Did your 9900K miss the silicon lottery?
Did your 9900K miss the silicon lottery?
Because of an unintended damage to the 8700k during maintenance, I switched to a 9900K on my Asus Maximus X Hero motherboard. I didn’t push the previous chip beyond its limits because of overheating problems, which I think stemmed from Asus’ automatic core voltage adjustment after encountering a similar issue with the 9900K. Temperatures would rise to around 75°C during startup for tasks like launching Chrome or Discord. The old chip stayed cooler, hovering near 95°C under load, while the new one reached 100°C and would slow down. It appears the chip was running close to its stock voltage at 1.38v, but I’ve managed to lower it to 1.27v and it’s now stable. I attempted a modest overclock, but it seems I’ve missed the mark—cannot safely add just 100Mhz to reach 4.8GHz without risking damage, as exceeding 1.3v triggers instability. My cooling setup is quite aggressive, using multiple radiators and fans. I’m hoping my suspicions are correct about this chip’s limits, or I need to explore other methods to push it. A 5GHz boost would be ideal, but at least hitting 4.8GHz would be a big win. I’m open to any advice or data that could help me progress. This is my first serious overclocking attempt, so any guidance would be greatly appreciated.
Typical overclocking needs more than 1.3V, but if your cooling isn't handling it, the issue might be with your loop or block. Check your package power on HwInfo during full load—are those values showing with the GPU fully loaded or just idle?
isn't the highest x value where ASUS really saved on the VRM (probably because it's a 4+2 design with boosters)? make sure it stays cool, even the poorest 9900k should handle 4.8-4.9.
GPU is running smoothly, even when under stress. The paste job on the CPU monoblock was completed recently, and everything has been cleaned and assembled properly. I'm puzzled about why the CPU block isn't functioning while the GPU stays cool. According to Hwinfo, my CPU package draws 135-140W while running folding@home. With the current settings (1.27v, stock speeds), core temperatures sit between 70-82°C and the package temperature reaches 83°C. What temperature thresholds would you consider too high?
Voltage and clock speeds align well for the 14nm chip. A 5GHz frequency around 1.35V is typical, similar to 4.7V at 1.3V which is also normal. The cooling system seems limiting performance. Most reliable custom loops can manage up to 1.4V. Could be high summer ambient temperatures affecting results?
Which CPU block is this? Is the flow accurate for that block? I understand that with advanced blocks, the fin layout is designed to move in a particular direction. If your inlet and outlet tubes are swapped incorrectly, it can cause significant resistance and heat problems.
For me it's fine as long as it stays under 100°C, but clearly a custom loop needs to handle 250-300W+ before that point and work smoothly at 5GHz all core. There must be an issue with how you connected the tubing between components—possibly a block flow problem or a faulty connection...