Looking for guidance on boosting 10700K and checking VIN3 voltage against Buildzoid specs?
Looking for guidance on boosting 10700K and checking VIN3 voltage against Buildzoid specs?
I adjusted many Buildzoids configurations during my first Intel CPU overclock, using a Gigabyte Z490. It seems this board performs better than mine, which is an Aorus Pro AX. My motherboard was priced at $220. My current overclock settings include all power-saving modes, turbo boost enabled, C-states off, 50x multiplier, a 1.320 vcore, override voltage mode on the Gigabyte Aorus Pro AX, standard load lines, and most voltages matched Buildzoids except for VIN3 which was around 0.3 or lower. His setup consistently stayed at 1.7, while mine aimed for 1.230v VR VOUT under load with override settings to avoid crashes. The video you shared shows his VIN3 consistently low, whereas mine fluctuated near 1.7. All other voltages were similar. The main variation was adjusting my VIN3 higher to reach the same output voltage. This difference might be due to specific tuning choices or a unique reading despite both being Z490 boards.
That's the HWmonitor? You can try HWinfo64. If you notice a 1.7v in HWI64, it might give you more details.
vdroop helps reduce vcore during heavy loads but may lead to crashes in some systems. Buildzoid's 10700k is a problematic model; consider switching to another LLC if you need high BIOS voltage. The 1.34 value might come from hwinfo or bios depending on your setup.
I focused on the 1.7V VIN3 voltage in HWINFO 64 and wondered if that could be a problem. It was lower than the 0.3V seen on a similar board. I considered reducing it to about 1.3V and using medium LLC. What voltage should I aim for on a 5.0OC? Maybe around 5.1? My VCore is set to 1.34, overriding it, but HWINFO shows it hovering near 1.28. During Cinebench it drops to roughly 1.21 VR VOUT.
uncertain about vin3 but if it doesn't appear in your bios it probably isn't a major concern—some z390 boards had similar settings that can't be adjusted. Regarding the voltages, you can use the values from this link: https://siliconlottery.com/collections/a...-i7-10700k. Begin at 5.0 and increase if stable, keeping temperatures in check. The 4.9 value seems very safe. In a side note, Buildzoid's 10700k OC was quite bold; I have a profile with idle vcore at 1.45v and it maintains precise voltages under all loads, though I don't run it continuously.
Never thought about it when I purchased it, back then overclockers were everywhere... I watched many benchmark clips and comparisons between Zen 2 and Coffee/Comet Lake, expecting anything from 9600K to 10900K to reach 5.1. Now I see 5.1 is essentially the ceiling for 10700K as well, especially on those silicon models with the override Vcore voltage option. Is this setting specific to Gigabyte boards like mine? Regarding OS and chipset compatibility, what socket and die information do you have? Also, are there any risks with enabling power-saving features like c-states, EIST, turbo boost for regular use? I’d prefer them off unless I’m planning a future upgrade. I’m thinking of moving to a 11900K soon so it lasts longer, but at least one or two upgrades would be nice. Buildzoid said it was safe, but I guess opinions differ.
Adjusted to 1.375vcore. Here are your results: HWinfo 64 temps stayed between 78-81°C. Cinebench R23 ran for 10 minutes and consistently exceeded 81°C. What do you think?
The performance has dropped noticeably between 1.39 and 1.29. I’d consider switching to another model if you can secure around 100MHz more. At idle it stays around 1.39 and then drops to about 1.34. I’d keep it unless the 80°C Cinebench meets your needs and you’re okay with that idle level. It’s hard to say for sure, but it seems the designers focused on the 10900k, while the 10700k performed better in real-world tests. The differences likely come from sensors on the motherboard, socket, and die, and the socket should reflect the actual voltage core. It looks like they prioritized the 10900k, so if you thought the 10700k was weak, check out the 10600k—you’d be surprised by how it performed.