Apple's footnotes don't even pretend to explain what these charts are.

> Performance was measured using select industry‑standard benchmarks.

The M1 announcement. They turned out to be pretty accurate.

So I’ll wait and see real benchmarks, but it wouldn’t surprise me if this does have incredible performance.

The problem was the implication that you'd get 3070 gaming performance. That was never going to be true because of the un-optimisation tax for games on Mac.

There doesn't exist a AAA game built for Metal and the Mac. The closest are games like World of Warcraft, Divinity Original Sin 2 – and even they are just "good ports" not originally designed for Mac (and are far from AAA graphics). This is why on Intel Macs, games under Bootcamp always ran 30%-50% faster, even though the hardware was the same.

Games on M1 Max run as you'd expect – about 30% slower than a 3070 for the same old reasons (and some new ones, like not being compiled for Apple Silicon at all). The GPU is about the same speed as a 3070 and it's doing what you'd expect, given the 30% unoptimization-tax workload.

- the workloads in games can vary a lot, vertex/fragment shaders imbalance, parallel compute pipelines, mixed precision (which the M1 gpu does not do), .. So another explanation is that you can get some 3070 parity on a cherry picked game, like a broken clock is right twice a day, but that does not make it generally true. Objective benchmarks have put the M1 gpus way slower than 3070 on average, and software support seems like an easy but false distraction given the Proton tax on Linux (which is not 30/50%)

- the M1 gpus are lacking a ton of hardware, matrix mul, fp16 again, ray tracing, VRR probably (not sure about this last one). These are used by modern games or applications, you may find a benchmark which skip them, but in the grand scheme of things it's something that the M1 gpu will have to emulate more often than not, and this has a cost

Waving all that as "the GPU is about the same speed" is technically wrong, or not really backed by facts at the very least

My understanding was that it was a reasonable comparison in some benchmarks, though when plugged in to power the mobile 3070 still had more headroom.

A lot of them, particularly the GPU benchmarks, were misleading because they only looked at performance that they had dedicated silicon for.

The advancements in such a short period of time in the amount of computing power, low power usage, size, and heat usage of these chips is unbelievable and game-changing.

The performance of x86 has been a leader for a while mostly because of the sheer amount of optimization work that has gone into them, but the cruft of the x86 instruction set and the architectural stuff you have to do to make the instruction set work is really showing it's age.

That being said, the GPU performance claims are incredibly misleading. The previous "relative performance" benchmarks that were done on the M1 Max for GPU performance were misleading as well, they definitely cannot keep up with a mid-tier modern discrete GPU.

The GPU claim isn't an ARM/x86 comparison like the CPU performance would be. This is comparing a 64 core 800GB/s GPU with a 10k core 900GB/s GPU and trying to make them look equivalent through misleading marketing.

None of this is to say that the M1 Ultra is bad necessarily, even if it performs roughly the same as a mobile GPU or powerful iGPU it would still be a very good chip, and I'd love to use one if I could use it in my environment properly. I'm just saying don't put too much faith in the GPU performance measurements provided here.

Without denying some good work and engineering having gone into some x86 chips, they are not the reasons for the x86 becoming a leader. The duopoly of Intel and Microsoft – coupled with the aggressive Intel strategy to undermine competitors on the pricing and the sheer production volume they could quickly ramp up – has squeezed every single other viable competitor out of the market and relegated very few to become niche players (e.g. POWER) and entrenched the duopoly as a unfortunate leader. And then complacency and arrogance had set in for years to come until recently.

This was probably further fueled by their soured relationship with Intel which was responsible for thermal issues on MBP's for years, poor performance increases across their entire Mac lineup, and poor cellular radio performance on some iPhone models -- forcing them to settle with Qualcomm and ditch Intel for mobile radios.

At least in the US You can get a 3090 for 2200 even with markup and an almost as good 3080 for $1150. If you wait until its in stock at a big box store you can get one for less even. A machine could be built with a 3080 for $2000.

Meanwhile a system based on on the 64 core GPU will run you $5000 and as such is affordable to nearly nobody thus few will get a chance to see it drastically under perform in the gaming arena on any of the games that don't support mac on arm.

With an absolutely invisible market share in the gaming desktop there will never be any incentive for anyone to change this insofar as direct support leaving you reliant on translation from x86 and from windows executables paying doubly in terms of compatibility and performance from an already very expensive and lackluster starting point.

The Studio is a more compact form factor than any modern 4K gaming console. If they chose to ship something in that form factor with tvOS, HDMI, and an M1 Max/Ultra, it would be a very competitive console on the market — if game developers could be persuaded to implement for it.

How would it compare to the Xbox Series X and PS5? That’s a comparison I expect to see someday at WWDC, once they’re ready. And once a game is ported to Metal on any Apple silicon OS, it’s a simple exercise to port it to all the rest; macOS, tvOS, ipadOS, and (someday, presumably) vrOS.

Is today’s announcement enough to compel large developers like EA and Bungie to port their games to Metal? I don’t know. But Apple has two advantage with their hardware that Windows can’t counter: the ability to boot into a signed/sealed OS (including macOS!), load a signed/sealed app, attest this cryptographically to a server, and lock out other programs from reading with a game’s memory or display. This would end software-only online cheating in a way that PCs can’t compete with today. This would also reduce the number of GPUs necessary to support to one, Apple Metal 2, which drastically decreases the complexity of testing and deployment of game code.

I look forward to Apple deciding to play ball with gaming someday.

They could always choose to remedy that with a generous buyout offer.

Edit: https://arstechnica.com/information-technology/2022/01/pluto... says

> Microsoft already used Pluton to secure Xbox Ones and Azure Sphere microcontrollers against attacks that involve people with physical access opening device cases and performing hardware hacks that bypass security protections. Such hacks are usually carried out by device owners who want to run unauthorized games or programs for cheating.

So initially you could have Pluton-only servers and down the line non-Pluton hardware will simply be obsolete.

They won't have the Ultra GPU, but Apple's been shipping for years and Microsoft is just now bringing Pluton to market. I do wish them luck, but that's a lot of PC gamer hardware to depreciate.

They also don't need the display, camera, microphone. And could sell it at a loss and make the margins with TV+ and game sales.

But they would need their own bundled controller/accessories and get serious about AAA gaming.

Also, might be cheaper a couple years down the line.

I wish.

But playing ball is more than hardware. It is spending billions to buy Activision or Bungie. And I can't honestly imagine Apple having the cultural DNA or leader aspiration to make a game like The Last of Us where the player is brutally beating zombies to bloody clumps.

In video games the business side demands having exclusives, or timed exclusives, to sponsor twitch streamers playing your game and cutting special deals with studios. This is very different to the App store where Apple emphasizes their role as a neutral arbiter and a dev having the same deal as any other dev. Can you imagine the complains here on hn if Epic Games would get a a special deal just because they are a bigger fish and Fortnite is popular?

I could definitely see the Max and Ultra with a beefier cooling system (like the Studio’s) having pretty respectable performance.

M1 Max struggles to keep up with an RTX 3060 mobile.

Now that's with the overhead of Rosetta 2 and all that, so it's of course "not fair" for the M1. But that's also the current reality of the market, so ya know.

So publishers/developers need to make more native games. Even though every Mac port will probably make 1/10th revenue of a Windows title I guess Mac users would be happy to pay more for better games. I certainly would.

In the PC space the average spend is $800 and a PS5 is $500.

Yet the iOS 'gaming' scene, despite being one of the major revenue drivers, consists mostly of low-quality F2P games.

In any case, it's a moot point. Apple clearly doesn't care about desktop gaming and it shows in both their hardware and software.

Though every video game company on the planet hates them because of App Store terms.

This criticism is something that is a positive to me. Opposing companies are often dependant on adverting money and the things this leads to are a whole lot worse in my view.

I remember people saying this about phones in 2006.

[0] https://www.youtube.com/watch?v=Tzrme9yWens

Apple has invested billions into their gaming division. The big thing they need right now is a new version of Metal that gets feature parity with Vulkan or DX.

Also of note, there are very persistent rumors of an upcoming VR headset. Their M1 alone would blow away competition like the Quest. A Pro or max chip with some disabled CPU cores wouldn't cost a ton due to being scavenged cores and would positively stomp the competition.

[1] https://www.youtube.com/watch?v=kjNaC0-hiPE

Proton on ARM Mac's would involve Rosetta and while that does a surprisingly good job of running x86 on ARM I'm not sure it's up to the job of running games at high speed...

Some games are a stable 4k120 and others are more like 4k75.

I feel the 3090 could feasibly drive 5k60 as a result.

GTAV on Win11 Arm VM - okay, but not great.

GTAV on Crossover - much much better, lack of joystick support (but that's a crossover issue)

Not to say it'll never happen, but its not a done deal basically, and to my knowledge the process hasn't yet started

I run VR games on the index at 144hz with high settings without issue on a 3060 Ti.

I've been on the market for a 3070-3090, but only because I want a card for which a water block is available, not because I need more power for any extant game.

Really looking forward to Apple's VR offering after seeing the performance of their compact SoC

So expect what that graph actually means is some extremely specific, cherry-picked benchmarks