It's the Roboto font in 300 weight, where 400 would be a normal weight. I believe Roboto was the Android default font, so probably lots of testing/tweaks on Android make it readable there.
You can turn on percentage in the menu bar, and there's several different usage graphs in multiple places.
If you want the actual milliamp-hours, load cycles, battery temperature, age, or even battery serial number, there's an api for that. You can use for example coconutBattery and get whatever data you want.
They reference future models having enough performance to be serious gaming machines, but in my experience the main reason Macs lose out on games is because Apple stops supporting older technologies very quickly, and many games rely on those. A good example is the recent end of 32-bit support, which removed about half my Steam library.
The games also need to be ported to metal and optimized for apple hardware, which just isn’t worth it. This was especially true when only high end macs had decent gpu’s. Now that even the mini and air have a gpu that can actually play a modern game the economics may change.
I have had good luck with dosbox games though. Many games on gog run fine on mac.
Not all of them. The AAA titles will need this, but there are a lot of games on the market built with Unity and have pretty basic graphics and will run fine on low-end laptops.
It really doesn’t make sense for MacOS (or iOS) to directly support Vulkan. A particular OS release would implement a limited range Vulkan versions (probably just one) which would always be out-of-date.
Apple would not be able evolve the OS, silicon and Vulkan API in unison (like they do with Metal) because they don’t control the API.
In the end it would be a poor experience for end users.
It would be nice if they supported open source efforts though.
(MS doesn’t maintain the Vulkan API in windows, do they? Similar issues would apply.)
Why exactly? Somehow AMD and Nvidia manage to keep up with the Vulkan releases on multiple platforms with their drivers for their hardware. Should be easier when you control the whole stack.
I don't think they update Metal in past OS releases either. You're lucky to get bug fixes.
> The games also need to be ported to metal and optimized for apple hardware, which just isn’t worth it.
The Macs now have the same GPU and Graphics API that developers already have to target for iOS devices, where there is a significant market share and amount of revenue in play.
On the other end - that's why it can do hardware that has this kind of battery life - it can freely drop older technologies and aggressively optimize for specific hardware.
Unless they want to make macs 3x thicker and throw out battery life it won't happen. People already don't want to develop games for linux, even with the same hardware as on PC, much less try to port to apples special APU's.
The power and size required for "serious gaming" GPU's on mobile is far too significant to fit into a mac size at this time. The term "gaming laptop" is a meme - by the time you've spent $2000 on a gaming laptop you've just bought a jet engine the shape of a brick that does half the work of a desktop at double the price and is terrible for productivity outside of specific use cases. (Like presenting 3d renders or something)
I suppose you could do it if apple were to support proton, and if amd/Nvidia were to standardize their external GPU implementations such that you can just plug in your egpu and have at it. But afaik they're very expensive and use case fitted.
This used to be true but it doesn't reflect reality. Yes if you want to play the latest and greatest games at 4k you do need a gaming laptop. But a lot of games coming out nowadays work pretty good on even Intel iGPUs at full hd medium graphics.
I'd wager you could easily game on an M1 if the game has support and you are willing to make some graphics concessions.
Apple could do very, very, very well if they targeted the nintendo switch level of 'serious' gaming. And there are many very good games for that segment. They also have all the kids already with iPads & Roblox. There are way more iPads & macbooks out there than there are nintendo switches, and if apple actually made a set of dedicated serious gaming hardware & efforts for the platform they would wipe nintendo on the floor.
TBH the macbook air would make a better switch than the switch as far as multiplayer portability goes. The switch screen is too small to play something like smash bros on it with friends, and the 'travel dock' solutions are too clunky IMO.
> There are way more iPads & macbooks out there than there are nintendo switches, and if apple actually made a set of dedicated serious gaming hardware & efforts for the platform they would wipe nintendo on the floor.
I really doubt it, Apple's model of constant updates requiring constant developer upkeep is incompatible with single purchase games. Only exploitative games as a service with recurring revenue can thrive on a model of endless upkeep. Its a big reason (besides Nintendo's first party quality, an attribute Apple has not demonstrated it is capable of) why the App store is defined by low quality p2w service games.
Its overkill for microsoft word, but its right in line with workstation applications - engineering,programming,content creation, etc all need beefy CPUs and GPUs. They're not really niche. Workstation laptops do sell quite a bit. Gaming laptops overlap with them.
Your view of gaming laptops is a couple of years out of date. There are plenty of capable gaming laptops now with slim profiles and while battery life is still not their forte, that's only when they are actually gaming which isn't usually something you engage in without at least being relatively close to an outlet.
My 2019 Intel MacBook Pro can run AAA games really well via Bootcamp. My Bootcamp partition runs windows faster and more stable than any of my windows machines.
Very quickly being a 10+ year period after 64bit was made standard. MacOS was never good for gaming, no point keeping dead tech on life support for literally decades if it isn't bringing much value.
The future of gaming on mac will probably be Apple Arcade and streaming services.
My M1 macbook pro battery lasts so long and doesn't get hot it's amazing. You don't have to use the laptop like a laptop any more. Range anxiety is totally gone. I can watch youtube videos without fear of burn or running out of battery fast. I thought it would be good, but not THIS good. It's actually cannibalized my iPad usage since that used to be the battery champ for me.
In comparison, my Thinkpad P1 Gen 3 I bought a month ago keeps the fans running while it is sleeping. The machine is warm to the touch all the time. From what I've read, Lenovo blames it on Microsoft's modern sleep requirements.
This morning I configured it to hibernate after 30 minutes, so I hope that fixes the problem.
I'm staying away from Lenovo ever since buying an IdeaPad laptop in 2018 with a fan so obnoxious that I literally could not get any work done. Lenovo needs to get their shit together.
Go into the bios and change the sleep setting so that it uses S3 sleep (traditional) rather than S5 sleep (modern). Now your computer will actually turn off and stay off, and it's only using a very small amount of power to keep the memory on.
It's comical how bad Windows S5 sleep is. Laptops waking up and running while they're in a bag should be the kind of bug that prevents an operating system from being released! But here we are in 2021, years after Windows 10 was released, and this is still common place.
S3 sleep has been removed from the options for devices. Dell actually pushed out a bios update that removed the option. I can see what MS wanted to do with modern sleep, the macbook is able do to things like beep when you ping it via find my while it is sleeping which is impossible in S3 sleep. But the current execution of it is horrible.
For me that’s been the biggest difference coming from my old laptop.
I no longer have any kind of battery anxiety, it probably lasts a couple of days or more (9 or 10 hours a day use IntelliJ web and python dev, bit of YouTube etc) I don’t have to panic to plug it in every night and I’ve taken it away for the weekend without the charger.
I hope they don’t use this performance to shrink the battery of later models to a day or less. You really need a couple days to keep margin for no battery anxiety over the life of a machine.
Docker is pretty bad. I also work out on the patio pretty often, cranking up the screen brightness will also give it a hit.
But still unimaginably better than the Intel notebooks. With Docker and the brightness up I'll get a measly 6.5 hours instead of the solid ten or so I get otherwise.
Docker is currently using qemu for x86-64 to arm translation so when I use it with an x86-64 container I only get 2-3 hours out of it, and the performance is atrocious bc of how slow qemu is. With arm64 containers I tend to get 5-8 hours even with cpu intensive tasks going most of the time.
I expect either qemu to get better or for apple to add some things to improve running x86-64 VMs. Rosetta 2's JIT portion (some of R2 is AOT) is worlds ahead of qemu's in terms of performance. So, the battery life and performance difference won't be forever.
Thankfully, building arm64 images for most containers I've used is pretty straightforward.
Leading up to and since the M1 Macs were released, the tech community went from skeptical, to intrigued, to excited, and if you were to completely remove Apple's hype machine, these machines stand on their own.
As an M1 user, I can attest to the ridiculously good battery life, and others in this thread are sharing similar anecdotes (I'm on an Air, so I'm not even on the best battery life available).
It is with this context that a statement from Apple's VP of marketing on the same topic is somewhat interesting. The performance and battery life surprised many of us in the tech community, and this anecdote seems to confirm that even at Apple, execs were/are downright surprised at how good this thing is.
If this statement was released before these machines launched, or if performance/battery weren't stellar, then a statement like this from a marketing VP might ring hollow. But in the context of this product's success, I think it's more interesting.
Among other things, the MacBook batteries are larger than the iPad batteries. Power consumption of the screen other components will obviously be different, but I would imagine that the MacBook Pro having a battery ~2x as large as that of the smaller iPad Pro would make a big difference.
I just got a 13" Macbook Pro with M1. My expectations were exceedingly high, based on all the reviews and stories like this.
After a few days of use, my take: It's good. It's not the messiah-of-tech that I was led to believe, but it's good.
Honestly, I think the hype is overblown. If it wasn't for the hype, I might have thought this machine was amazing. But the expectations being set by all this hype is crazy town.
Not trying to take away from the real technical accomplishment. I'm sure if I put it through the ringer with benchmark tests, it would beat everything with flying colours.
But my day-to-day experience using it hasn't been much different than my previous 2016 Macbook Pro. Yet so many reviews make it seem like everything is so fast that you simply have to experience it to believe it, that every action you take feels faster and better. And that, in my own experience, is proving to be a load of B.S.
To me it’s not the performance that’s the “messiah-of-tech” part of the M1 laptops (have an Air myself).
It’s that I can do the same things at the same speed as an Intel MacBook, but:
* The machine is dead silent (literally no sound coming from it)
* The battery life is incredible
I’m not sure if it’s as simple as this, but I guess if they said: we don’t mind blaring fans and 6 hours of battery life, then they could clock the M1 so high it would have truly insane performance.
I think the mini and iMac both serve as indicators that this isn’t true - they have identical performance. I’m willing to guess that thermal transport for the SoC isn’t all that good and while a fan would help likely not as much as you’re crediting. ASIC teams design for a specific envelope.
I'm on the Air, and I had medium expectations because my previous work laptop (16" Pro, i7, 32GB RAM) was already incredibly snappy and rarely struggled except occasionally with Docker.
I was still very impressed even with my tempered expectations. For the kind of work I do, it's just as capable as the 16" MBP was - but with no heat, no noise, and real all-day battery.
My experience exactly mirrors yours. Battery life is really good. Processing speed is good, but it's not like stuff compiles instantly now. There are, contradicting reviews, waiting times. I've seen many beach balls using my MBP and it comes free with a couple of issues:
- After a few hours of developing an Angular app using ng serve kernel_task consumes insane amounts of CPU time, resulting in extreme lag and completely locking up the device on an attempted restart (not even the touchpad clicks anymore).
- Sometimes MPEG4 videos just would not play until after a reboot
That said: Besides some minor UI glitches it's really, really solid. x86 compatibility is perfect. The GPU is good (again, not amazing, but Minecraft runs at good settings).
Over all it's not revolutionary. Slow software is going to be slow on these too. What they pulled off is amazing ecosystem wise, but as an end user you can be happy about some non-revolutionary pretty good decent spec bumps.
I just worry about the SSD's life. I'm already about 20TB down after a few weeks.
I'm on the 12.0 beta, though. The article also doesn't put any hard numbers on the actual expectancy. If the smartmontools reading is wrong, what's my estimated actual TBW? Because at that speed it won't take all that long to reach the rated amount for a typical SSD (even at a multi X factor we'd still be talking about a low number of years).
That's my understanding. It's how the OS accounts for the overhead that's unavailable. So if kernel_task is using 20% leaving 80% for userland, what's actually happening is your cpu is throttled to 80%.
It's also why it's the go-to to blame for performance issues - it's the symptom not the cause. Your laptop isn't running warm because kernel_task is at 80% - but vice versa.
Over 8 cores this figure can start to look crazy. (eg 80% kernel_task can be 8 cores * 10%)
That's the main source of high kernel_task CPU usage on Intel macs, but M1s don't really get that warm, especially not with active cooling like my device has. So that's probably not it, though a glitching temp sensor might be at fault. What tends to speak against it is that the issue seems to mostly, but not fully, disappear once I stop ng serve (running on nodejs v16.4.0).
I feel like one thing that seldom gets mentioned is that there is a difference in battery life between the Pro and the Air. I have an M1 MBA and while the battery life is very impressive I still go through about 10% of my battery per hour of simultaneous VMWare Horizon Client and Firefox (10 tabs, one playing music) with a little Zoom sprinkled in. If I just have Safari open I can definitely get a lot more out of the battery, but I feel like that's a little unrealistic.
That being said it's still the best laptop I've ever owned and I don't have regrets about getting the Air, but when these battery life articles come up it's just assumed that everyone has the Pro.
I use AlDente to keep my battery between 80% and 20%. Zoom probably gets 30-60 min of usage during the day. If I start at 9 I usually have to plug in (hit 20%) around 2, so that's 60% in 5 hours. So probably 8-9 hours if I let it run all the way from 100% to 0%. It's not really a complaint about longevity, like I said it's above and beyond any other laptop I've owned, I just see many anecdotes of "I Zoom for 6 hours and my battery is at 90%!" and I find that hard to believe.
One of Zoom’s most intensive CPU tasks is continuous H.264 encoding/decoding, which Apple has hardware accelerated and presumably made power efficient; so it isn’t valid to compare a VMware/Firefox workload to a Zoom workload in a vacuum, especially if the attendee is mostly facemuting (which saves further power in both encoding and webcam) or screensharing video with audio (which makes my 8-core Apple desktop turn on the heat-venting fans). Both workloads are possible, even just within a Zoom-only scenario, and anecdotes are generally insufficient to tell which is the case.
The five relevant questions are, as best as I can remember right now:
1) Is the user's camera on, transmitting an H.264-encoded stream?
2) Is the user using virtual backgrounds, which have a CPU minimum requirement beyond that of the rest of Zoom?
3) Is the user screen sharing, transmitting an H.264-encoded stream, with or without the full-motion video checkbox enabled?
4) How many attendees are shown on the user's screen and are being H.264-decoded to the display?
5) Does Zoom use Apple's A/V interfaces for H.264 encoding and decoding, directly or indirectly, and has Apple (not Zoom!) improved the watts used for those operations on the M1 CPU?
#5, at least, I can consider. I'm looking inside the Zoom macOS binary now and I see a variety of functions in aomhost that are, essentially, image and video stream processing functions, with custom code for _avx2 and _sse2, on my Intel Mac. Zoom supposedly released an M1 native update in December, but I don't (yet) know how to get the M1 client on my Intel Mac to inspect further.
EDIT: Found the M1 client: https://zoom.us/download, "Apple Silicon". In the M1 client, the framework aomhost, which contained x64-optimized function names, is absent; as well as libmpg123. Framework aomagent replaces ApplicationServices with ColorSync,CoreGraphics. So at least for #5, I suspect that Zoom is using Apple's encoding/decoding interfaces, since there's little chance they ported the third-party open source libraries to native macOS interfaces. They're still bundling OpenSSL 1.1.1k rather than using Apple's SSL interfaces, so presumably they only rewrote the MPEG interfaces because they had SSE/AVX optimizations.
"When the first test chips came back from the lab on the 26 April 1985, Furber plugged one into a development board, and was happy to see it working perfectly first time.
Deeply puzzling, though, was the reading on the multimeter connected in series with the power supply. The needle was at zero: the processor seemed to be consuming no power whatsoever.
As Wilson tells it: “The development board plugged the chip into had a fault: there was no current being sent down the power supply lines at all. The processor was actually running on leakage from the logic circuits. So the low-power big thing that the ARM is most valued for today, the reason that it's on all your mobile phones, was a complete accident."
Wilson had, it turned out, designed a powerful 32-bit processor that consumed no more than a tenth of a Watt."
are those subtle marketing materials produced for apple, or can't apple estimate how long a battery life will be without seeing it in the indicator first.
I finally realized I use it more like a Kindle. Unfortunately I'm busy and I'm almost always on my work MBP (Intel) so my usage of the M1 is light.
But we have a Surface Pro (i5) that gets the same style of "light" usage. The difference with the M1 is crazy. I don't bother to plug it in very often. It uses so little battery in light usage that it doesn't drain much, then it's suspend logic is so well done that it doesn't lose much for days and days sitting there. The Surface Pro goes right back on the charger pretty much immediately when you're done with it by comparison, cause you can't trust it's going to have battery if you don't, and when you use it even for something light it burns through lots of battery by comparison. My i7 MBP for work of course chugs down battery at an incredible velocity compared to both, but it gets harder usage.
So in effect it gets treated like a kindle, I plug it in every week or two. That's very new, and it seems weird and took adjustment.
What I don't get is that the intel based macbook pros allow you to get 32G of RAM and 4TB hard drive. While the M1s only allow a max of 16G of RAM and 2TB hard drive. I'm not taking a hit on these other, in my opinion, much much more important specs to gain an extra battery life.
The real test will be the M1X or whatever they put in the 16" (hopefully) this year. Rumors say up to 64GB RAM among other things. I'd settle for 32 again (I'm on 2018 15", i9, 32GB, and whatever the best graphics card was) and I'll buy the 16" M1* immediately when it goes on sale as long as I can run 3-4 monitors (ideally 4 but I'll settle for 3).
This is still a first-gen product. It's like saying that you don't get why the first iPhone had 32GB+ of storage or 3G at launch.... it will be refined and improved over time.
Well my Mac book pro (2018) battery has started to expand and bend the case. To replace the battery according to ifix it is very difficult. So I effectively have a paper weight with a high performance i9 in it because Apple couldn't make the battery removable in a device that isn't even waterproof. Even Apple them selves say a battery is a consumable. Then why do we have to deal with this garbage?
you should run (run) that thing to an Apple Store. They tend to be pretty generous with this kind of defect, "house burned down due to macbook battery" is kind of bad press.
I'm just so terrified that for the next generation they use this low-power advantage to slim down the battery/machine down back to their typical target of "8 hours of wireless web" (=2 hours of coding in Xcode Swift) in favor of being able to say how thin and light-weight the machine is. With my Intel 16" MBP I currently have to carry around one of those gigantic Jackery-style 220 Wh Li-ion batteries to get 6 hours of work out of it
Yeah, I just got an M1 Air after hearing people talk about it as a sort of religious experience, and I can confirm that even with high expectations I'm incredibly impressed with the battery life.
It's worth noting that the stories about crazy battery life (getting through 4+ full days on one charge) come when you use the Apple apps and don't have the brightness too high. If you have the brightness high and use Chrome/lots of non-Apple stuff then you'll have good but not insane battery life. If you set the brightness to a medium-low level and use Apple apps then you'll get ridiculous battery life - I managed to use 17% battery for a full working day the other day (Safari, Sublime, a node server, Postgres, the notes app, and two terminals open.)
When I bought the thing I told myself that using Apple apps was what came with the territory, and I'm actually not missing much. The Safari devtools are great. I do miss uBlock Origin (the alternative for Safari seems to be AdGuard which is not great.) But Safari is a solid browser - certainly snappier than Firefox on Windows or Linux, competitive with Chrome, and I haven't run into any compatibility issues yet.
I think it's interesting to hear the number of developers switching from Linux to macOS as they get M1 Macs. Personally I was a diehard Arch-on-a-beefy-Thinkpad guy (two extended batteries, three SSDs, etc etc) and in the first day of using this Mac I was completely won over. It all just works so seamlessly, performance is great, battery is incredible, screen is gorgeous, speakers are by far the best I've ever heard on a laptop, and shockingly I actually like the keyboard. The webcam is your average laptop quality (aka not great, and a disappointment in contrast to the rest of the hardware.) I do miss the pageup/dn keys, but aside from that, as almost all the M1 reviews end, "this is the best computer I've ever used."
88 comments
[ 3.1 ms ] story [ 124 ms ] threadBut it's an interesting story.
https://github.com/linpengcheng/PurefunctionPipelineDataflow...
You can turn on percentage in the menu bar, and there's several different usage graphs in multiple places.
If you want the actual milliamp-hours, load cycles, battery temperature, age, or even battery serial number, there's an api for that. You can use for example coconutBattery and get whatever data you want.
I have had good luck with dosbox games though. Many games on gog run fine on mac.
Apple would not be able evolve the OS, silicon and Vulkan API in unison (like they do with Metal) because they don’t control the API.
In the end it would be a poor experience for end users.
It would be nice if they supported open source efforts though.
(MS doesn’t maintain the Vulkan API in windows, do they? Similar issues would apply.)
I don't think they update Metal in past OS releases either. You're lucky to get bug fixes.
The games or the game engines?
If Apple gave a bunch of money to port Unreal, Unit, etc, would that help the situation?
The Macs now have the same GPU and Graphics API that developers already have to target for iOS devices, where there is a significant market share and amount of revenue in play.
(There are exceptions.)
The power and size required for "serious gaming" GPU's on mobile is far too significant to fit into a mac size at this time. The term "gaming laptop" is a meme - by the time you've spent $2000 on a gaming laptop you've just bought a jet engine the shape of a brick that does half the work of a desktop at double the price and is terrible for productivity outside of specific use cases. (Like presenting 3d renders or something)
I suppose you could do it if apple were to support proton, and if amd/Nvidia were to standardize their external GPU implementations such that you can just plug in your egpu and have at it. But afaik they're very expensive and use case fitted.
I'd wager you could easily game on an M1 if the game has support and you are willing to make some graphics concessions.
TBH the macbook air would make a better switch than the switch as far as multiplayer portability goes. The switch screen is too small to play something like smash bros on it with friends, and the 'travel dock' solutions are too clunky IMO.
I really doubt it, Apple's model of constant updates requiring constant developer upkeep is incompatible with single purchase games. Only exploitative games as a service with recurring revenue can thrive on a model of endless upkeep. Its a big reason (besides Nintendo's first party quality, an attribute Apple has not demonstrated it is capable of) why the App store is defined by low quality p2w service games.
But isn't that good thing for game companies? They don't have to compete with those old games anymore.
The future of gaming on mac will probably be Apple Arcade and streaming services.
This morning I configured it to hibernate after 30 minutes, so I hope that fixes the problem.
It's comical how bad Windows S5 sleep is. Laptops waking up and running while they're in a bag should be the kind of bug that prevents an operating system from being released! But here we are in 2021, years after Windows 10 was released, and this is still common place.
My next laptop will absolutely not be a PC.
But yeah seriously, Modern sleep is one of the primary reasons I'll never get another Windows laptop after this.
I no longer have any kind of battery anxiety, it probably lasts a couple of days or more (9 or 10 hours a day use IntelliJ web and python dev, bit of YouTube etc) I don’t have to panic to plug it in every night and I’ve taken it away for the weekend without the charger.
But still unimaginably better than the Intel notebooks. With Docker and the brightness up I'll get a measly 6.5 hours instead of the solid ten or so I get otherwise.
I expect either qemu to get better or for apple to add some things to improve running x86-64 VMs. Rosetta 2's JIT portion (some of R2 is AOT) is worlds ahead of qemu's in terms of performance. So, the battery life and performance difference won't be forever.
Thankfully, building arm64 images for most containers I've used is pretty straightforward.
Leading up to and since the M1 Macs were released, the tech community went from skeptical, to intrigued, to excited, and if you were to completely remove Apple's hype machine, these machines stand on their own.
As an M1 user, I can attest to the ridiculously good battery life, and others in this thread are sharing similar anecdotes (I'm on an Air, so I'm not even on the best battery life available).
It is with this context that a statement from Apple's VP of marketing on the same topic is somewhat interesting. The performance and battery life surprised many of us in the tech community, and this anecdote seems to confirm that even at Apple, execs were/are downright surprised at how good this thing is.
If this statement was released before these machines launched, or if performance/battery weren't stellar, then a statement like this from a marketing VP might ring hollow. But in the context of this product's success, I think it's more interesting.
https://www.apple.com/ipad-pro/specs/
https://www.apple.com/macbook-pro-13/specs/
After a few days of use, my take: It's good. It's not the messiah-of-tech that I was led to believe, but it's good.
Honestly, I think the hype is overblown. If it wasn't for the hype, I might have thought this machine was amazing. But the expectations being set by all this hype is crazy town.
Not trying to take away from the real technical accomplishment. I'm sure if I put it through the ringer with benchmark tests, it would beat everything with flying colours.
But my day-to-day experience using it hasn't been much different than my previous 2016 Macbook Pro. Yet so many reviews make it seem like everything is so fast that you simply have to experience it to believe it, that every action you take feels faster and better. And that, in my own experience, is proving to be a load of B.S.
It’s that I can do the same things at the same speed as an Intel MacBook, but:
* The machine is dead silent (literally no sound coming from it)
* The battery life is incredible
I’m not sure if it’s as simple as this, but I guess if they said: we don’t mind blaring fans and 6 hours of battery life, then they could clock the M1 so high it would have truly insane performance.
I was really hoping the iMac would do as you say.
I was still very impressed even with my tempered expectations. For the kind of work I do, it's just as capable as the 16" MBP was - but with no heat, no noise, and real all-day battery.
- After a few hours of developing an Angular app using ng serve kernel_task consumes insane amounts of CPU time, resulting in extreme lag and completely locking up the device on an attempted restart (not even the touchpad clicks anymore).
- Sometimes MPEG4 videos just would not play until after a reboot
That said: Besides some minor UI glitches it's really, really solid. x86 compatibility is perfect. The GPU is good (again, not amazing, but Minecraft runs at good settings).
Over all it's not revolutionary. Slow software is going to be slow on these too. What they pulled off is amazing ecosystem wise, but as an end user you can be happy about some non-revolutionary pretty good decent spec bumps.
I just worry about the SSD's life. I'm already about 20TB down after a few weeks.
If my understanding of how mac works is correct, this is caused by overheating and basically is throttle mode.
It's also why it's the go-to to blame for performance issues - it's the symptom not the cause. Your laptop isn't running warm because kernel_task is at 80% - but vice versa.
Over 8 cores this figure can start to look crazy. (eg 80% kernel_task can be 8 cores * 10%)
That being said it's still the best laptop I've ever owned and I don't have regrets about getting the Air, but when these battery life articles come up it's just assumed that everyone has the Pro.
1) Is the user's camera on, transmitting an H.264-encoded stream?
2) Is the user using virtual backgrounds, which have a CPU minimum requirement beyond that of the rest of Zoom?
3) Is the user screen sharing, transmitting an H.264-encoded stream, with or without the full-motion video checkbox enabled?
4) How many attendees are shown on the user's screen and are being H.264-decoded to the display?
5) Does Zoom use Apple's A/V interfaces for H.264 encoding and decoding, directly or indirectly, and has Apple (not Zoom!) improved the watts used for those operations on the M1 CPU?
#5, at least, I can consider. I'm looking inside the Zoom macOS binary now and I see a variety of functions in aomhost that are, essentially, image and video stream processing functions, with custom code for _avx2 and _sse2, on my Intel Mac. Zoom supposedly released an M1 native update in December, but I don't (yet) know how to get the M1 client on my Intel Mac to inspect further.
EDIT: Found the M1 client: https://zoom.us/download, "Apple Silicon". In the M1 client, the framework aomhost, which contained x64-optimized function names, is absent; as well as libmpg123. Framework aomagent replaces ApplicationServices with ColorSync,CoreGraphics. So at least for #5, I suspect that Zoom is using Apple's encoding/decoding interfaces, since there's little chance they ported the third-party open source libraries to native macOS interfaces. They're still bundling OpenSSL 1.1.1k rather than using Apple's SSL interfaces, so presumably they only rewrote the MPEG interfaces because they had SSE/AVX optimizations.
"When the first test chips came back from the lab on the 26 April 1985, Furber plugged one into a development board, and was happy to see it working perfectly first time.
Deeply puzzling, though, was the reading on the multimeter connected in series with the power supply. The needle was at zero: the processor seemed to be consuming no power whatsoever.
As Wilson tells it: “The development board plugged the chip into had a fault: there was no current being sent down the power supply lines at all. The processor was actually running on leakage from the logic circuits. So the low-power big thing that the ARM is most valued for today, the reason that it's on all your mobile phones, was a complete accident."
Wilson had, it turned out, designed a powerful 32-bit processor that consumed no more than a tenth of a Watt."
https://www.theregister.com/Print/2012/05/03/unsung_heroes_o...
I finally realized I use it more like a Kindle. Unfortunately I'm busy and I'm almost always on my work MBP (Intel) so my usage of the M1 is light.
But we have a Surface Pro (i5) that gets the same style of "light" usage. The difference with the M1 is crazy. I don't bother to plug it in very often. It uses so little battery in light usage that it doesn't drain much, then it's suspend logic is so well done that it doesn't lose much for days and days sitting there. The Surface Pro goes right back on the charger pretty much immediately when you're done with it by comparison, cause you can't trust it's going to have battery if you don't, and when you use it even for something light it burns through lots of battery by comparison. My i7 MBP for work of course chugs down battery at an incredible velocity compared to both, but it gets harder usage.
So in effect it gets treated like a kindle, I plug it in every week or two. That's very new, and it seems weird and took adjustment.
There is a recall program for an earlier model, probably doesn't cover yours (yet) but worth checking anyway: https://support.apple.com/15-inch-macbook-pro-battery-recall
It's worth noting that the stories about crazy battery life (getting through 4+ full days on one charge) come when you use the Apple apps and don't have the brightness too high. If you have the brightness high and use Chrome/lots of non-Apple stuff then you'll have good but not insane battery life. If you set the brightness to a medium-low level and use Apple apps then you'll get ridiculous battery life - I managed to use 17% battery for a full working day the other day (Safari, Sublime, a node server, Postgres, the notes app, and two terminals open.)
When I bought the thing I told myself that using Apple apps was what came with the territory, and I'm actually not missing much. The Safari devtools are great. I do miss uBlock Origin (the alternative for Safari seems to be AdGuard which is not great.) But Safari is a solid browser - certainly snappier than Firefox on Windows or Linux, competitive with Chrome, and I haven't run into any compatibility issues yet.
I think it's interesting to hear the number of developers switching from Linux to macOS as they get M1 Macs. Personally I was a diehard Arch-on-a-beefy-Thinkpad guy (two extended batteries, three SSDs, etc etc) and in the first day of using this Mac I was completely won over. It all just works so seamlessly, performance is great, battery is incredible, screen is gorgeous, speakers are by far the best I've ever heard on a laptop, and shockingly I actually like the keyboard. The webcam is your average laptop quality (aka not great, and a disappointment in contrast to the rest of the hardware.) I do miss the pageup/dn keys, but aside from that, as almost all the M1 reviews end, "this is the best computer I've ever used."
I've never had a Li-Ion device that didn't exhibit a non-linearity in terms of power draw vs percentage estimation.
100% lasts much longer than 99%, which lasts longer than 98%, down to about 92%, after which point the draw-per-percentage seems to be fairly linear.
Anecdata, I know.