The other day we published a post explaining that Intel’s 1st party benchmarks were skewed as they had a much higher stock (PL1) power draw compared to the Ryzen competition (105W). While it’s true that the in-game power consumption for the latter is going to be a fair bit higher than 105W, it still makes this an unfair comparison. Today (courtesy of @Wolsame on Weibo), we have the Cinebench R20 benchmarks of the Core i7-12700K, 12900K, and the 12600K with the PL1 set to 125W and 241W:
As you can see, the top-end Core i9-12900K sees a performance gain of 35% (from 7.4K to 10K) upon switching from 125W (PL1) to 241W (PL1). The gains are lesser as you move down the stack. The Core i7-12700K gains 30% upon switching to the higher power state while the Core i5-12600K gains only 10%. This indicates that the impact of a higher PL1 is primarily limited to heavily threaded workloads like content creation.
Gaming should see smaller deltas, but considering that already bleak performance advantage over the Ryzen 5000 parts, these could be the difference between a win or a loss. The Alder Lake-S reviews should land on the 5th of November, and that’s when things will become clear for the better.