would be HWinfo. Run "Sensors only" and just Sensors only. It will be an option when opening the program. Put a check next to that setting alone. Core Temp is also...
Initially, I considered putting in a fresh CMOS battery. The existing one is likely at least three years old, and possibly older, having spent time in storage before being placed on a board. CMOS batteries from motherboard makers might already be several years past their prime before installation. This could easily reach five to seven years, though it's not common. I've experienced this situation before.
This issue would definitely lead to lost settings. It’s also possible there’s a configuration error and fast boot plus DRAM fast boot are disabled. I’d ensure both are turned on. If fast boot isn’t active, the system might try to retrain itself each time. Another possibility is that your overclock isn’t stable, causing the system to reset during cold starts—this usually happens even after a restart if it was the root cause.
I own the exact board and a Hero VIII with a 6700kHz, which feels unusually low for that frequency unless the LLC setting is very high. Every chip behaves differently, so it might be fine at 1.315v. It would help to run Realbench for eight hours to confirm stability before making any changes.
Initially, I considered replacing the CMOS battery. The existing one is likely more than three years old, possibly even older since it's been stored in a standard bin before being installed on the board. Manufacturers often supply CMOS batteries several years prior to installation, so it could be anywhere from five to seven years old. While this isn't typical, I've encountered such cases before.
This change would likely explain the loss of settings. It might also indicate a configuration issue—fast boot and DRAM fast boot are probably not enabled. I recommend ensuring both are activated. If fast boot remains disabled, the system may attempt to retrain itself each time. Another possibility is that your overclock isn't stable; cold boots can trigger resets, which usually occur even after a restart if it was the root cause.
I own the exact board and a Hero VIII with a 6700kHz, which feels unusually low for that frequency unless the LLC setting is very high. Each chip behaves differently, so it might be stable at 1.315v. It would be wise to confirm this by running Realbench for eight hours—the most reliable test for verifying CPU overclock stability.
Thanks for your prompt reply.
I'll start with the battery check first. The overclock is fresh, and I had to remove the CMOS because of an incorrect setting.
I've been using Prime95 and Intel CPU burn for stress testing, but I'm seeking a more effective tool. I plan to try Realbench next.
This is my first attempt at overclocking with this BIOS; the LLC value was unclear, offering only high, standard, and no clear guidance on what each did. I left it on auto for the overclock but kept an eye on vcore via HWiNFO64. It's stable enough to show only a 0.06v fluctuation. The confusing part is the random vcore spikes that appear as errors, reaching up to 2.6v. This appears to be a monitoring software issue—during stress tests and normal use, voltage stays consistent.