Are the new nodes for real on the performance and efficiency? The "nm" doesn't mean much since a while but at least we were getting considerable improvements from node to node, is that still the case?
Apple has been unusual in that they have not pursued "performance no matter what it does to power draw and heat" like other companies have in recent years.
> On the CPU side of things, Apple’s initial vague presentation of the new A15 improvements could either have resulted in disappointment, or simply a more hidden shift towards power efficiency rather than pure performance. In our extensive testing, we’re elated to see that it was actually mostly an efficiency focus this year, with the new performance cores showcasing adequate performance improvements, while at the same time reducing power consumption, as well as significantly improving energy efficiency.
https://www.anandtech.com/show/16983/the-apple-a15-soc-perfo...
Others may make different choices.
The early reporting said the A17 improved the performance on the big cores by 10%, but not the little cores. If they had used the node change oooortunity for performance increases, you would expect to see the gains on both the big and little cores.
> Apple says the performance cores are 10 percent faster than they were before, a relatively mild improvement, while the efficiency cores are more efficient rather than being faster.
https://arstechnica.com/gadgets/2023/09/apples-new-iphone-15...
Indeed. For a long time, new nodes could be read as: more ICs/wafer = lower cost/IC = lower cost/transistor in those ICs. Or bigger ICs with more transistors.
Leaving it up to designers to choose between cheaper/smaller/lower power, or use extra transistors to achieve higher performance at similar cost/IC.
Maybe that lower cost/transistor 'law' doesn't hold anymore for these latest nodes? Or not as much as it used to?
Not saying Moore's law is dead! But it talks transistors/IC. Not so much about the cost of those transistors.
So maybe we're seeing some decoupling here between Moore's law & cost/transistor gains for newer nodes? If so, that would certainly affect market segmentation between products built on older vs. built on newer nodes. As opposed to the 'newer = better everything' we're used to.
Note this is just speculation on my part.
Seems like N3B is a bit of a flop: simply too expensive for anyone but Apple.