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

 

[Thala]

If you think the ARM ISA has some sort of an advantage over x86, you're simply wrong.

The ISA makes a big difference both in terms of performance as well as efficiency. x86 ISA is as worse as it possibly gets - because base concepts are inherited from the 70s.

 

[Eug]

The other thing I'm a bit surprised hasn't been talked about more. Apple is now all in on ARM, and Nvidia just bought ARM. That might actually be a blessing though, as Apple has the legal resources to take on Nvidia should the latter try something. But, it also means that Apple doesn't get ARM to itself. Had Apple bought ARM then rolled out these, I bet the talk would be the death of Windows, and that Android might not be far behind. Perhaps we'd be looking at one or more of the PC 3 (AMD, Intel, Nvidia) partnering up.

IIRC, Apple has a perpetual ARM architectural licence.

The ARM Diaries, Part 1: How ARM’s Business Model Works

www.anandtech.com

Plus NVIDIA as owner of ARM would make money with every iPhone, iPad, Apple Watch, Mac, etc. sold with ARM, so it's not as if they would want to withdraw that licence even if it could.

I've read that it would be very difficult for Apple to own ARM from a legal/regulatory point of view, and this way they don't have to, and can avoid all the headache that would entail.

It's a little ironic though, since Apple was one of the founding members of ARM thirty years ago.

 

[dmens]

I don't understand your thought If you think the ARM ISA has some sort of an advantage over x86, you're simply wrong. The ISA hasn't made any difference since the 90s, once transistors became so plentiful that translating CISC instructions into microops became reasonable the ISA became irrelevant from a performance perspective. AMD would have shot themselves in the foot if they tried to go ARM instead of sticking with x86.

This is 100% wrong.

 

[LightningZ71]

Oh sweet baby (insert diety of your choice here)!

Here we go with yet another near religious argument over ISA pros and cons...

 

[dmens]

What is there to argue. Go look at a die shot of a modern x86 CPU and figure out the size of the decoding front-end, then you know the statement that transistors being so plentiful that uop translation is free is absurd. It isn't even just the area cost either, for example, variable length decoding creates design limitations that are very difficult to overcome. Go ask any CPU designer who has had to deal with x86, they will all say the same thing.

 

[mikegg]

A brand new, few weeks old MBA is already $900.

I predicted many times that Apple would release a $700 Macbook in the future to recapture their iPhone/iPad customers who bought a Windows laptop, and to take on the huge $600-$800 laptop market.

Still think it's impossible?

It's quite obvious by now that Apple will release a Macbook SE, like they've done for the iPhone, Watch, and iPad (no official SE naming). Maybe it's an older MBA Air. Maybe it's the return of the 12". Whatever it is, it's quite obvious to me, but it seems like many in this forum think it's impossible, still stuck to the mindset that Apple will never change from their luxury and "poor value" ways.

Apple is pivoting from being mostly a hardware company. Listen to an earnings call for god's sake. They've repeatedly said they intend to be serious players in subscriptions and services. And how do they sell subscriptions and services? Unify their ecosystem so Apple One can work in all Apple devices, and take more hardware marketshare by releasing cheaper devices.

A 2 month old Apple Watch SE, which is mostly the same as the Series 6, is selling for $230. It's by far, the best Smart Watch under $300. Still think Apple is only for the rich?



Apple has changed. They're gunning for marketshare. They've dominated the tablet market (65%) by releasing cheap iPads that are far superior to Android tablets. They've dominated the Smart Watch market (55% pre-Watch SE, expect much higher in Q4 reports) by releasing and lowering the price of their Watches to the point where no one else can really compete. And they've focused on the lower end of the phone market by releasing the iPhone SE, and then a cheaper low-end model of the Pros. Three years after launching Airpods, Apple took 71% of total wireless headset revenue.

Still think a cheap Mac is impossible?

The only difference between a low-end Mac and a $280 iPad is a bigger SoC, a few extra GB of RAM, a bit bigger storage, and a keyboard. RAM and SSD storage is cheap. Mac SoCs are more expensive to manufacture but their design costs are shared with iPhones and iPad and more expensive Macs. A keyboard is low tech and cheap.

$700 Macbook is definitely doable, especially if it uses 5nm SoCs while the more expensive Macs move to 3nm.

If so, my prediction of Apple taking 50% PC shipment market share in 5 years isn't so far fetched. They've done it in the phone, tablet, and watch markets with the exact same strategy.

A $400 iPhone SE is faster than a $1500 Android phone. One day, a $700 Macbook SE might be faster than your $3000 Windows laptop.

 

[beginner99]

The ISA makes a big difference both in terms of performance as well as efficiency. x86 ISA is as worse as it possibly gets - because base concepts are inherited from the 70s.

True but rumors since years are intel is working on a new uarch that will ditch a lot of the crux. I hope it will actually materialize at some point.

Plus this.

So yeah, ARM-bases CPU will always win in efficiency. Given Apples move, intel and AMD should really think about sitting together and drop this non-sense.

 

[teejee]

Gaming is going streaming, so I'm not sure how much the local hardware is going to matter. Plus AAA games are already on mobile. PUBG had mobile versions pretty quickly, and CoD and Fortnite are there. Epic did mobile before they did their own service. Even Nintendo has multiple non-Nintendo hardware games. Its just a matter of time til we get the rest of the way there.

I actually think that's what sold Nintendo on Nvidia is the future server/streaming tech and not Tegra (well not directly, as I think Tegra is important for being cheap and low power).

Which, aren't mobile games already making as much as PC and console combined?

Apple doesn't really have to do much of anything to become big in gaming, gaming is already moving the direction Apple is moving. As we see with the M1, Apple has no problem biding its time and letting things naturally move that direction. Some of you also seem oblivious to the fact that Apple is following Microsoft, Amazon, and Google, and are moving their brand to "aaS" (as a service) model.

Well, consoles are selling extremely well but gaming streaming is not yet very succesful.

And the latency issues with gaming streaming will probably make it a very non-premium gaming solution for many years.

 

[moinmoin]

True but rumors since years are intel is working on a new uarch that will ditch a lot of the crux.

You sure you didn't accidentally look at old news about Itanium?

 

[Eug]

View attachment 34837

A brand new, few weeks old MBA is already $900.

I predicted many times that Apple would release a $700 Macbook in the future to recapture their iPhone/iPad customers who bought a Windows laptop, and to take on the huge $600-$800 laptop market.

Still think it's impossible?

You're sounding more and more ridiculous.

1) $900 is not $700.
2) Apple computers go on sale EVERY FRICKIN' YEAR on Black Friday/Cyber Monday. This is not new.
3) The list price is still $1000. $1000 is 43% more than $700.

Come back when you actually have something real to argue with.

If so, my prediction of Apple taking 50% PC shipment market share in 5 years isn't so far fetched. They've done it in the phone, tablet, and watch markets with the exact same strategy.

Whatever you're smokin', I want some!

 

[beginner99]

You sure you didn't accidentally look at old news about Itanium?

I think they learned from that. I guess you know which rumors I mean but for those who don't, with new uarch I still meant x86 but with many legacy instructions no one uses in new code, removed and hence making the core design easier and the cores themselves smaller and hopefully more power efficient (or much faster)

 

[Commodus]

I'm with Eug. As transformative as Apple's switch to ARM is poised to be, I don't think Apple is about to claim half of the PC market in a few years. Gain market share? Quite possibly — I can see Apple capturing more of the premium market, particularly those borderline customers around the $1K mark.

Getting 50 percent of the market would not only require selling Macs at much lower prices (think $400-500), it would entail either destroying major PC vendors or expanding the overall PC market by a gigantic amount.

Me, I'm just happy that Apple isn't going the route of Sony and giving up on its computers at a time when other businesses are more appealing.

 

[nxre]

If Apple was serious about wanting to attract the AAA game market, they would have put more effort into the Apple TV. It was their chance to put performance hardware in living rooms where game consoles live and people play games. A few years ago, I played a demo game on an Apple TV in an Apple store. It was a solid effort and gave the expected level of visuals and game play and seemed to be a good foundation to build on. All Apple would need to do is to make an Apple TV based around the M1 8GB SOC package and release a comprehensive design kit for it and we could see where it goes. The current one is based on the A10X. While it's not an awful SOC, it isn't exactly state of the art. The problem for Apple is that the Xbox and PS4 will represent a price ceiling for them that is substantially less than the cheapest Mac mini, which is going to be not too far off the Apple TV in terms of cost of goods sold.

!!!
Apple TV has so many chances to be a huge sucess it pains me to see how it is left aside.
It could act as a console, apple certainly can match the best consoles out there with no problem.
It could act as a TV Player.
It could act as a hub for all Apple Devices.
It could act as a Wi-Fi extender or even router.
It could even act as a computer like a mac mini or even work by extending the processing power of connected devices by performing the most intensive backs and streaming them back, kind of like cloud gaming services but with the benefit of little to no latency.
Meanwhile it seems the only Apple TV apple cares about is the service, which I very much doubt has any legs to stand on despite the massive ammount of cash being spent on it.

 

[nxre]

Getting 50 percent of the market would not only require selling Macs at much lower prices (think $400-500), it would entail either destroying major PC vendors or expanding the overall PC market by a gigantic amount.

I think Apple's answer to the sub $1000 market is iPad. A $599 iPad can pretty much do everything a PC in its range can, and do it better. For surfing the web and web-apps, iPad is leagues ahead. The problem is despite their attempts at framing iPad as a laptop replacement, they still havent been able to suceed in changing the publics mind about that. If you need a computer for college or work, most people assume it has to be a laptop. Part of the problem is obviously lack of software available, but I bet 90% of the apps regular people use on their sub $1000 laptops are already available or easily replaceable. Apple is also guilty in this, not porting their professional software for iPad such as FCP or Logic or even xcode. But in the end, most people just use those cheap laptops for web surfing and text editing and occasional light gaming. And an iPad can do those things, you just need to actually convince people that it can.

 

[Heartbreaker]

!!!
Apple TV has so many chances to be a huge sucess it pains me to see how it is left aside.
It could act as a console, apple certainly can match the best consoles out there with no problem.
It could act as a TV Player.
It could act as a hub for all Apple Devices.
It could act as a Wi-Fi extender or even router.
It could even act as a computer like a mac mini or even work by extending the processing power of connected devices by performing the most intensive backs and streaming them back, kind of like cloud gaming services but with the benefit of little to no latency.
Meanwhile it seems the only Apple TV apple cares about is the service, which I very much doubt has any legs to stand on despite the massive ammount of cash being spent on it.

Basically you are describing a product that is completely antithetical to what Apple builds. Almost no one builds anything this functional, outside of general purpose PC for much higher price, and if anyone ever did, it certainly wouldn't be Apple.

Heck they don't even allow a Web Browser on Apple TV.

 

[Heartbreaker]

Getting 50 percent of the market would not only require selling Macs at much lower prices (think $400-500), it would entail either destroying major PC vendors or expanding the overall PC market by a gigantic amount.

Yeah, this part of the discussion is a lost argument that is dragging out. I can see Apple knocking $100 off the price of a laptop, but that is about it. A large part of the PC market is low end trash, that Apple want's nothing to do with.

Apple is no hurry for market share, they can just keep building a great product and gaining more market share that way. If they just hold the line on pricing, inflationary forces will make them better priced positioned over time.

Slow and steady is the way here. You can imagine how compelling the lineup will looking in 2 years when it's all Apple Silicon, including much more powerful, high core count versions, and critically nearly all SW will be Native ARM, and there will be likely be multiple Windows virtualization options.

Mac market share future looks VERY bright, but it will be steady growth, not a mass takeover.

 

[lobz]

The "app" model assumes that people are paying the true cost of the app, or something like it, just like they used to when they bought a game for their PC or console previously. Back when people were buying physical discs for games at brick and mortar stores the developer of the title was lucky to get 30%, let alone the 70% they get from app stores.

The problem is that app makers moved on to "game is free, we'll nickel and dime you for every little thing within the game and/or make our money via advertising" and now want to move to "web shell platform for streaming game is free, we'll charge you for the games either per title or per hour or via the same in-app nickel and diming".

So Apple's resistance is understandable, these new models take a free ride on everything Apple built, and require Apple to host and provide updates for their stuff without collecting a cent in revenue. It would be like if someone used your garage to have a garage sale, and made you foot the bill for the electricity they used as well, while they collected all the profit.

Apple and 'true cost' in the same sentence makes me puke.

 

[lobz]

I think Apple's answer to the sub $1000 market is iPad. A $599 iPad can pretty much do everything a PC in its range can, and do it better. For surfing the web and web-apps, iPad is leagues ahead. The problem is despite their attempts at framing iPad as a laptop replacement, they still havent been able to suceed in changing the publics mind about that. If you need a computer for college or work, most people assume it has to be a laptop. Part of the problem is obviously lack of software available, but I bet 90% of the apps regular people use on their sub $1000 laptops are already available or easily replaceable. Apple is also guilty in this, not porting their professional software for iPad such as FCP or Logic or even xcode. But in the end, most people just use those cheap laptops for web surfing and text editing and occasional light gaming. And an iPad can do those things, you just need to actually convince people that it can.

This iPad nonsense again. It looks like the last 5-6 years gone past by you and you were either sleeping or high. Tablets are great complementary accessories for a few (or to a certain group of people, many) things PCs suck at. In all other things, a tablet sucks 10 times worse than a PC sucks at reading an e-book.

 

[moinmoin]

I think they learned from that. I guess you know which rumors I mean but for those who don't, with new uarch I still meant x86 but with many legacy instructions no one uses in new code, removed and hence making the core design easier and the cores themselves smaller and hopefully more power efficient (or much faster)

You mean that rumor from 4 years ago that the "Core" legacy would come to an end after Tiger Lake (then planned for 2019)? Keep in mind that unlike AMD, which regularly deprecated instructions no longer in use, Intel is keeping the whole baggage since 8086 alive. The more recent talk about Alder Lake doesn't sound like such a clean break anymore tough (unless they deem it not an important enough talking point).

!!!
Apple TV has so many chances to be a huge sucess it pains me to see how it is left aside.
It could act as a console, apple certainly can match the best consoles out there with no problem.
It could act as a TV Player.
It could act as a hub for all Apple Devices.
It could act as a Wi-Fi extender or even router.
It could even act as a computer like a mac mini or even work by extending the processing power of connected devices by performing the most intensive backs and streaming them back, kind of like cloud gaming services but with the benefit of little to no latency.
Meanwhile it seems the only Apple TV apple cares about is the service, which I very much doubt has any legs to stand on despite the massive ammount of cash being spent on it.

For some time it seemed like Apple TV would lead to some actual TVs by Apple. Now we at least got a service by the same name.

 

[amrnuke]

This iPad nonsense again. It looks like the last 5-6 years gone past by you and you were either sleeping or high. Tablets are great complementary accessories for a few (or to a certain group of people, many) things PCs suck at. In all other things, a tablet sucks 10 times worse than a PC sucks at reading an e-book.

I think the simplest answer would be a MacBook Mini, with an A14-derivative and an 11.6" screen (though Apple haven't been reluctant to put the latest chip in cheap phones, so perhaps they just use an M1-derivative with cut down GPU). That could probably be sold for $799 and still make a sizeable profit. I don't expect Apple to do consumers any favors and price it with low margins or anything. But I also don't see an impetus for Apple to do this - even though it'd likely be a solid machine for reviewing lecture slides, writing papers, browsing the internet, and doing light photo editing.

 

[Commodus]

I think Apple's answer to the sub $1000 market is iPad. A $599 iPad can pretty much do everything a PC in its range can, and do it better. For surfing the web and web-apps, iPad is leagues ahead. The problem is despite their attempts at framing iPad as a laptop replacement, they still havent been able to suceed in changing the publics mind about that. If you need a computer for college or work, most people assume it has to be a laptop. Part of the problem is obviously lack of software available, but I bet 90% of the apps regular people use on their sub $1000 laptops are already available or easily replaceable. Apple is also guilty in this, not porting their professional software for iPad such as FCP or Logic or even xcode. But in the end, most people just use those cheap laptops for web surfing and text editing and occasional light gaming. And an iPad can do those things, you just need to actually convince people that it can.

It is Apple's answer to the low-cost market, but I don't think it's truly ready yet. Remember, an iPad Air with a Magic Keyboard is just shy of $900... and there is no Magic Keyboard for the base iPad. You don't need a keyboard, but we're talking about PC replacements, and many people aren't ready to go touch-only.

You're also overselling the iPad's capabilities a bit. It's much better than it was even a couple of years ago, but it's not at the point where it's great for serious productivity yet. I can't do my job on one, and not because of apps — our web content system won't play nicely with it. I don't think Apple needs to do "macOS with touch," but iPadOS needs to be a more generalized platform to really get people to ditch their PCs.

 

[Heartbreaker]

I think the simplest answer would be a MacBook Mini, with an A14-derivative and an 11.6" screen (though Apple haven't been reluctant to put the latest chip in cheap phones, so perhaps they just use an M1-derivative with cut down GPU. That could probably be sold for $799 and still make a sizeable profit. I don't expect Apple to do consumers any favors and price it with low margins or anything. But I also don't see an impetus for Apple to do this - even though it'd likely be a solid machine for reviewing lecture slides, writing papers, browsing the internet, and doing light photo editing.

Just because people want a less expensive laptop, doesn't mean people want ridiculously tiny <12" screens. There is a niche market for this size, but it won't drive market share.

 

[amrnuke]

Just because people want a less expensive laptop, doesn't mean people want ridiculously tiny <12" screens. There is a niche market for this size, but it won't drive market share.

Sure, let's give it a 13.3" screen and we can call it the, um, MacBook Air!

Apple have done 11.6" screens before, for years. My wife owns one, actually, and has been resistant to getting a bigger one because the smaller form factor fits in her bags, is lighter, and easier to move around.

How would you differentiate a cheaper MacBook, out of curiosity?

 

[Heartbreaker]

Sure, let's give it a 13.3" screen and we can call it the, um, MacBook Air!

Apple have done 11.6" screens before, for years. My wife owns one, actually, and has been resistant to getting a bigger one because the smaller form factor fits in her bags, is lighter, and easier to move around.

How would you differentiate a cheaper MacBook, out of curiosity?

I wouldn't sell a cheaper Macbook. Apple either has to sacrifice on features, margins, or both to move into a lower price bracket. A 12" screen might save $10 on the BOM. That doesn't translate into a $200 retail price drop.

 

[Roland00Address]

If so, my prediction of Apple taking 50% PC shipment market share in 5 years isn't so far fetched. They've done it in the phone, tablet, and watch markets with the exact same strategy.

Just a reminder with desktop and laptops combined the ASP for the entire computer industry, averaging celeron laptops with gaming computers and ultrabooks, all computers sold the mean is about $630 per computer sold. (It dip below $600 a few years ago, but has increased but at the same time total units sells decreased in the same timeframe so what happen was less computers were being sold but better computers.)

I bring up average ASP for the industry for you made the claim apple is going to gain 50% market share and not a more reasonable number like 20% or even 40%.

To achieve the goal you laid out then apple needs much cheaper computers, they will need computers that are cheaper than $600 to gain 50% marketshare.