Well take any review you like,youll never see a 7950x consuming less power in st tasks than a 13900k. It's not even close actually, that's how intel is beating amd by a lot in taks like the whole Adobe suite. Because in mixed usage they sip power.
The 12900k tuned for efficiency consumes exactly half the power in every single game compared to a 13900k tuned for efficiency. That for me is a big big difference. Ofcourse the 13900k is faster by 15% or more,but double power draw is pretty nuts.
Actually its a commonality among all mainstream reviewer results
---------------
I'll give you an example of TPU's Wizard shenanigans.
After reviewing the RTX 3080 he used the RTX 2080 to review the 5950X for gaming performance just a few weeks later, predictably TPU's bar charts were a solid block with little to no variation in performance and Wizard concluded Zen 3 was not much better for gaming that Zen 2.
That was very obviously contrived to arrive at that result.
Later testing the 5800X3D he used the 3080 10GB, he resulted +9% over the 5950X, it gave him a reason to say a similar thing, again.
HUB used the 3090TI and resulted +19%, later with the 4090 +23%.
Its Ryan Shrout levels of nonsense.
You get the impresstion Wizard often asums the reader is stupid and wont notice he has his own agenda.