Modern SoCs throttle in a fanless design...

[[AUT] Thomas]

edit:
that said, i'd strongly suggest: don't buy a 256 GB machine irrespective of speed. Its capacity you should be worried about. You're crippling your shiny new mac (of any type) with a meagre amount of local storage which will be quickly eaten by a few applications and time machine local backups in short order, never mind if you try and back up your phone with it or try store any actual work on it.

Yeah 512 GB or more is more money, but you'll get a far more usable machine for a relatively small percentage increase in total cost vs. the benefit.

I would generally not suggest buying the combo of 8GB memory and a 256GB SSD. That doesn't play well no matter if macOS or Windows. Open a browser and it will start swapping to the SSD which given its small size already be at least 50% full, leaving only few space left. The less space left, the worse the swap will perform. The TBW on small SSDs are also significantly less, due to the overall size of the SSD.
Unless this is grandma's Air I strongly suggest against the 8GB/256GB combo.

 

[darngooddesign]

People just want more bang for the buck, especially after a price bump.

Bigger and brighter screen, better speakers, better webcam, Magsafe so both usb ports are free for data, ultra wide band chip, 15-30% faster on all tasks not just long sustained, better headphone jack, higher ram limit, dedicated media engine to accelerate video tasks pulled from the M1-Pro/Max, fast charge with the upgraded plug, etc.

Seems like a lot of bang for the buck to me provided you aren't running long sustained benchmarks or transferring 100GB of data to a small internal drive which probably only has 150GB free.

 

[Tyler O'Bannon]

Lots of level headedness in this thread. The thing is, a fanless design is going to throttle, period. This is the purpose of the M2 MBP, to have a fan that cools the chip. The design is not the same.

I’m very thankful that we live in a day and age where we can have an incredible design like the M2 Air, and have a ton of power and incredible battery life in a thin and light enclosure.

It’s not the computer for anyone who needs “sustained loads”, especially often.

It’s probably the only reason there is an M2 MBP.

Now, I REALLY hope that the 13 MBP gets replaced with a 15” MacBook Studio (or just MacBook?) that has a single fan design.

But nonetheless, we are fortunate to have a machine like the new Air that can function WELL day to day without a fan. And even crank out a lot of power in bursts.

 

[doboy]

Bigger and brighter screen, better speakers, better webcam, Magsafe so both usb ports are free for data, ultra wide band chip, 15-30% faster on all tasks not just long sustained, better headphone jack, higher ram limit, dedicated media engine to accelerate video tasks pulled from the M1-Pro/Max, fast charge with the upgraded plug, etc.

Seems like a lot of bang for the buck to me provided you aren't running long sustained benchmarks or transferring 100GB of data to a small internal drive which probably only has 150GB free.

Don't need to convince me since I already own a 14" M1 Pro

 

[aevan]

I almost feel like the 'thermal throttling' gate thing is to steal attention away from the much more significant SSD 'issue'

Both of these issues are severely overblown. As is the case with every major Apple launch.

 

[hans1972]

Only Apple can make a fanless laptop because their SoCs are more energy efficient than other SoCs, but should they?

Yes, because you can never take back fan noise you have heard.

Reducing computing performance is just a delay. Every two or three billionth of second you gain some new computing power you can use.

 

[mikethemartian]

I’ve got something to say:

 

[hans1972]

Yeah, but an unthrottled M1 is still faster than a throttled M2, isn't it?

So, if the M2 would throttle let's say 50% more often than the M1 doing similar tasks, the question is how much of that advantage of the M2 remains.

What if the M1 throttled 0% with my usage and the M2 throttled 0% with my usage. Should I care?

 

[hans1972]

You're misunderstanding the issue. Yes, we know that this is how a modern chip is designed to work. If it gets too hot, it throttles by slowing down the clock speed. We know this. We've known this for many years.

The point is that Apple combined the new M2 chip that can use a LOT more power than the M1 chip with a very thin chassis that DOES NOT have an actual proper heatsink, according to iFixit's teardown.

The previous M1 MacBook Air actually had a decent heatsink, which allowed the temps to stay lower.

This time around, Apple not only got rid of the proper heatsink and replaced it with a thin metal shield, but they also put in a chip that uses more power.

This leads to throttling that happens sooner than it did before.

So yes, the throttling is working as intended, but it could've been avoided if put the M2 chip over to the left side and connected a proper heatsink that fills in that large open area on the left side that isn't filled up with the logic board. (see image)

That's what people aren't happy about. We're not mad at throttling. It works as intended like it always has. We're not happy with Apple's passive cooling design which is vastly worse than the previous M1 MacBook Air's passive cooling design.

View attachment 2032243

But wouldn't this lead the heat to the exterior shell and thus increase the likelihood of feeling the heat?

 

[Wizec]

But wouldn't this lead the heat to the exterior shell and thus increase the likelihood of feeling the heat?

This has been tested.

Simply adding a cheap thermal pad made a very large difference in performance under heavy loads, and the chassis only reached 47° C.

Without the thermal pad, the chassis still reached 45 ° C.

2 degrees C more in the same ambient temp, but drastically better performance under load.

The apologies are unwarranted. Apple could have done better and should have done better. Appeal to authority (Apple engineers know best!!) is a logic fallacy. Let the facts speak for themselves.

 

[doboy]

This has been tested.

Simply adding a cheap thermal pad made a very large difference in performance under heavy loads, and the chassis only reached 47° C.

That may still be uncomfortable and can be higher depending on ambient temp.

 

[Pugly]

My MBA from 2015 throttled all the time with a fan...I used it to compress DVDs and ran it with a box fan blowing on it to improve export times. I think some websites looked at it the wrong way, and the fan would go full blast and probably throttle.

This is basically the design of all thin laptops... the Air doing the same without a fan isn't surprising. Isn't it still remarkable that it can get this performance without a fan? And the throttled performance is still top notch, so just wait your 10 mins for the longer export... or time is money and get a MBP or Mac Studio.

 

[Wizec]

That may still be uncomfortable and can be higher depending on ambient temp.

It’s only 2 degrees higher than the measured chassis temp of 45 C under the same loads with stock throttling behavior in the same ambient temps.

 

[exoticSpice]

35C ambient is 95F. You're confusing Steam Deck throttling due to high ambient temperature vs M2 throttling under normal ambient temperature or air conditioned environment.

M2 Air has no fan. It is a fanless device of course it'd going to throttle under normal conditions as well.

Take a look at ipads and iphones they behave the same as they are fanless too.

 

[1096bimu]

It’s only 2 degrees higher than the measured chassis temp of 45 C under the same loads with stock throttling behavior in the same ambient temps.

They didn’t do the complete test. If you only stress the CPU or only the GPU yes it’s about the same.
I just tried this with max CPU+GPU stress, the bottom maxed out to 62 degrees, which is definitely in danger territory.

where as without this mod you’d never have this danger.

 

[exoticSpice]

When you have a laptop with a fan dust also gets sucked in to the laptop during the intake and dust gets stuck in the fan.

Fanless laptops do have benefits. For the casual user M2 or M1 Air is fine.

 

[exoticSpice]

They didn’t do the complete test. If you only stress the CPU or only the GPU yes it’s about the same.
I just tried this with max CPU+GPU stress, the bottom maxed out to 62 degrees, which is definitely in danger territory.

where as without this mod you’d never have this danger.

Yes adding thermal pads will make the bottom case hotter. What max tech suggested is bad. Honestly most of these tech youtubers are misleading people.

 

[tmoerel]

This has been tested.

Simply adding a cheap thermal pad made a very large difference in performance under heavy loads, and the chassis only reached 47° C.

Without the thermal pad, the chassis still reached 45 ° C.

2 degrees C more in the same ambient temp, but drastically better performance under load.

The apologies are unwarranted. Apple could have done better and should have done better. Appeal to authority (Apple engineers know best!!) is a logic fallacy. Let the facts speak for themselves.

This was a test that tested one thing and changed the thermal design. Now that we have a new thermal design with a different spread of heat inside the mac, what will happen to the rest of the system? Will the batteries suffer differently? Will the SSD suffer differently? Are there other parts that now get a different thermal load? Are there any side effects to these different thermal loads?
You see, just adding a thermal pad can totally change the thermal design to something that hasn't been tested. And it is always a good principle that "What hasn't been tested doesn't work". So as long as we don't test this we have to assume that Apple did test and has good reason for doing what they did. Otherwise we have no idea what side effects have been introduced!

 

[Richu]

Making a fanless laptop is a marketing ploy that favors form over function to demonstrate how energy efficient Mx SoCs are. I doubt that Mx SoCs would lose appeal if they had a fan.

I wonder if future MBAs will go back to having a fan.

Like the iPhone and iPad?

Throttling is a tool, not necessarily a problem. Sure, they could also design a weaker CPU for the air form factor. But that increases supply chain and development complexity. Why not just throttle the design and achieve the exact same thing?

In this case throttling allows them to design a single CPU that maximizes two different form factors.

 

[1096bimu]

Yes adding thermal pads will make the bottom case hotter. What max tech suggested is bad. Honestly most of these tech youtubers are misleading people.

Well I don’t plan on running both cinebench and 3dmark at once while on the move with it on my lap so I should be fine.
any normal usage, including exporting videos won’t stress up the machine that much.

however if you’re Apple you can’t have it be at 60 degrees at all times.

 

[camotwen]

It’s already the most powerful computer in its class.

It is also quite more expensive than the previous model, and not that more powerful than that. Considering that right now one can get the previous model at higher specs at a lower price than the newer basic one, these details matter in taking an informed decision.

 

[leman]

You're misunderstanding the issue. Yes, we know that this is how a modern chip is designed to work. If it gets too hot, it throttles by slowing down the clock speed. We know this. We've known this for many years.

Vadim, with all respect, this issue only exists because it has been artificially created by YouTubers like yourself. The new M2 chassis hits the same sustained power dissipation target as the old one, M2 produces more performance at the same power consumption than M1 — and if M2 has a higher peak power, well, that's great since it means it will perform better in a desktop chassis (a common point of criticism of M1 Mini and iMac). You are basing the entirety of your argument on the fact that M2 has a larger dynamic performance range than the M1, but that's a fallacy. M2 does not have to behave exactly like M1. What's important is the relative performance improvements of M2 over M1 — and your tests demonstrate that it's there and sometimes quite substantial (with the only thing really spoiling the picture is the slower SSD).

The point is that Apple combined the new M2 chip that can use a LOT more power than the M1 chip with a very thin chassis that DOES NOT have an actual proper heatsink, according to iFixit's teardown.

The previous M1 MacBook Air actually had a decent heatsink, which allowed the temps to stay lower.

This time around, Apple not only got rid of the proper heatsink and replaced it with a thin metal shield, but they also put in a chip that uses more power.

This leads to throttling that happens sooner than it did before.

[...]

That's what people aren't happy about. We're not mad at throttling. It works as intended like it always has. We're not happy with Apple's passive cooling design which is vastly worse than the previous M1 MacBook Air's passive cooling design.

How do you know that the new solution is inferior to the old one? Did you do thermal conductivity tests or did you just hold the new heatsink in your hand and decided that it's not adequate?

Now, I don't have a M2 MBA in hand to check this, but all the tests I have seen suggest that the sustained power draw of the new MBA is around 10W. This is identical to — maybe even better than — the old design. If a cooling solution can dissipate its target wattage then it's performing as it should. I mean, how can you argue that the cooling solution is somehow worse if its thermal performance is exactly the same?

Besides, we see results of M2 outperforming the M1 by 15-20% in CPU tasks in sustained operation, which is not a bad result at all. Graphics performance is more hit and miss as it seems, but still better than M1 Air overall.


P.S. By the way, thanks for joining the discussion. I might not always agree with your analysis or methods but I appreciate that you are engaging with the enthusiast community. This shows integrity and deserves respect.

 

[henrikhelmers]

And at this point, it just shows how good the M1 was. Thus if one was looking for base model, the M1 Macbook Air imo is the better deal.

I think that’s the real issue here. The M1 was such a fantastic chip. Makes you wonder how the M2 will work in a new iPad Pro.

On the other hand I’m glad Apple doesn’t hold the Air back. Most workloads are bursty anyway.

 

[darngooddesign]

They didn’t do the complete test. If you only stress the CPU or only the GPU yes it’s about the same.
I just tried this with max CPU+GPU stress, the bottom maxed out to 62 degrees, which is definitely in danger territory.

where as without this mod you’d never have this danger.

If you are regularly stressing the CPU and GPU at 100%, you shouldn't be considering a MBA because even with the thermal pad mod its still slower than a MBP at the tasks which stress the CPU+GPU like that.

If you are rarely stressing it like that, then the M2Air is fine.

 

[Wizec]

Vadim, with all respect, this issue only exists because it has been artificially created by YouTubers like yourself. The new M2 chassis hits the same sustained power dissipation target as the old one, M2 produces more performance at the same power consumption than M1 — and if M2 has a higher peak power, well, that's great since it means it will perform better in a desktop chassis (a common point of criticism of M1 Mini and iMac). You are basing the entirety of your argument on the fact that M2 has a larger dynamic performance range than the M1, but that's a fallacy. M2 does not have to behave exactly like M1. What's important is the relative performance improvements of M2 over M1 — and your tests demonstrate that it's there and sometimes quite substantial (with the only thing really spoiling the picture is the slower SSD).




How do you know that the new solution is inferior to the old one? Did you do thermal conductivity tests or did you just hold the new heatsink in your hand and decided that it's not adequate?

Now, I don't have a M2 MBA in hand to check this, but all the tests I have seen suggest that the sustained power draw of the new MBA is around 10W. This is identical to — maybe even better than — the old design. If a cooling solution can dissipate its target wattage then it's performing as it should. I mean, how can you argue that the cooling solution is somehow worse if its thermal performance is exactly the same?

Besides, we see results of M2 outperforming the M1 by 15-20% in CPU tasks in sustained operation, which is not a bad result at all. Graphics performance is more hit and miss as it seems, but still better than M1 Air overall.


P.S. By the way, thanks for joining the discussion. I might not always agree with your analysis or methods but I appreciate that you are engaging with the enthusiast community. This shows integrity and deserves respect.

The lady doth protest too much, methinks…

A lot of people had success with this mod on the M1 gen machines for almost 2 years now, and I can’t find a soul on the interwebz who has had any fallout from it. All this weeping and wailing and gnashing of teeth over the findings and the thermal mod is literally worse than the maligned YouTubers‘ findings and worse than any perceived damage that might occur.

No matter how you slice it, the chassis isn’t going to get hotter than my old 2018 MacBook Pro which got unbearably hot when running heavy loads. A known thermal problem that Apple shipped for YEARS with no compunction. It was at LEAST 60° C during 20 minute Visual Studio builds. And to answer another poster, the heat is being drawn away from the battery in the front toward the back. Also, internal temps are still being monitored exactly the same as before and the SOC will throttle once it gets too high.

Anyway. Peace! I’m out, these quixotic apology threads are a waste of time.