Missing the most important metric: processing power per watt.
I have some server CPUs laying around that can also beat Apple CPUs. But putting them in a compact, densely packed, sealed box (aka a phone) would result in a fireball and almost no battery life.
> I have some server CPUs laying around that can also beat Apple CPUs
In multicore, sure. In most single threaded workloads as well? That'd be interesting enough on its own, regardless of wattage (though that'd make it even more interesting), as my A19 device gives my overclocked 9800X3D a run for its money (and some losses) in single threaded workloads. Considering that's the only thing the 9800X3D is really targeting, it's impressive beyond just whether or not it does so with less power.
Even aside from battery life, high TDP means thermal throttling under sustained load. Which most of these benchmarks don't really test, or if they did, the ambient temperature affects the results.
They do test continuous (1-hour) gaming performance in controlled environment regularly. Thermal throttling can be easily spotted on the frame rate vs time graph.
The goalposts have been at "performance-per-watt" for a very, very long time. That was why Apple pivoted from PowerPC to Intel, believe it or not. PowerPC laptops couldn't match the battery life of Intel, so Apple pivoted. How times change!
I mean, they're in a thread on my comment whose point is "let's see power per watt".
The goalposts started there. A comment responded to me by saying some Chinese youtubers found it to be good. The person responded saying "but what about battery life" which was the original topic.
The person you're responding to also has never said anything about any other metric.
I have some server CPUs laying around that can also beat Apple CPUs. But putting them in a compact, densely packed, sealed box (aka a phone) would result in a fireball and almost no battery life.