Got a tip for us?
Let us know
Become a MacRumors Supporter for $50/year with no ads, ability to filter front page stories, and private forums.
Key iOS Chip Architect Gerard Williams III Departs Apple
- Thread starter MacRumors
- Start date
- Sort by reaction score
The 80s were great because there were a lot of competing platforms and you weren’t limited to IBM and Apple.
[doublepost=1554086347][/doublepost]
So for no particular reason then.
Art is subjective. I found the play exceptionally boring and preachy. Feel free to enjoy it if you’d like. Doesn’t change the point that the guy everyone is pining over has accomplished nothing in this field since apple canned his ass.
I understand that, they have separate teams for both, but what i meant was that both teams had handful of people.
The chip design team is actually pretty large. Bigger than we had at AMD (at least the California division that did the even-named chips like K6, K8, K10)
No they do not. There are other reasons besides salary level that impact a person’s desire to go or stay.
Yes he created Swift, he and his team yet it's HIS brainchild and gift to Apple and coders before he left to goto Tesla which really didn't work out but he got his feet wet in unknown territory and got to lead a very different kind of team.
The architecture of A12 currently has a number of things that separates it from standard cortex based ARM cores,
* First of all Apple has a architectural license from ARM that allows them to use the same instruction set and design a completely different architecture from scratch.
* Starting from A11, apple chips can do per core process handling unlike the Snapdragons or Huawei chips.
* Apples chips contain separate Neural Engine and Image processing Engine cores. Cortex A76 cores have a single AI Accelerator core, marked at 4500 Pictures per minute, compare this to 5 Trillion Ops/second of Apple Neural engine.
* Mali G76 GPU even with 12 cores is atleast a Yr or two behind Apples 4 core GPUs, also Apple's GPU Core is their Internal Design and not based on Mali architecture.
* Apples chips are big, the die size of A series chips on average is twice the size of a competing CPU.
Yes, I know all that. But, sticking to the CPU (and not the other application blocks like neural engine), the CPU core architecture isn’t very different than everything else that is out there. But your bullet points conflate design and architecture. Some bullets refer to design and some to architecture. And, again, my point is that most of the reason that the A-series chips beat the competition has to do with their design, not their architecture. For example, there’s a very good reason that their die are much bigger than the competition. I can’t prove it, but my theory is that this is because they are using the n-of-m technology from Intrinsity (who they bought long ago). The people they hired are also people who designed chips at exponential and AMD, for example. AMD physical design was always much better than Intel’s.
Chris has worked at Google for a couple of years. He is bring Swift to Tensorflow, Google's Deep Learning Framework.
He is pretty busy and has some really bright people on his team. Making Swift work on everything from small PCs to DataCenters. Very challenging project.
I think his work with LLVM was even more impressive than his Swift work.
More than likely he just finished his work on that chip. In my experience people tend to leave at the end of a major project.
Apple cant keep people just by waving huge sums of cash in front of them. Employees who want to move will move on. The reasons for moving on could be pretty much anything, it might be about how they feel their work is being used or the direction a company wants to go in or if they see a better path with some fresh air behind it elsewhere, and maybe less stress for the same money. Apple is not a penal colony for the rich and people with integrity can't be bought. Your post is really bizarre as cash is not the only incentive to stay at a job, no matter how prized you are within a firm. Apple can't just keep people by plastering the exit in bank notes.
Just because companies jump on a bandwagon doesn't make the original idea a good one...
For example 3D TVs (where are they now?)
As a HW engineer with a strong aesthetic design sense I still find the notch "ugly" and a bit of a bodge.
Regarding the SoC of apple's A-series, yes the gap is closing, but this is inevitable. Same thing happens on all other platforms as well. Eventually, the one on top can do so much to evolve, while others are catching up. However, both apple and competitors are already manufacturing very fast mobile chips for some time now. Actually, too fast for anyone to care for any small gap between them.
Apple should worry most on other aspects of their phones that already fall behind.
