Become a MacRumors Supporter for $50/year with no ads, ability to filter front page stories, and private forums.
again this is digitmes...couple days ago...they come back with their mistake on the imac with no mini-led
So again, expect this 3nm to be in late devices like iphones, and ipad pro
The 3nm will be armv9 and it will be another big step overall....Intel just released ander lake...and for 2-3 years they will not have another big step .....
 
M2 will have, an increase in cpu around 15% , but for the gpu we will see an increase of over 40%
Something that we dont see generation from generation anywhere else
 
It depends on whether Apple skips A-series core generations with each M-series generation. More recent A-series chips see about a 10-15% boost in CPU speed per generation, but with larger boosts in graphics. If we go with the assumption that each M-series generation matches the A series generation-for-generation then an M2 Mac would have a single core Geekbench score of about 1800-ish and an M3 would potentially be pushing 2100. But if Apple decides to do an 18-24 month refresh cycle on M processors and only use even numbered A series chips then the M2 would be closer to 1900-2000 single core and the M3 could hit 2500, but the catch would be the M3 might not materialize until 2024 or later with the A18.

The biggest performance boost will be going from Intel to M-series just by virtue of Apple’s optimizations and sheer efficiency gains, with lesser gains coming from each successive M-series generation (just like where the iPhone is at now). Going from an M1 to an M3 will certainly give a performance boost but the M1 is already faster than most average consumers need half the time.
Thanks! That just above covers it I think!
 
Impressive. I just hope they can keep it up and have the 4nm process ready in time for M4
what? we are already talking about 3nm for M3...and you are talking about regressing to 4nm for M4?
Maybe you wanted to say 2nm for M4
 
  • Like
Reactions: SFjohn
It seems each year this is reduced by 1. Once they do hit 1nm what happens next. I know its not zero, it must become a decimal of the size or is there a smaller unit of measure to switch to. Must be option two now that I type it out.
You are correct, and the industry is working on it. Intel went with the later, switching from nanometers to Angstroms. 1 A = 0.1 nm. Intel is shooting for 2024 for release of 20A branded chips which would be 2 nm.

It’s a marketing headache, because 20A sounds larger than 2 nm. But, Intel is banking on consumers recognizing the change in units as smaller but not doing the math to see they are the same size.
 
  • Like
Reactions: gsmornot
I don't expect iPad Pro to use M2, but rather a 3nm M2X.

iPad Pro already has a higher base price than MacBook Air and I doubt Apple will continue using a high wattage design in their most expensive tablets.

Makes 0 sense. A "M2x" would be an M2 with more cores so thats a nogo for the power limits of a tablet.
A smaller "M2" would just be A15(x) or A16(x) which also doesn't make sense.

-> iPP will get the base M2 (assuming it has a similar power draw to the M1) which will be overkill for iPadOS and people will pay the price to the same features the M1 iPP has over an M1 MBA.
 
2nd Quarter 2023

m3-feature-black.jpg
 
  • Like
Reactions: JMacHack
Makes 0 sense. A "M2x" would be an M2 with more cores so thats a nogo for the power limits of a tablet.
A smaller "M2" would just be A15(x) or A16(x) which also doesn't make sense.

-> iPP will get the base M2 (assuming it has a similar power draw to the M1) which will be overkill for iPadOS and people will pay the price to the same features the M1 iPP has over an M1 MBA.

No, not simply more cores. But rather more E-cores. A different mix similar to how we see more P-cores in the M1 Pro/Max.
 
IBM produced the planet's first-ever 2nm chip in early May.

Intel is a Tech Partner of IBM.

I suspect the two will be offering 2nm manufacturing to third-parties by mid-2022.

Would NOT be at all surprised if Apple is already talking to them about trying it out with a Test Chip.

Ideally, that Test Chip would be a 12 Mpx 2um 10-bit image sensor with super-low ADC read noise !

Tim, are you seeing this ?
 
  • Haha
Reactions: jdb8167
It seems each year this is reduced by 1. Once they do hit 1nm what happens next. I know its not zero, it must become a decimal of the size or is there a smaller unit of measure to switch to. Must be option two now that I type it out.
Next generation may be quantum computing...
 
Sounds weird to me, M1 is being dragged to 2 years with M1 Pro and M1 Max release recently hsving atleast 1 year of shelf life. How come within 1 year they will release M2 and then go on for M3 by end of same year. I think it will make sense to have m2 in September/October as usual followed by M2 Pro and M2 Max
M2 in 2022 and M3 in 2023, I don't get how that's too fast?

TSMC will ramp up productions months and months before any products are released. A late 2022 ramp will only see new chips in 2023.

N5P in 2021, N4 (N5 refinement) in 2022, N3 (full node) in 2023.
 
By the time I get around to replacing my 2013 22nm Haswell iMac the improvements are going to be OFF the charts.
 
IBM produced the planet's first-ever 2nm chip in early May.

Intel is a Tech Partner of IBM.

I suspect the two will be offering 2nm manufacturing to third-parties by mid-2022.

Would NOT be at all surprised if Apple is already talking to them about trying it out with a Test Chip.

Ideally, that Test Chip would be a 12 Mpx 2um 10-bit image sensor with super-low ADC read noise !

Tim, are you seeing this ?
That is pure fantasy.

If you've followed chip lithography for as long as I have you'd know new 'lab' nodes are years, maybe even a decade ahead of what is commercially possible to produce. An Intel 2nm node i.e. what would be comparable to a TSMC or Samsung 2nm is half a decade away at best. Intel can't even catch up to where TSMC was with N7/N7P with their failed 10nm node, the idea they are going to leapfrog TSMC's N7, N5, N3 nodes in this next year is impossible.

Even Intel's own highly optimistic roadmaps (completely unrealistic) are lucky to have a 2nm node competitor by late 2026 and that's if everything goes perfectly and they face no delays. Realistically it's 2027 or 2028.

Intel is in 3rd place in chip manufacturing and whilst I would never write them off, I would put money on them failing to catch TSMC this decade. Given their CEO has been in talks with TSMC about using them as a fab partner tells you how little confidence Intel has in its' own manufacturing road map.

 
By the time I get around to replacing my 2013 22nm Haswell iMac the improvements are going to be OFF the charts.
I jumped from a 14nm Skylake m3 MacBook to a M1 Pro MacBook Pro and the improvements were massive. Before I was constantly hitting the limits of the CPU, now I have far more power than I'll ever need. I haven't heard the fans spin yet after several months of ownership and the battery life is great.

I also went from an iPhone 6S to a 13 Pro and the improvements there were just as dramatic.
 
Register on MacRumors! This sidebar will go away, and you'll see fewer ads.