Published in PC Hardware

Intel stretched the truth in its AMD comparison

by on04 November 2021


Fancy that

The US tech press which loves telling positive stories about everyone appears to be somewhat shocked that Intel might have not being telling the full truth when it compared its Core i9-12900K and Ryzen 9 5950X.

Intel compared the Core i9-12900K against the Ryzen 9 5950X, currently AMD's most comparable processor but Intel allowed the Core i9-12900K to consume 2.4x the power of its AMD competitor and benchmarked the Ryzen 9 5950X using an older version of Windows 11 with AMD performance issues.

Last week, Intel published the official specifications, performance statistics and prices for its Alder Lake desktop CPUs. While everyone expected  Intel's preferred benchmarks showed the Core i9-12900K in a favourable light, it outscored the Ryzen 9 5950X in seven of the nine gaming benchmarks.

But Intel apparently forgot to mention how it tested the tested the processors, which when looked at more closely made the whole test rather silly.

Intel allowed the Core i9-12900K to consume 241 W at PL1, with the Ryzen 9 5950X restricted to 105 W, also PL1. Correspondingly, the Core i9-12900K could operate at significantly higher clock speeds than the Ryzen 9 5950X, influencing benchmark results.

Additionally, Intel benchmarked the Core i9-12900K and Ryzen 9 5950X using Windows 11 Pro 22000.9, a version known for performance issues with AMD Ryzen processors. For reference, Microsoft fixed the L3 caching issue for AMD Ryzen processors in mid-October with Windows 11 Build 22000.282.

When the results came out, they were thought to be a little odd as they were a little close given that Intel is flogging older technology. But it seems that not only did Intel fudge the results a little, it did its best to cast its rival in the worst possible light.  It is probably better to wait for independent benchmarking before deciding who is the real king of the chips.

Last modified on 04 November 2021
Rate this item
(9 votes)

Read more about: