10850k weird bench is a curious project or item with a unique value.
10850k weird bench is a curious project or item with a unique value.
Hello everyone, after updating to BIOS 2004 on the Maximus 12 hero Wi-Fi card, my 10850k reaches around 7200 on CPU Z at a fresh start. Cinebench R23 scores 16200. All limits are cleared with MCE enabled, but no OC is applied. Once I use the machine again, CPU Z drops to 6900 and CB R23 hits 15300. HWinfo64 shows no throttling detected. Max temperature stays at 73°C with all cores running at 4.8GHz during both tests. How can this be? Please share your thoughts and any insights you have. Your feedback is greatly appreciated.
Run consistently at high frequencies across all cores without thermal throttling. With MCE, expect around 4.9ghz or higher consistently. It seems MCE is similar to the Hero XII experience—most of the time I hit 4.9ghz and only a few times 5ghz. These numbers don’t matter much if your CPU stays cool enough for TVB. My i9-10900K performs reliably at 5.2 and 5.3ghz with stable temps, and CPUz shows 7463, which aligns with your expectations for 4.8ghz. I’m not using MCE because TVB outperforms it in most games, keeping temperatures under 70°C, but bench results don’t justify it. Currently, TVB works well for me, though I plan to push to 5.2 or 5.3ghz manually when needed.
At the standard configurations, my 10850K operates smoothly at 4800 MHz and I receive these CPU-Z readings. A variation of 100 points in the Multi Thread test is typical fluctuation. If your data drops after running the machine for a while, launch Task Manager, navigate to the Details section and check for any unusual background processes. I maintain steady outcomes since System Idle Process consistently displays 99% at the start of testing.
Thank you guys for the feedback. So in essence, i should reset CMOS and press F1 upon boot? Or just change MCE to auto? FYI, on the maximus, upon first boot ir after bios update, we are presented with an option to F1 for default intel or F3, all limits removed. However, settings in extreme tweaker remains the same in terms of MCE.
What should your CPU be doing? Should it maintain the same performance (4800 MHz) whether one core or all ten are running? Would you like it to use the standard turbo settings so it reaches up to 5200 MHz when just one core is active? Here’s an illustration of a 10850K using its default turbo settings during a single-thread test. These processors offer great flexibility—you can configure them to operate in any way you prefer. First, decide your desired performance level. You can also reduce the voltage and power usage, which will lower temperatures significantly. Here’s what you might achieve.
I haven't located any decimation settings for this function. It's configured as F3 because I viewed this tutorial right before assembling the system.
You're asking if running MCE at Auto mode is better than enabling it, and you want a CPU with 4.8GHz, all cores, and solid cooling from a Kraken X63.
The 10850K supports a 52x multiplier under light load conditions. This setting isn't considered overclocking. These parameters are the standard configuration for this processor. Intel developed it to automatically select a higher multiplier when lightly loaded. You won't damage the CPU by operating it in this manner. If you prefer not to run your CPU at its default settings, enable Sync All Cores in the BIOS, adjust it to 48, and your CPU will consistently use the 48x multiplier regardless of how many cores are active.
I’m not sure about the issues with my 10900k. At standard settings I rarely see 5.1ghz, and game temperatures stay in the low 50s. Mostly around 4.9ghz about 99% of the time. I haven’t noticed 5.2 or 5.3ghz for more than a brief moment. TVB is turned on by default. Manual 5.1ghz tuning gives better performance. Uncertain why stock settings aren’t working as expected. Right now I’m running CBR23 single core at defaults—core 3 briefly hit 5.3ghz before settling between 4.9 and 5.1ghz. The CPU package is 44c.
High-speed performance metrics are tracked inside the CPU, operating at billions of cycles each second. These timers help precisely gauge the CPU’s multiplier activity. Many monitoring tools compromise on accuracy, failing to reveal the true workload a processor handles. By skipping the Intel-recommended approach, they can’t display how much time a 10900K spends with the 53 multiplier. For a simple check, I adjusted my 10850K to match the default turbo settings a 10900K would use. At standard configurations, it activates the 53 multiplier when one or two cores are running. This drops to 51 when three or more cores are active and reaches 50 with up to five cores. When five to ten cores are engaged, the 10900K switches to the 49 multiplier. Below are the findings during a single-thread benchmark. It’s clear the CPU predominantly uses the 53 multiplier. Windows launches this one-core test on the first core, making it straightforward to observe peak performance. A single-core operation on a 20-thread capable CPU means the processor dedicates 5% of its time in the C0 state to handle this task, with the remaining 0.1% for background processes. Intel’s designs are reliable. Unfortunately, most monitoring programs mislead users into thinking their chip is faulty.