AnandTech Reviews Apple's M1 Pro and M1 Max Chips (anandtech.com) 207
AnandTech reviews the recently unveiled M1 Pro and M1 Max chips : The M1 Pro and M1 Max change the narrative completely -- these designs feel like truly SoCs that have been made with power users in mind, with Apple increasing the performance metrics in all vectors. We expected large performance jumps, but we didn't expect the some of the monstrous increases that the new chips are able to achieve. On the CPU side, doubling up on the performance cores is an evident way to increase performance -- the competition also does so with some of their designs. How Apple does it differently, is that it not only scaled the CPU cores, but everything surrounding them. It's not just 4 additional performance cores, it's a whole new performance cluster with its own L2. On the memory side, Apple has scaled its memory subsystem to never before seen dimensions, and this allows the M1 Pro & Max to achieve performance figures that simply weren't even considered possible in a laptop chip. The chips here aren't only able to outclass any competitor laptop design, but also competes against the best desktop systems out there, you'd have to bring out server-class hardware to get ahead of the M1 Max -- it's just generally absurd.
On the GPU side of things, Apple's gains are also straightforward. The M1 Pro is essentially 2x the M1, and the M1 Max is 4x the M1 in terms of performance. Games are still in a very weird place for macOS and the ecosystem, maybe it's a chicken-and-egg situation, maybe gaming is still something of a niche that will take a long time to see make use of the performance the new chips are able to provide in terms of GPU. What's clearer, is that the new GPU does allow immense leaps in performance for content creation and productivity workloads which rely on GPU acceleration. To further improve content creation, the new media engine is a key feature of the chip. Particularly video editors working with ProRes or ProRes RAW, will see a many-fold improvement in their workflow as the new chips can handle the formats like a breeze -- this along is likely going to have many users of that professional background quickly adopt the new MacBook Pro's. For others, it seems that Apple knows the typical MacBook Pro power users, and has designed the silicon around the use-cases in which Macs do shine. The combination of raw performance, unique acceleration, as well as sheer power efficiency, is something that you just cannot find in any other platform right now, likely making the new MacBook Pro's not just the best laptops, but outright the very best devices for the task. It's a comprehensive review, and Intel should be panicking.
On the GPU side of things, Apple's gains are also straightforward. The M1 Pro is essentially 2x the M1, and the M1 Max is 4x the M1 in terms of performance. Games are still in a very weird place for macOS and the ecosystem, maybe it's a chicken-and-egg situation, maybe gaming is still something of a niche that will take a long time to see make use of the performance the new chips are able to provide in terms of GPU. What's clearer, is that the new GPU does allow immense leaps in performance for content creation and productivity workloads which rely on GPU acceleration. To further improve content creation, the new media engine is a key feature of the chip. Particularly video editors working with ProRes or ProRes RAW, will see a many-fold improvement in their workflow as the new chips can handle the formats like a breeze -- this along is likely going to have many users of that professional background quickly adopt the new MacBook Pro's. For others, it seems that Apple knows the typical MacBook Pro power users, and has designed the silicon around the use-cases in which Macs do shine. The combination of raw performance, unique acceleration, as well as sheer power efficiency, is something that you just cannot find in any other platform right now, likely making the new MacBook Pro's not just the best laptops, but outright the very best devices for the task. It's a comprehensive review, and Intel should be panicking.
What would "panicking" help them with? (Score:5, Funny)
Re: (Score:2, Insightful)
Games are still in a very weird place for macOS and the ecosystem, maybe it's a chicken-and-egg situation, maybe gaming is still something of a niche that will take a long time to see make use of the performance the new chips are able to provide in terms of GPU.
Maybe game developers see Apple suing the snot out of Epic Games and thus know better than to bother with supporting Apple? I know I don't give two shits about Apple's processors, even if they are a little faster.
Don't shit on developers if you want them to help you sell shit to the masses.
Re: (Score:3)
Not being able to update, means major risks of security breach. And, if something breaks and you need a patch, work up until that point becomes throw-away, especially if it’s a scaling issue. If a 3d scene stops being able to ha
Re: (Score:3)
Re: What would "panicking" help them with? (Score:2)
Re: What would "panicking" help them with? (Score:5, Insightful)
Re: What would "panicking" help them with? (Score:5, Insightful)
Do you really think Epic would have 'picked a fight' over the app store if Apple didn't price gouge them for using it?
By "price gouge" you mean the same 30% that Apple has charged everyone since the inception of the App Store, then you would need to learn what the term price gouging [wikipedia.org] means. Terms do not mean what you want them to mean based on your preferences.
when there is no alternative available.
A fact that existed since the beginning of the App Store. Apple did not suddenly surprise Epic with this fact. Epic agreed with this fact since the beginning. That is like me objecting to my wife practicing her lifelong Catholic faith after being married to her for 7 years.
Re: (Score:3)
Do you really think Epic would have 'picked a fight' over the app store if Apple didn't price gouge them for using it?
By "price gouge" you mean the same 30% that Apple has charged everyone since the inception of the App Store, then you would need to learn what the term price gouging [wikipedia.org] means. Terms do not mean what you want them to mean based on your preferences.
So, price gouging can be avoided by simply imposing the onerous terms from the very beginning on everyone. That's an interesting definition of price gouging.
Re: What would "panicking" help them with? (Score:2, Flamebait)
Yes, by its very definition. Gouging is when a price is X and people consider this fair and something happens (doesn't have to) and the vendor raises the price to a level that most of its customers consider unfair.
You can't call something over priced if you voluntarily accept it with zero demand and based on your own forecast of the demand. Especially so when it's a percentage of demand rather than a fixed or stepped cost that may change your cost model.
Re: (Score:2)
Re: (Score:2)
The nuance being that the customers should consider more the substance of the disagreement than the technicality of how the agreements are and who moves first.
If you believe that you want Apple to gatekeep and never ever for code to run on your device/outside of browser without going through Apple first or for all your financial transactions to be arbitrated by Apple, then you should be upset with Epic for trying to open up alternatives.
If you believe that you'd like a bit more control over extending your f
Re: (Score:2)
Re: (Score:2)
. I do not want to give 200+ developers my payment info. I do not want to open as many accounts
Strange, I don't give 200+ developers my payment info or open a lot of accounts despite Android allowing it to possibly happen.
Further strange, despite the fact that Android supports third party app stores, they don't seem to have any shortage of people going to the Google one, and developers seem to still feel it is important to be available via that path.
The Android model seems to be a good compromise, by default a locked down ecosystem with the ability to opt out, with that barrier being enough to encour
Re: (Score:2)
https://www.forbes.com/sites/t... [forbes.com]
https://www.bleepingcomputer.c... [bleepingcomputer.com]
https://threatpost.com/android... [threatpost.com]
https://www.forbes.com/sites/t... [forbes.com]
The list goes on, and on, and on. So, sorry, but your argument does not invalidate mine, it proves it. Apple's store may not be perfect. But Google's? Fuck that. Ill
Re: (Score:2)
One may make arguments about whether Google needs to do more to be better at vetting, but at the conceptual level, nothing in those articles says that the Apple store users would suddenly suffer if other users were permitted to add other sources at their informed risk.
Re: (Score:2)
It's not like Apple are the only ones that could possibly make those types of CPUs.
Re: (Score:2)
Actually, they pretty much are the only ones that make those types of CPUs. ISA compatibility does not mean implementations are comparable performance wise. The benchmark results of an M1 Max tell you nearly nothing about a Qualcomm part. Further, the way macOS and the M1 cpus are exclusive to each other, it limits ability to do direct hardware comparisons anyway.
If the ARM licensees took a reference design and barely changed it, sure. However, that's not the case among the most prolific ARM vendors.
Re: (Score:2)
Yes, Apple does make the M1. But they don't have a patent on the bigLITTLE ARM concept.
Nvidia (also not the most beloved company) is acquiring ARM for a reason. They want to get into the PC CPU market and have the money to buy the people that can make it happen. Hence generating some interest for x86 alternatives is a positive development in my eyes, even though everyone at Apple can burn and rot in Hell for all I care.
Re: (Score:3)
57 billion transistors on a single chip? You can only make this at a single fab, and that fab production is sold for 5 years ahead.
So yes, Apple is basically the only one that could make those types of CPUs.
Not to mention all the research and development into that 57 billion transistors project... there are maybe 10 companies able to invest enough into this kind of projects.
Re: (Score:2)
Didn't Epic bring the lawsuit?
Re: (Score:2)
Re: (Score:2, Interesting)
Maybe game developers see Apple suing the snot out of Epic Games and thus know better than to bother with supporting Apple?
Why? Developers make a shitton of money on Apple's platform. What the greedy fucks at Epic think is largely irrelevant to them. Apple isn't shitting on developers. They are providing them an incredibly large platform full of potential users with demonstrably open wallets which Apple built for them. You'd have to be a pretty special form form of entitled twat (like Tim Sweeny) to think that this isn't a good deal.
Re: (Score:2)
Sticking to the Epic case narrowly, if the studio used Unreal thten they would see that Apple tried to smack down the vendor's ability to support on MacOS (though court intervened and blocked Apple from inflicting that retribution).
Broadly speaking, the fact that a dispute over Apple's cut of all revenue that goes through iOS spilled over into Apple trying to hit Epic with every hammer they had, including trying to cripple their macOS support speaks to a potentially rough business relationship.
You may want
Re: (Score:2)
I said they tried and a court blocked the reprisal:
https://www.pcgamer.com/apple-... [pcgamer.com]
Apple absolutely tried to tank Epic on Apple ecosystem across the board until a court forced them to narrow their response to the issue at hand.
Re:What would "panicking" help them with? (Score:4, Informative)
No Need to Panic Yet (Score:2)
Re: (Score:2)
As pointed out in another Slashdot story [slashdot.org] Intel do not need to panic yet: they still outperform the new M1 Max. However, it is good that they are close since this will no doubt spur Intel to start improving their CPUs more rapidly which would be good for all of us.
The Intel chip outperformed the M1 Max by the skin of its teeth. The mere fact that a (in terms of the CPU industry) small time player like Apple decided that "To hell with Intel, let's design our own CPU" and was able to give Intel a run for their money should indeed have Intel panicking. If Apple can do this so can any number of others.
Re: (Score:3)
Re: (Score:2)
Re: (Score:2)
If any number can, I don't know why no one else has. Even die-hard Android fans have grudgingly admitted for the past few years that Apple's chips are top-notch. I wish others would step up. CPUs are starting to get exciting again, but only if you're in the Apple ecosystem.
The real question with Intel's CPU is the power:performance ratio.
Re: (Score:2)
A bunch of MBAs putting an engineer in charge sounds like panicking to me.
No games (Score:2, Interesting)
TFA complains that there are no real games on the Mac.There will never be, as each OS upgrade will likely break them.
Some Games (Score:2)
TFA complains that there are no real games on the Mac.There will never be, as each OS upgrade will likely break them.
News to me as there are quite a few games on Steam that support the Mac, including some AAA titles - I've never had any break due to an OS release, why would they? If games were breaking it would mean lots of other apps were breaking also.
The only exception is 32 bit games in the transition to Monterey. But that is not ALL the games.
Re: (Score:2)
TFA complains that there are no real games on the Mac.There will never be, as each OS upgrade will likely break them.
News to me as there are quite a few games on Steam that support the Mac, including some AAA titles - I've never had any break due to an OS release, why would they? If games were breaking it would mean lots of other apps were breaking also.
The only exception is 32 bit games in the transition to Monterey. But that is not ALL the games.
Dropping 32 bit was before Monterey. We also know the next batch to be cut - when Apple drops their Intel compatibility layer.
Re: (Score:3)
We also know the next batch to be cut - when Apple drops their Intel compatibility layer.
I'm not sure I'm too concerned about soemthing that MAY happen in ten years or more.
10 years is a lot longer than they supported Rosetta1 and PPC after the Intel transition, so why do you think that the compatibility layer would last so much longer? Rosetta wasn't installed by default in 10.6 (roughly 4 years after its introduction, as well as dropping PPC support completely from the OS) and wasn't supported at all in 10.7 (roughly 5 years after its introduction).
Re: (Score:2)
We also know the next batch to be cut - when Apple drops their Intel compatibility layer.
I'm not sure I'm too concerned about soemthing that MAY happen in ten years or more.
Rosetta was dropped a lot quicker than that - it was introduced alongside Intel Macs in the OS X 10.4 series, and removed from OS X 10.7. Since there is a lot of work to transition to Arm for older releases that no more work is being done on at all, I expect my game library on Steam and GOG to be unplayable in a couple of years.
Re: (Score:2)
> News to me as there are quite a few games on Steam that support the Mac, including some AAA titles - I've never had any break due to an OS release, why would they?
Really?
I just opened my Steam. I have 40 games in my library, 27 stopped working with the 64-bit only release. That's just Steam, I had lots of others that stopped too. All the Halos, which I would still fire up time to time, and especially annoying is my rainy-weekend-beer-n-pretzels Age of Empires III.
Overall, the vast majority of games sto
Market Size (Score:2)
This makes porting games to Macs far more of a challenge and you then have to support two parallel streams of development for very little (by comparison) extra revenue. You'd be better putting those people to work porting to consoles or developing your next PC game than working on a mac port.
Re: (Score:3)
You mean like OpenGL which nobody uses? The only thing Apple could do was create their own graphics API so that software can take full advantage of MacOS and the underlying hardware.
Re: (Score:2)
Apple tend to be better known in the graphics world and multimedia market. If you can raytrace faster than a pile-of-PCs (the usual clustering method) can use basic shading and other simplistic rendering, then you catch the eyes of studios for whom time is a whole lot more expensive than a stack of Apples. Likewise, for music, if faster processing helps in studios with software effects, then why would they care about games? Far as recording studios are concerned, they ARE the game. And, again, I can't see m
Which is it? (Score:5, Funny)
Competition is good all around and Intel definitely needs pressure from all sides (...they got a little bit complacent for a few years before AMD started catching up to them, I think) but I doubt that they are "panicking".
Re:Which is it? (Score:4, Insightful)
Comparing laptop CPUs is a waste of time because so much depends on the particular laptop they are installed in.
All modern CPUs scale speed based on thermal and power limits, which are dictated by the design of the laptop. Same with GPUs.
At idle the screen backlight is the biggest consumer of power, and thus battery life is mostly dependent on how bright you set it to. You can tell if the review is half baked or not by if they bothered to set the screen to a standard brightness level before testing.
Re:Which is it? (Score:5, Insightful)
That Apple is even competitive with Intel's newest offerings is a massive technical accomplishment for Apple. And hacks hard at the foundations of Intel. It demonstrates how much they have lacked in true R&D for far too long.
A tick/tock between Apple and Intel trying to stay at the top of the performance pyramid would be good for all of us. Regardless of which chips we personally prefer.
What Intel should be panicked over is: Apple has fully legitimized ARM. It's not just something for embedded computing, then maybe mobile computing. It's now a legitimate offering across the whole spectrum of all computing needs. And legitimate in a way that Intel IS NOT. Intel has no real presence in mobile and embedded computing.
What's making ARM win over Intel is this: ARM licenses their core design. You can build a SoC with ARM. Intel is two decades in the past and will only sell you completed chips. You can't add the 3-8 things you need into an intel CPU to get an Intel SoC. Intel will not partner with you to do that. And it will be the demise of X86 and possibly Intel.
Re:Which is it? (Score:5, Informative)
That Apple is even competitive with Intel's newest offerings is a massive technical accomplishment for Apple. And hacks hard at the foundations of Intel. It demonstrates how much they have lacked in true R&D for far too long.
Actually it's a massive technical accomplishment for ARM. There's a key difference. Apple didn't ground up invent something and catch up with Intel. They are standing on the shoulders of giants who have for years been pouring money into R&D, and nothing drives this home more than Apple not being the only one eating Intel's lunch. There are whole server farms powered by ARM. The world's fastest single chip is ARM, and as of last year so is the world's most powerful supercomputer Fugaku, powered by Fujitsu A64FX ARM chips.
Apple has fully legitimized ARM
Also Apple didn't legitimise ARM, they popularised it for the unwashed unknowning masses. ARM has been a legit competitor for x86 for a while. Heck we've had ARM encroaching on x86 for close to a decade now, I'd actually say ARM was "legitimised" by Gigabyte 7 years ago when they started shipping off the shelf server motherboards along side their x86 offerings.
Apple's achievements is great, but let's not over exaggerate.
Re: (Score:2)
Actually it's a massive technical accomplishment for ARM. There's a key difference. Apple didn't ground up invent something and catch up with Intel.
Actually, Apple did design their chip from the ground up. It might support ARM instructions but the chip is not based on ARM designs. The design was generated by P.A. Semi [wikipedia.org] - or what was P.A. Semi after they were purchased by Apple. They were working on power efficient PowerPC before the purchase so if anything, this new chip has been built up from the PPC design.
This is not a technical accomplishment for ARM but it does demonstrate that the ARM ISA can be designed to go fast. So overall it is good f
Re: Which is it? (Score:2)
Apple used ARM in the Newton in the early 90â(TM)s. Intel never could compete with ARM on mobile. What the heck does legitimized mean? What is an illegitimate processor?
Re: (Score:2)
Apple used ARM in the Newton in the early 90â(TM)s.
We're talking about popularising it, not creating some weird technical geeky toy. What next, Nintendo Virtual Boy popularised VR? ;-)
What the heck does legitimized mean? What is an illegitimate processor?
Let's go with definition 2b from the Merriam-Webster dictionary: "genuinely good, impressive, or capable of success". What's illegitimate in this context? Well Intel's answer to ARM: The Intel Atom.
Re: (Score:2)
Apple used a shitty trade-off to increase the speed of their chips - they put the full system memory inside the CPU. Sure, it makes things faster, but now you need to buy an entire CPU if you want to upgrade your memory (IF Apple even allows you to upgrade at all, they currently don't). And yes, people do want upgradeable computers. But most Apple customers probably won't upgrade their memory o
Re: (Score:2)
There you'll find a list of the countries in which Apple offers either a monetary trade-in on the device you have for the one you're going to purchase, or if it's too old and of no value you can simply send it in to them or drop it off for free recycling.
And to save everyone a bit of click-through:
Does Apple offer recycling?
Yes. Apple Trade In lets you recycle any Apple device (including devices from Apple-owned brands) at any Apple Sto
Re: (Score:2)
Well to be fair, having memory in a single chip makes it not only smaller, but also not dependent on gold and friends for card edges.
No DIMMs, no plastic tabs, etc
You also assume that people literally throw away older computers "because they are slow". Yeah no. Ever heard of "hand-me-down"? Ever gave away a perfectly usable computer?
My beloved HP dv2990nr was obsolete in just a few years. The same amount of time it takes for a mac to become obsolete.
Let's not fool ourselves. The common folk is extremely unl
Re: (Score:2)
Benchmarks for unreleased products must always be taken with a grain of salt (yes, Apple too). That's especially true when the release of those benchmarks is timed to coincide with the release of a competing product, because it's pretty much a guarantee that those numbers will be slanted to the greatest degreepossible.
In the case of the Alder Lake benchmarks from yesterday, while they did post some impressive numbers, we don't know any of the details. We don't know if it was overclocked, we don't know if it
Re: (Score:2)
Intel is playing PR games with its "steal the lead back" BS; it's a repeat and shameless offender in that regard.
Great point on power, but there's a lot more that Intel also isn't talking about: Intel produces strong CPUs with weak graphics. Apple's silicon proves it can do both. I think we can discard the game benchmarks as anomalous - once Apple and game companies start optimizing for each other that full graphics power should be unleashed. The only thing that will keep games performance low on the pla
Intel has nothing to panic over (Score:5, Interesting)
Intel has nothing to worry about. First off: yes, the M1 chips are truly impressive. Apple has managed to take a very powerful mobile chip and make impressive laptops out of them. (Or maybe they're notebooks. Apple kept on referring to the M1 Pro/M1 Max MacBook Pros as "notebooks." Which tend to be slower than "laptops.")
The thing is - they only run macOS software. By moving to ARM, they've abandoned the ability to run Windows (either via emulation or Boot Camp), which means even if Rosetta 2 is capable of running your older x86 macOS apps, it can't help you if you need actual Windows. (Or, in my case, actual Linux running on x86. AFAIK you can't emulate ARM Linux on M1 yet either but I haven't checked recently.)
And there just isn't a lot of macOS software out there these days. Because Apple insists on "thinking differently" porting an app to macOS is non-trivial: if you want to use their native OS UI libraries, you have to use what's essentially their own programming language. (Either literally in the case of Swift or effectively in the case of their dialect of Objective-C.) Part of the reason they moved to ARM appears to be to leverage the iOS software library on macOS, but the problem with that is that developers have to flag iOS apps as working on macOS, and not a lot of developers have bothered to do that.
In short, Intel is still providing CPUs to the largest laptop market. Apple's chips have impressive performance, but it hardly matters if the software you want to run doesn't run on them.
Re:Intel has nothing to panic over (Score:4, Informative)
You can run windows for arm under virtualization, and the only thing stopping it from running natively is a lack of drivers. It can run x86 binaries through emulation the same as macos can.
You can also run linux under virtualization (and natively, but with only partial drivers for now), but the need for x86 binary emulation is much less as pretty much all open source linux software can already compile natively for arm.
Pretty much all open source that runs on native macos can also be compiled for arm.
Re: (Score:3, Interesting)
Being able to run ARM Linux won't help me, though - I specifically need to run x86 Linux because what I'm doing is running a VM image that will eventually run on x86-based cloud platforms locally. It's fairly common to do this, too - you run an effectively local copy of the server so you can be fairly sure that what works locally will run on the actual VMs when you publish it. (And I only just realized I said "emulation" when I meant "virtualization.")
You can run windows for arm under virtualization, and the only thing stopping it from running natively is a lack of drivers.
My understanding is that the other stopping point is tha
Re: (Score:2)
So while it's possible to get ARM-based Windows running on M1 in some fashion, you can't do it legally.
Why would you want to do that as Windows on ARM is terrible even from MS. Apple has demonstrated that Windows x86 running on their ARM works well enough for general every day tasks.
re: Windows on ARM (Score:3)
I think the hope, at least, is that the availability of all of these Apple ARM based machines might convince Microsoft to put some actual effort towards the ARM version of Windows? It's not surprising it performs poorly right now, since I can't think of almost any real-world use-cases for it on computers where performance is needed?
In the past, Microsoft was known to make multiple versions of Windows for other platforms, such as their DEC Alpha compatible versions of Windows Server.
They could potentially d
Re: (Score:3)
Man, you've got a lot of unsupported assertions going on.
"Notebooks" vs "laptops" is semantics, not specifications or benchmarks. There is no implied performance differential between them.
"They only run MacOS software" - as the other poster noted, almost any open source UNIX-based software can be successfully compiled and run on MacOS.
"No Windows via emulation" - false, you can totally run ARM Windows via emulation. The licensing of this is another question.
"Not a lot of MacOS software out there" - I don't
Re: (Score:2, Informative)
The term "notebook computer" came about because in the early days "laptops" were often very bulky, and since there were no mobile specific parts were just made of standard desktop parts crammed together.
Manufacturers, especially Japanese ones, started using "notebook" to refer to slim and light (for the time) machines. To this day they are universally called notebooks there, and Apple picked up the nomenclature way back when. In fact Apple was one of the few US hardware companies to have much of a presence
Re: (Score:2)
US manufactures like HP have traditionally called systems targeted at the 'road warrior' who will actually attempt to run the system on aircraft tray table for example -notebooks- and larger usually more powerful systems more designed to be used when you arrive somewhere laptops, and even bigger systems that are not really designed for regular travel at all just "portables" or "desktop replacements" that was the conventional sales-guy usage of the term from the late 90s-2000s.
Too bad for intel most software runs on macos (Score:3)
Apple's chips have impressive performance, but it hardly matters if the software you want to run doesn't run on them.
What can you not run on MacOS beyond games? I run desktop linux, macos, windows 10 in different contexts. There are things I like about each. However, everything one would pay me to do runs fine on MacOS. All of my development tools run on it. All of my photo editing tools run on it. Every creative tool (that anyone pays you to use) I know of runs on it. All the office suites run on it.
I fucking love my video games, but if I do a random sample of coworkers, neighbors, people I encounter daily, onl
Re: (Score:2)
>Intel has a lot to worry about.
People don't need to run out and buy an overpriced M1-based Apple computer to use a browser for visiting facebook. They just don't. And they certainly don't need to pay the "Apple tax" to do it. Only Apple fanbois are the ones thinking M1 is going to somehow crush intel and turn everyone into mac users. Sorry, that simply isn't going to happen.
Re: (Score:2)
At $750, the MBA Air is not overpriced, and no other Intel laptop can beat it in battery life, weight, noise, and even performance per dollar. The "Apple tax" is really no longer a thing. iPhones are simply "top end" phones, and when you go to the top end of Android, they are as expensive as Apple devices.
Apple doesn't sell you any 32GB eMMC/4GB RAM machines. Acer does. Want something comparable to a Macbook Pro? Go and look at Dell XPS price tags, and compare them to apple's.
Are there cheaper "same spec" m
Re: (Score:2)
Intel has a lot to worry about. They lost one of their best customers. You may not be Apple's target market, but the macbook and macbook pro is probably the most lucrative computing market....people who want luxury computers with Apple fit and finish, but in the end, they're just browsing the web with it. They pay top dollar and the replace their machines often...to keep up with the jones. They're irrational big spenders and every company's dream customer.
Not sure how many Mac users you know, but as a Mac user for 21 years I can tell you a big reason many people that I know buy a Mac is because the lifecycle is so long. They aren't buying a new Mac every year to "keep up with the Joneses" because the "Joneses" aren't buying a new Mac every year. Most Mac users I know keep their Macs for 5 years at least. It's been proven time and time again that the Mac is TCO king when compared to a PC. The Apple tax is a myth.
Re:Intel has nothing to panic over (Score:5, Insightful)
"Apple has managed to take a very powerful mobile chip and make impressive laptops out of them."
It's very Slashdot that this whole conversation is framed as if performance was Apple's goal.
Intel maybe doesn't need to worry about the M-series chips, but their customers selling laptops certainly will. Most consumers buying laptops use them as internet/email/light office task machines, so performance benchmarks are pretty much irrelevant. Once the battery life is up to 15-20 hours (as it is with these machine) even that isn't a big deal. My guess is that Apple's goal is product design:
What this hardware does make possible is smaller, lighter laptops than ever before, because Apple can use a smaller battery (the largest single component in the envelope) for acceptable battery life and performance. I'm really quite excited about the new MacAir in 2022.
Re:Intel has nothing to panic over (Score:4, Insightful)
Panic? Probably not yet. But remember what Andy Grove used to say? "Only the paranoid survive." Intel has not been paranoid for quite some time. For at least the last decade or so, probably closer to 15 years; they've been resting on their laurels and peddling fairly underwhelming incremental improvements. The only really mindblowing things they did in that time was the switchover to the Core duo and i lineup and hyperthreading. Compare that to the 10-15 years previous which saw them go from 286s all the way up to Pentiums and Xeons; with many more comparative "knock your socks off" leaps in performance and features. They've gotten complacent. And what, traditionally, happens to tech companies that get complacent?
As others have pointed out, the M1 Pro and Max don't just provide Apple with some very good laptops. They legitimize ARM as a powerful mainstream computing platform no longer relegated embedded systems and comparatively low-power mobile kit. Maybe Intel shouldn't be panicking... yet. But they bloody well should go back to being paranoid.
(And, really, an ARM-ascendant Apple and a properly paranoid Intel could be the best thing to happen for both of them, and for all of us. A renewed arms race in the CPU space could be what it takes to get *everyone* off their complacent butts and get us back to the pace of improvements that were being delivered back in the 1990-2005ish years.)
Re: (Score:3)
I am yet own an M1 Pro/Max machine but I do already have an M1 Mac mini I bought for desktop use and to see how well the transition was handled. Things that have stunned me are that it runs Intel native code faster than a real Intel i7 and it does so without the fan running like a leaf blower and the CPU temps sitting comfortably around 40C. Intel chips typically run in the high 60-90C range - my i7 runs around 80C when under heavy load even after I did the thermals again because the fans were running so lo
Re: (Score:2)
Deserves some Insightful, too.
Re: (Score:2)
No reason for panic (Score:2)
There is no reason for Intel to panic. The M1 architecture isn't direct competition for Intel's x86 line in anything but Apple computers. At most, this is 10% of Intel's annual sales. Between Apple pursuing the ARM architecture and AMD pushing hard with Zen on the x86 side, Intel is facing additional pressure, but it's the kind of pressure they felt during the Pentium 4 years. The outcome of that was the Core architecture.
Apple's ARM architecture has been putting this pressure on Intel and AMD for a few yea
Re: (Score:2)
Re: (Score:3)
Notebooks and tablets are a big chunk of the processor market, and the only one that's growing. ARM dominates the tablet and phone market because its efficient. x86 dominates notebooks because it's fast. And then there comes Apple making ARM processors that are both efficient and fast.
Panic might be too strong a word, but the other notebook manufacturers are watching.
It's PowerPC all over again. (Score:4, Interesting)
Sure, Apple has a more sustainable market these days. But when the M1 came out, they were benchmarking it against Intel Ice Lake mobile CPU, not the more recent Tiger Lake (which was already shipping, albeit in only a handful of laptops). Later it came out, in at least in one benchmark, that Tiger Lake was faster than the M1.
Seems like this is the case with the M1 Pro and Maxy. As already pointed out, just yesterday we on Slashdot were told that Intel's next chip, due in a month or two, will be faster than Apple's. But how much faster? I can't believe it's a big enough difference to generate all this discussion.
Perhaps the most important consideration is software. Not the OSX vs Windows debate, but the desktop-and-web-software-developers-will-bloat-to-the-edge-of-usablity factor, which means that we the end user, independent of platform, will see little of the advantage of these faster processors.
Re:It's PowerPC all over again. (Score:4, Informative)
Re: (Score:3)
Re: (Score:2)
a lot of people don't want OSX, they want Windows.
And a lot more people don't care what OS they use as long as they have their email, a browser and MS Office.
Re: (Score:3)
It's the opposite of PowerPC all over again.
When Apple was using Power chips, they needed to design their systems around what IBM and Motorola were able to deliver; this meant they were were way behind on the power / performance curves. Their systems were slower than Intel system and they hated that.
More recently, Intel has had a terrible time delivering laptop chips on schedule, and Apple's release product releases have been dictated largely by that. I'm gonna guess Apple hated that, too.
For better or wors
Re: (Score:2)
on Slashdot were told that Intel's next chip, due in a month or two, will be faster than Apple's.
The cores might be, but they're going to have a hard time coming even anywhere close to the 400GB/s memory bandwidth in the top-of-the-line M1 Max, never mind the 200GB/s available in the M1 Pro.
Yaz
Panic??? why? (Score:2)
Re: Panic??? why? (Score:3)
Re: (Score:2)
Whats wrong with Anandech? (Score:2)
Anandtech used to be one of the best, with quite technical and thorough reviews, even though they did get things wrong some times. In the last years they seem to take forever to do some reviews of major products and in this case, while they did a review quickly, they admit to using their old scores of the M1 on an older OS because they "no longer have an M1" - seriously, just go out to the store and pick one up they are very cheap (I even bought one myself for fun - tried some perl benchmarks too [perl.org]). Then, mo
Re: (Score:2)
By "very cheap" do you mean $700 for a Mac Mini?
Re: (Score:3)
Intel's existential problem simplified (Score:3)
If you can take an ARM core and make it mostly comparable to an Intel chip with a lower power budget and a lower cost, why wouldn't you?
This is Intel's existential crisis. 95% of computer applications don't need the power of an i7/i9. If Microsoft Office and Windows 11 works perfectly on ARM, why bother with Intel chips at all? The market can migrate to Chromebook pricing for i5-class machines. No games? IT doesn't care. Does it run office? Can we just recompile our windows stuff for ARM? Can we manage it using our current tools? If the answer to all of this is "yes" then businesses will move.
If this happens Intel is Fucked with a capital F.
This is what Intel is scared of, and is trying to head off. But they're basically fucked at this point.
Intel Should be panicking? Why? (Score:5, Funny)
Why should Intel (or AMD for that matter) be panicking?
Is not like Apple will start selling M1Max chips to HP, Dell, Lenovo, ASUS and Acer to put in Windows for ARM PCs or on Chromebooks. Or am I missing something.
Also, I seriously doubt that Apple will sell M1Max chips to the five mentioned above + Supermicro, Inspur, and others to do departamental serves.
I also do not think that Apple will be selling M1Max chips to Nokia, Ericsson, Huawei, Samsung and ZTG to build 5G equipment and Edge computing....
But guess what: All those use cases use X-64 processors. Yes, some of those workloads can also be done on ARM, but those ARM processors WILL NOT BE M1max chips.
So, please, tell me again why oh why Intel and AMD should be panicking?
As compelling and clearly cutting edge... (Score:2)
Re:Benchmarks? (Score:4, Funny)
Ignore me, multiple pages on that article, there's tons of benchmarks. :D
Re: (Score:2)
Most of them synthetic though. I wonder if Apple had a hand in this, they seem to love synthetic benchmarks and tune for them. Or maybe Anandtech just knows that and wanted to write a clickbaity article.
The detail is sadly lacking here. For example, they didn't test how badly the GPU can affect CPU memory access speeds. With a shared memory system that's important. They are all single benchmarks too, which Gamers Nexus has shown to be rather useless. GN always shows you a graph over time so that you can see
Re: (Score:3)
Most of them synthetic though. I wonder if Apple had a hand in this, they seem to love synthetic benchmarks and tune for them. Or maybe Anandtech just knows that and wanted to write a clickbaity article.
These just started arriving today. I suspect Anandtech used mostly synthetics to get the article out quickly and get a jump on traffic from people looking for reviews. Even the article's author mentions the time crunch, and posted to twitter about a screen lag issue that didn't make into the article: https://twitter.com/andreif7/s... [twitter.com]
You should see more detailed reviews come out over the week as reviewers get to spend some actual time with them.
Re: (Score:2)
Ah yeah that makes sense. Seems to have worked too.
Re: (Score:2)
They almost certainly do throttle if needed, else they have a problem in unfavorable ambient conditions or if there's some degradation of the cooling solution.
A single graph by one article doesn't mean much. The cooling solution in question may simply 'catch up' when you get to that level of temperature, the 'speed' may be clock frequency but have a lot of throttle induced halts, or the app could be measuring 'speed' in some other oddly specific way.
The alternative to no throttling would be it would damage
Re: (Score:2)
Apple has an absolute shit record of designing products that overheat. They've been sued in class actions for this (I was a plaintiff, and won). I have no doubt they won't cool these chips properly, or there will be some kind of severe performance benefit once you use this at max performance longer than 5 minutes, or it will just burn out the chip, and Apple will try to charge
Re: (Score:2)
I still remember the nvidia 8400M fiasco.
But at least apple fixed laptops with that chip even out of warranty.
My "dream machine" 14 years ago (HP dv2990nr) died early because of this terrible chip.
Re: (Score:2)
Or maybe Apple is simply better at designing chips that do not need to thermal throttle along with systems that have much better than minimum spec cooling systems
Physics is physics. If you have a 40C ambient, that's going to be tough for a cooling solution to dissipate even a modest wattage. If heat compound deteriorates and degrades contact with the processor, then that will cause an issue. If there is a fan involved and said fan seizes, then that will happen.
They may not *normally* throttle but it's a pretty confident thing that they *can* throttle if need be. To design otherwise would be terrible. Frankly, that articles picture suggests *somethting* happened wh
Re: (Score:2)
Looks like a bug in the application, because it's clearly thermally throttling if you look at the temperature graph. There is an upward trend and then it stops at 99C, in a machine with no active cooling.
Maybe they changed the way the M1 reports its core speed, or just pause cores instead of reducing the frequency, or maybe it's just a bug and it's reading the maximum possible frequency. The graph doesn't make sense otherwise, what stopped the CPU from continuing to heat up?
Re: (Score:2)
If it reached the point where the maximum temperature was reached simply due to passive cooling then the curve would smoothly level off.
The fact that it hits a ceiling shows that it's thermally throttling.
Re:M1 Don't (Score:4, Insightful)
M1 chips definitely do throttle. It's the reason the M1 Macbook Pro (the 13", not the new ones) outperforms the Air at high loads. The Pro has a fan, and the Air doesn't. It's also why you can get Pro speeds from the air if you do a small hardware modification to turn the entire bottom of the case into a heatsink.
Re: (Score:2)
multiple pages on that article
I know right? It's 2021. I expect everything to scroll infinitely, even after the article is finished directly into adverts. I got to the bottom of the screen and was actively confused. Is this site from the 90s like Slashdot? Wait... Does this site even support Unicode!!!!?
Re:Benchmarks? (Score:4, Interesting)