And now we will look at the second footnote, regarding the “up to 45% longer battery life” claim:
“[ii] Based on testing by AMD as of 4/1/22. Battery life evaluated in hours using a nine participant Microsoft Teams video conference with camera on, 200 nit brightness, slider position AC#2 (Balanced), with 95% utilization. Battery life results normalized for battery capacity differences. System configuration for Intel Core i7 1260P CPU/GPU performance: Lenovo ThinkPad X1 Carbon, 57 watt hour battery, Intel Iris Xe Graphics, 2X8 GB RAM (LPDDR5 5500), 1TB SSD, BIOS version N3AET45W (1.10), Windows 11 Pro. System configuration for Ryzen 7 PRO 6860Z: Lenovo ThinkPad Z13, 50 watt hour battery, 2x16GB LPDDR5 6400, Windows 11 Pro, 1TB SSD, AMD Radeon 680M graphics, GPU driver 30. 0, BIOS N3GET12WE (0.12). Actual battery life will vary based on several factors, including, but not limited to: product configuration and usage, software, operating conditions, wireless functionality, power management settings, screen brightness, and other factors. The maximum capacity of the battery will naturally decrease with time and use.”
Of note when comparing these results is the line “battery life results normalized for battery capacity differences”, as the specific models of laptops being compared have different battery capacities. In other words, the Intel laptop , which has a slightly larger battery, was not allowed to use its full installed battery capacity in the rundown test. As noted, the Intel-powered ThinkPad X1 Carbon has a 57 WHr battery, while the AMD-powered ThinkPad Z13 only has a 50 WHr battery.
It’s also worth noting that the Intel machine has a larger screen, 14 inches, which would presumably consume a bit more power than the 13.3-inch AMD laptop – but a bigger issue is that the display resolutions were not disclosed. I would hope that both systems were tested at the same resolution, regardless of the native resolution of the panel, but this is not clear.
I doubt any changes to the test setup would erase the massive 45% margin of victory for the AMD system, and the claim of 17% faster performance from the 8-core Ryzen 7 PRO 6860Z processor vs. the Intel Core i7 1260P and its 4 P-core / 8 E-core configuration probably comes down to parallelization in testing, given the performance core deficit from Intel here (I’m treating all of AMD’s core as “P-cores” here). Regardless of the test methodology, which still seems a bit odd to me, the larger issue is that these vague disclosures leave something be desired.
Wow. Questioning test methodology and disclosures after PR performance claims? I feel like I’m channeling Charlie Demerjian. It’s exhilarating.