Can we talk about the fanlessness

[G5isAlive]

AS Mac can outdo that even with no fan, there just doesn't seem to be any advantage in shoving one in, just so those who love the whining noise can continue to enjoy it.

You've given me a great idea for an App!

"Are you disappointed by how quiet your new MacBook Air is? Outraged Tim Cook would cheat you out of white noise? Is the silence messing with your anxiety? Decreasing your productivity while you wait for the whine to ramp up?

We at Bogus Software are here for you with our newly released App, "Fan Buddy aka FUD. " For the low low price of $19.84, FUD is a fan emulator that emits a whiny sound depending on how many CPU cycles you use. The more work you do, the louder it gets! But wait there's more, for the full experience and only $6.66 a month we will unlock "intel mode" where FUD both whines AND throttles your CPU!

Bogus Software, Apps you deserve but won't get from Tim Cook. We innovate for you!"

 

[TechnoMonk]

You've given me a great idea for an App!

"Are you disappointed by how quiet your new MacBook Air is? Outraged Tim Cook would cheat you out of white noise? Is the silence messing with your anxiety? Decreasing your productivity while you wait for the whine to ramp up?

We at Bogus Software are here for you with our newly released App, "Fan Buddy aka FUD. " For the low low price of $19.84, FUD is a fan emulator that emits a whiny sound depending on how many CPU cycles you use. The more work you do, the louder it gets! But wait there's more, for the full experience and only $6.66 a month we will unlock "intel mode" where FUD both whines AND throttles your CPU!

Bogus Software, Apps you deserve but won't get from Tim Cook. We innovate for you!"

Need special discount for Macrumors members.

 

[G5isAlive]

Need special discount for Macrumors members.

I can think of a very 'special' discount for any Mac Rumors member that used 'fan boy' in a message.

 

[6749974]

Need special discount for Macrumors members.

MacRumors members use special code at checkout: 8GB-ISNT-ENOUGH

 

[Colpeas]

I think you misinterpret the design intention of the product.

Apple didn't omit the fan(s) in a MacBook Air to "save 1 or 2 bucks", they did it because A) it would add size/weight/bulk in a very thin design and B) the target audience of the Air will rarely push the machine that would benefit from active cooling under sustained loads.

Don’t forget the extra battery life that you’re getting thanks to not having a fan spinning inside…

Mbair really is a marvelous machine. I could barely get the m1 warm in scenarios, where my intel-based pro was fanning like crazy to keep it within the temp limit. And i’m talking about the Air with the more powercul chip from m1 pro (which has a fan for some reason)

 

[jclardy]

I'm the opposite - I love using my M3 Air. I've flip flopped - from M1 Air, to M1 Pro, to eventually and M1 Max, and now an M3 Air for iOS/visionOS developmenet. The fans never really bothered me, I would only hear them in a quiet room if I was doing something processor intensive, and even still they weren't bad at all. But there was one huge problem I had with the Pro/Max machines - I was always "surprised" by low battery warnings.

What I mean is Xcode is one of the buggiest apps on the platform, and often the simulators will get stuck doing something in the background and will just sit at 100% on a single pcore. You have a bunch of other cores, so you don't notice performance issues. And you also don't notice the fans. But what you do notice is that you were at 80% battery and now 30 minutes later you are at 25%.

With the fanless Air - this isn't a problem because it has a built in indicator of stuck processes - the heat. If I'm just browsing the web and I feel warmth at my fingertips I know something is up and I can go find the stuck process and kill it.

Also just the battery life I get on the Air, doing the same tasks I do on the pro (visionOS dev and running the 3d simulator) I still get incredible battery life, it feels like nearly 2x the Pro. Add in the slightly smaller footprint and the much thinner & lighter chassis & of course much lower cost. Of course - you are limited to 24GB of RAM - so that could be your limiting factor. But for iOS/web/api development...IMO 24GB is just fine.

 

[SABunting]

Apple's approach was always to make the quietest machine. Not just in one product, but them all - so long as it still is a marketable, appealing and competitive product. The first MacBook was an attempt, but Intel's offerings were too hot, too low powered and expensive - it was just about possible to sell it / use it. Now, the current Air is able to offer a big step forward in power / usability than its Intel counterpart and omit the fan altogether.

 

[Ma2k5]

My M3 Air for my development is outperforming my 16” M1 Max. Happy to have a lighter, thinner, cheaper machine than a fan and slightly more performance in certain workloads. It’s fast enough. Probably diminishing returns on fan if you aren’t doing prolonged workloads too.

 

[Jwinnin]

I have been using Mac laptops for 25 years. The worst laptop I ever owned was an intel 2019 MBP. Not only did it have the lousy keyboard, and (for me) the useless Touch Bar, it also had fans that sounded like jet engines. Inexplicably, when sitting idle on my desk, it would sometimes spin up.

I use my laptops primary for basic productivity and portability. I ditched the 2019 MBP and now have a fanless M3 MBA. I could not be happier with this upgrade. For my needs, fanless is a huge upgrade.

 

[zach-coleman]

This thread is how I learned the M3 MBA was still fanless. I swear I read a review that said they added a fan. Happy that’s incorrect…

 

[Rhyalus]

With the fanless Air - this isn't a problem because it has a built in indicator of stuck processes - the heat. If I'm just browsing the web and I feel warmth at my fingertips I know something is up and I can go find the stuck process and kill it.

Also just the battery life I get on the Air, doing the same tasks I do on the pro (visionOS dev and running the 3d simulator) I still get incredible battery life, it feels like nearly 2x the Pro. Add in the slightly smaller footprint and the much thinner & lighter chassis & of course much lower cost. Of course - you are limited to 24GB of RAM - so that could be your limiting factor. But for iOS/web/api development...IMO 24GB is just fine.

Not sure why you should wait to feel the heat and lose some battery time... I have Stats installed in the menu bar which shows the heat, CPU and network activity... it is a great, free program.

It also provides a ton of additional information if you click on one of the indicators.

I need to see what the laptop is doing...I guess it is a throwback to the old days when laptops used to have hard drive activity lights.

With respect to this thread, of course it is a bit silly. As many have mentioned, the Air is meant to be a slim and silent laptop.

R

 

[sunapple]

We know the M1 Air shared the chassis with the Intel version for cost/production savings.. now that we have bespoke M2/M3 Airs, they are thinner, and as such, no room (or need) for a fan.

The 12" retina MacBook did not have a fan!

If you really need the difference a fan will make, either do the thermal pad mod, or buy the base M3 14" Pro.

Apple has got you covered The rest (99%) of the Air users really don't need as it only adds size and complexity.

What are you talking about, my MacBook 12" had a fan!

 

[Mike Boreham]

You're contradicting yourself.

• You "don't really need (or want)" a more computationally performant device
• But you're mad the Air isn't a more computationally performant device

Problem solving this is not challenging. Lets walk through this:

• Want an M3 chip and don't mind throttling? → Buy a MacBook Air with M3
• Want an M3 chip and do mind throttling? → Buy a MacBook Pro with M3
• Want to compress the time it takes to process large tasks? → Buy a MacBook Pro with M3 Pro/Max

Either way, the Air's throttling isn't bad at all. Let's take a look:

View attachment 2370011

It's slowing down only by 6% at the 10 minute mark, 11% at the 20 minute mark, and then staying at about 15% slow-down at 23 minutes to the 30 minute mark.

See? It's not a big deal.

Or if it is—buy something with more CPU/GPU cores because that will double or triple your speeds—something fans can't do.

That looks like one of mine! thanks for digging it out. Note that the y-axis is not a zero origin making the degree of throttling look greater than it is.

 

[6749974]

That looks like one of mine! thanks for digging it out. Note that the y-axis is not a zero origin making the degree of throttling look greater than it is.

Yes it is! Thank you. I failed to link to your post.

EDIT: added source url to original post

 

[G5isAlive]

That looks like one of mine! thanks for digging it out. Note that the y-axis is not a zero origin making the degree of throttling look greater than it is.

Aye. You'd be amazed how many people can't understand a graph... to make one's point one is allowed to chose whether to focus on the difference, as your original graph did with the non zero origin, or highlight the overall performance with a zero origin as I did in this replot of your data.

Same data, different emphasis. It also needs to be emphasized this is under extreme conditions of stressed repeat testing, not at all to be expected with more routine use. But the maximum effect itself is just not that large.

I also put the two data sets on the same graph to show the difference between the MBA 1 and 2. I suppose I could dig out ancient data on intel MBA's to highlight the vast improvement of the fanless models over the fanned ones, though of course that's slightly biased as fan is not the only variable but AS versus Intel.

Bottom line is The MBA is well balanced for performance and fanless operation. If you need closer to 100% over long sustained use, buy a MBP.

But eh, why bring facts to an opinion fight?

 

[Torty]

Aye. You'd be amazed how many people can't understand a graph...

-snip-

View attachment 2370794

Maybe it's because the graph shows MBA1 vs. MBA2. If it's about throttling I would expect MBA vs. MBP.
X axis show "Run#" what is this? I would expect "time" for the X axis.

I think that's why people don't understand it well cause the graph isn't good to show fanless vs. fan.

 

[TechnoMonk]

-snip-

Maybe it's because the graph shows MBA1 vs. MBA2. If it's about throttling I would expect MBA vs. MBP.
X axis show "Run#" what is this? I would expect "time" for the X axis.

I think that's why people don't understand it well cause the graph isn't good to show fanless vs. fan.

Why would it be MBA vs MBP? Why not Mac mini, Mac Pro, iMac and so on. MBP and MBA are two different device categories, fan isn’t the only difference. If you need to compare the thermal blueprint of MBP to MBA, then you are probably looking at the wrong device.

The original graph by @Mike Boreham and updated graph by @G5isAlive show that even higher throttling on M2 is still faster than M1 which is relatively cooler. Compare M1 MBA with Intel MBA with fan, there is no comparison even with Throttling and no fan in either M2 or M1 MBA.

 

[Torty]

Why would it be MBA vs MBP? Why not Mac mini, Mac Pro, iMac and so on. MBP and MBA are two different device categories, fan isn’t the only difference. If you need to compare the thermal blueprint of MBP to MBA, then you are probably looking at the wrong device.

The original graph by @Mike Boreham and updated graph by @G5isAlive show that even higher throttling on M2 is still faster than M1 which is relatively cooler. Compare M1 MBA with Intel MBA with fan, there is no comparison even with Throttling and no fan in either M2 or M1 MBA.

Cause people can chose between fanless MBA and MBP with fans. IMO it doesn’t make sense to compare a desktop with a notebook cause if you plan to buy a notebook you are not interested in a desktop and vice versa.

 

[TechnoMonk]

Cause people can chose between fanless MBA and MBP with fans. IMO it doesn’t make sense to compare a desktop with a notebook cause if you plan to buy a notebook you are not interested in a desktop and vice versa.

It would be useful if the only difference between MBP and MBA was fan, it isn’t. You don’t need graphs to know, if you are going to run sustained load, get a MBP. Neither of the graphs had anything to do with MBP. It was OPs post about being fanless that makes latest MBA bad.

 

[AgeOfSpiracles]

I built a HTPC 15 years ago, and was super excited to include a fancy new Seasonic fanless PSU. Sadly, sticking the thing into my entertainment center meant that the CPU cooler sounded like a Dyson. Tried a few other solutions, such as supposedly "silent" Noctua coolers that ended up being not-so-silent, and a fanless heatsink that just couldn't cope with the heat buildup in the cabinet. The rise of streaming services and the purchase of an Apple TV means that I no longer have a permanent white noise generator in my living room.

 

[Mike Boreham]

-snip-

Maybe it's because the graph shows MBA1 vs. MBA2. If it's about throttling I would expect MBA vs. MBP.
X axis show "Run#" what is this? I would expect "time" for the X axis.

The x-axis is actually labelled "Cinebench Run #". You can set Cinebench to run consecutively. Each run takes a bit over 2 minutes and the next one starts straight after the last. Yes I could have plotted it against time but I recorded the data for each run so plotted it that way.

 

[G5isAlive]

-snip-

Maybe it's because the graph shows MBA1 vs. MBA2. If it's about throttling I would expect MBA vs. MBP.
X axis show "Run#" what is this? I would expect "time" for the X axis.

I think that's why people don't understand it well cause the graph isn't good to show fanless vs. fan.

The graphs show the decline in performance for both the MBA1 and MBA2 as a function of thermal throttling under intense loads.

I see you have been reading this thread.. so I am not sure why you are dragging in the MBP. The discussion is on whether to add a fan to a MBA or not. Some claim Apple is being cheap, and think there would be a huge benefit to a fan in the MBA. They cite throttling concerns of the MBA. Mike Boreham provided data to quantify and illustrate the effect of throttling under extreme load conditions of repeated Cinebench Runs. This is the standard way most people have been testing these devices for throttling as it eliminates variables of what load someone is running. Cinebench 23.

Sure, I agree, its known even laptops with fans throttle so perhaps if we were discussing the varying degrees (pun intended) of throttling on the MBP versus the MBA a better graph would be comparing the two... but its not needed for this discussion because the original graphs are a worst case scenario. fans don't make them better than baseline. If they have any effect at all, it's to reduce performance degradation. What these graphs show that in the worst case scenario you would have to be really concerned to worry about performance degradation on a fanless air, so if you are, get the pro.

Smiles.

 

[Torty]

It would be useful if the only difference between MBP and MBA was fan, it isn’t. You don’t need graphs to know, if you are going to run sustained load, get a MBP. Neither of the graphs had anything to do with MBP. It was OPs post about being fanless that makes latest MBA bad.

Thanks for explanation. I still don’t understand maybe cause I didn’t read complete thread. If you don’t like the fanless mba you can get the M3 MBP. So I would be interested how they perform against each other.
But yes everyone is different 😁

 

[G5isAlive]

Thanks for explanation. I still don’t understand maybe cause I didn’t read complete thread. If you don’t like the fanless mba you can get the M3 MBP. So I would be interested how they perform against each other.
But yes everyone is different 😁

Okay. Fair enough. Let's try this another way. The base model of the MBP has the exact same chip as the MBA. There are other differences (screen for example), but the performance is the chip, so same chip, similar performance. That's the assumption for a lot of the Apple Silicon Macs. The MBA does not have a fan. The MBP does. The argument is this makes the MBP so much better, because the chance of it thermal throttling is lower. Note the presence of a fan does NOT add cpu cycles, so performance can not be better than 100%. I assumed people could visualize 100%. Okay my bad.

In this graph, the green shaded area represents available performance over extended stress conditions. It gets less, but not by much. The red, is the theoretical maximum that could be 'reclaimed' IF the MBA had a fan (like the MBP does). Point is it's not a lot. A typical MBA customer never causes the machine to get warm, so it performs close to 100% all the time. But even when pushed hard, the performance degradation (the red area) just isn't a lot (compared to the green area). Does it fall enough to warrant one spending money on the MBP? Might for some, doesn't for a lot.

So. Let the customer decide if they want a fan or not. I have previously outlined the advantages of NOT having a fan. The idea of pushing fans on all of us, as the OP might want and some others, is an opinion not shared by many that currently own a MBA, and I dont think is supported by the facts. And we all like lower costs.

My apologies in advance that the graph isnt as pretty but I ran out of time. I hope it gets the point across. And my thanks again to Mike Boreham for doing the real work in providing the raw data. Data talks in my world.

 

[Torty]

Okay. Fair enough. Let's try this another way. The base model of the MBP has the exact same chip as the MBA. There are other differences (screen for example), but the performance is the chip, so same chip, similar performance. That's the assumption for a lot of the Apple Silicon Macs. The MBA does not have a fan. The MBP does. The argument is this makes the MBP so much better, because the chance of it thermal throttling is lower. Note the presence of a fan does NOT add cpu cycles, so performance can not be better than 100%. I assumed people could visualize 100%. Okay my bad.

In this graph, the green shaded area represents available performance over extended stress conditions. It gets less, but not by much. The red, is the theoretical maximum that could be 'reclaimed' IF the MBA had a fan (like the MBP does). Point is it's not a lot. A typical MBA customer never causes the machine to get warm, so it performs close to 100% all the time. But even when pushed hard, the performance degradation (the red area) just isn't a lot (compared to the green area). Does it fall enough to warrant one spending money on the MBP? Might for some, doesn't for a lot.

So. Let the customer decide if they want a fan or not. I have previously outlined the advantages of NOT having a fan. The idea of pushing fans on all of us, as the OP might want and some others, is an opinion not shared by many that currently own a MBA, and I dont think is supported by the facts.

My apologies in advance that the graph isnt as pretty but I ran out of time. I hope it gets the point across. And my thanks again to Mike Boreham for doing the real work in providing the raw data. Data talks in my world.

View attachment 2370917

Thanks I found it helpful to know the cinebenchmark runs ca. 2 mins so the graph is now very understandable to me.