'M2 Max' Geekbench Scores Surface Again Online Ahead of Release in 2023

[MacRumors]

Last month, Geekbench scores for an unannounced Mac running the upcoming M2 Max chip surfaced online, showing only minor performance increases compared to the M1 Max. Now, another set of scores claiming to be for the M2 Max chip has surfaced online, showing a larger jump in performance.

In the Geekbench scores last week, the M2 Max chip scored 1,853 in single-core and 13,855 in multi-core, representing only a minor jump compared to its predecessor. Now, in a new set of scores alleged for the M2 Max, the chip scored 2,027 in single-core and 14,888 in multi-core. For reference, the M1 Max chip achieves 1,755 in single-core and 12,334 in multi-core.

The new Geekbench scores offer no further details over any upcoming Macs, which we expect first to be 14-inch and 16-inch MacBook Pros. The scores list the chip as running on a Mac with an identifier "Mac14,6" with 96GB of memory and running macOS Ventura 13.2. The only difference between today's scores and last month's ones is that the updated scores show an M2 Max chip with a higher base frequency of 3.68 GHz compared to 3.54 GHz, which could explain the higher scores.

Apple was initially expected to announce updated Macs with M2 Pro and M2 Max chips in November this year, but the company pushed plans until early 2023. Apple has multiple new Macs in the work, including an updated iMac and MacBook Pro.

Article Link: 'M2 Max' Geekbench Scores Surface Again Online Ahead of Release in 2023

 

[Leon Ze Professional]

Now we're talking! Breaking the 2000 barrier for single core.

 

[playtech1]

Interesting: by comparison I see the i9 13900k scores about 2,206 on single core and 23,786 in multi-core, so is about 10% faster in single core and about 60% faster in multi-core.

This would still be very impressive for the M2 Max, particularly in a laptop, if the power consumption remains as low as it has been in the M1 Max.

Still perhaps a bit of a worry on the desktop if this is the fastest desktop chip outside of ultra-expensive M2 Ultra. Power efficiency on the desktop is simply not as big a deal.

 

[Tankmaze]

If this is the 3nm, then it will be worth the wait for M2 Pro !!!

 

[David1986H]

That's more like a score I was hoping for, perhaps not as high as I would like but way better then last months score.

Maybe they had something else running in the background last time plus over time progress in software should increase the score even further.

Knowing that 13.2 isn't even out to developers yet I still have high hopes for the M2 PRO and MAX even if its still on 5nm

 

[Eagle_Eye]

Interesting: by comparison I see the i9 13900k scores about 2,206 on single core and 23,786 in multi-core, so is about 10% faster in single core and about 60% faster in multi-core.

This would still be very impressive for the M2 Max, particularly in a laptop, if the power consumption remains as low as it has been in the M1 Max.

Still perhaps a bit of a worry on the desktop if this is the fastest desktop chip outside of ultra-expensive M2 Ultra. Power efficiency on the desktop is simply not as big a deal.

Don't be fooled by Intel benchmarks. Mostly run on overclocked and overcooled CPU's, consuming hundreds of Watts of power. If you'd put such a chip in a laptop, it would overheat instantly and than throttle like crazy. The battery wouldn't last for 30 minutes either...

 

[Isengardtom]

Seems a bit odd as I would expect single core of M2 - Pro - Max to be more or less the same as the normal M2

Could this be M3?

 

[David1986H]

Seems a bit odd as I would expect single core of M2 - Pro - Max to be more or less the same as the normal M2

Could this be M3?

As the reports says the new score had a faster base clock at 3.68ghz, where as the normal M2 has a 3.49 ghz base clock.

 

[xxray]

So 20.7% faster in multicore and 15.5% faster in single core. Doesn’t seem minor to me.

 

[philcourage]

Can't wait for this chip to get to the iPad Pro. Imagine how much faster browsing the web while on the toilet would be. Amazing.

 

[TheYayAreaLiving 🎗️]

I think at this point it is worth waiting for the M2 Max chip till next year. Pure Performance!

 

[iBluetooth]

If this is the 3nm, then it will be worth the wait for M2 Pro !!!

It's the 4nm TSMC process

 

[playtech1]

Don't be fooled by Intel benchmarks. Mostly run on overclocked and overcooled CPU's, consuming hundreds of Watts of power. If you'd put such a chip in a laptop, it would overheat instantly and than throttle like crazy. The battery wouldn't last for 30 minutes either...

I absolutely agree - in a laptop. There's a reason I own an M1 MBA, M1 Pro 14 and M1 Max 16, but no Windows laptops!

On the desktop though I have an i9 12900K / 4900 for a bit of work and gaming and in normal desktop use it's not even all that power hungry, but can really chew through the watts when you need them.

I would love to see Apple put out a chip that is as unconstrained, as I just don't think desktop power consumption is that big a deal.

 

[TheYayAreaLiving 🎗️]

Can't wait for this chip to get to the iPad Pro. Imagine how much faster browsing the web while on the toilet would be. Amazing.

I'm guessing it will handle Stage Manager just fine. Oh wait! It will own Stage Manager 🤪

 

[iBluetooth]

Seems a bit odd as I would expect single core of M2 - Pro - Max to be more or less the same as the normal M2

Could this be M3?

They are probably tuning each CPU for the intended usage and a Mac Pro will have a much better cooling budget compared with a laptop. I am getting excited for next year!

 

[t0rqx]

Cool that the performance is higher, but will this be the end of the noiseless and cool experience which people have had with the M1?

 

[enterthemerdaverse]

So 20.7% faster in multicore and 15.5% faster in single core. Doesn’t seem minor to me.

It’s 20% faster with current software and then all the apps and system gets updated and after a year everything is slower again 😂😜

 

[Zest28]

So 20.7% faster in multicore and 15.5% faster in single core. Doesn’t seem minor to me.

3.7ghz (m2 max) / 3.2ghz (m1 max) = 16%.

So it's a very small "upgrade" since it is mostly from overclocking. This M2 Max is going to run hotter and sip more energy under stress.

If you needed M2 Max performance, you could have gotten it with a simple overclock of the M1 Max if Apple allowed it through firmware settings, rather than buying a whole new laptop.

 

[tobybrut]

That's more like a score I was hoping for, perhaps not as high as I would like but way better then last months score.

Maybe they had something else running in the background last time plus over time progress in software should increase the score even further.

Knowing that 13.2 isn't even out to developers yet I still have high hopes for the M2 PRO and MAX even if its still on 5nm

It was an unreleased OS version, so it was probably the initial benchmarks before Apple had the opportunity to optimize the OS for the new chip. The later scores are probably with optimizations on a later OS version. The final release may be even better if Apple plans additional optimization.

 

[triton100]

Come to papi my delicious M2 Max

 

[Two Appleseeds]

It was an unreleased OS version, so it was probably the initial benchmarks before Apple had the opportunity to optimize the OS for the new chip. The later scores are probably with optimizations on a later OS version. The final release may be even better if Apple plans additional optimization.

This. The 13.5% single core and 17.2% multicore gains are more like the optimization we see from the m2 over the m1. Now the graphics from the base m1 8 core (2.6 teraflops) to the base m2 10 core (3.6 teraflops) had a 28% jump. If this kind of gain for the m2 pro/max chips can be had or better with optimized software and power efficiency, I wouldn’t call this a small update at all. Sure, people with the m1 pro/max chips won’t need to upgrade. But for those like myself, this update might be pretty nice indeed. As Intel, Nvidia, and Amd put out faster cpus and gpus hotter and more power hungry, Apple is keeping its silicon cool and efficient. Low key impressive indeed.

 

[bradman83]

It was an unreleased OS version, so it was probably the initial benchmarks before Apple had the opportunity to optimize the OS for the new chip. The later scores are probably with optimizations on a later OS version. The final release may be even better if Apple plans additional optimization.

This. The 13.5% single core and 17.2% multicore gains are more like the optimization we see from the m2 over the m1. Now the graphics from the base m1 8 core (2.6 teraflops) to the base m2 10 core (3.6 teraflops) had a 28% jump. If this kind of gain for the m2 pro/max chips can be had or better with optimized software and power efficiency, I wouldn’t call this a small update at all. Sure, people with the m1 pro/max chips won’t need to upgrade. But for those like myself, this update might be pretty nice indeed. As Intel, Nvidia, and Amd put out faster cpus and gpus hotter and more power hungry, Apple is keeping its silicon cool and efficient. Low key impressive indeed.

The M2 uses the same instruction set as the M1 and isn’t overly different architecturally, so there really isn’t anything optimization that could be done for the M2 Max that wouldn’t also benefit an M1 series chip.

As the article noted the performance increase is seemingly due to a higher clock speed.

 

[Denzo]

All that power.. for safari and pages

 

[RamGuy]

With modern CPUs, raw performance numbers aren't all that valuable or exciting anymore unless you want something to stick inside a moderately large desktop with great cooling. I have an Intel Core i9-12900K on my desktop, and its performance will vary greatly depending on what system you are running it on. It's running unrestricted in terms of power usage, so depending on where you set your limitations for cooling and noise is going to have a significant effect on what performance numbers you are going to be getting. In my case, I have limitations as I don't want any noise, thus limiting how far the performance can extend even though I have great cooling and could have let it run rampant but at the cost of having active noise from my desktop.

In most scenarios, even on desktops, you will have cooling restrictions when looking at the top-of-the-line Intel SKUs. Whereas the M-series from Apple is far more efficient and constrained, so you know what to expect, and you know you will be getting it without having to have a large desktop with extremely active cooling to achieve its performance.

Especially for laptops, efficiency is king. With such limited space for active cooling and the fact that most people would prefer a combination of excellent performance and great battery life, it's all about how much performance you get per wattage. Without knowing how these M2 Max numbers are in terms of power usage, we lack critical information to make any judgement. A 10% improvement in performance is not impressive, but if you are gaining 10% performance while you are reducing the power usage by 30% at the same time, then it's suddenly more impressive.

With the M-series chips, there are additional things to consider as well. As both memory and GPU are on the same SoC, you also have to take into consideration what kind of improvements it might bring on these areas as well, not just looking at the CPU numbers alone.

 

[bradman83]

3.7ghz (m2 max) / 3.2ghz (m1 max) = 16%.

So it's a very small "upgrade" since it is mostly from overclocking. This M2 Max is going to run hotter and sip more energy under stress.

If you needed M2 Max performance, you could have gotten it with a simple overclock of the M1 Max if Apple allowed it through firmware settings, rather than buying a whole new laptop.

Yes and no. The Blizzard and Avalanche cores in the A15 and M2 series are generally more efficient than the the M1/A14 equivalents despite being clocked higher. The biggest design changes with the A15 and M2 is that the efficiency cores are considerably faster, meaning slightly fewer tasks need to rev up the much more energy-hungry performance cores. That’s not to say significantly upping the clock speed won’t cause the M2 Max to run hotter or consume more energy, only that the changes Apple made to the cores can offset that; it’s not like simply overclocking an M1 Max and getting the same result.