Discussion Apple Silicon SoC thread

[Eug]

M1
5 nm
Unified memory architecture - LP-DDR4
16 billion transistors

8-core CPU

4 high-performance cores
192 KB instruction cache
128 KB data cache
Shared 12 MB L2 cache

4 high-efficiency cores
128 KB instruction cache
64 KB data cache
Shared 4 MB L2 cache
(Apple claims the 4 high-effiency cores alone perform like a dual-core Intel MacBook Air)

8-core iGPU (but there is a 7-core variant, likely with one inactive core)
128 execution units
Up to 24576 concurrent threads
2.6 Teraflops
82 Gigatexels/s
41 gigapixels/s

16-core neural engine
Secure Enclave
USB 4

Products:
$999 ($899 edu) 13" MacBook Air (fanless) - 18 hour video playback battery life
$699 Mac mini (with fan)
$1299 ($1199 edu) 13" MacBook Pro (with fan) - 20 hour video playback battery life

Memory options 8 GB and 16 GB. No 32 GB option (unless you go Intel).

It should be noted that the M1 chip in these three Macs is the same (aside from GPU core number). Basically, Apple is taking the same approach which these chips as they do the iPhones and iPads. Just one SKU (excluding the X variants), which is the same across all iDevices (aside from maybe slight clock speed differences occasionally).

EDIT:



M1 Pro 8-core CPU (6+2), 14-core GPU
M1 Pro 10-core CPU (8+2), 14-core GPU
M1 Pro 10-core CPU (8+2), 16-core GPU
M1 Max 10-core CPU (8+2), 24-core GPU
M1 Max 10-core CPU (8+2), 32-core GPU

M1 Pro and M1 Max discussion here:

Page 78 - Discussion - Apple Silicon SoC thread

Page 78 - Seeking answers? Join the AnandTech community: where nearly half-a-million members share solutions and discuss the latest tech.

forums.anandtech.com

M1 Ultra discussion here:

Page 109 - Discussion - Apple Silicon SoC thread

Page 109 - Seeking answers? Join the AnandTech community: where nearly half-a-million members share solutions and discuss the latest tech.

forums.anandtech.com

M2 discussion here:

Page 127 - Discussion - Apple Silicon SoC thread

Page 127 - Seeking answers? Join the AnandTech community: where nearly half-a-million members share solutions and discuss the latest tech.

forums.anandtech.com

Second Generation 5 nm
Unified memory architecture - LPDDR5, up to 24 GB and 100 GB/s
20 billion transistors

8-core CPU

4 high-performance cores
192 KB instruction cache
128 KB data cache
Shared 16 MB L2 cache

4 high-efficiency cores
128 KB instruction cache
64 KB data cache
Shared 4 MB L2 cache

10-core iGPU (but there is an 8-core variant)
3.6 Teraflops

16-core neural engine
Secure Enclave
USB 4

Hardware acceleration for 8K h.264, h.264, ProRes

M3 Family discussion here:

Page 215 - Discussion - Apple Silicon SoC thread

Page 215 - Seeking answers? Join the AnandTech community: where nearly half-a-million members share solutions and discuss the latest tech.

forums.anandtech.com

M4 Family discussion here:

Page 263 - Discussion - Apple Silicon SoC thread

Page 263 - Seeking answers? Join the AnandTech community: where nearly half-a-million members share solutions and discuss the latest tech.

forums.anandtech.com

 

[smalM]

Do we know for sure that it doesn't support TB? It isn't like Apple publishes chip specs, or would expose TB connection ability in a phone.

If you look very closely, you may spot a difference.

 

[poke01]

What's the point here? That's why they have it. It isn't free, it's their chosen way of increasing MT performance and increasing utilization of execution resources.

It has downsides, but barely any in the benchmarks shown.

That was me being hyperbolic. SMT is a part of makes Zen5 great. It doesn't even come at the cost of increased power consumption.

What I meant was even with the P-core/thread deficiency, the M4 Pro is able to keep with Strix Halo.
Anyway, this was brought up cause M4 doesn't win benchmarks cause of SME but because Apple did make notable gains in INT and FP and is able keep its execution units fed with the M4 while maintaining the perf\w lead.

I don't think the M3 uarch would been able to keep up with Zen5 despite the node advantage with N3B. Node advantages mean nothing if the design sucks or is not up to par.

 

[johnsonwax]

Wrong in so many details.
For A4 and A5 Apple's contribution was to the SoC design (how the various IP blocks were wired together, in particular how they all accessed memory and how they all implemented QoS features). The CPU was from Samsung.
For A6 a "baby" CPU design was implemented, probably based on an existing design inside PA Semi, with
A7 as the "real" Apple CPU that resulted (I am guessing) from right after Apple acquired PA Semi they were told "OK, go wild, design the best CPU you can imagine with only this power level and this area as your constraints".

This matches my understanding. The A7 was explained to me less as a 'go wild' project as it was Apple wanting to get the 64 bit transition over with ASAP now that they had an in-house design team and that was a pretty big project that needed a bit more runway than the A6 which also gave them a bunch of opportunities.

 

[Jan Olšan]

A node advantage does not enable you to keep up with a 16 Zen5 core setup while Apple setup has 10P cores in the example above.

I'd say that is exactly what node advantage does, the reduction in power consumption by far helps most in multithread tasks.

 

[Eug]

This page claims a new Mac Pro is coming in the next few months. That makes sense, as the current model is now over two years old and is still stuck on M2 Ultra. I wonder what chip the new Mac Pro will get. I hope it's not the old M3 Ultra.

Exclusive: How Apple Product identifiers have leaked every Mac release through 2026

Apple is developing many new M5 and M6 Mac configurations, and we now have exclusive details on exactly what the company is developing.

appleinsider.com

The current M4 Pro Mac mini actually beats the M2 Ultra Mac Pro for CPU speed.



Of course, M2 Ultra has a much better GPU and more media accelerator engines though. Still, it'd be nice to see the Mac Pro get an M5 Ultra this year in order to actually justify its existence. I'm guessing there will be no Mx Extreeeeeeme.

 

[poke01]

I'd say that is exactly what node advantage does, the reduction in power consumption by far helps most in multithread tasks.

Yes but will 10 N3B Lion Cove cores with zero e-cores keep up with 16 N4 Zen5 cores at 50-70 watts. I very much doubt it.

Why cause? Intels P-core design warrants high power, that is how they scale their P cores which is horrible design for laptops and servers.

I’m tired of people saying it’s the node advantage. The node advantage is not an easy way to excuse the design of Apple’s cores and their architecture.

We can see this if we take a look at the mobile N3E CPUs available in the market now, Apple’s P cores lead by a notable percentage in INT and FP.



And compared to Xiaomi’s take on the X925 P core and which is clocked ~3% lower at 3.9 GHz compared to A18 Pro’s 4.05 GHz, A18 Pro offers ~15-17% better INT
and ~10% better FP.

That’s like a generational uplift for some companies and Apple did on the same fudging node!! Also the X925 is one of the highest IPC cores on the market, so to be even better than that is crazy good.

Now add in the M4 Pro P core which is clocked higher and has faster RAM than A18 Pro and it’s no wonder why Apple matches Strix Halo Zen5 with fewer P cores and threads.

 

[Io Magnesso]

Yes but will 10 N3B Lion Cove cores with zero e-cores keep up with 16 N4 Zen5 cores at 50-70 watts. I very much doubt it.

Why cause? Intels P-core design warrants high power, that is how they scale their P cores which is horrible design for laptops and servers.

I’m tired of people saying it’s the node advantage. The node advantage is not an easy way to excuse the design of Apple’s cores and their architecture.

We can see this if we take a look at the mobile N3E CPUs available in the market now, Apple’s P cores lead by a notable percentage in INT and FP.

View attachment 126701

And compared to Xiaomi’s take on the X925 P core and which is clocked ~3% lower at 3.9 GHz compared to A18 Pro’s 4.05 GHz, A18 Pro offers ~15-17% better INT
and ~10% better FP.

That’s like a generational uplift for some companies and Apple did on the same fudging node!! Also the X925 is one of the highest IPC cores on the market, so to be even better than that is crazy good.

Now add in the M4 Pro P core which is clocked higher and has faster RAM than A18 Pro and it’s no wonder why Apple matches Strix Halo Zen5 with fewer P cores and threads.

No, the X925 has not yet surpassed Apple in terms of IPC. Certainly, it has evolved to a different level than the previous ARM IP core, and the difference has been reduced.
Still, Apple's P core is the strongest IPC at the moment

 

[poke01]

No, the X925 has not yet surpassed Apple in terms of IPC. Certainly, it has evolved to a different level than the previous ARM IP core, and the difference has been reduced.
Still, Apple's P core is the strongest IPC at the moment

That’s not what you should conclude from that essay I wrote. It’s basically saying the node advantage is meaningless if the design sucks.

 

[Io Magnesso]

That’s not what you should conclude from that essay I wrote. It’s basically saying the node advantage is meaningless if the design sucks.

There was certainly a difference in nodes around the time of M1, but... We now know that other companies can't beat Apple even if they use the same process.

 

[Io Magnesso]

I think the M5 will have a higher clock.
Increasing the clock frequency is the quickest way to improve performance.
The ARMs hadn't thought much about the clock frequency until now., I think I've been attacking there recently.

 

[johnsonwax]

The A18 Mac rumor is interesting from a different perspective.

Apple is very judicious when going downmarket, because while it potentially adds revenue it also potentially cannibalizes higher revenue/margin offerings, and Apple has always been about breaking the traditional value power curve by offering the (arguably) best consumer product on the market and doing so at relatively high volume. They want to sell 100 million Ferraris a year. My sense is that Apple is considering an A18 Mac either because they think they can capture a new market, like the Chromebook market that is large enough and new to them, possibly up for grabs depending on if the courts force Google to divest ChromeOS, or because they think that consumer sentiment will be soft (tariffs, recession, etc.) and they'll need a more affordable option or risk losing their current entry level customers to cheaper Windows machines.

My guess is that this is not an optimistic move by Apple but a protective one. Apple's primary marketing focus is customer retention - if you pick up more customers from your competitors than you lose to your competitors, you're winning the market regardless of other factors. My guess is this is a move to retain customers in anticipation of worsening consumer sentiment. Keeping withered tech in the lineup is kind of an easy way to be flexible for downmarket, but keeping low-volume M1 production rolling for so many years has a cost. Using the A series in this role has a lot of margin advantages for them because those fixed costs can be better distributed in exchange for having to maintain another variant of MacOS.

 

[johnsonwax]

Of course, M2 Ultra has a much better GPU and more media accelerator engines though. Still, it'd be nice to see the Mac Pro get an M5 Ultra this year in order to actually justify its existence. I'm guessing there will be no Mx Extreeeeeeme.

I don't think there's any possibility it'll ship with anything but the M3 ultra. The Mac Pro is a service product to keep certain markets on the platform which drive other sales - it's not really a product that makes Apple money. Its going to languish as a 'good enough' product.

 

[Io Magnesso]

The A18 Mac rumor is interesting from a different perspective.

Apple is very judicious when going downmarket, because while it potentially adds revenue it also potentially cannibalizes higher revenue/margin offerings, and Apple has always been about breaking the traditional value power curve by offering the (arguably) best consumer product on the market and doing so at relatively high volume. They want to sell 100 million Ferraris a year. My sense is that Apple is considering an A18 Mac either because they think they can capture a new market, like the Chromebook market that is large enough and new to them, possibly up for grabs depending on if the courts force Google to divest ChromeOS, or because they think that consumer sentiment will be soft (tariffs, recession, etc.) and they'll need a more affordable option or risk losing their current entry level customers to cheaper Windows machines.

My guess is that this is not an optimistic move by Apple but a protective one. Apple's primary marketing focus is customer retention - if you pick up more customers from your competitors than you lose to your competitors, you're winning the market regardless of other factors. My guess is this is a move to retain customers in anticipation of worsening consumer sentiment. Keeping withered tech in the lineup is kind of an easy way to be flexible for downmarket, but keeping low-volume M1 production rolling for so many years has a cost. Using the A series in this role has a lot of margin advantages for them because those fixed costs can be better distributed in exchange for having to maintain another variant of MacOS.

That's a good opinion I like it

 

[Doug S]

My guess is that this is not an optimistic move by Apple but a protective one. Apple's primary marketing focus is customer retention - if you pick up more customers from your competitors than you lose to your competitors, you're winning the market regardless of other factors. My guess is this is a move to retain customers in anticipation of worsening consumer sentiment. Keeping withered tech in the lineup is kind of an easy way to be flexible for downmarket, but keeping low-volume M1 production rolling for so many years has a cost. Using the A series in this role has a lot of margin advantages for them because those fixed costs can be better distributed in exchange for having to maintain another variant of MacOS.

I dunno. Not discounting what you're saying but when Apple has chosen to go downmarket with stuff like the iPhone SE and Watch SE it wasn't about protecting their markets but expanding them.

Now sure Apple doesn't want to cannibalize MBA sales by having millions of those people buy the new cheaper option. If that's all it does and it hardly increases total sales then it would be a failure. With the iPhone SE they used the "old design" and less expensive parts to keep the margins under control. They knew they would cannibalize some sales of base iPhones (or the older models they kept around at a discount) but they did it to grow the installed base. For the iPhone that's a services play - they will make money off the App Store post sale. Ditto with the long software support timeline, which helps the used market thrive, and iPhones get second and third owners making their installed base percentage much higher than their unit sales percentage would indicate.

Now with the Mac there's not really the same sort of post sale services income. It isn't zero, but its small though maybe they could grow it with cloud or AI or something, but I think that's not what they'd be banking on. The iPhone doesn't have a problem attracting developers - it is usually the first platform developed for even though it has a minority of worldwide sales. The Mac could use some help with the developer community though. There is a lot of software that's developed only for Windows, and not for the Mac, or if it is "ported" to the Mac it is a low quality effort.

If Apple can grow the market share and installed base of the Mac, and gain developer attention, that could create a virtuous cycle - more developers developing for the Mac, more consumers and small businesses considering it as a worthy alternative to a Windows PC, which increases sales and attracts more developer attention.

Unless this is targeted specifically at the edu market, it won't have anything to do with Chromebook.

 

[johnsonwax]

I dunno. Not discounting what you're saying but when Apple has chosen to go downmarket with stuff like the iPhone SE and Watch SE it wasn't about protecting their markets but expanding them.

Now sure Apple doesn't want to cannibalize MBA sales by having millions of those people buy the new cheaper option. If that's all it does and it hardly increases total sales then it would be a failure. With the iPhone SE they used the "old design" and less expensive parts to keep the margins under control. They knew they would cannibalize some sales of base iPhones (or the older models they kept around at a discount) but they did it to grow the installed base. For the iPhone that's a services play - they will make money off the App Store post sale. Ditto with the long software support timeline, which helps the used market thrive, and iPhones get second and third owners making their installed base percentage much higher than their unit sales percentage would indicate.

Now with the Mac there's not really the same sort of post sale services income. It isn't zero, but its small though maybe they could grow it with cloud or AI or something, but I think that's not what they'd be banking on. The iPhone doesn't have a problem attracting developers - it is usually the first platform developed for even though it has a minority of worldwide sales. The Mac could use some help with the developer community though. There is a lot of software that's developed only for Windows, and not for the Mac, or if it is "ported" to the Mac it is a low quality effort.

If Apple can grow the market share and installed base of the Mac, and gain developer attention, that could create a virtuous cycle - more developers developing for the Mac, more consumers and small businesses considering it as a worthy alternative to a Windows PC, which increases sales and attracts more developer attention.

Unless this is targeted specifically at the edu market, it won't have anything to do with Chromebook.

Yeah, but I'm not sure Apple has faith in the expansion of the Mac market. They have significant structural disadvantages as you note due to Windows being the dominant platform, a problem that doesn't exist really with iPhone or at all with Watch. Apple's been making very modest market share progress with Mac over the last two decades, but it's VERY modest, and I don't think they believe moving downmarket is a remedy to that problem. They've rejected that approach so many times in the past. Their strategy to solve this problem, which I think they've clearly leaned into even harder is that their path to PC share is via iPad. They've been pushing iPad as the real 'computer for the rest of us' for a decade now, and the substantial changes to iPadOS of late seem to only reinforce that. So if iPad is the better Mac-like experience, then what message does a Mac laptop that competes with it on price? It seems like one effort is undermining the other in that regard. It makes more sense to me that it's designed to capture legacy Mac users that are getting priced out, rather than convert new buyers which is what the iPad appears to be positioned to do.

 

[Eug]

I don't think there's any possibility it'll ship with anything but the M3 ultra. The Mac Pro is a service product to keep certain markets on the platform which drive other sales - it's not really a product that makes Apple money. Its going to languish as a 'good enough' product.

I hope you're wrong but I suspect you could be right.

 

[Doug S]

Yeah, but I'm not sure Apple has faith in the expansion of the Mac market. They have significant structural disadvantages as you note due to Windows being the dominant platform, a problem that doesn't exist really with iPhone or at all with Watch. Apple's been making very modest market share progress with Mac over the last two decades, but it's VERY modest, and I don't think they believe moving downmarket is a remedy to that problem. They've rejected that approach so many times in the past. Their strategy to solve this problem, which I think they've clearly leaned into even harder is that their path to PC share is via iPad. They've been pushing iPad as the real 'computer for the rest of us' for a decade now, and the substantial changes to iPadOS of late seem to only reinforce that. So if iPad is the better Mac-like experience, then what message does a Mac laptop that competes with it on price? It seems like one effort is undermining the other in that regard. It makes more sense to me that it's designed to capture legacy Mac users that are getting priced out, rather than convert new buyers which is what the iPad appears to be positioned to do.

I dunno. Windows has had structural advantages for a long time but those have been fading over time. Kids used to be indoctrinated into Windows from grade school, now they are using Chromebooks. Windows used to be the way they accessed the internet and chatted with their friends, now that's all smartphone, and they ignore the boring PC their parents still use. They aren't using Windows for either homework or as their primary means of accessing the internet. Apple and Google's brands are in their face in those, and Microsoft's is nowhere to be found.

It is probably difficult to get very many Gen Zers interested in buying ANY PC but I'd hate to be Microsoft in that situation when they are associated with "old tech", "work and boring meetings" and "stuff my parents use". They have no experience with let alone allegiance to Windows unless they are PC gamers, and Gen Z games much less with a PC than older people - they are much more likely to game on a phone or a Switch. Its all handhelds and on the go with them.

So I think Apple has some good odds to make inroads here. As does Google. Microsoft has a really difficult task getting anyone under 25 to ever become a Windows user aside from when they're at work and have no choice.

There was a report going around recently claiming that Windows had "lost 400 million users". That's ridiculous of course, but the underlying story isn't. Microsoft has always reported on the size of the Windows installed base as it increased over the years. They had reported 1.4 billion Windows PCs a few years ago. Recently they claimed "over a billion Windows PCs". Why the change? Because they're trying to hide the ugly fact that it has either stopped growing, or more likely has begun the inevitable process of slowly shrinking.

 

[Eug]

I dunno. Windows has had structural advantages for a long time but those have been fading over time. Kids used to be indoctrinated into Windows from grade school, now they are using Chromebooks. Windows used to be the way they accessed the internet and chatted with their friends, now that's all smartphone, and they ignore the boring PC their parents still use. They aren't using Windows for either homework or as their primary means of accessing the internet. Apple and Google's brands are in their face in those, and Microsoft's is nowhere to be found.

I'm not sure how accurate your assessment of Windows adoption is, but it does remind me of the Samsung problem in Korea. Samsung phones are viewed as dad phones, whereas the kids want iPhones, despite the fact Apple Pay doesn't even work in most places there from what I understand, and Samsung Pay is everywhere.

Young Koreans favor iPhones over Samsung Galaxy: survey

In South Korea, 69 percent of adult smartphone users use a Samsung device, while only 23 percent use iPhones, a Gallup survey of 1,001 adults found. But the sur

www.koreaherald.com

 

[kingsleyopara]

I don't think there's any possibility it'll ship with anything but the M3 ultra. The Mac Pro is a service product to keep certain markets on the platform which drive other sales - it's not really a product that makes Apple money. Its going to languish as a 'good enough' product.

My money would be on the M3 Ultra here too, but don’t you think Apple’s recent server chip ambitions have changed the game? There seem to be a lot of synergies with a hypothetical M5 or M6 Extreme.

 

[mvprod123]

It would be hilarious if the M3 Ultra Mac Pro were presented alongside the M5 MacBook Pro. I can't imagine that Apple's most expensive computer would be two generations behind in terms of architecture. If the rumours and leaks about the identifiers of future Macs are true, then I think the Mac Pro will get the M5 Ultra.

 

[LightningZ71]

Maybe using an A series chip on the bottom of the stack is Apple just finally replacing the M1 and M2 base chips? It should easily outperform both of them across the board. In addition, I think that this generation of A series chips will all feature Apple modem IP, reducing Apple's cost for including a modem in the devices.

 

[okoroezenwa]

Maybe using an A series chip on the bottom of the stack is Apple just finally replacing the M1 and M2 base chips? It should easily outperform both of them across the board. In addition, I think that this generation of A series chips will all feature Apple modem IP, reducing Apple's cost for including a modem in the devices.

I’d guess this was the reason too. No point continuing to ship M1/2 Macs if they can just use A-series chips to cover these machines at the margin they’d want. My assumption is the A18 will just be a drop-in replacement for the M1 in that Walmart (is it) old-chassis MBA they’re still selling instead of reviving the 12” MacBook.

 

[Eug]

Maybe using an A series chip on the bottom of the stack is Apple just finally replacing the M1 and M2 base chips? It should easily outperform both of them across the board. In addition, I think that this generation of A series chips will all feature Apple modem IP, reducing Apple's cost for including a modem in the devices.

I’d guess this was the reason too. No point continuing to ship M1/2 Macs if they can just use A-series chips to cover these machines at the margin they’d want. My assumption is the A18 will just be a drop-in replacement for the M1 in that Walmart (is it) old-chassis MBA they’re still selling instead of reviving the 12” MacBook.

Well, Apple itself doesn't sell M1 or M2 anymore anyway. I wonder how much Apple is really re-supplying M1s to Wal-Mart, and how much is just old stock now, since they're actually 8/256 GB machines. The M2s and M3s got bumped up to 16 GB last year.

For Apple retail:

MacBook Air - M4 only
iMac - M4 only
Mac mini - Starts at M4
iPad - A16 only
iPad Air - M3 only
iPad Pro - M4 only

The only one with an M2 in the name isn't a base M2. It's the M2 Ultra in the Mac Pro (which is sad). I did see a blowout sale in 2025 for the M2 MacBook Air 16/256 GB at CA$899 (US$661) at Costco here in Canada, but that was many months ago, sold out quickly, and I haven't seen it since.

 

[johnsonwax]

I dunno. Windows has had structural advantages for a long time but those have been fading over time. Kids used to be indoctrinated into Windows from grade school, now they are using Chromebooks. Windows used to be the way they accessed the internet and chatted with their friends, now that's all smartphone, and they ignore the boring PC their parents still use. They aren't using Windows for either homework or as their primary means of accessing the internet. Apple and Google's brands are in their face in those, and Microsoft's is nowhere to be found.

It is probably difficult to get very many Gen Zers interested in buying ANY PC but I'd hate to be Microsoft in that situation when they are associated with "old tech", "work and boring meetings" and "stuff my parents use". They have no experience with let alone allegiance to Windows unless they are PC gamers, and Gen Z games much less with a PC than older people - they are much more likely to game on a phone or a Switch. Its all handhelds and on the go with them.

So I think Apple has some good odds to make inroads here. As does Google. Microsoft has a really difficult task getting anyone under 25 to ever become a Windows user aside from when they're at work and have no choice.

There was a report going around recently claiming that Windows had "lost 400 million users". That's ridiculous of course, but the underlying story isn't. Microsoft has always reported on the size of the Windows installed base as it increased over the years. They had reported 1.4 billion Windows PCs a few years ago. Recently they claimed "over a billion Windows PCs". Why the change? Because they're trying to hide the ugly fact that it has either stopped growing, or more likely has begun the inevitable process of slowly shrinking.

Yeah, if GenZ has interest in a PC, it's in a gaming PC, where Apple is even weaker and where SteamOS could really capture share. I could see Microsoft having lost 400M users given the number of people I know who have put their occasionally used computer in the closet and just use their phone now. They still have a PC but Microsoft can see that it's not getting updated any longer.

 

[Jan Olšan]

Yes but will 10 N3B Lion Cove cores with zero e-cores keep up with 16 N4 Zen5 cores at 50-70 watts. I very much doubt it.
Why cause? Intels P-core design warrants high power, that is how they scale their P cores which is horrible design for laptops and servers.

You realize that the effect is iso-architecture so dragging different architectures into it is entirely unproductive. Unless it's to muddy the waters.
Also, trying to prove something about AMD cores by pointing at Intel cores, that is supposed to work how? It's nonsense as a rule, not to mention it's kind of well known that current Intel cores have issues and AMD seems to be doing quite a bit better.

I’m tired of people saying it’s the node advantage. The node advantage is not an easy way to excuse the design of Apple’s cores and their architecture.

So the answer to that is to pretend newer nodes don't reduce power consumption?
The sensible approach is to realize that at least some part of any advantage apple processor would have HAS to be thanks to using newer node. We don't really know how much just by glance, would be tough to properly analyze/isolate to have a good guess.

This is not an all or nothing question ("NOOO it's all 3nm" "NNOOOOO it's all MAFIC ARCHITECTURE"). As almost always...