Is there anything better than another leaked benchmark of upcoming products? This time we won’t have to settle for Geekbench or some other useless benchmark, as PRC user bilibili posted a video where he tested the upcoming Intel Core i5-13600K and Core i7-13700K processors in 10 different games, plus 3DMark Fire Strike and Time Spy. This was done at 1080p, 1440p and 2160p using a GeForce RTX 3090 Ti graphics card. Both processors are QS or Qualification Samples, which means they will be virtually identical to the retail chips, unless there are any last minute issues. The processors were tested using the ASRock Z690 Steel Legends WiFi 6E motherboard, or rather two, since both DDR4 and DDR5 versions were used. The DDR4 RAM ran at 3600 MHz with slow timings of 18-22-22 in gear 1, while the DDR5 memory ran at 5200 MHz, most likely with timings of 40-40-40, although the modules were designed for 6400 MHz, in both cases we are dealing with 32 GB.
Thanks to @harukaze5719 we have some graphs that are much easier to read than those provided by the person who tested the two CPUs. Each processor was compared to its current equivalent SKU from Intel, and in many of the games tested, the gain was only three to four percent. However, in some games-at certain resolutions-especially when paired with DDR5 memory, performance gains were as high as 15-20 percent. In some of the games tested, such as FarCry 6 at 4K resolution, the game appears to be GPU-limited, so a faster CPU won’t help here, as you can see from the charts below. There are also some odd results where DDR5-equipped systems showed a regression in performance, so it’s hard to draw any definitive conclusions from this test. Nevertheless, both CPUs should provide decent performance gains if the game is not GPU limited, around five percent at 1440p when paired with DDR5 memory.
