Next iPhone faster than M1 Ultra Mac Studio

[UBS28]

The iPhone 13 already has the same single-core performance as the M1 Ultra.

Hence the next iPhone 14 is going to be faster in single-threaded operation than the most expensive M1 Ultra.

That is going to be funny.

 

[senttoschool]

Yes, achieving max ST speed requires very little power with Apple Silicon. Even an iPhone could max the power.

That plus Apple has chosen not to do more overclocking on the MBPs and Studio. M1 ST is already overclocked compared to A14. Apple could go higher with the MBPs or Studio but I guess they want a consistent experience across all their Macs.

 

[Mr.Blacky]

Which tasks, do you think, will complete faster on an iPhone 14 compared to a Mac Studio M1 Ultra?

 

[senttoschool]

Which tasks, do you think, will complete faster on an iPhone 14 compared to a Mac Studio M1 Ultra?

Depends. If the iPhone 14 has faster ST speeds than the M1, and you're doing a short, pure ST task, then the iPhone 14 could actually be faster.

In most workloads, the Mac Studio would embarrass the upcoming iPhone 14.

Note that the OP is a bit of a troll. But the point is that Apple has chosen not to push ST speeds more on Macs when they could.

 

[PowerPCFan]

The iPhone 13 already has the same single-core performance as the M1 Ultra.

Hence the next iPhone 14 is going to be faster in single-threaded operation than the most expensive M1 Ultra.

That is going to be funny.

But isn't the M1 Ultra, like, 32 cores? It's way faster than the iPhone 13! (Maybe the same on 1 core, though.)

 

[leman]

In modern consumer-oriented computing, excellent single-core performance is essential to achieving smooth and responsive user interaction. So it makes a lot of sense to try and maximise ST in a phone. Apple's approach with predictable performance and linear horizontal scaling is a very nice touch IMO, especially compared to the usual anti-consumer business practices.

That plus Apple has chosen not to do more overclocking on the MBPs and Studio. M1 ST is already overclocked compared to A14. Apple could go higher with the MBPs or Studio but I guess they want a consistent experience across all their Macs.

What makes you think that M1 could go any faster in ST? It is very much possible that the design of Firestorm prevents it from achieving higher clocks. Maybe 3.26 ghz is as fast as it can go while remaining stable. That said, it is entirely possible that some individual chips could be clocked higher (binning), but Apple indeed chose to bin their chips horizontally rather than vertically.

 

[senttoschool]

What makes you think that M1 could go any faster in ST? It is very much possible that the design of Firestorm prevents it from achieving higher clocks.

This doesn't make sense. You can argue that the Apple wants to keep M1's clock speed within the sweet spot of TSMC's 5nm but you can't argue that M1 can't go higher in clock speed if Apple wants to. Yes, there will be diminishing returns because clock speed is not linear with performance. I'm guessing Apple can boost the clock speed to 4ghz+ if they decide to throw out the perf/watt advantage and probably approach 2000 points in Geekbench5.

For a Mac Studio, this would be welcomed.

While Zen4 is a narrower design compared to the M1, AMD is boosting it to 5.5ghz on TSMC's 5nm.

I'm guessing that Apple hasn't done it yet because of the following reasons:

* Consistent single-core performance from an iPad Air all the way to Mac Studio
* Apple hasn't had experience with boosting technology on desktop because they've been building mobile chips only. They could add an Intel or AMD-like boost to their MBP/desktop chips in the future.
* Wants to stay within the sweet spot of the power curve in order to tout maximum efficiency benefits

 

[leman]

This doesn't make sense. You can argue that the Apple wants to keep M1's clock speed within the sweet spot of TSMC's 5nm but you can't argue that M1 can't go higher in clock speed if Apple wants to. Yes, there will be diminishing returns because clock speed is not linear with performance. I'm guessing Apple can boost the clock speed to 4ghz+ if they decide to throw out the perf/watt advantage and probably approach 2000 points in Geekbench5.

Why doesn't Intel or AMD boost their CPUs to 6 or 7ghz then, power consumption be damned? It's not like high-end enthusiast PCs will have any problems with consuming 200W for single-threaded burst tasks and the cooling can definitely handle it (works just fine for multicore).

What I am trying to say is that peak frequency can be limited by the very design of the circuitry. My reasoning:

- Firestorm is a shorter core than x86 designs, and we know that synchronisation becomes increasingly more challenging with higher clocks for such designs
- Tests done by Anandtech on previous Apple CPU designs shows that power consumption grows exponentially close to the peak frequency, which suggest that there is not much space to go
- If Apple could easily boost frequency on M1 desktop configs, they would have done it already. Currently, x86 has a decent enough lead over M1 in ST, and if Apple would increase the ST power consumption by a factor of two or even three to catch up, they would still be ahead in power efficiency.

So yeah, I believe there are technical reasons why Firestorm is limited to 3.26Ghz. It probably simply can't go faster than that, as it just won't work.

 

[Kazgarth]

Which tasks, do you think, will complete faster on an iPhone 14 compared to a Mac Studio M1 Ultra?

Loading a web page.

 

[boak]

In single-core, yes. However, the M1 Ultra is mostly bought by customers that care about multi-core performance a lot more.

 

[ChromeCloud]

I don’t see any problem with that.

 

[Unregistered 4U]

The iPhone 13 already has the same single-core performance as the M1 Ultra.

Hence the next iPhone 14 is going to be faster in single-threaded operation than the most expensive M1 Ultra.

That is going to be funny.

Funny? Well, mostly to those who find the obvious funny

 

[Unregistered 4U]

If Apple could easily boost frequency on M1 desktop configs, they would have done it already.

Why? Their current offerings outperform all previous Macs and that’s all they need to do. The next Apple Silicon processors will be faster than the current ones. Some non-Apple processors will be faster, most will be slower, but Apple makes the only ones that run macOS so it doesn’t matter.

 

[MauiPa]

it is funny that anyone finds this funny

 

[Juraj22]

Which tasks, do you think, will complete faster on an iPhone 14 compared to a Mac Studio M1 Ultra?

breaking front glass

 

[bobcomer]

The iPhone 13 already has the same single-core performance as the M1 Ultra.

Hence the next iPhone 14 is going to be faster in single-threaded operation than the most expensive M1 Ultra.

That is going to be funny.

It just goes to show that single core performance is pretty irrelevant in a normal computing environment. I have no idea why some people think it's so important.

 

[pubb]

Which tasks, do you think, will complete faster on an iPhone 14 compared to a Mac Studio M1 Ultra?

Phone calls.

 

[JouniS]

Why? Their current offerings outperform all previous Macs and that’s all they need to do. The next Apple Silicon processors will be faster than the current ones. Some non-Apple processors will be faster, most will be slower, but Apple makes the only ones that run macOS so it doesn’t matter.

The people who really care about performance are rarely "Mac users". They are more likely to be people who happen to be using a Mac for the moment, because it's the best available option. If another platform is consistently better for a while, they can switch to it. If they prefer macOS, they can usually continue running the frontend on a cheaper Mac while offloading the computation to another device.

macOS is an important factor in consumer devices, where subjective user experience is the key. But when we are talking about computers as professional tools, Apple is in constant competition against other hardware manufacturers on raw performance.

 

[ian87w]

The iPhone 13 already has the same single-core performance as the M1 Ultra.

Hence the next iPhone 14 is going to be faster in single-threaded operation than the most expensive M1 Ultra.

That is going to be funny.

...for maybe a few minutes/hours at best and the iPhone will throttle down very quickly.
The higher number of performance cores and GPU along with the hardware encoders will keep the M1 Ultra separate from the iPhone. Different chips tailored for different devices.

 

[Unregistered 4U]

Apple is in constant competition against other hardware manufacturers on raw performance.

Apple has lost and will continue to lose to other hardware manufacturers on raw performance due to the flexibility of non-Apple platforms. Anyone right now that REQUIRES the utmost in raw performance over all else are not using macOS by default.

Because, it’s guaranteed that macOS systems don’t currently and will never provide the utmost in performance in any metric but one. And that’s running macOS and macOS applications.

 

[Mr. Dee]

I suspect Apple will bring the M series to the iPhone Pro to differentiate the A series iPhones. They did it with the iPad Pro, I see why not for the iPhone too.

 

[JouniS]

Apple has lost and will continue to lose to other hardware manufacturers on raw performance due to the flexibility of non-Apple platforms. Anyone right now that REQUIRES the utmost in raw performance over all else are not using macOS by default.

Because, it’s guaranteed that macOS systems don’t currently and will never provide the utmost in performance in any metric but one. And that’s running macOS and macOS applications.

Being competitive does not mean being the best on any single metric. It means offering good trade-offs that beat the competition in some niches. If you want to be competitive, you can't ignore easy gains, because such mentality leads to mediocrity. If the M1 can be clocked higher but Apple chooses not to, even in situations where it's not constrained by heat and power, it's a sign that Apple is no longer interested in producing competitive hardware, and the future of Macs looks uncertain. If the M1 just doesn't run reliably much above 3.26 GHz, we can still assume that Apple is trying to make the best hardware it can.

Mac apps are no longer that important, as it's not the 2000s anymore. Among the people I work, Keynote is pretty much the only Mac app they care about. Almost everything else is cross-platform, can be easily replaced, or runs in the cloud or in a browser. They don't want to buy new high-end MBPs or Mac Studios because those are Macs but because they want cost-effective performance in a convenient form.

 

[EntropyQ3]

The people who really care about performance are rarely "Mac users". They are more likely to be people who happen to be using a Mac for the moment, because it's the best available option. If another platform is consistently better for a while, they can switch to it. If they prefer macOS, they can usually continue running the frontend on a cheaper Mac while offloading the computation to another device.

macOS is an important factor in consumer devices, where subjective user experience is the key. But when we are talking about computers as professional tools, Apple is in constant competition against other hardware manufacturers on raw performance.

I fully agree.
However I’d like to add that raw performance is very far from the only metric of importance for computers used professionally.
I’d even go so far as to say that the performance deltas that seems to excite the computer tech media today (10-20%) are pretty much irrelevant in any setting I can think of. My experience is that you need at least a factor three before it makes an appreciable difference to workflows - well beyond what you can achieve by creeping upward on the frequency vs power curve for instance or (unfortunately) what you can expect from a couple of generations of lithographic process updates.

 

[leman]

Apple has lost and will continue to lose to other hardware manufacturers on raw performance due to the flexibility of non-Apple platforms. Anyone right now that REQUIRES the utmost in raw performance over all else are not using macOS by default.

Because, it’s guaranteed that macOS systems don’t currently and will never provide the utmost in performance in any metric but one. And that’s running macOS and macOS applications.

This is entirely untrue. One of the main reasons I use a Mac is because it’s the fastest portable computer (by a wide margin) for my type of work out there.

 

[senttoschool]

What I am trying to say is that peak frequency can be limited by the very design of the circuitry. My reasoning:

- Firestorm is a shorter core than x86 designs, and we know that synchronisation becomes increasingly more challenging with higher clocks for such designs
- Tests done by Anandtech on previous Apple CPU designs shows that power consumption grows exponentially close to the peak frequency, which suggest that there is not much space to go
- If Apple could easily boost frequency on M1 desktop configs, they would have done it already. Currently, x86 has a decent enough lead over M1 in ST, and if Apple would increase the ST power consumption by a factor of two or even three to catch up, they would still be ahead in power efficiency.

So yeah, I believe there are technical reasons why Firestorm is limited to 3.26Ghz. It probably simply can't go faster than that, as it just won't work.

In order to do this on, let's say the Mac Studio SoC, Apple would have to do extra work testing, validating, and binning chips for higher frequencies. In addition, they'd probably have to make some changes to MacOS specifically to handle boosting in various situations. For example, no boosting if Low Power mode is on.

It's a lot for Apple to do on the very first generation.

While Firestorm is a wider design (already mentioned by me), it can certainly boost much higher on a single-core. Even if the power is exponential, we're still talking very little power here. When running Geekbench5 ST, my M1 Pro mostly stays around 0.3w - 1w of power. Meanwhile, AMD and Intel boost to probably 10x or higher on their laptop chips.

I've been overclocking chips for 15 years. Any chip can be overclocked and you'd likely see extra performance. IE. Back in the days, my AMD Athlon XP would run at 1.8Ghz but you could overclock it to 2.5ghz. I doubt Firestorm is an exception.

But doing official binning is extra work for the chip makers.