M3 Max Chip Has Hidden Change That Could Affect Future 'M3 Ultra' Chip

[MacRumors]

Apple's M3 Ultra chip may be designed as its own, standalone chip, rather than be made up of two M3 Max dies, according to a plausible new theory.

The theory comes from Max Tech's Vadim Yuryev, who outlined his thinking in a post on X earlier today. Citing a post from @techanalye1 which suggests the M3 Max chip no longer features the UltraFusion interconnect, Yuryev postulated that the as-yet-unreleased "M3 Ultra" chip will not be able to comprise two Max chips in a single package. This means that the M3 Ultra is likely to be a standalone chip for the first time.

This would enable Apple to make specific customizations to the M3 Ultra to make it more suitable for intense workflows. For example, the company could omit efficiency cores entirely in favor of an all-performance core design, as well as add even more GPU cores. At minimum, a single M3 Ultra chip designed in this way would be almost certain to offer better performance scaling than the M2 Ultra did compared to the M2 Max, since there would no longer be efficiency losses over the UltraFusion interconnect.

Furthermore, Yuryev speculated that the M3 Ultra could feature its own UltraFusion interconnect, allowing two M3 Ultra dies to be combined in a single package for double the performance in a hypothetical "M3 Extreme" chip. This would enable superior performance scaling compared to packaging four M3 Max dies and open the possibility of even higher amounts of unified memory.

Little is currently known about the M3 Ultra chip, but a report in January suggested that it will be fabricated using TSMC's N3E node, just like the A18 chip that is expected to debut in the iPhone 16 lineup later in the year. This means it would be Apple's first N3E chip. The M3 Ultra is rumored to launch in a refreshed Mac Studio model in mid-2024.

Article Link: M3 Max Chip Has Hidden Change That Could Affect Future 'M3 Ultra' Chip

 

[chucker23n1]

This is all possible, but the entire thread hinges on “the M3 Max looks, in a blurry, cropped photo, like it might not have an interconnect”, which also hinges on “the M3 Ultra’s interconnect looks exactly like the M1 and M2’s”.

 

[falkon-engine]

Given that the m4 generation will likely have big modifications to the neural engine to support AI, and likely thunderbolt 5 /usb4v2, I would be very wary of an m3 ultra chip if it doesn’t have those features. In fact, I would skip it.

 

[mcled53]

(I'm not a chip designer.) I wonder if the M3 Ultra could have the ultra connection on both sides allowing a 3+ configuration.

 

[Thunderstrike76]

If there’s one YouTubers videos I roll my eyes constantly at it’s MaxTech.

 

[SoldOnApple]

Put it in a MacBook you cowards!

 

[leman]

When will we finally stop confusing popularity with competence? It is entirely possible that this theory is correct, however, it is unlikely to be correct simply because Yuryev said so. He knows a great deal about video editing, general computing and semiconductor technology are not really his forte though.

 

[imtoretto]

Whatever new M series comes out later on, JUST PLEASE DO YOURSELF A FAVOR AND DONT WATCH IJUSTINE FOR THE REVIEW.

 

[PotentPeas]

I'm skeptical. The M3 Max die is already huge! They're going to have a 2× bigger single die for M3 Ultra? That seems crazy.

There are a number of options that could pan out.
* The interconnect is there but not clearly visible in this blurry image.
* The interconnect has changed in some way and it is not clear what we should be looking for.
* There is a different variation of the chip that has the interconnect that we haven't seen yet.
* M3 Ultra will be significantly scaled back from what we expected based on M1/M2 Ultra, being a more modest upgrade over M3 Max.
* M3 Ultra isn't even going to make an appearance this generation.

 

[EugW]

When will we finally stop confusing popularity with competence? It is entirely possible that this theory is correct, however, it is unlikely to be correct simply because Yuryev said so. He knows a great deal about video editing, general computing and semiconductor technology are not really his forte though.

"not really his forte"

Heh. You are being very, very generous with that description.

 

[hoorayforhollywood]

Oh, hello!! So if the Ultra is standalone, and - as the article suggests - has the POSSIBILITY of a new UltraFusion connection.... Then that would be the mythical (but reported on) M3 Extreme.

That would not only allow the MacPro to be a valid machine (if Extreme was limited to it exclusively), but presumably that redesign would allow for say - cards and stuff? Which would justify keeping the Mac Pro chassis in production in the first place. #MacPro

 

[sunny5]

SoC design is a failure and that's why Apple can't make beyond Max chip. They really need to use chiplet design so that they can make whatever they want.

Whoever disagree means you know NOTHING about it.

 

[ph001bi]

I read this on X earlier and was vaguely excited. Now I realise it's from Max Tech and the excitement. If there is one YouTuber who gets it wrong 100% of the time, it's Max Tech. Pass

 

[Bobbuilds69]

Whatever new M series comes out later on, JUST PLEASE DO YOURSELF A FAVOR AND DONT WATCH IJUSTINE FOR THE REVIEW.

You are bout to experience the wrath of the iJustine lovers 🤣

 

[manumacs]

Should I buy the M2 Mac Mini now or is there a real hope for WWDC, there are zero news on it

 

[EugW]

Should I buy the M2 Mac Mini now or is there a real hope for WWDC, there are zero news on it

There's always hope, but no guarantees.

 

[boak]

What a reliable source!

/s

 

[HudsonSteele]

When will we finally stop confusing popularity with competence? It is entirely possible that this theory is correct, however, it is unlikely to be correct simply because Yuryev said so. He knows a great deal about video editing, general computing and semiconductor technology are not really his forte though.

Agreed! I watch his videos from time to time and it reminds me why I don’t watch his videos. They are usually completely worthless in terms of usable information for me.

 

[gatorvet96]

The M series of chips is further evidence that Apple keeps innovating despite all the cries on here from naysayers. People complain that the iPhone hasn't innovated, yet no one gives examples of what they could do. Just a mature product.

 

[leman]

SoC design is a failure and that's why Apple can't make beyond Max chip. They really need to use chiplet design so that they can make whatever they want.

They've been using chiplet designs for years. What do you think Ultra is?

 

[Analog Kid]

If MaxTech says it, it's probably wrong.

 

[Contact_Feanor]

according to a plausible new theory.

​

The theory comes from Max Tech's Vadim Yuryev,

Plausible new theory from Max Tech. yeah right... Max Tech is usually over-the-top clickbait when it's reporting about anything Apple related.

 

[blastdoor]

I'm skeptical. The M3 Max die is already huge! They're going to have a 2× bigger single die for M3 Ultra? That seems crazy.

There are a number of options that could pan out.
* The interconnect is there but not clearly visible in this blurry image.
* The interconnect has changed in some way and it is not clear what we should be looking for.
* There is a different variation of the chip that has the interconnect that we haven't seen yet.
* M3 Ultra will be significantly scaled back from what we expected based on M1/M2 Ultra, being a more modest upgrade over M3 Max.
* M3 Ultra isn't even going to make an appearance this generation.

I sure hope it isn’t the last one, but that was the first thought that occurred to me.

Another thought is that Apple could move to an entirely different approach for the high end desktop. Have one die that has lots of CPU cores (plus other basic SOC stuff) and move the GPU to its own die that connects via ultrafusion. That would allow for more customization in terms of cpu to gpu balance.

It would probably not make sense to develop such a highly custom design for the existing high end Mac desktop market. But if Apple were to also use such a design for AI model training in their own data centers, then it might make sense

 

[AndiG]

ultra. extreme, max …. What matters today is the number of ML/Cuda Cores.

If you want to know what innovation looks like:

 

[novagamer]

Given that the m4 generation will likely have big modifications to the neural engine to support AI, and likely thunderbolt 5 /usb4v2, I would be very wary of an m3 ultra chip if it doesn’t have those features. In fact, I would skip it.

Not to mention the inherent security flaw (GoFetch) in all M series chips. The M3-only mitigation is not sufficient because it requires developers to opt-in and change their code when performing specific operations, and Apple didn't even disclose this functionality existed until last week. Javascript can execute an attack that steals data, I am not satisfied that Apple is only disabling the Prefetcher when performing encryption operations (which only works on M3 chips), and also that it doesn't (so far?) seem like a bug bounty was paid out for this is which is incredible given how significant the flaw is.

Unless the Ultra has some logic redesign specifically addressing this I would absolutely not spend a ton of money on a decked-out pro machine which was my plan. I'll either hobble along with my old Mac for another year or get a base level M3 Studio and replace it as soon as M4 ones are available.

I completely agree about your points also, the neural engine is likely to be massively upgraded in M4.

Edit: since people are challenging this (which is understandable, it is a complicated issue) I'll attach the conclusion of the research work. You can also see my follow-up posts in this thread for a bit more vague background. The most severe issue is getting security keys, which is what news coverage has focused on rightly because it is the biggest problem, but the scope of the flaw is not limited to that.

This part in particular, which is not in the screenshot below but is on page 1 of the abstract elucidates my concern:

"Undergirding our attacks is a new understanding of how DMPs behave which shows, among other things, that the Apple DMP will activate on behalf of any victim program and attempt to “leak” any cached data that resembles a pointer."