New MacBook Pro chips deliver desktop performance with better power efficiency
Apple's new M1 Pro and M1 Max chips in the MacBook Pro have reached new heights in terms of performance and power efficiency, according to an in-depth analysis and review of the silicon.
Credit: AppleInsider
The upgraded Apple Silicon chips, which power the recently announced 14-inch MacBook Pro and 16-inch MacBook Pro, are updates on the M1 platform that bring a host of pro-focused changes and improvements.
Ahead of the Tuesday launch of the new MacBook Pro models, AnandTech has published their in-depth breakdown of the new system-of-chips.
The M1 Pro is a 10-core chip that features increased memory bandwidth and interfaces, and AnandTech notes that it's been designed from the ground up to feature more performance for professional users. The M1 Max is "essentially identical," except for the massive 32-core GPU.
AnandTech highlighted the "huge memory bandwidth" of the new chips, as also pointed out the fact that both chips are well ahead of the competition in terms of power draw and efficiency.
As far as CPU performance, AnandTech notes that Apple hasn't talked much about increased CPU performance, which is likely because not much has changed from the M1. Despite that, the M1 Max still lands as the top-performing laptop chips in AnandTech's ranking. It ranked below the AMD Risen 5950X as far as best overall CPU, but by a close margin.
The site calls the M1 Max's CPU multi-thread performance "a real monster." AnandTech says it's never seen such a design before, but that the M1 Pro and M1 Max will likely "vastly outperform" any other system out there for content creation and productivity tasks like video editing, audio mastering, or code compiling.
As far as GPU performance, the M1 Pro and M1 Max "should prove very welcome" because of significant real-world gains to graphical processing power. "There is a massive amount of pixel pushing power available in these SoCs, so long as you have the workload required to put it to good use," the site wrote. It did note that gaming is a poorer experience overall compared to similar Windows systems.
All in all, AnandTech concluded that the new MacBook Pro models are tailor-made for the specific workflows and use cases in which Macs shine. They will likely be the best laptops for those jobs on the market, the site added.
"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," the site wrote.
Read on AppleInsider
Credit: AppleInsider
The upgraded Apple Silicon chips, which power the recently announced 14-inch MacBook Pro and 16-inch MacBook Pro, are updates on the M1 platform that bring a host of pro-focused changes and improvements.
Ahead of the Tuesday launch of the new MacBook Pro models, AnandTech has published their in-depth breakdown of the new system-of-chips.
The M1 Pro is a 10-core chip that features increased memory bandwidth and interfaces, and AnandTech notes that it's been designed from the ground up to feature more performance for professional users. The M1 Max is "essentially identical," except for the massive 32-core GPU.
AnandTech highlighted the "huge memory bandwidth" of the new chips, as also pointed out the fact that both chips are well ahead of the competition in terms of power draw and efficiency.
As far as CPU performance, AnandTech notes that Apple hasn't talked much about increased CPU performance, which is likely because not much has changed from the M1. Despite that, the M1 Max still lands as the top-performing laptop chips in AnandTech's ranking. It ranked below the AMD Risen 5950X as far as best overall CPU, but by a close margin.
The site calls the M1 Max's CPU multi-thread performance "a real monster." AnandTech says it's never seen such a design before, but that the M1 Pro and M1 Max will likely "vastly outperform" any other system out there for content creation and productivity tasks like video editing, audio mastering, or code compiling.
As far as GPU performance, the M1 Pro and M1 Max "should prove very welcome" because of significant real-world gains to graphical processing power. "There is a massive amount of pixel pushing power available in these SoCs, so long as you have the workload required to put it to good use," the site wrote. It did note that gaming is a poorer experience overall compared to similar Windows systems.
All in all, AnandTech concluded that the new MacBook Pro models are tailor-made for the specific workflows and use cases in which Macs shine. They will likely be the best laptops for those jobs on the market, the site added.
"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," the site wrote.
Read on AppleInsider
Comments
Windows devices are still better in the vast majority of workflows and applications, most of that is due to the fact that the applications needed for these workflows just don't exist on macOS.
If you're going to buy one of these new MacBook Pro laptops, you're probably doing it for something like Final Cut Pro or similar. Even Premiere Pro might still be faster on a Windows laptop.
These are very specific workflows that Apple is targeting, but they're doing an exceptional job at it.
I think Dave Lee put it well, the new MacBook Pros are professional tools and they're expensive. Similar to the Mac Pro, unless you have a very specific workflow, you don't need it.
A challenge for gaming on the Mac has long been that the vast majority of Macs sold are laptops, and the vast majority of those laptops had crappy (for gaming) GPUs. Another challenge (pre-Metal) was a lack of software coherence.
Until those two technological problems could be solved, there probably wasn't much point trying to tackle the big problem of developer interest.
It's taken a *long* time, but now the technological problems appear solved. The vast majority of Macs being sold by Apple today have the hardware and software capacity to be good gaming computers. If Apple were to ever want to address the issue of game availability, this is the context in which to do it. Fingers crossed!
But, the chip itself doesn’t do more than be a partial determinant of the size of the package, which also includes the RAM, which are in separate chips on the same package.
In theory, both M1 Pro and M1 Max chips could be put on silicon with the same size and shape, but defects would have a larger impact on yield and drive up costs of the M1 Pro.
https://www.notebookcheck.net/NVIDIA-GeForce-RTX-3060-Mobile-GPU-Benchmarks-and-Specs.497453.0.html
This is the level most gaming computers are at and handles 1080p gaming at highest quality in almost every available game. This is pretty much PS5 level and can give very high quality output:
The M1 Max Macbook Pro is the most power efficient high-end gaming laptop. It's the software side that has always suffered with lack of games, poor ports etc. These Anandtech tests use the Rosetta versions of x86 games. Some overhead is ok though, the main thing is compatibility. For games that are over 3 years old, they can be handled with a compatibility library. For newer games, a lot of them are using DirectX 12, which isn't supported on Mac with the compatibility libraries.
https://www.pcgamingwiki.com/wiki/List_of_DirectX_12_games
Assassin's Creed Valhalla, Battlefield 2042, Black Ops Cold War, Call of Duty Vanguard, Cyberpunk, Death Stranding, Far Cry 6, Forza 3/4/5, Gears of War, Halo, Hitman 3, Horizon Zero Dawn, Red Dead Redemption 2, Resident Evil Village would need ports.
They run on PS4/5 without DirectX 12 so the developers have ported to the PS5 APIs and that's because the platform has a significant volume of gamers.
https://www.vgchartz.com/article/450848/ps5-vs-xbox-series-xs-vs-switch-sales-comparison-charts-through-september-18/
PS4 - 116m, PS5 - 11m
XBox One - 50m, XBox Series X - 7m
Switch - 91m
Apple would either have to build that size of gaming audience (just now Mac gamers would be around 2 million and mostly involved with casual gaming like The Sims), which wouldn't happen quickly, or they have to make the games as easy to port as possible and they so far have been doing the opposite. What would be useful to a lot of game developers is a translation layer that games build on top of. This is what the game engines do but for big AAA publishers, they often use custom engines so a code translation layer would help them port to other platforms. This is what Apple is trying to do with Web GPU to allow browsers to run the same code but target different lower level APIs like DirectX, Metal, Vulkan etc.
Porting to Metal manually is a no go for a lot of developers and we can see this from how long it took to port Redshift. They need better tools to make this process as easy as possible. Aspyr/Feral probably have tools to help with this.
If Apple wants to sort gaming out, they'd need to do a few things. Firstly get a compatibility layer up and running for older games, that should cover the majority of games 3 or more years old. For newer games, either commission ports from Aspyr/Feral/publishers or bring some of those in-house and make porting tools available. They can sort the situation out quite quickly but it seems that they just don't want to get involved. Valve has said this in the past:
http://archive.videogamesdaily.com/features/gabenewell_valve_iv_sep07_p1.asp
I expect Apple has changed a bit since then and their moves with Arcade and directly contacting game devs suggest they are more interested in it but they are more focused on iOS casual gaming. AAA desktop gaming is a much more costly and competitive environment than mobile and they won't be able to attract gamers if they don't get most or all of the good games.
I for one would be happy with an M1 Max Mac Mini. I already have an XDR display, and the M1 Max (maxed out) already outperforms my 2017 iMac Pro. I don't need another iMac, not a large tower machine. I guess we'll see sometime in the first quarter next year what's in store for the desktop.
Here's what I'm thinking:
Much like how AppleTV+ (service) has "billions" being thrown at original programming to create quality programming to watch on that service, the same kind of thing that shows big game companies how to make the most money coding something(s) exclusive for Apple Silicon would get big games coded for Apple Silicon.
But Apple seems to value that about like they value massive libraries of video content that sometimes is up for sale: near nill. Without some kind of artificial monetary incentive to develop for Apple Silicon, any game development motivations must be organic. The most likely path to Apple Silicon games for PRO & MAX are iPhone games being coded with some extra advantages when run on PRO & MAX. iPhone is where the money is for game development in any Apple tech, so it is "least common denominator" for target platform.
Why are many Mac games just iPhone games scaled up? Because that's where the money is.
How do brand new consoles launch with big, cutting edge games? Console makers pay up huge to motivate game developers to create them. Does Apple pay up huge to any game developers for exclusives? As far as I know, the biggest news of Apple and big game developers has been Apple vs. Epic... mostly fighting over about 15% profit, mostly for a VERY popular game from a few years ago. When it comes to game profits, it appears Apple wants their cut FIRST and right off the top even at the expense of banning what was the #1 game from the platform.
Some of us celebrate the fight, celebrate the lack of access to the banned game and celebrate the win(?)... like we won something too. But what message does that fight and win (or loss?) send to other big game developers?
What all businesses want is what Apple wants: more money, more money, more money. Show big game developers how to maximize profits coding for Apple Silicon and the games will roar onto the platform. Expect some kind of magic, organic uptake because programmers want to make much less money than they would building a big game for Windows or consoles... and not many are going to feel much motivation to go that way.
Apple seems to have the most applicable key to overcoming this issue. Will they ever turn that key?
Apple sells mostly laptops, and the integrated iris graphics of the past, and the Radeon mobile discrete GPUs in 15 and 16 inch MBP probably meant it wasn’t worth chasing gaming developers before this who target Nvidia. Maybe it will change now Apple is getting the hardware capability, although looks like we will have to wait for M2 or even further if we are hoping for hardware raytracing.
Microsoft put an axe in Mac gaming in the 90’s with the development of DirectX and getting GPU vendors to support it. And not creating a version of DirectX for Macs. That role reversed on mobile, with the popularity of iOS devices and then Apple’s Metal… But Metal was at first limited to those devices which were mobile. AAA games that needed hardcore GPUs had no reason to port to Metal and they still don’t as the Mac doesn’t have a gaming ecosystem. So most Metal games are just mobile games as that’s the market.
As Anandtech pointed out, most, if not all, higher-end games on Macs are ports. Until a studio comes along and creates a game from the ground up for Macs, they’re just not going to be gaming systems. If Apple wants to get serious about gaming (they don’t), they’ll have to buy or start a gaming studio to develop Metal first games and possibly even a new gaming engine and have it backed up by custom accelerators on their SOCs. Creating a gaming platform around an TV XL ($399 with M2 Max) wouldn’t hurt either as those games would be instantly available on Macs as well. Does the world want a new gaming platform though? Probably not.
This is typical of the Mac market though, always has been. Just too small to consider compared to Windows. Apple is going to have to step in and develop and support their own market and gaming platform.
I'd buy that ASAP.
As rumored, I think there will be two more M1 variants for the desktop; M1 "Ultra" with 16+4/64, and an M1 "Extreme" 32+8/128. The Ultra will end up in the iMac "pro" and will be released at WWDC. The Extreme will end up in the Mac Pro, announced at WWDC, but released later in the year.
To keep up excitement in the base M1 systems, We could see a redesigned Air in new colors this Spring. And we may see the lowest end M1 Pro (6+2/14) in a high-end model of the mini and 24" iMac (and possibly the 13" MBP).
The M2 will come in the Fall and be based on the A16 cores (completely skipping the A15 generation). This will give it a substantial performance/efficiency boost over the M1. It will make its first appearance in an updated iPad Pro and then in new 12" and 14" MacBook models.
One thing to consider about the Mac SoC's, is that Apple is going to push all these chips into as many different systems as possible and use them as long as possible. Unlike the A-series where there is enough volume for yearly turn around, the Mac sells far fewer numbers of units. This especially going to be true with the higher-end variants.
Unfortunately it's the chicken and egg scenario... A platform needs titles. You can't have one without the other. Microsoft had an advantage with the XBox, there was a huge library of Windows games that were easily ported over to it. (More like, moved to it, as the hardware was just a modified WIntel+nVidia computer.)
I do think, eventually Apple will set its sights on games, but not until after the transition is over and there's large enough base of higher end Apple silicon based Macs out there. Then start releasing titles for those systems (maybe even in an Apple Arcade+ type package). Then bring out a "cheap" console that can play those titles; like I mentioned, a $399 TV "XL" with an M2 Max. This allows those who don't need a $2000+ computer to play those titles. And then just roll with it.
The hardware is the easiest part and is basically already in place. The difficult part is the titles. Creating your gaming engine - they basically have the pieces for it already (SpriteKit, SceneKit, Metal, ARKit, etc), but would have to be scaled up. Then start developing your own titles or exclusively contract/finance 3rd party studios to develop them.
Apple needs to develop an entire gaming ecosystem running on Swift+Metal, essentially from zero. It's always "all of the above". The game studio talent is arguably more important then hardware. They have to throw hundreds of millions to iOS developers to make macOS games to spur new game development, then buy the promising talent and bring them inhouse. They should get classic games running, with modern polish, using native APIs.
Still seem quite uncertain that Apple wants gaming. They know they need games, and that it's a big weakness for Macs, but they've never actually committed to paying for it. They are committed to Apple TV+ with billions invested per year. Games will require the same level of investment, and are more important as they reinforce the platform in a much more complete way than streaming.