We present a more in-depth analysis than just average FPS to show how the framerate changes over time, which helps highlight FPS drops. Minimum FPS at both the 95th and 99th percentile are reported in these charts, too. A second chart, a histogram, shows shape and spread for the frametime data—how tightly grouped the measurements are. The "IQR" result is called "Interquartile Range," which is an outlier-resistant statistical value that tells us the range in the middle of the frametime distribution.
In Counter Strike 2 on the Intel Core Ultra 285K, the XPG Lancer Blade RGB 5600 is actually edging ahead of the same-frequency JEDEC DDR5-5600, and with 200S Boost enabled it is narrowly beating JEDEC DDR5-5600 + 200S Boost. What stands out here is simply margin of error and just how tightly packed the entire chart is. Without 200S Boost, nearly every kit from DDR5-6000 up through DDR5-9466 clusters within roughly 10 FPS of each other, regardless of frequency or capacity. This is because Arrow Lake's die-to-die (D2D) interconnect between the compute tile and the I/O die acts as a bottleneck that caps how much memory bandwidth can actually reach the cores, flattening the differences between kits that would otherwise scale much further apart on a monolithic design. The 200S Boost profile helps push past some of that ceiling, which is why boosted kits separate out at the top, but without it, most memory kits perform within a narrow range of each other on Arrow Lake, making the XPG kit's reduced IC count far less impactful here than it is on AMD Ryzen or even Raptor Lake.
The relative view confirms it. At the 100% baseline, the XPG Lancer Blade RGB 5600 actually sits ahead of same-frequency JEDEC DDR5-5600 and stays within about 1-2% of most kits up to DDR5-8800. Only by the D2D-limited ceiling being pushed by enabling 200S Boost and having the bandwidth will it be separated from the pack. This reinforces that on Arrow Lake, the D2D interconnect bottleneck compresses performance across nearly all memory kits, making the XPG's reduced IC count largely irrelevant unless you enable 200S Boost on higher frequency memory.
As we increase the resolution, the gap between memory kits narrows, with the average FPS remaining consistent while the top-to-bottom variance is only a 5 FPS difference for the 1% lows and 95th percentile. This occurs because the system becomes GPU-bound, bringing all the memory kits closer together on the performance scale.