How hot does your m2 MBA run? Doing what?

[jdb8167]

How do you find the m2 air to be faster than the m1pro when web browsing? I would have thought the pro-motion etc would have made the 14” feel superior? Do you get lag with 100s of tabs open on the m1pro but not on the m2 air? Is there lag when resizing windows or switching tabs?

Web browsing is heavily dependent on single thread performance. The M2 has better single thread performance than the M1 Pro/Max/Ultra. Any speed issue with a high number of tabs is more likely dependent on the amount of RAM available rather than multiple cores. ProMotion will make scrolling smoother but won't affect the actual speed of the browser.

Having said all that, I doubt that any difference in browser performance is noticeable between any Apple silicon Mac with equal amounts of RAM.

 

[0906742]

How do you find the m2 air to be faster than the m1pro when web browsing? I would have thought the pro-motion etc would have made the 14” feel superior? Do you get lag with 100s of tabs open on the m1pro but not on the m2 air? Is there lag when resizing windows or switching tabs?

I think those are mostly referring to one single page opening speed maybe blink of an eye faster. Anyway, I find that completely moot point, since I can only read couple lines of text in a second. It does not make any difference to me if some pages like for example this forum opens few milliseconds faster as long as machine does not get laggy with multiple tabs or resizing window type of things and jumping between apps. My MBP 14" has absolutely never any lag what so ever what ever I do.

If single web page opening speed is really important to some, here is a free tip - iPad Air 5 (M1) is faster in that than MBA M2 or MBP 14". Also iPad Pro 11" would be the same. Anyway, I find it hard to find any reason why that would make any difference to me, I don't browse adult sites, maybe something like that or similar pages is where this could matter.

 

[jdb8167]

How does the user know when throttling occurs? Is there a visual indicator such as a menubar icon or, perhaps, a notification? Is there a fixed threshold--for internal temperature--beyond which throttling occurs? Thank you!

I've now done some testing and the results are interesting. The M2 MacBook Air shows the throttling flag from the command line powermetrics tool much sooner than the M1 MacBook Air does when not using a fan running Cinebench or Wild Life Extreme. With a 30 minute Cinebench stress test the throttle flag shows up in about 4 minutes on the M2 where it takes over 10 minutes on the M1. But the M2 seems to maintain a higher performance with the throttle flag showing than the M1 does without.

The really interesting thing is using the cheap laptop cooler I can run a 30 minute Cinebench benchmark and the throttle flag never shows and I get a significantly higher score than without the cooler on the M2. On the M1 MBA on a 30 minute test without the cooler shows 6888. The M2 without the cooler shows 7342. The Cinebench results with the cooler are M1 7032 and the M2 MBA is 8106.

The M1 MBA shows almost no difference between using the laptop cooler and not. About a 2% speedup. The M2 MBA shows that throttling is more severe with a 10% speedup with the cooler. Even so, the M1 is still slower in all cases and even when throttled the M2 is faster than the unthrottled M1.

A graphics benchmark was also interesting with and without the cooler. The 10 GPU core M2 MBA without the cooler running the Wildlife Extreme Stress Test (20 minutes) showed throttling (again with powermetrics) after about 3 minutes. The final score was peak 6702 with the lowest 4365 with the average over 20 loops of about 4800. With the cooler, over 20 loops again, the range was 6711 to 6136 with the average about 6300. Interestingly, the throttling flag was set after about 5 minutes even with the cooler despite the test only showing a decrease of 8% vs 34% without the cooler.

With the M1 MBA on the cooler, the throttle flag never came on. Without the cooler the throttle flag came on after about 8 minutes. The 8 GPU core M1 MBA without a cooler showed a range of 4962 to 4105 with an average about 4380. With the cooler the M1 MBA showed a range of 4961 to 4786 with an average of about 4800.

Final thoughts, the M2 MBA with thermal throttling is at least as fast as the M1 MBA without throttling and in most cases faster for both CPU and GPU. For brief bursts, it looks like the extra 2 GPU cores help a lot but when the M2 is heat saturated, it looks like the 10 GPU core benefit is marginal at best. I don't have an 8 GPU core M2 to test on but I would love to see someone with an 8 CPU core M2 and a laptop cooler do similar tests. Ultimately, if you want to use your M2 MacBook Air for heavy GPU or CPU tasks, a laptop cooler can help a lot to keep your performance high over long tasks.

 

[0906742]

I've now done some testing and the results are interesting. The M2 MacBook Air shows the throttling flag from the command line powermetrics tool much sooner than the M1 MacBook Air does when not using a fan running Cinebench or Wild Life Extreme. With a 30 minute Cinebench stress test the throttle flag shows up in about 4 minutes on the M2 where it takes over 10 minutes on the M1. But the M2 seems to maintain a higher performance with the throttle flag showing than the M1 does without.

The really interesting thing is using the cheap laptop cooler I can run a 30 minute Cinebench benchmark and the throttle flag never shows and I get a significantly higher score than without the cooler on the M2. On the M1 MBA on a 30 minute test without the cooler shows 6888. The M2 without the cooler shows 7342. The Cinebench results with the cooler are M1 7032 and the M2 MBA is 8106.

The M1 MBA shows almost no difference between using the laptop cooler and not. About a 2% speedup. The M2 MBA shows that throttling is more severe with a 10% speedup with the cooler. Even so, the M1 is still slower in all cases and even when throttled the M2 is faster than the unthrottled M1.

A graphics benchmark was also interesting with and without the cooler. The 10 GPU core M2 MBA without the cooler running the Wildlife Extreme Stress Test (20 minutes) showed throttling (again with powermetrics) after about 3 minutes. The final score was peak 6702 with the lowest 4365 with the average over 20 loops of about 4800. With the cooler, over 20 loops again, the range was 6711 to 6136 with the average about 6300. Interestingly, the throttling flag was set after about 5 minutes even with the cooler despite the test only showing a decrease of 8% vs 34% without the cooler.

With the M1 MBA on the cooler, the throttle flag never came on. Without the cooler the throttle flag came on after about 8 minutes. The 8 GPU core M1 MBA without a cooler showed a range of 4962 to 4105 with an average about 4380. With the cooler the M1 MBA showed a range of 4961 to 4786 with an average of about 4800.

Final thoughts, the M2 MBA with thermal throttling is at least as fast as the M1 MBA without throttling and in most cases faster for both CPU and GPU. For brief bursts, it looks like the extra 2 GPU cores help a lot but when the M2 is heat saturated, it looks like the 10 GPU core benefit is marginal at best. I don't have an 8 GPU core M2 to test on but I would love to see someone with an 8 CPU core M2 and a laptop cooler do similar tests. Ultimately, if you want to use your M2 MacBook Air for heavy GPU or CPU tasks, a laptop cooler can help a lot to keep your performance high over long tasks.

I believe all those you mentioned have been said in the many reviews already.

Anyway, you really find this more nicer looking and convenient over MBP 14" with build in very silent fan?
From what I have seen these laptop coolers, they are IMHO ugly as hell.
I mean if you really do something that needs active cooling with your MBA M2, to me sounds like you actually need Pro.

 

[jdb8167]

iPad Air 5 (M1) is faster in that than MBA M2 or MBP 14".

Why would that be the case? Do you have Speedometer numbers for the M1 iPad Air? As far as I know, the M2 has the fastest Speedometer V2 numbers ever recorded.

 

[jdb8167]

I believe all those you mentioned have been said in the many reviews already.

Anyway, you really find this more nicer looking and convenient over MBP 14" with build in very silent fan?
From what I have seen these laptop coolers, they are IMHO ugly as hell.
I mean if you really do something that needs active cooling with your MBA M2, to me sounds like you actually need Pro.

I haven't seen any reviews comparing an unthrottled (mostly) M2 MBA to a throttled one along with the M1 MBA. But I like to confirm things for myself anyway.

For me the laptop cooler was just for this experiment. I might use it at some later date again if I decide to play a GPU intensive game but I'm not much of a gamer. The cooler was $20. Cheap enough that I could throw it out after this experiment if I wanted to.

 

[0906742]

Why would that be the case? Do you have Speedometer numbers for the M1 iPad Air? As far as I know, the M2 has the fastest Speedometer V2 numbers ever recorded.

Because I've seen it myself. I don't give a s**t what that theoretical benchmark shows. Obviously it is not just CPU that makes the difference here. Go test it yourself if you don't believe. All web pages and every animations and all just open like in fast forward in iPad Air 5, it feels almost ridiculous fast in some cases.

 

[jdb8167]

Because I've seen it myself. I don't give a s**t what that theoretical benchmark shows. Obviously it is not just CPU that makes the difference here. Go test it yourself if you don't believe. All web pages and every animations and all just open like in fast forward in iPad Air 5, it feels almost ridiculous fast in some cases.

OK, I'm glad you are happy with the M1 iPad Air performance but without numbers, I'm going to doubt than any M1 is faster on any mostly single thread task than an M2.

 

[0906742]

I'm going to doubt than any M1 is faster on any mostly single thread task than an M2.

It ain't. That's what I'm telling you! It is just faster in real life, so CPU speed is one thing, OS is another, as well as screen resolution, browser etc.... Again, go test is yourself and you see. Don't just blindly trust benchmarks.

 

[jdb8167]

It ain't. That's what I'm telling you! It is just faster in real life, so CPU speed is one thing, OS is another, as well as screen resolution, browser etc.... Again, go test is yourself and you see. Don't just blindly trust benchmarks.

Like I said, I need numbers. Confirmation bias is a real thing. I don't have an iPad Air and I'm not going to an Apple store just to test an iPad that I would never buy so there isn't an easy way for me to test for myself.

 

[0906742]

Like I said, I need numbers. Confirmation bias is a real thing. I don't have an iPad Air and I'm not going to an Apple store just to test an iPad that I would never buy so there isn't an easy way for me to test for myself.

I need to get my stopwatch ready. It is noticeable faster. But obviously benchmark numbers are your thing over real world performance, so why bother.

 

[jdb8167]

I need to get my stopwatch ready. It is noticeable faster. But obviously benchmark numbers are your thing over real world performance, so why bother.

I use an M2 MacBook Air every day for real work. Benchmark numbers aren't my thing over real world use but they can be useful. I was interested in exactly how much throttling affects the more extreme use cases for the M2 MacBook Air. I don't have any typical workload that causes any throttling so my real world use isn't really relevant.

This thread is titled, "How hot does your M2 MBA run? Doing what?". I was interested in answering more than, my M2 MBA is never hot for my normal development work. Sorry if you don't find that interesting but then I wonder why you are on this thread at all? Thread hijacking is a thing on MR forums so I guess it doesn't matter.

 

[Johnnyseeth]

Mine got hot last night. Lots of tabs open in Safari. I will have to do more testing. I am running a Beta: OS Ventura.

 

[Tagbert]

…

 

[Tagbert]

Mine got hot last night. Lots of tabs open in Safari. I will have to do more testing. I am running a Beta: OS Ventura.

It is usually not the number of tabs, but specific websites running in un-optimized ads and javascript frameworks that cause performance problems. If you look at Performance Monitor you can see what RAM and CPU individual tabs are consuming. It can get rediculous. Forums.macrumors.com will slowly eat up RAM if you let it sit. I regularly see 2GB for that domain before I close the tab. Haven’t checked the CPU.

 

[Johnnyseeth]

It is usually not the number of tabs, but specific websites running in optimized ads and javascript frameworks that cause performance problems. If you look at Performance Monitor you can see what RAM and CPU individual tabs are consuming. It can get rediculous. Forums.macrumors.com will slowly eat up RAM if you let it sit. I regularly see 2GB for that domain before I close the tab. Haven’t checked the CPU.

Yes I know this. I regularly have 16 GB of RAM wired in at any given time. I have ad block activated as well shutting down many ads.

 

[0906742]

I use an M2 MacBook Air every day for real work. Benchmark numbers aren't my thing over real world use but they can be useful.

I see, you could have fooled me since it looked like the benchmark score was the main thing.

Sorry if you don't find that interesting but then I wonder why you are on this thread at all? Thread hijacking is a thing on MR forums so I guess it doesn't matter.

Looked to me this topic was hijacked already.
It was some other persons in this thread earlier who spoke about single score speed and keep bringing up this web pages load faster in MBA M2 argument, so I just thought to comment on this ”important” web page load faster issue that iPad Air/Pro with M1 beat all MacBooks in this.

 

[Johnnyseeth]

I see, you could have fooled me since it looked like the benchmark score was the main thing.


Looked to me this topic was hijacked already.
It was some other persons in this thread earlier who spoke about single score speed and keep bringing up this web pages load faster in MBA M2 argument, so I just thought to comment on this ”important” web page load faster issue that iPad Air/Pro with M1 beat all MacBooks in this.

People hijack threads here and incessantly argue in the context of their worship for a fruity company.

My M2 Air is holding up well but does have its performance limitations.

 

[Chateaunole-du-Pape]

I need to get my stopwatch ready. It is noticeable faster. But obviously benchmark numbers are your thing over real world performance, so why bother.

Come on, now. In another thread, I presented you with real-world video tests from an apparently well-known reviewer demonstrating that the M2 Air does load web content noticeably faster than the 14" M1 Pro, and you dismissed them out-of-hand. "Oh, maybe she pushed the button on the M2 first." And now you want us to believe that you care about real-world performance?

For whatever reason, you seem to derive pleasure from trashing the M2 Air. That's entirely your prerogative. But don't dismiss multiple real-world tests and then tell us that benchmarks are nonsense and you only care about real-world tests.

 

[0906742]

Come on, now. In another thread, I presented you with real-world video tests from an apparently well-known reviewer demonstrating that the M2 Air does load web content noticeably faster than the 14" M1 Pro, and you dismissed them out-of-hand. "Oh, maybe she pushed the button on the M2 first." And now you want us to believe that you care about real-world performance?

First of all, as all can see the difference between the two in the video you was referring to is so small that you had to really pay attention to even notice it and differences such small are already affected by which machine had button pressed first. I don't question your reviewer, but personally I've never heard of this Ella before, but I take your word this is a well-known reviewer.

However since some MBA M2 owners keep bringing up this only thing where MBA M2 has blink of an eye difference in single web page loading speed compared to M1 counterpart MacBooks like MBA/MBP all models.
Due to this obviously important feature of single page loading time being the most important thing in notebooks for some people, I wanted to give them a free tip that M1 based iPad Air and Pro actually beat MBA M2 easily and noticeable, you don't need even stopwatch to noticed that.

Since it does not matter obviously what I say, I just post a link to the TechRepublic site where they already measured the difference between iPad Pro M1 vs MBP 13" M1 single not previously cached web page loading time and the difference was huge, iPad Pro M1 loaded page in 5.6 seconds where MBP 13" M1 took 8 seconds. That is like night and day difference.
Now considering that MBA M2 is only blink of an eye faster loading single page than M1 MBA/MBP counterparts, we can safely see from that link that M1 based iPads are superior in web page loading speed.
Also in app starting and machine starting/powering up times, the difference is huge, iPad Pro M1 being many times faster, so obviously this is important to people who single page web loading test being blink of an eye faster is already required.
Here is link to this test: https://www.techrepublic.com/articl...ipad-pro-which-is-quickest-in-the-real-world/

Maybe you could ask Ella make this test between MBA M2 and iPad M1 Air or Pro, so we can have official result between those two?

For whatever reason, you seem to derive pleasure from trashing the M2 Air. That's entirely your prerogative. But don't dismiss multiple real-world tests and then tell us that benchmarks are nonsense and you only care about real-world tests.

Again, this was just to put in to perspective this MBA M2 owners trashing even MBP 14"/16" being slow because they are blink of an eye slower loading single web pages, or when in some synthetic benchmark can be seen difference between the two, when in everything else MBP 14"/16" beat easily MBA M2. Even in web pages where happens to be some scripts taking CPU power and putting MBA M2 on its knees due to throttling it goes when CPU gets used for a bit longer time. There are much more benchmark tests where MBP 14"/16" is much faster than MBA M2, unlike you try make it sound like. I know truth is sometimes hard to accept.

 

[LeeW]

Mine is great from a heat perspective. Always cool unless I play a bit of World of Warcraft but even then it's warm, not hot.

 

[ascender]

How hot does the M2 MBA get? What kind if work is it doing when it gets hot?

Do you find rhe heat annoying or tolerable? Or hardly noticeable?

Cool to touch apart from if Windows VMs are doing system updates and I've felt it get a wee bit warm to touch. It can get slightly warm when playing Football Manager 2022 on the match engine, but that seems to have stopped - not sure if that was because I was playing that in the first few days of purchase when its indexing etc.

But normal use with lots of apps, tabs, Citrix, Parallels VMs etc its cool to the touch.

 

[Technerd108]

How do you find the m2 air to be faster than the m1pro when web browsing? I would have thought the pro-motion etc would have made the 14” feel superior? Do you get lag with 100s of tabs open on the m1pro but not on the m2 air? Is there lag when resizing windows or switching tabs?

No I don’t have a hundred tabs open on either. Lol.

Promotion is good when slowly scrolling a web page and for animations feeling a little smoother. However in real world usage the animations feel the same on both.

Honestly I often wondered what the 120hz was doing. Like was Promotion using the higher refresh rate and when since it is a dynamic system that adjusts in real time depending on usage. A lot. Of the times I felt like it was scaling back to save power. So in day to day usage I really didn’t notice a difference from my 60hz windows laptop or M1 MBP Scrolling would be the only time I saw a difference but if I scrolled fast both displays would blur text so in my opinion the difference is minimal. Really the difference shows mostly on a tablet or phone.

No lag when resizing windows or switching tabs on either pro or air at all.

The lag is when loading up a webpage. Sometimes using the mail application and other applications where you are loading content. This is where the M2 air was noticeably faster. Webpages load almost instantly with the air. While the M1 MBP 16“ M1 max would not.

Again I am not saying the M1 was slow but when you buy a very expensive and supposedly pro laptop you expect everything to be instantaneous. Maybe that is a little unrealistic but that was my expectation.

When using the M2 everything is pretty much instantaneous including app switching.

I definitely find the M2 air to be faster while browsing but I don’t ever have more than 5 or so tabs. The M1 Max with 32 gb ram was slower in browsing than my M2 air with 16gb ram. Now neither is slow so let’s make sure that is stated.

You got to ignore the trolls. They be trolling, trolling, trolling……Lol. It is crazy on this forum how people just want to argue instead of help.

 

[GooseInTheCaboose]

No I don’t have a hundred tabs open on either. Lol.

Promotion is good when slowly scrolling a web page and for animations feeling a little smoother. However in real world usage the animations feel the same on both.

Honestly I often wondered what the 120hz was doing. Like was Promotion using the higher refresh rate and when since it is a dynamic system that adjusts in real time depending on usage. A lot. Of the times I felt like it was scaling back to save power. So in day to day usage I really didn’t notice a difference from my 60hz windows laptop or M1 MBP Scrolling would be the only time I saw a difference but if I scrolled fast both displays would blur text so in my opinion the difference is minimal. Really the difference shows mostly on a tablet or phone.

No lag when resizing windows or switching tabs on either pro or air at all.

The lag is when loading up a webpage. Sometimes using the mail application and other applications where you are loading content. This is where the M2 air was noticeably faster. Webpages load almost instantly with the air. While the M1 MBP 16“ M1 max would not.

Again I am not saying the M1 was slow but when you buy a very expensive and supposedly pro laptop you expect everything to be instantaneous. Maybe that is a little unrealistic but that was my expectation.

When using the M2 everything is pretty much instantaneous including app switching.

I definitely find the M2 air to be faster while browsing but I don’t ever have more than 5 or so tabs. The M1 Max with 32 gb ram was slower in browsing than my M2 air with 16gb ram. Now neither is slow so let’s make sure that is stated.

You got to ignore the trolls. They be trolling, trolling, trolling……Lol. It is crazy on this forum how people just want to argue instead of help.

Thats interesting about the speed differences with webpages. Why do you think it is? Do you think it is the number of efficiency cores? (Iirc the m2 has more of them compared to the m1pro; the n1pro has more performance cores instead)

Do you think the future m2pro mbp will be as fast/faster than the m2 air in this regard? (Web browsing) Or do you think the Air will still have the edge?

 

[Technerd108]

Thats interesting about the speed differences with webpages. Why do you think it is? Do you think it is the number of efficiency cores? (Iirc the m2 has more of them compared to the m1pro; the n1pro has more performance cores instead)

Do you think the future m2pro mbp will be as fast/faster than the m2 air in this regard? (Web browsing) Or do you think the Air will still have the edge?

It has to do with single core speed. M2 Pro should be just imperceptibly faster than M2 as they should have the same single core speed but the Pro will have more bandwidth in the ram. Anything that uses multi core will also be a lot faster than M2.

I am really excited to see how powerful the new M1 Pro/Max will be. Obviously it will be nothing like the transition from Intel but I don't discount the new a15 cores and new process. Also I think the gpu on the new M2 Pro series will be a lot better than M1. More significant than the cpu gains.