M2 fan to fanless takes 32% performance hit on 30 minute sustained load vs 22% for M1

[Gerdi]

For what it's worth, Apple Silicon doesn't seem to have anything akin to PL1/PL2. The one behavior which maps well onto Intel power/frequency control concepts is that maximum P core clock speed is permitted only while just one P core is active, and it drops off as more P cores wake up. However, unlike Intel, this is governed on a per-cluster basis rather than the whole chip, and the worst case frequency penalty with all cores active is quite small - only about 6% for M1 family chips.

Well, therefore I was just applying the terms conceptionally and not literally - where PL1 describes the power limit, which is sustainable under current conditions - which in case of Apple depend on the current thermal conditions. In case of Intel with a pre-defined PL1 it may happen, that the SoC jumps to PL1 prematurely. But for my argument, these differences did not matter anyway.
Still I do expect, that the MBA stabilize at below 10W under nominal ambient temperatures of 20 degree Celsius - like any other passively cooled device in a laptop form factor.

 

[Jumpthesnark]

Keep sustained load under 30 minutes or get the version with fan.

View attachment 2032028

According to the posted screengrab, which purports to show one gaming-based benchmark:
M1 Air: 18fps < M2 Air: 19fps
M1 Pro: 23fps < M2 Pro: 28fps

So what you're saying is M2 MBAs and MBPs are both faster than the M1 versions, and that's somehow a bad thing? 🤔

 

[Zest28]

Once again people are using their computers wrong and then getting all pissed off at the manufacturer.

The MacBook Air is designed for basic users doing light tasks. Anyone pushing the Air into thermal throttling is on the wrong laptop.

Even without a fan, the M1 and M2 have more processing headroom than PRO COMPUTERS THAT ARE STILL IN USE. My Classic Mac Pro is more processing power than most people ever need (myself included), and the M1 and M2 are even more powerful than that.

The M2 under maximum throttling is still faster than a cold M1.

There is no thermal problem. End of line.

That is not true.

The iPhone and iPad Pro are Apple their gaming computers. And the iPad Pro is also fanless with the M1, just like the MBA.

The M1 iPad Pro runs AAA games like Aliens Isolation faster than the 16” M1 Max MacBook Pro even.

 

[leman]

The M1 iPad Pro runs AAA games like Aliens Isolation faster than the 16” M1 Max MacBook Pro even.

This is a very odd claim to make. How would a slower GPU run games faster?

 

[Zest28]

This is a very odd claim to make. How would a slower GPU run games faster?

Run both games on the M1 iPad Pro and the 16” M1 Max MacBook Pro and see for yourself.

All AAA games on the M1 iPad Pro are ARM optimized while on Mac you need to use Rosetta 2.

 

[leman]

Run both games on the M1 iPad Pro and the 16” M1 Max MacBook Pro and see for yourself.

All AAA games on the M1 iPad Pro are ARM optimized while on Mac you need to use Rosetta 2.

Sure, but that’s the developers fault for not providing optimized software. If Feral were to take their iPad port and adapt it to the Mac it would run much faster.

And of course games on the iPad are ARM optimized, it’s not like you can submit Intel binaries to the App Store. Which is also why modern AAA games on iPad can be counted on a single finger. Not that it is much better on the Mac of course, but at least you get access to many PC titles.

 

[MajorFubar]

I'm usually the first to bitch but i'm not seeing the problem here. Happy to be explained to if I'm missing something, but what I'm reading from this is that when you persistently push it to its limits, the M2 in a fanless computer throttles-down more as a percentage of its overall maximum performance than the M1 did, but even when throttled down, it's still faster than a throttled M1.

So, what's the issue?

Happy to wear the dunce's hat if I've missed something totally obvious. But I don't think I have.
This doesn't at all appear to be SSD-gate V2, and appears in fact to be 'no news'.

 

[leman]

I'm usually the first to bitch but i'm not seeing the problem here. Happy to be explained to if I'm missing something, but what I'm reading from this is that when you persistently push it to its limits, the M2 in a fanless computer throttles-down more as a percentage of its overall maximum performance than the M1 did, but even when throttled down, it's still faster than a throttled M1.

So, what's the issue?

Happy to wear the dunce's hat if I've missed something totally obvious. But I don't think I have.
This doesn't at all appear to be SSD-gate V2, and appears in fact to be 'no news'.

Exactly. Peak performance is higher, sustained limit is the same, sustained performance is higher. But somehow it’s “bad” because you get a bigger range between the peak and the sustained performance. Go figure…

 

[Moka Akashiya]

So the passively cooled M2 loses 30% off its peak performance after 30 minutes and it’s a disaster and an engineering fail. But the 6800U loses over 40% off it’s peak performance after seconds and it’s a marvelous product and the best CPU ever. Double standards and misdirection like usual.

When nobody compared M2 to 6800U or intel in the thread yet, but you really want to scream about double standarts somewhere.
First post only shows how bad it can be with such hot cpu and so privitive cooling solution, idk why people want to shut up somebody with worst case scenarios.
Personally i would not buy M2 air for performance boost, M1 temps can be better for internals on long distance, and there is no much to pay 300-500$ more on chip side.

 

[Xiao_Xi]

For what it's worth, Apple Silicon doesn't seem to have anything akin to PL1/PL2.

Regarding their review of the 13" MBP M2, Notebookcheck wrote:

We can see a maximum package power of 35W for the M2 SoC during combined CPU/GPU load. However, this value can only be maintained for a couple of seconds before there is a drop to around 25W while the fans speed is slowly increasing. The package power will stabilize at 28-30W once the fan has reached its maximum speed, but the chip temperature is still very high at 98 °C. This means the cooling solution cannot utilize the full potential of the M2 chip, despite the fan.

Apple MacBook Pro 13 2022 M2 Laptop Review – Debut for the new Apple M2

Notebookcheck reviews the updates 2022 Apple MacBook Pro 13 with the brand-new Apple M2 SoC and 16 GB RAM.

www.notebookcheck.net

If true, that is exactly Intel's concept of PL1/PL2.

 

[diamond.g]

Regarding their review of the 13" MBP M2, Notebookcheck wrote:

Apple MacBook Pro 13 2022 M2 Laptop Review – Debut for the new Apple M2

Notebookcheck reviews the updates 2022 Apple MacBook Pro 13 with the brand-new Apple M2 SoC and 16 GB RAM.

www.notebookcheck.net

If true, that is exactly Intel's concept of PL1/PL2.

Technically PL2 state is time based not temperature based, though very few manufacturers actually adhere to the time part. Apples power states appear to be temp based.

 

[Xiao_Xi]

Technically PL2 state is time based not temperature based, though very few manufacturers actually adhere to the time part. Apples power states appear to be temp based.

You are right, as Anandtech explains:

Tau is a timing variable. It dictates how long a processor should stay in PL2 mode before hitting a PL1 mode. Note that Tau is not dependent on power consumption, nor is it dependent on the temperature of the processor (it is expected that if the processor hits a thermal limit, then a different set of super low voltage/frequency values are used and PL1/PL2 is discarded).

Why Intel Processors Draw More Power Than Expected: TDP and Turbo Explained

www.anandtech.com

 

[kevcube]

Or buy a gaming laptop AKA one with Windows?.... I bought this really expensive wrench but it just sucks for hammering in nails. Do you think it's defective and I should return it? That's how these reviews with silly benchmarks are coming off as.

you should've gotten the wrench with the fan like I told you.

 

[russell_314]

you should've gotten the wrench with the fan like I told you.

Or I can just use the wrench for turning bolts? I already have a hammer for driving nails 😂

 

[darngooddesign]

When I discovered that the M1 Air couldn't handle 30 minute renders without heat-related throttling, instead of forcing it to fit my needs I bought a MBP which did fit my needs. Of course, this issue is irrelevant for people who don't do 30 minute renders.

 

[JCCL]

Or buy a gaming laptop AKA one with Windows?.... I bought this really expensive wrench but it just sucks for hammering in nails. Do you think it's defective and I should return it? That's how these reviews with silly benchmarks are coming off as.

Yeap, for gaming, you still need a PC or a console, why would one want a Mac. A couple of games like Resident Evil and Shadow of the Tomb Raider don't come close to the huge library a PC or a console have. Maybe one day Apple will come closer to match PC offerings, but that day is not here yet.

 

[mi7chy]

I'm usually the first to bitch but i'm not seeing the problem here. Happy to be explained to if I'm missing something, but what I'm reading from this is that when you persistently push it to its limits, the M2 in a fanless computer throttles-down more as a percentage of its overall maximum performance than the M1 did, but even when throttled down, it's still faster than a throttled M1.

So, what's the issue?

Happy to wear the dunce's hat if I've missed something totally obvious. But I don't think I have.
This doesn't at all appear to be SSD-gate V2, and appears in fact to be 'no news'.

Unless you keep your CPU+GPU workloads less than 30 minutes at a time and let it cool in-between there's no benefit in upgrading from M1 fanless to M2 fanless since performance is a wash. Apple would've been better off dropping the M2 MBP and adding a fan to the M2 MBA.

 

[bcortens]

Unless you keep your CPU+GPU workloads less than 30 minutes at a time and let it cool in-between there's no benefit in upgrading from M1 fanless to M2 fanless since performance is a wash. Apple would've been better off dropping the M2 MBP and adding a fan to the M2 MBA.

Wouldn't that just make it a redesigned M2 MBP? The M2 gives more peak performance than M1 and (worse case scenario) ever so slightly exceeds M1. The M2 MBA is a perfectly fine machine for people who don't need the sustained performance of the MBP, it is exactly the right successor to the M1 MBA in that regard.

The problem is that it's overpriced and continues the Tim Cook trend of introducing the new thing at a higher price and continuing to sell old hardware at the old price. I think if the M2 MBA had been a drop in replacement for the M1 MBA there would be a little less controversy. The continued existence of the M1 MBA confuses the lineup and makes the M2 MBA hard to recommend since it is so close to the M2 MBP in price.

Edit: I should clarify - at the same 10GPU core spec there is about $50 CAD separating thee M2 MBA and M2 MBP which is ridiculous considering the MBP has better battery life and won't throttle ... this is terrible product segmentation.

Edit 2: because of this ridiculous price overlap I would recommend the M1 MBA or the M2 MBP to a family member even though the M2 MBA is nicer looking

 

[mi7chy]

even though the M2 MBA is nicer looking

That's debatable. Wedge design is unique unlike the Pixelbook Go Chromebook look.

 

[bcortens]

That's debatable. Wedge design is unique unlike the Pixelbook Go Chromebook look.

I'll give you that, I actually quite like the way the tapered edges of the 2016-2020 MBPs (and I suppose the current 2022 M2 MBP) give them a sleeker look while still providing a good amount of internal volume.

 

[russell_314]

That's debatable. Wedge design is unique unlike the Pixelbook Go Chromebook look.

I'm guessing that's meant to be a jab at how the Air looks but I kind of like how the Pixelbook Go looks. I almost bought one. Looks is in the eye of the beholder so what looks nice to me might look ugly to you. I didn't get the Midnight color because of how it fingerprinted and I've always liked the classic look of a silver MacBook.

 

[thadoggfather]

Stop with the toxic benchmarks, geez.

 

[kevcube]

I'm guessing that's meant to be a jab at how the Air looks but I kind of like how the Pixelbook Go looks. I almost bought one. Looks is in the eye of the beholder so what looks nice to me might look ugly to you. I didn't get the Midnight color because of how it fingerprinted and I've always liked the classic look of a silver MacBook.

Yeah pixelbook go is a very strange comparison, I nearly bought one too because that was back when I had faith in Google’s product strategy, but most non-techies have never heard of that notebook. He could’ve compared it to like a razer blade.

 

[camotwen]

So you’re showing that it’s faster?

1 more fps for 200$ more does not seem like great value for money either.

 

[NT1440]

1 more fps for 200$ more does not seem like great value for money either.

Sure…if you were simply buying a CPU, but you’re not.

Got any other oversimplifications for us?