The holistic efficiency comparison definitely favours Intel on desktop, as Ryzen has heavy idle power usage, due to its 12nm I/O die from GloFo.
In laptops idle power draw and heat are far more important because they effect system endurance. In fact most gains in mobile device power management are with lowering idle consumption not load consumption.
It certainly is if you're a major corporate entity deploying tens of thousands of desktops idling 24/7. The 30-40W IO die idle cost adds up.
If you're doing distributed rendering you're better off buying Ampere Arm CPUs anyways as they're more power efficient than any x86 offering.
That money can dry up very quickly.
Intel's in a much worse position now. Their process woes aren't issues that the rest of the industry will hit in a year or so like the end of Dennard scaling was.
>One of the big takeaways from our initial Core i7-11700K review was the power consumption under AVX-512 modes, as well as the high temperatures. Even with the latest microcode updates, both of our Core i9 parts draw lots of power.
The Core i9-11900K in our test peaks up to 296 W, showing temperatures of 104ºC, before coming back down to ~230 W and dropping to 4.5 GHz. The Core i7-11700K is still showing 278 W in our ASUS board, tempeartures of 103ºC, and after the initial spike we see 4.4 GHz at the same ~230 W.
There are a number of ways to report CPU temperature. We can either take the instantaneous value of a singular spot of the silicon while it’s currently going through a high-current density event, like compute, or we can consider the CPU as a whole with all of its thermal sensors. While the overall CPU might accept operating temperatures of 105ºC, individual elements of the core might actually reach 125ºC instantaneously. So what is the correct value, and what is safe?
https://www.anandtech.com/show/16495/intel-rocket-lake-14nm-...
Perhaps it could downlock more to address the wattage while still coming well ahead in terms of performance, although some older CPUs doing this gave AVX512 a bad reputation.
Few workloads are as well optimized to take advantage of AVX512 as the 3d particle movement benchmark, so both the increases in performance and wattage seen in that benchmark are atypical. If they were typical, then AVX512 would be much more popular.
FWIW, I'm a big fan of wide CPU vector instructions. The real reason it was disabled is probably to push people like me to buy Saphire Rapids, which I would've favored anyway for the extra FMA unit. Although I'll also be waiting to see if Zen4 brings AVX512, which some rumors have claimed (and others have contradicted).
>Electromigration is the movement of atoms based on the flow of current through a material. If the current density is high enough, the heat dissipated within the material will repeatedly break atoms from the structure and move them. This will create both ‘vacancies’ and ‘deposits’. The vacancies can grow and eventually break circuit connections resulting in open-circuits, while the deposits can grow and eventually close circuit connections resulting in short-circuit.
Chips that get very hot are expected to be the first to show this sort of failure.
>In Black’s equation, which is used to compute the mean time to failure of metal lines, the temperature of the conductor appears in the exponent
https://www.synopsys.com/glossary/what-is-electromigration.h...