Become a MacRumors Supporter for $50/year with no ads, ability to filter front page stories, and private forums.

TDTOMW

macrumors newbie
Original poster
Oct 16, 2020
17
6
Apple recently updated the 27" iMac to include a 10th gen Intel CPU and SSD storage (https://www.apple.com/shop/buy-mac/imac/27-inch). I don't see Apple in its current form releasing a machine with a sub-6 month shelf life, which is what your scenario would require. I know people point to the PowerPC - Intel transition as an example of Apple doing just that, but Apple as a company is in a significantly different place in 2020 than it was in 2005-6. That's not thinking like an "old PC window clone head", that's looking at how the company and the market has changed in the last 15 years. In fact, replacing a model so soon after release as you "predict" is more representative of the PC model, as PC manufacturers usually refresh their lineups three times a year. As far as the DTK is concerned, the benchmarks are meaningless, since that is a unit meant for development of AS applications rather than actual production hardware of the AS lineup of Macs.

Where Apple needs to compete head to head with Intel and AMD up front is not in the high end - they need to go head to head with the Core i3/i5 notebooks and provide at a minimum equivalent performance with significantly increased battery life. If the new AS Macs get improved performance and battery life from the jump, that positions Apple better going forward. I will seriously consider replacing my i3 MBA with an AS Mac if the performance is equal to or greater to this machine with better battery life. That is the combination that is going to hit Intel and AMD the hardest. Look at Intel's 11th gen presentation, where the only processors they are actually replacing are the Y and U mobile SKUs - not even the slightest mention of desktop parts. That's because mobility is the driving force in the computer industry, so performance per watt (although the average user doesn't use that term) and battery life are driving the purchasing decisions of many businesses and individuals.

Intel is scrambling right now because AMD just formally announced their 5th gen Ryzen parts, in which they were able to get big boosts in performance without increasing the CPU power draw (still 105W). There are two reasons for this: AMDs move to the 7nm process (which puts them approximately 2 years ahead of Intel in that regard), and the adoption of "Big Navi" and the Zen 3 core for their processors. Meanwhile, Intel is releasing spec bumps as new processor generations while misleading their audience by only comparing their new iGPU to 3-year old AMD iGPUs and workstation class nVidia cards (MX250, not even the current MX350). In order to regain some of their clout, Intel would have to either take massive risks (which its risk-adverse business model doesn't allow) or bite the bullet and outsource fabrication to TSMC or a similar company who has already solved the yield issues that have bottlenecked Intel for the last few years. Microsoft pledging to improve Windows on ARM is another shot across Intel's brow, as moving away from x86 would by necessity further shrink Intel's market share and revenue streams.
 

Ballis

macrumors 6502a
May 27, 2008
961
915
Oslo, Norway
Yea one can only hope, but Apple wont do that to their recent iMac customers. The best we can hope for is a 24 right now, and Id be surpriced if even that is the first arm Mac released.
 

OldCorpse

macrumors 68000
Dec 7, 2005
1,758
347
compost heap
There's literally ZERO chance of that. They JUST released the Intel version of the 27" iMac. How do you imagine the customers would feel if barely a few months later they released an all new 27" AS iMac??? That's insane from a marketing point of view. An insult and a slap in the face.

Z E R O chance.

Besides the best way to demonstrate the advantage of AS is to play to its strengths, which is battery life. A desktop will not demonstrate battery life. To demonstrate the advantage of AS, they'll release a laptop with fantastic battery life and bettter performance at the same time. A 27" iMac is irrelevant to that. It's a very poor idea to think that chip performance is what Apple will focus on, instead of what people most clamor for, which is battery life in a strong laptop. Also, laptops are where it's at, desktops being a small fraction of the market. And if Apple wants to convert more Windows users (this is where the growth potential lies), they'll address the laptop space more than the desktop space.

Also, it's widely anticipated, rightly so, that a new 27" iMac will sport a re-design. Given the thermal envelope performance of AS, you can do a completely different design. It takes time to do that, and they are not going to come out with a new design so soon after the Intel 27" came out. NO CHANCE.

Finally, there's been zero leaks about the AS 27" iMac. There's been 24" speculation from analysts, based at least in part on the 21" not being updated, but there's NOTHING about the 27". Why does that matter? Because Tim C. said the first AS machine will be released this year - that leaves only about 2 months to do so. And that means that the designs are not only finalized but they've been released to the production teams and supply chains which is where the leaks happen - and yet NOTHING has come out... telling us that no such events have taken place. Ergo, this is very strong evidence that the 27" iMac is nowhere in sight.

All in all, I find it fascinating anyone can come to such a wrong-headed conclusion. Whenever one speculates, one takes into account all considerations for one's hypothesis FOR and AGAINST. All you seem to have come up with is one single PRO, a very, very weak one may I add, and zero in the way of against. The chances that a hypothesis will be right when it has undergone no tests of counterarguments is vanishingly small. You need to re-think how you come up with theories in the future, because there are literally dozens of CONS to this hypothesis that seems never even occurred to you and you hung your entire thesis on this one weak, weak, weak PRO. Not good.
 

retta283

Suspended
Jun 8, 2018
3,180
3,482
24" iMac will come before the larger one, I don't think they're going to keep the size at 27". But MacBook will be first. It is possible that a 24" iMac and an AS MacBook of some brand will release at the same time, but the iMac won't be first, really.
 

deconstruct60

macrumors G5
Mar 10, 2009
12,493
4,053
Apple is not yet ready to compete in larger desktop space. We won’t see high performance Apple Silicon until sometime next year.

Maybe 2021. Wouldn't be surprising if that class isn't pragmatically 2022. Apple has zero 3rd party GPU support at this point in time. Maybe something will come with macOS 11.1

Pretty good chance that Apple is going to use most 2021 to push the agenda on there own integrated GPUs and that the higher end of the performance ( top BTO iMac , iMac Pro , Mac Pro ) continue to coast on Intel/x86-64 remnants for extended period of time. Those are not the bulk of Mac sales. Not even close. Flipping the laptop line up is probably much higher priority than the top 15% of the desktop line up.

70+% of Mac sales are laptops. Apple Silicon is primarily about laptops. The top end desktop stuff will be probably the last stuff that Apple tries to cover with the new Apple Silicon. The Mini and possibly even the 24" iMac could get covered by baseline design that are targeting the laptops ( all the more so if Apple is going to attempt to collapse the MBP 16" into on single processor package (drop the dGPU). )
 

deconstruct60

macrumors G5
Mar 10, 2009
12,493
4,053
I thank you for liking my enthusiasm but I didn’t say the 27 iMac would be the only AS mac released you have to think Apple is trying to change the world, don’t think like a old PC window clone Head, Apple have to come out with a “WOW” punch in order to make naysayers believe in there Apple silicon product line as you know for years everyone always doubt Apple.

They will , but far more likely it will be "Wow that is a great laptop". Most Mac sales are laptops. So it is primarily a very proven laptop they primarily need . ( even more so if Apple wants to continue on their 'Captain Ahab', ' ever thinner' laptop quest. )

GPU wise Apple doesn't have anything to compete with the BTO GPU options in the 27" models. Not even remotely close. There is humongous gap they would need to fill there if they were going to exclude all third parties from filling that. There is a more modest gap they would need to fill there if still going to use a 3rd party GPU. ( but that gap includes both software and hardware I/O. ) .

Highly unlikely Apple is going to close either one of those gaps any time soon. Apple's inability to walk and chew gum at the same time across large product categories , casts some doubt they could close that during most of 2021 also. The notion that Apple is going to "big bang" covert the whole Mac line up extremely isn't backed up by much of the evidence of allocation to Mac resources over the last 5-7 years. 3-4 years to fix keyboard. Mac Pro lost in Rip van Winkle slumber for 3-6 years . Mac Mini comatose for many years at a time. etc.

P.S. The Home pod mini shipping with the Watch S5 chip is yet another example of the limited chip biodiversity that Apple is likely to engage in for the Macs. Apple is not likely at all to quickly produce a broad line up of Mac targeted SoCs. Mac products that need a SoC that is detached from laptop limitations is likely on a largely separate and very likely delayed development path. Most of the Apple products that depend upon Apple SoC are largely "hand me down" usages. There is no track record for the Siilcon part of Apple doing a concurrent, board base of product roll outs at all.
 
Last edited:

Fomalhaut

macrumors 68000
Oct 6, 2020
1,993
1,724
All in all, I find it fascinating anyone can come to such a wrong-headed conclusion. Whenever one speculates, one takes into account all considerations for one's hypothesis FOR and AGAINST. All you seem to have come up with is one single PRO, a very, very weak one may I add, and zero in the way of against. The chances that a hypothesis will be right when it has undergone no tests of counterarguments is vanishingly small. You need to re-think how you come up with theories in the future, because there are literally dozens of CONS to this hypothesis that seems never even occurred to you and you hung your entire thesis on this one weak, weak, weak PRO. Not good.

If only people applied the same level of critical thinking to their political choices :cool: <sigh>
 

Fomalhaut

macrumors 68000
Oct 6, 2020
1,993
1,724
GPU wise Apple doesn't have anything to compete with the BTO GPU options in the 27" models. Not even remotely close. There is humongous gap they would need to fill there if they were going to exclude all third parties from filling that. There is a more modest gap they would need to fill there if still going to use a 3rd party GPU. ( but that gap includes both software and hardware I/O. ) .

Highly unlikely Apple is going to close either one of those gaps any time soon. Apple's inability to walk and chew gum at the same time across large product categories , casts some doubt they could close that during most of 2021 also. The notion that Apple is going to "big bang" covert the whole Mac line up extremely isn't backed up by much of the evidence of allocation to Mac resources over the last 5-7 years. 3-4 years to fix keyboard. Mac Pro lost in Rip van Winkle slumber for 3-6 years . Mac Mini comatose for many years at a time. etc.

The GPU roadmap is the big question in my mind. As you say, there is (currently) a large gap between the A-series SoC GPU and the AMD GPUs found in the MBP16, iMac & Mac Pro (the AMD GPUs score at least double in the Geekbench Metal benchmarks). Even the Intel Tiger Lake Xe iGPU seems to score about double the A12Z in OpenCL benchmarks.

AMD, NVidia and even Intel won't stop moving, so Apple needs to work on a massive improvement in this space over the next couple of years. I am optimistic that they are putting the effort in, and that we won't see the long delays you quoted for other hardware.
 

eric89074

macrumors 6502
Sep 19, 2012
292
570
We’ll get one or two laptops. Maybe a new Mac mini if we’re lucky. No iMacs or pro models until the second generation AS chip.
 

theorist9

macrumors 68040
May 28, 2015
3,880
3,060
Apple's first three AS Macs will be a 17" gaming laptop, followed by a blade server designed for data centers, and finally a headless x-Mac. Yes, I am kidding.

Consistent with the consenus view, I expect their first AS Macs to be versions of their less-powerful current/past products (13" MBP, MBA, MacBook, 21" iMac, and possibly even Mac mini). I suspect making AS versions of the 27" iMac or 16" MBP will be a tougher nut for them to crack because their experience is with lower-powered products, and thus it makes sense for them to get experience reaching for intermediate power before they reach for the high end.

Here is the conventional view (which may or may not be right, but sounds reasonable) on why: First, the chip in the first AS Macs will probably be a souped-up A14, which takes advantage of the Macs' larger thermal envelope in part by having higher maximum clock speeds. But there's a limit to how far they can push this before reaching diminishing returns. Thus they will probably need a different chip line (the A15?) to thoroughly beat the CPU performance of the the highest-end Intel chips in the 27" iMac and 16" MBP. Second, unlike with any of their lower-end models, these two only come in dGPUs-equipped configurations (the 21" iMac is available with or w/o a dGPU, and the 13" MBP, MBA, and MB are [were, for the MB] iGPU-only). Thus there is a higher expectation for GPU performance with these machines, which again calls for changes too extensive to be achievable with a modified A14.

Anyways, that's the standard speculation.
 
Last edited:

deconstruct60

macrumors G5
Mar 10, 2009
12,493
4,053
Apple appears to be working on a 5nm discrete GPU code-named "Lifuka".


That may not be a discrete chip at all. That could be a chiplet. "Lifuka" is an island that is part of Tonga. Since the two names are coupled if the relative relationship is part of the codeword encoding that can easily indicate that Lifuka is potentially a "part of" something that is Tonga. ( that would also line up with Apple's usage of "Unified Memory" for their GPUs; as a chiplet hanging off the same memory controller the the application (ARM) cores are coupled to. ) . Apple going on and on and on about how great Unified Memory is basically says they are not interested in doing a actual discrete GPU.

Fairly likely Lifuka allows an Apple SoC to cover what the MBP 16" combo does. Not really talking "big Navi" , "3090" , CDNA or A100 Ampere like competitor. Or even Pro Vega II . Apple has a broad scale of dGPU performance to cover. The MBP 16" and iMac 24" are at one end. The more extreme dGPU area as in add-in cards is a vastly larger jump. Given Apple's laptop focus trying to cover the MBP 16" is probably a priority.
 
Last edited:

Fomalhaut

macrumors 68000
Oct 6, 2020
1,993
1,724
Apple appears to be working on a 5nm discrete GPU code-named "Lifuka".


Yes, I read about that a few weeks ago, but also heard that it is not necessarily a discrete GPU, but could be the code-name of the GPU sub-system on the new SoC. Certainly, you can make powerful GPUs on SoCs/APUs, as the new PlayStation and XBox demonstrate - but these are 180-200W TDP chips, so unlikely to appear in a laptop.

It will be fascinating to see what Apple comes up with!
 

Fomalhaut

macrumors 68000
Oct 6, 2020
1,993
1,724
That may not be a discrete chip at all. That could be a chiplet. "Lifuka" is an island that is part of Tonga. Since the two names are coupled if the relative relationship is part of the codeword encoding that can easily indicate that Lifuka is potentially a "part of" something that is Tonga. ( that would also line up with Apple's usage of "Unified Memory" for their GPUs; as a chiplet hanging off the same memory controller the the application (ARM) cores are coupled to. ) . Apple going on and on and on about how great Unified Memory is basically says they are not interested in doing a actual discrete GPU.

Fairly likely Lifuka allows an Apple SoC to cover what the MBP 16" combo does. Not really talking "big Navi" , "3090" , CDNA or A100 Ampere like competitor. Or even Pro Vega II . Apple has a broad scale of dGPU performance to cover. The MBP 16" and iMac 24" are at one end. The more extreme dGPU area as in add-in cards is a vastly larger jump. Given Apple's laptop focus trying to cover the MBP 16" is probably a priority.

Interesting. I see a fairly wide gap in the graphical capability of SoCs/APUs between the likes of Apple A12Z/A13/A14 or AMD Ryzen 5 & 7, and the high-performance SoCs found in gaming consoles.

I'm hoping that Apple will fill this with Apple Silicon that matches mid-tier mobile dGPUs such as those used in the MBP16 - at least getting close to the AMD Radeon Pro 5300M or 5500M, if not the 5600M / 5500XT / 5700XT.
 

Boil

macrumors 68040
Oct 23, 2018
3,477
3,173
Stargate Command
Part of the allure of the Apple silicon transition is supposed to be better battery life, which only applies to portable Macs, aka laptops, so I could see a laptop as the first Mac making said transition...

But I could also see Apple offering the same hardware in the three entry level Macs; MacBook, 24" iMac, & Mac mini...

They get to tout the battery life dealio with the laptops, they cover both budget ends of the Mac desktop (smaller iMac & Mac mini), and they have standardized hardware across the three products, which should make for an easier first launch of Apple silicon Macs...?

2021 brings us the larger 27" iMacs, & MacBook Pros; and maybe even a return of the Cube...!

2022 brings us 30" iMac Pros, & Mac Pros; closing out the transition...!
 

thingstoponder

macrumors 6502a
Oct 23, 2014
916
1,100
They’re not going to base their release schedule on proving the few people who don’t think they will be faster wrong. Besides, it would be more impressive if the 24 inch iMac beat the Intel
I believe that if Apple comes out of the gate with a processor that competes with Intel processor for the top end of iMac product line, then there’s no doubt that they have processors that will outperform intel in their laptops. as we have seen a two year old iPad processor can outperform most of their own laptops now. Then They will have time to to deliver a Mac Pro processor over the next two years, and there will be no need to continue to produce intel Mac, as Tim said they will continue to support macOS operating system for intel systems for years to come.

They said there would be new Intel Macs and have only released one so far. I’ve heard murmurs of a new 16” Pro on twitter.
 

thingstoponder

macrumors 6502a
Oct 23, 2014
916
1,100
You do realize that:

- Apple tried to get rid of the Air moniker and it failed which is why both the MacBook Air and iPad Air got new life breathed back into them in 2018 and 2019 after almost disappearing entirely in 2017.

- The 12" MacBook design was deeply flawed for reasons that go beyond the Y-series Intel chips and the keyboard

right?

Also, you can't revive that same 12" form factor and put two USB-C ports without killing the headphone jack. Getting rid of the headphone jack on the phone is one thing.

What problems did it have other than speed and keyboard? The people who bought it loved it.

Also, I do hope they remove the headphone jack for a second USB port. The MacBook should have never had a headphone jack. What a waste of a port. Buy a 9 dollar adapter if you need it but the rest of us shouldn’t be punished with a non versatile single use port.

People won't go for it on the Mac, especially people who actually do work on their Mac and not use it as an over-glorified Facebook/e-mail machine.

Huh? Headphone jack is about the most casual user feature there is. People who do “real work” would much rather have a second data and accessory port, not a headphone jack.
 

iPadified

macrumors 68020
Apr 25, 2017
2,014
2,257
Apple can compete with intel on CPU tomorrow if they want. GPU needs more development to be competitive. However, the postprocessing workflow of a picture the in A14 is interesting as it goes from GPU to neural engine/ML to GPU before it is displayed to save energy because the GPU is not efficient enough. It is clear that strong GPU for data processing is less necessary if coprocessors are used. I doubt the picture processing is the only thing the neural engine/ML can be used in terms of compute.

Strange that people think the a new shiny iMac 27 will be obsolete when a new ASi iMac will arrive. The new Intel iMac will likely be usable until Apple does not update the intel based MacOS anymore, say 7 years in my opinion.
 
  • Like
Reactions: lysingur

mr_roboto

macrumors 6502a
Sep 30, 2020
856
1,866
My guess is an ARM Mac Mini first because they have already shipped that to developers.

The DTK doesn't accelerate the timeline for the mini. Aside from parts borrowed from Intel minis (some of which can be reused again), there's nothing inside which is useful as a base for designing the real product. The motherboard will need to be completely different.
 
  • Like
Reactions: Jorbanead

MysticCow

macrumors 68000
May 27, 2013
1,564
1,760
My guess for the first ASMacs are:

1. A MacBook Not Pro, probably under the MacBook Air moniker

2. ASMac mini

Both of those would likely perform better than the current generation of Airs and minis. Pretty much, if it has Intel Iris Integrated Extreme GMA HD graphics or whatever they call it now, it's dead in the water with an ASMac.

The next step is to decide whether or not it is worth it to make a gaming graphics chip in an ASMac or just run with AMD again. Right now, I cannot see a situation where ASMac Integrated Graphics are going to outperform AMD right now. I'm not saying they won't in the near future, but right now...no.
 

OldCorpse

macrumors 68000
Dec 7, 2005
1,758
347
compost heap
My guess is an ARM Mac Mini first because they have already shipped that to developers.

I see many people speculating that the mini will be the first AS Mac released. I highly, highly, doubt it. In fact, I'm pretty much 100% sure it won't be. Think about it. Of all macs, it's probably the one with the smallest share - it's smaller than any laptop of course, but it's also smaller than the iMac, only the MacPro might be smaller. Why would Apple release to such a small user market? Is that going to make any noise in the media?? It's nuts. Now, you release a laptop with amazing battery life and specs and you've created waves. But a mini? What in the world are people thinking? SMH.

Apple wants the whole world to take notice - not just Mac users... other Apple product users, windows users and so on. Literally anything else Apple releases will be more relevant in that context than a mini! Who gives a rip about a mini apart from a tiny group of special requirement folks? Madness. A spectacular ASi LAPTOP will be heard around the world! But a mini?? What planet do you folks live on??

How does a mini showcase the great battery life ASi can achieve?? How does a mini showcase ANYTHING??

A mini is uniquely badly suited to showcase - because people use a whole range of peripherals, keyboards, monitors, speakers etc. There is literally no BEFORE and AFTER comparison you can make for the average person. Even an iMac you can clearly compare, because they all share the same peripherals, but how are you comparing an old mini or new mini when Bob's got a Staples Monitor and Barry's got a top of the line monitor?? Is it the mini, or is it the monitor?

Why would Apple be anxious to showcase their first AS Mac by releasing it into a sea of a bewildering and uncontrollable peripherals??

And how is Apple getting feedback when every setup is different? At least when they release a laptop or an iMac, they know every single component and how they interact. HOW IN THE WORLD are they doing that with a mini? And monitoring the performance of THEIR VERY FIRST AS is super important!! They don't want a bunch of reports or bad press because people's random peripherals are not working properly? THAT IS INSANE!!!

I just lose faith in humanity when so little thought goes into forming a hypothesis. When you speculate about something, you really, really, really must think of both PRO and CON arguments. Instead folks just pick one thing, like "developers already have a mini", or "27" iMac will be STRONG" and run with that, giving ZERO thought to if it makes sense from ANY other point of view.

C'mon friends. Think. Think. Don't just blurt out whatever first pops into one's mind. We've got to make better quality rumors and speculation on Macrumors.
 
Register on MacRumors! This sidebar will go away, and you'll see fewer ads.