I find it odd that Apple couldn't design w better chip just because its top talent left. I mean, they have been doing this for years now, so I am sure there are people in the chip team who have a pretty good idea regarding the chip's architecture and how to improve it further.
So, this story, linked to SemiAnalysis, written by Dylan Patel, apersonal speculation that Apple's CPU design has been impacted by loss of talent, is now debunked simply by the release of the latest GeekBench results that demonstrate a CPU performance increase at about what would be expected for an evolved 5nm node design that was likely finalized in the spring of this year?
My initial thought was that the key CPU designers were focusing more on the M1X, M2, and ultimately what will become the Mac Pro Silicon… a third processor series even? MP1? And then let the iPhone processor chill for one year. But if key talent has left the company, it sounds as if it might be a tougher blow. Apple Silicon has been their number one flex for many years now. Let's hope the CPU team is in good hands again.
Well this article didn’t age too well. Up to 55% jump in GPU, over 15% CPU is pretty good IMHO.
it’s almost like Apple was hoping praying for such an article to generate angst ridden press, then repeat another bunch of press once the benchmarks start to dribble out.
in Australian parlance it is called feeding the chooks (the chickens being the media).
This is what I was thinking.
Besides, Apple has always been about overall bottom-line device performance, not seeking out a championship to crow about for every line on the spec sheet. The real question is whether the performance is quick for the current iteration of device hardware and OS, and is there enough room to take on the next several years' OS and software upgrades?
There's more to the hardware than the CPU and clock speed. SoC's are about systems, and it's been decades since the only "processing" unit was the CPU.
Comments
I'm Shocked!
Besides, Apple has always been about overall bottom-line device performance, not seeking out a championship to crow about for every line on the spec sheet. The real question is whether the performance is quick for the current iteration of device hardware and OS, and is there enough room to take on the next several years' OS and software upgrades?
So much for this theory, eh?