I can imagine people at Qualcomm are not super excited to see themI was wrong and Maxtech was right. A bitter pill to swallow. Now curious to see benchmarks.
Hard to say since they also increased the number of cores and the numbers given are relative to M2 (not M3), but the M3 is not 50% faster than the M2 (CPU wise), so I'd expect there's some improvements in the CPU side on top of the added E cores.Confused about what has or hasn’t improved vs the M3. Any ideas?
Edit: what concerns me is, if either cpu or gpu haven't improved, we have to wait another year probably.
In GB6, the M3 scores ~11,400 points vs the ~9,600 points of the M2. That's a 18% multicore improvement from M2 -> M3. Then, if as said on the keynote M2 -> M4 is 50% improvement, I don't think two additional E cored can explain the difference.I was reading Apple marketing pages about the M4. They mention a "second-gen more efficient 3nm process". Could this refer to N3E? I really would like to know how the planning works behind the scenes. These must be incredibly tough business decisions. I know of no other company that would switch nodes this quickly. Also, I was under impression that N3E is only available later this year? Could it be that M4 is iPad Pro only simply because the current manufacturing capability is currently poor?
Some other thought: I wouldn't be surprised if the CPU cores are actually slower than in M3. Increasing E-core counts sounds like a good move for some additional sustained compute. The only architectural changes in the CPU seems to be the updated AMX. I'd really like to hear more details.
The 50% relates to Affinity Photo’s benchmark. Which is an interesting choice for a cpu benchmark.In GB6, the M3 scores ~11,400 points vs the ~9,600 points of the M2. That's a 18% multicore improvement from M2 -> M3. Then, if as said on the keynote M2 -> M4 is 50% improvement, I don't think two additional E cored can explain the difference.
In GB6, the M3 scores ~11,400 points vs the ~9,600 points of the M2. That's a 18% multicore improvement from M2 -> M3. Then, if as said on the keynote M2 -> M4 is 50% improvement, I don't think two additional E cored can explain the difference.
N3E is indeed already available, but typically Apple stockpiles chips for a fall release - especially for the iPhone. They must really want something to test new features on after WWDC.I was reading Apple marketing pages about the M4. They mention a "second-gen more efficient 3nm process". Could this refer to N3E? I really would like to know how the planning works behind the scenes. These must be incredibly tough business decisions. I know of no other company that would switch nodes this quickly. Also, I was under impression that N3E is only available later this year? Could it be that M4 is iPad Pro only simply because the current manufacturing capability is currently poor?
yeah, that’s my thinking. Something is coming at WWDC that will benefit mightily from something that M4 has that M3 doesn’t.N3E is indeed already available, but typically Apple stockpiles chips for a fall release - especially for the iPhone. They must really want something to test new features on after WWDC.
Huh you're right I missed that part. It's not very clear in their website either.They didn't use GB6, they used Affinity Photo. I doubt the GB6 improvements will be as high.
@Jimmyjames beat me to it ^^
i am guessing that’s artificial and not due to fallout, but who knows.Strangely binned chips too: https://x.com/markgurman/status/1787864987110682921?s=46&t=AVo4Ae4rwcqD3xOi5WvXyg
This site uses cookies to help personalise content, tailor your experience and to keep you logged in if you register.
By continuing to use this site, you are consenting to our use of cookies.