First M2 benchmarks

[Santabean2000]

Yes I do agree with that. Notice how M2 did not gets its own event because it was not a good CPU increase...

A12Z to M1 was 50% faster.

M1 to M2 was 18% faster.

I think M3 will get its own event in 2023 like M1 did because new node, and new architecture.

The A12Z was an iPad chip, (more or less the same as the iPad Pro's 2018 A12X [+ 1 GPU core]). The M1 is the first Mac chip, so...

Let's be honest M2 was underwhelming for Apple standards. M3 should be more exciting.

The CPU possibly, but if you look more closely you'll see that Apple focused mainly on the power of the e-cores [rather than P-cores] and the gains there are more in the 30-35% range, so actually quite good.

The GPU is also a lot better. And now ProRes video engine too.

The end result being we'll have a stronger machine overall .

 

[l0stl0rd]

The M2 alone is still a good evolution over the M1 thanks to its lpddr5, bigger bandwidth , over 30-40% gpu
But yes, when it comes to the M2 pro/max IF they will be just scaled up M2, that will be a less of an upgrade
In pc era, i dont see over course of 18 months over 30-40% improvements in gpu department
So for now, the M2 is good enough for the new Macbook Air, its a complete package

I thing it is funny when people praise 30-40 % better gpu but it has 25% more cores, so without those it would not be much better then M1 and we would have 5-10%.

Also if you want to have a laugh check the fan noise of the M2 in Lukes video when he is running Cinebech.
Sure will be interesting how they do in the 14 and 16”.

 

[Pressure]

M2 is just an overclocked and bigger M1 as the M2 apparently thermal throttles in the MacBook Pro with a fan.

Better wait for the M3 MBA next year which will be much cooler on 3nm.

Guess this also answers all the complaints about why Apple kept the 13” MBP around, the M2 needs a fan.

This is just plain false. The entire core architecture is using the newer cores from the A15 Bionic in the M2.

The M1 uses the Firestorm (high-performance) and Icestorm (high-efficiency) cores first found in the A14 Bionic.

The M2 uses the Avalanche (high-performance) and Blizzard (high-efficiency) cores as found in the A15 Bionic.

This gives M2 the better Neural Engine (+43% faster performance) and better media engine with support for ProRes and ProRes RAW. The biggest gainers in that iteration are the high-efficiency cores and the neural engine though with only a modest improvement for the high-performance cores.

 

[UBS28]

This is just plain false. The entire core architecture is using the newer cores from the A15 Bionic in the M2.

The M1 uses the Firestorm (high-performance) and Icestorm (high-efficiency) cores first found in the A14 Bionic.

The M2 uses the Avalanche (high-performance) and Blizzard (high-efficiency) cores as found in the A15 Bionic.

This gives M2 the better Neural Engine (+43% faster performance) and better media engine with support for ProRes and ProRes RAW. The biggest gainers in that iteration are the high-efficiency cores and the neural engine though with only a modest improvement for the high-performance cores.

Might be, but it behaves as a bigger overlocked M1 as it draws more power and heat up to the point that the M2 even thermal throttles in the 13” MBP.

The M1 in the 13” MBP doesn’t thermal throttle.

So the real successor to the M1 will be the 3nm M3 next year.

 

[UBS28]

According to what data?

Look at Max Tech his video. The M2 cannot keep it’s base frequencies in the 13” MacBook Pro with a fan.

The M2 MacBook Air will be fun to see. That laptop will catch fire with the M2 as it has no fan.

 

[Bodie CI5]

Does anyone else have any more information on the SSD speeds on the M2 MBP? Alex Ziskind shows Blackmagic’s Disk Speed Test With some appalling figures. (Apologies if I’ve posted this in the wrong thread)

 

[UBS28]

Does anyone else have any more information on the SSD speeds on the M2 MBP? Alex Ziskind shows Blackmagic’s Disk Speed Test With some appalling figures. (Apologies if I’ve posted this in the wrong thread)

Even if it was 3000MB/s, that would still be a joke, due to how much Apple charges for SSD’s.

You can buy internal 1TB SSD’s that do 7000MB/s like the M1 Max does for only $150 from Samsung.

 

[JimmyjamesEU]

I don't see why this is relevant. They're not comparing Apple to Intel, but M2 to M1.

Because if cinebench isn’t efficient on a platform, you can throw all the cores you want at it and you won’t see much improvement. That Just tells you cinebench is leaving performance on the table for both the M1 and M2. There’s a reason it’s one of the few apps not to show much performance increase.

https://twitter.com/i/web/status/1328777333512278020

 

[Bodie CI5]

Even if it was 3000MB/s, that would still be a joke, due to how much Apple charges for SSD’s.

You can buy internal 1TB SSD’s that do 7000MB/s like the M1 Max does for only $150 from Samsung.

Indeed. I’m a little scandalised to be honest.

 

[jeanlain]

Because if cinebench isn’t efficient on a platform, you can throw all the cores you want at it and you won’t see much improvement.

I believe cinebench performance scales very well with the number of threads on Apple Silicon.
The fact the cinebench doesn't use the SIMD units very well is another matter.

 

[jeanlain]

Does anyone else have any more information on the SSD speeds on the M2 MBP? Alex Ziskind shows Blackmagic’s Disk Speed Test With some appalling figures. (Apologies if I’ve posted this in the wrong thread)

I don't know, but I wish people stopped using that tool to measure SSD performance. SSD is all about random read/write, which this tool doesn't measure.

 

[JimmyjamesEU]

I believe cinebench performance scales very well with the number of threads on Apple Silicon.
The fact the cinebench doesn't use the SIMD units very well is another matter.

What is that belief based on? The link I provided shows it doesn’t utilise all the available resources. It’s multi-threaded, but under utilises each core.
what would you say is the explanation for both using less power and less increase thank other benchmarks? For example Spec and geekbench.

 

[killawat]

I'm just happy to get some figures for the M2. The perf bump is nice. The media engine adds are very nice. Those of us still on Intel Macs have plenty of choices.

 

[Bodie CI5]

I'm just happy to get some figures for the M2. The perf bump is nice. The media engine adds are very nice. Those of us still on Intel Macs have plenty of choices.

Not so fast as one might think in all metrics:

 

[UBS28]

Indeed. I’m a little scandalised to be honest.

Seems testing from Max Tech also showed the same thing.

So it doesn’t seem to be a faulty unit if an other reviewer has found the exact same thing.

Guess Apple got their hands on some cheap ass SSD’s due to their current production issues I guess.

 

[jav6454]

Not so fast as one might think in all metrics:

So it's getting hotter? I was so right.

 

[Bodie CI5]

So it's getting hotter? I was so right.

Yep, it seems so. Not sure if this video has made the rounds here yet:

 

[jav6454]

Yep, it seems so. Not sure if this video has made the rounds here yet:

It hasn't, MR has been suspiciously quiet when it comes to independent reviews. Which means the MBA will throttle so much faster.

 

[CWallace]

Even when you run the fans at over 7000RPM to drop the CPU temp by 20 degrees C, the machine doesn't run any faster per Max Tech's testing so I don't believe it is correct to say that the M2 thermally throttles.

As with Max Tech's testing of the entire M1 family, it doesn't looks like Apple allows any M series SoC to draw maximum power (as measured by wattage) to run the clock speed at the maximum rated speed, but instead keeps it a few hundred MHz below that maximum both single and multi core.

This seems to be to allow the SoC to run with either no fans or at the minimum speed possible for as long as possible. And with the number of folks on these forums complaining that any fan speed sounds like a box of anvils being dropped next to their ears, it should not be surprising Apple is following this tack.

 

[jav6454]

Even when you run the fans at over 7000RPM to drop the CPU temp by 20 degrees C, the machine doesn't run any faster per Max Tech's testing so I don't believe it is correct to say that the M2 thermally throttles.

As with Max Tech's testing of the entire M1 family, it doesn't looks like Apple allows any M series SoC to draw maximum power (as measured by wattage) to run the clock speed at the maximum rated speed, but instead keeps it a few hundred MHz below that maximum both single and multi core.

This seems to be to allow the SoC to run with either no fans or at the minimum speed possible for as long as possible. And with the number of folks on these forums complaining that any fan speed sounds like a box of anvils being dropped next to their ears, it should not be surprising Apple is following this tack.

Agreed. But can you imagine how hot an MBA will be with the M2?

 

[MayaUser]

Like i said...impressive gpu gain
cpu was already impressive...the gpu takes the cake now

 

[MayaUser]

Agreed. But can you imagine how hot an MBA will be with the M2?

it cannot be any hotter than 104-105C
Since the body is around 40C while my Intel or amd laptops are all over 44C or even 46C...this is heaven, not to mention to keep the same battery life...Cannot wait to test mine with Maya

 

[jav6454]

it cannot be any hotter than 104-105C
Since the body is around 40C while my Intel or amd laptops are all over 44C or even 46C...this is heaven, not to mention to keep the same battery life...Cannot wait to test mine with Maya

It can't, I'm just exaggerating. However, the fact is, the MBA will be hotter for longer and rapid reach that state given the passive arrangement.

 

[CWallace]

Agreed. But can you imagine how hot an MBA will be with the M2?

It looks like the fans did not kick in until the CPU core temps reached 100°C and with the fans at around half RPM, temps peaked at around 104°C.

My guess is that the Air would allow temps in the 100°C range before it started reducing wattage to keep it around that temp. And even at 100°C, it was still running over 3GHz so we will likely not see severe throttling (where we lose a GHz or so).

 

[jav6454]

It looks like the fans did not kick in until the CPU core temps reached 100°C and with the fans at around half RPM, temps peaked at around 104°C.

My guess is that the Air would allow temps in the 100°C range before it started reducing wattage to keep it around that temp. And even at 100°C, it was still running over 3GHz so we will likely not see severe throttling (where we lose a GHz or so).

The reviewer did state it seems like Apple's fan curve is broken on the M2 as on the M1 is revs correctly. Also, likely the MBP didn't see throttling as the heat pipes were actively taking heat away to the fan fins which even at extremely low speeds helped.