The Use of Lapping on a 5C CPU: A Experimentation and Exploration of Performance Gains
As I sit here with my 5C CPU, I find myself pondering the possibilities of performance gains that could be achieved through custom modifications. In particular, I'm interested in exploring the use of lapping, a technique used to shave off tiny amounts of metal from the CPU die, potentially leading to improved thermal conductivity and therefore increased clock speeds.
I started by checking my current scores, which read 5.14 and 5.069 for each of the two cores. While not bad, I'm aware that these values are likely limited by the PPT (Power Phase Terminal) limit, which restricts the maximum voltage that can be applied to the CPU. To overcome this limitation, I considered enabling the PBO (Performance Boosting Option) setting, which would increase the voltage and potentially unlock higher clock speeds.
However, before proceeding with the PBO setting, I wanted to verify if there were any improvements in performance at the current temperature level. Unfortunately, temperatures are still relatively high, hovering around 88-89 degrees Celsius, compared to the initial reading of 94 degrees during a previous test run. While this is not ideal, it's worth noting that temperatures have indeed decreased slightly since then.
I also experimented with using different thermal pastes on my CPU, including MX4 and CryoNut Extreme, in an effort to improve heat transfer and, consequently, performance. However, I didn't see any significant differences between these two materials, suggesting that the issue lies elsewhere. To identify the root cause of this problem, I decided to use a kryonautic stream to visualize the thermal interface between the CPU die and the heat sink.
After running some tests with different setups, including using CryoNut Extreme thermal paste, I noticed an improvement in performance, dropping from 5.14/5.069 to 4.92/4.76. However, this was only achieved through a single run, and it's unclear if this result will hold up over time.
In light of these findings, I decided to investigate further by lapping the CPU die using a lap tool, aiming to remove tiny amounts of metal from the surface. This process involves carefully removing microscopic layers of material from the die to create a smoother surface for heat transfer.
As I progressed through the lapping process, I noted that temperatures continued to drop, with my scores improving slightly in each subsequent run. The most recent reading showed a score of 4.96/4.72, indicating that the performance gain is indeed real and tangible.
The use of a lap tool allowed me to shave off an additional 0.2 millimeters from the die surface, potentially contributing to the observed improvement in thermal conductivity. I'm eager to explore further the effects of this modification on overall system performance and stability.
After completing the lapping process, I took some time to discuss the results with a fellow enthusiast who had previously experimented with custom cooling solutions for his CPU. His input was invaluable, providing insights into the challenges associated with lap-lapping and the importance of proper tooling.
Throughout this experiment, it's essential to remember that the process of lapping is not without risks. If not done correctly, it can result in damage to the CPU or other components within the system. As such, I must emphasize the importance of caution and careful planning when considering custom modifications like lap-lapping.
As I wrap up this experiment, I want to emphasize that the use of a lap tool is not for everyone, particularly those who value their warranty above all else. However, for those who are willing to take on the risks associated with custom modifications, the potential rewards can be significant. In my case, while the results may not have been earth-shattering, I'm convinced that further experimentation and refinement will yield even more impressive gains in performance.
In conclusion, this experiment has shown me the potential benefits of lap-lapping on a 5C CPU, including improved thermal conductivity and increased clock speeds. As I move forward with further experimentation and exploration of custom cooling solutions, I'll be sure to document my progress and share insights gained along the way.