Apple M1 Hands-On Comparison: MacBook Air vs. MacBook Pro vs. Mac Mini

[Unregistered 4U]

The best example I can bring up is the i9 MacBook Pros. Obviously they were faster than the i7 ones in bursty tests, but (at least the first year) they were actually slower in long-running tests.

If Geekbench showed that the i9’s were faster and they were actually faster at the same type of work, then that’s not really that different. Vastly different would be more like if the i7 was faster than the i9 in ALL tests, but Geekbench said it was slower.

Bursty tasks are common too, so the Geekbench results are useful for that. The biggest reasons I can see to pick Geekbench as THE test is 1. they're clearer about the differences between versions 2. long-running test results will depend more on the room temperature and other factors, making them less clean. But I think it's still too far from the average use case.

It’s not just that bursty tasks are common, bursty tasks are the way the majority of folks experience their computers every day. I’d guess that for over 90% of the tests on Geekbench, that’s the heaviest load those computers have been placed under and will ever be placed under.

When you consider this, long-running tests are the edge cases that don’t really reflect how most folks use their systems on a day-to-day basis. Geekbench shows that the Air, Mini, and Pro are virtually the same performance for the type of tasks that your average person will experience. I TOTALLY understand those that may need ADDITIONAL tests above and beyond this, but Geekbench appears to mirror the use cases of most folks really really well.

 

[hot-gril]

If Geekbench showed that the i9’s were faster and they were actually faster at the same type of work, then that’s not really that different. Vastly different would be more like if the i7 was faster than the i9 in ALL tests, but Geekbench said it was slower.


It’s not just that bursty tasks are common, bursty tasks are the way the majority of folks experience their computers every day. I’d guess that for over 90% of the tests on Geekbench, that’s the heaviest load those computers have been placed under and will ever be placed under.

When you consider this, long-running tests are the edge cases that don’t really reflect how most folks use their systems on a day-to-day basis. Geekbench shows that the Air, Mini, and Pro are virtually the same performance for the type of tasks that your average person will experience. I TOTALLY understand those that may need ADDITIONAL tests above and beyond this, but Geekbench appears to mirror the use cases of most folks really really well.

"Long-running" here doesn't mean exporting something in Final Cut for hours, just loading the machine for long enough that the temperature stabilizes. That's pretty common in the days of inefficient software like Google Meet and Docs running side-by-side, which present a thermal throttling issue for average people doing their jobs from home. Geekbench only runs for about 2 minutes, and I'd say a long-running test takes at the very least 5.

Not sure what you mean by i7 vs i9. In the real world, the i9 MBP is slower if you let it heat up, that's why people complained.

 

[ValueArb]

err firebase ? I'm not google staff nor qualify to be their staff. if they can change simulator to emulator a god send. My dream is I can test directly XCode in iPad.

Sorry, wasn’t implying you work on Firebase. Just a shout out to them to please fix their problems.

 

[FryerTuck18]

Again, more false information. There is a 7 core GPU version with the 512 GB SSD, and there is an 8 core GPU verison with the 512 GB SSD There is no "automatic" upgrade to the 8 core version if you select the 512 GB SSD as a build option on the 7 core GPU machine. This is just like last week when you said the Pro is only $50 more, but neglected to mention the $200 needed to bring the storage back up to the same amount as the Air.

View attachment 1678113

Seems to be a bit of difference on the web site

 

[Unregistered 4U]

Geekbench only runs for about 2 minutes, and I'd say a long-running test takes at the very least 5.

Not sure what you mean by i7 vs i9. In the real world, the i9 MBP is slower if you let it heat up, that's why people complained.

When you combine fast low power processor with light jobs, are there any of the Apple Silicon systems that will heat up in 5 minutes or less? If so, the Geekbench is going to become more and more accurate where it relates to Apple’s M series. (Though I’m sure folks will say that the benchmarks for the M2 being better than the M1 are just that, BENCHMARKS! We’ll need real world testing like 24 hour 100% loads!)

Oh, and by i7 vs i9, I meant that Geekbench wasn’t VASTLY incorrect and I don’t think any of their performance expectations have been wildly incorrect. When they indicate a system is faster than another system, it’s generally true and any differences are due to system and environment variables. That’s a pretty accurate test, generally speaking.

 

[hot-gril]

Oh, and by i7 vs i9, I meant that Geekbench wasn’t VASTLY incorrect and I don’t think any of their performance expectations have been wildly incorrect. When they indicate a system is faster than another system, it’s generally true and any differences are due to system and environment variables. That’s a pretty accurate test, generally speaking.

Well they were wrong about which is faster, but the question is also how much faster. So, looking at the sources...

https://www.computerworld.com/artic...ook-pro-core-i9-throttling-story-updated.html The issue was the 2018 15" i9 MBP was slower than the 2017 i7 one. Turns out the test wasn't good, but the complaint was accurate. Also, Apple said it was a software problem with the throttling, but their fixes didn't really fix it.

MacBook Pro Performance (July 2018) - Geekbench Blog

www.geekbench.com

Multicore 22439 for the i9, 15548 for the older i7. So it was supposed to be 44% faster. If it's indeed slower, the Geekbench results are very misleading here.

https://www.notebookcheck.net/MacBo...k-results-after-software-update.318234.0.html supports the claim of thermal problems. The 2018 15" gets 1064 on Cinebench R15 multicore. Now, I can't find a 2017 15" MBP Cinebench R15 results anywhere, and I've already spent too much time looking. But if anyone else finds one, I'm curious whether Cinebench would've caught this discrepancy.

 

[Born Again]

bring on the touch screen macbooks!!!!!!!!

sorry you disagree but isn't the macbook just a big ipad with keyboard?

 

[abtechgroup]

Apple's M1 Macs are out in the wild now, but ahead of the holidays, you might still be trying to figure out which one to pick up, either for yourself or as a gift for someone else. We've got all three of the new Macs available, so we thought we'd give MacRumors readers a hands-on overview of each machine in our latest YouTube video.

Subscribe to the MacRumors YouTube channel for more videos.​

Price wise, the Mac mini is the cheapest of the bunch at $699, while the MacBook Air starts at $999 and the MacBook Pro starts at $1,299. With the Mac mini you do, of course, need to provide all of your own peripherals, including display, mouse, and keyboard at minimum.

If you already have those on hand, the Mac mini is potentially the way to go. For the notebooks, the MacBook Air is lighter, but there's not a huge amount of difference in size between the Air and the Pro.

The entry-level MacBook Air has a 7-core GPU while the entry-level MacBook Pro has an 8-core GPU, which is why the price is a bit higher. If you get the 512GB SSD with the MacBook Air, you get upgraded to that 8-core GPU automatically.

There are some thermal differences between the M1 Macs, but most people may not be able to distinguish between them when it comes to performance. The Mac mini has the most space for heat dissipation, while the MacBook Pro has a fan for better thermal management.

The MacBook Air, meanwhile, has no fan. You may see some slightly better sustained performance when doing system intensive tasks on the Macs with better cooling, but there's not a whole ton of difference between them. In most benchmark tests, they're all just about the same, and they're all huge upgrades over their Intel counterparts. Watch our M1 MacBook Pro vs. 2020 MacBook Pro test below to see.

As for upgrades, there's not a lot available. You can bump up the SSD storage, which is a good idea if you plan to use onboard storage and want a machine that's going to have enough space for years to come, but you can also work with external SSDs if you opt for the 256GB minimum.

You can also upgrade the RAM on each M1 Mac from 8GB to 16GB. If you have the extra $200, it's a good idea to do so for futureproofing and because you're never going to go wrong having extra RAM. This year's Macs have unified memory architecture which basically means all of the chip components can draw from the same memory pool, bringing some speed and efficiency improvements.

Apple is excellent at optimizing RAM usage and there's no exception here. You may not even need 16GB because in our testing, the M1 Macs are super quick even with the built-in 8GB.

All of the M1 Macs offer incredible performance at their price points, and in our testing, they've all outshone their Intel counterparts. macOS Big Sur runs great on the M1 Macs, and apps built for Apple Silicon are fast and efficient. Even apps not built for Apple Silicon and running under the Rosetta 2 translation layer are close to as fast as they are on Intel machines.

If you rely on specific Windows apps and regularly use Boot Camp, the M1 Macs are not for you. There is no way to run Windows on an M1 Mac at this time. CrossOver does allow for some Windows apps to be used, but this solution may not be worth the effort for heavy Windows users.

Touch ID also seems to be quicker on the M1 Macs in comparison to Intel Macs, and even small tasks like waking from sleep or switching resolutions are instantaneous. All of the Macs are close to whisper quiet. The MacBook Air makes no sound at all, while the Mac mini and the MacBook Pro rarely activate their fans even under intensive tasks.

Apple has made some serious battery life promises with the M1 Macs, and the MacBook Pro can last for up to 20 hours. We've seen that impressive battery life in day to day usage. Intel Macs have never been able to hold up to editing a video with a decent amount of battery life to spare, but the M1 Macs have no problem.

The M1 Macs are amazing so far, but it's worth keeping in mind that these are first generation machines. Apple is upgrading the entire Mac lineup to Apple Silicon chips, a process expect to take two years. These are low-end Macs that have been refreshed, and there are some more exciting, higher-end Mac refreshes in the works which could be one reason to hold out on a purchase right now.

One other consideration is design. Apple did not update the design of the new Macs, and there are some design changes rumored to be in the works for the MacBook Pro at a minimum. A fresh design could also be worth the wait.

If you need an affordable Mac for everyday tasks, the M1 Macs are a solid choice. The Mac mini is the one to get if you want a desktop, and if you need a notebook, the MacBook Air is going to work for most people. The MacBook Pro has an extra GPU core in the base model, a brighter display, and better speakers, which is a consideration. We have a full guide on the MacBook Air vs. the MacBook Pro if you're trying to choose between them.

Do you have an M1 Mac or are you going to buy one? Are you holding out for next year's Apple Silicon offerings? Let us know in the comments.

Article Link: Apple M1 Hands-On Comparison: MacBook Air vs. MacBook Pro vs. Mac Mini

There is no way to run windows on an M1? I picked up the Pro with M1 and it has bootcamp... so why not? And what about parallels? Anyone running parallels on an M1? Windows working on it?

 

[haruhiko]

There is no way to run windows on an M1? I picked up the Pro with M1 and it has bootcamp... so why not? And what about parallels? Anyone running parallels on an M1? Windows working on it?

As Craig said, it’s all up to Microsoft.

 

[Tagbert]

I never use touch on my work Thinkpad that has touch. Hardly anything is optimized for it.

I used to have a Windows laptop with touchscreen. It was sometimes nice to use touch for scrolling or touching buttons or moving things on screen. Some things just seemed more natural with touch. I would never expect touch to be the main interaction method for such a device. Most interactions were keyboard or trackpad but it was good to have an alternative sometimes.

I also use an iPad in laptop mode with a keyboard and mouse. Even there it often feels better to use touch for some things. Touch can be more interactive and engaging. Yes, there are some smudges on the screen and I clean it daily with a microfiber cloth. Not something that I fret over.

One thing about having multiple interaction modes is that it helps avoid RSI by switching between keyboard, trackpad, and touch and not always using the same movements. Touch does not have to be primary to be useful.

 

[Tagbert]

Oh it'll throttle, but it'll be faster than a throttling Intel Mac.

It will also run non-native apps in Rosetta 2 faster than Intel even while throttling

 

[patent10021]

Basically my dream laptop right now. I could live with 32gigs of RAM.

Yeah, can guarantee you most if not all pro users want MagSafe back. At the same time, trash this BS iPhone MageSafe business.

 

[tallscot]

Raise of hands - how many are video editing where there is no power?

 

[cyberslammer]

Thanks for the information - Haven't slept for a few nights wondering if you would upgrade.


Really feel Apple nailed the transition WAY better than what I expected. Just wondering if laptop or Mini setup is the way to go next - the combination of an iPad Pro and Mini might be great - both look amazing.

No problem, glad you think someone gives a rip what you own:

| 2015 Space Grey MacBook | iPhone 11 Pro Midtnight Green | Apple Watch Series 4 Stainless Steel 44 mm Saddle Brown Classic Leather Buckle & Stainless Steel Link Bracelet |

 

[Tech198]

"extra money to spare" = yes i do, but i'd rather spend it elsewhere

I dunno about anyone else. but costs still always comes to play despite anything else. The only exception is if you MUST have it, not just because everyone else says you must.

 

[xDKP]

No problem, glad you think someone gives a rip what you own:

| 2015 Space Grey MacBook | iPhone 11 Pro Midtnight Green | Apple Watch Series 4 Stainless Steel 44 mm Saddle Brown Classic Leather Buckle & Stainless Steel Link Bracelet |

It’s not really to satisfy your curiosity, but if we discuss products, a perspective might make more sense if you know what products the person is uses currently.

 

[sup123a]

I'll just wait for 12" MacBook M1.
I want a <1kg laptop.

 

[09872738]

I'll just wait for 12" MacBook M1.
I want a <1kg laptop.

I guess you‘ll be waiting for a looong time. I doubt there‘ll be a Mac below the MBA