I think it would be fair to perform battery test like: - low power mode on all M4, then - automatic on M4 Pro/Max and low power off (default) on M4 - high power on M4 Pro/Max
I might be confused, but it seems like you ran your m4 pro and max battery tests on balanced mode, but the performance tests on low power mode. Outside of running the fans more aggressively I don’t think high performance mode changes much from balance mode. So as long as you’re not thermal throttling you should see very similar performance and battery life between those two modes. What we want to see is what the battery life is lot on low power mode. It should be substantially longer.
@@AZiskSorry but where exactly? At 5:15 is the high power mode for the Pro and Max, and at 6:20 is the Automatic mode for Pro and Max. I combed through the video multiple times but can’t see a battery life chart for Pro and Max low power mode.
hey there Alex! really appreciate you doing all these battery tests on so many different models of laptops. I think you are one of the only people I've seen that don't just use a lazy video playback test and call it a day. you have made a really sophisticated test that represents daily use. just wanted to ask if it's possible for you to publish your results in some sort of blog post or something similar? I would really like to compare the battery life across a wide range of products and that would make it easier to compare all of the ones that are not included in one video together. cheers
I love the diagrams, but as pretty color blind, I can’t tell the difference between the lines - so would it be possible to indicate which machine is which line? Using arrows, text, or symbols?
at least there is some realistic battery test now. But I am a bit confused about low power mode; where is the battery drain graph for that? (did you forget to put that)
The screen is one a substantial power draw and creates a notable difference in battery endurance (in my rudimentary tests my Mx series laptops). While I agree with you that I love a bright screen, it is shocking how many more hours I’ll get on battery when I drop the screen brightness.
I use low power some times, but I’ve become a ProMotion snob, and now it feels like the computer is lagging once it’s in 60fps. If it wasn’t for that I probably wouldn’t notice it’s in low power.
M4 Max: 12P + 4E = 156 seconds M4 Pro: 10P + 4E = 163 seconds M4: 4P + 6E = 247 seconds This one's missing but I reckon... M4 Pro: 8P + 4E ~ 176 seconds Having a great time @Alex thanks!
Love that you are doing this massive test. But how is it possible that the m4 pro lasts longer in high power mode than in automatic mode? And did the m4 achieve double battery life by switching to low power mode? Seems like a crazy difference.
I wish creators made a comparison between same models different ram configurations, this is one of the most sought questions. m4 pro macbook pro 24 vs 48gb
I've got a M4 Max and keep mine in low power mode most of the time. everything's still faster than my old M1 Pro! only if I need to do anything more demanding do I crank it up to high power.
And where are the Macbooks Pro 14 and 16 low powered test? You checked them with high powered and automatic, but not low powered (which uses more the e-cores) - in the first part of your video and the "coding simulation" ?
I got the base M4 model in part because battery life was a concern. I'm not compiling code or encoding large videos; I am, however, determined to keep using my laptop during a long flight or meeting. And the performance is still good enough that it's walloping high-end MacBooks from just a year or two ago.
Hi Alex. How long did the M4 Max 16 inch last in low power mode? Did I miss that? You tested high power mode and automatic but not low power mode, right?
Correct me if I’m wrong, but having an M1 Pro, I really wish they’d just give me a high power mode. I paid launch prices, so just give it to me! Not complaining about the current performance, but put that choice in my hands, Apple!
Vdry interesting video. Perhaps it' z s bit out of topic but which one go choose between zM4 Pro 12 Core CPU and 14 Core CPU? I'm using Adobe Crestive Suite programs in particular Lightroom and Photoshop but not for a living. I als do some encoding using Handbrake but again not in a very intensive mode. I can afford the 14/20 but I think for my needs It would be a waste of money.
My 14 inch M4 max lasts around 6 hours with video editing, 85% brightness and 8 hours with camera tethering to capture one (auto apply presets while shooting).
Coffee ☕ I'm going to go with 33% :) - What's the BTP, Battery:Time to Performance ratio? :| How would you measure that? I have a Max.... Do a "CRUNCH test" Crunch numbers and see how many numbers you can get before it dies.... lol 😂 Find pi numbers or something. You can do more stuff in less time!
nice vid, would love some graph showing the battery life at the end of the first test in hours. Instead of showing the clock that shows 12 houra and then additional 2 hours or what was that?
the only channel you need! 👍 thanks so much for this test, I still not sure about keeping my m4 max, I see a drop of 20%+ of battery drain running cinebench 24 in about 10m, which is quite ... impressive^^ ... sure, no real life test, but nevertheless ... and not amazed by the keyboard of the 16" m4 max, do you also get different sound and haptic on keys in the middle of the keyboard or near the trackpad, compared to ones at the edges? really annoying imho 🙄
Great video! Its just s shame you didnt test how long the pro and max chips last in low power mode. In my limited experience (the few days I had the m4 pro) the automatic mode is very similar to high power mode. Maybe apple can tweak this in a future update but it doesn’t work as I would expect it. I did some light coding but the battery was draining much faster than it does in low power mode so I just keep it in low power mode all the time.
Guys & Gals, I always use low power or automatic mode and include the switch to and from high power mode in my build, export, run etc. scripts. That can be achieved with "sudo pmset -Y lowpowermode X", where Y is "b" for battery and "a" for all powersources and X is 0 for automatic, 1 for low power and 2 for high power. Note: This will result in a warning about disksleep, which I would ignore when using SSDs. Otherwise the pmset manual has options to deal with that. None of the Apple Silicon Macs need more than lowpower mode when editing, surfing, email watching etc. And even a bunch of VMs don't matter much, unless they always load the CPUs.
Alex, you're doing these machines a disservice by always leaving them in high performance mode. Apple's E-cores are so good that the Low Power Mode could literally double your battery life whilst doing general tasks.
Alex can you do a video with a more in-depth explanation for Glossy vs Nano texture. Because opinions online seems to be very mixed. You briefly mentioned it in a previous video.
@@AZisk Okay makes sense. Quick question how big are the performance difference between 16 inch M4 Max MBP with 64 GB of ram vs 16 inch M4 Max MBP with 128 GB? In terms of work flow and performance? because I bought the 64 GB and am having buyers remorse lol
Why didn't you do the test with all computers in low power mode to compare apples to apples? it wouldn't make sense to compare the default modes in all computers, except on the m4.
i have a weird question. which macbook would have their fans spin up less for undemanding work? the m4 or m4 pro. i live in a dusty environment and want to limit fans spinning
I think the real test you should do is xcodebench on high power vs automatic to see how much extra perf you’re getting for that faster battery drain. Also did you factor in battery size into that efficiency graph??
Alex I’m torn between the m4 pro binned version but 48gb ram or m4 pro 20core 16 core version with 24gb ram. What do you recommend? I need one for school work, 3D modeling, editing and I game sometimes.
It is strange that the Elite X and Lunar Lake seem to be much more efficient however in real world use my MacBook M4 Max 36gb 14 core lasts way longer for me for my workloads and is much faster. I am a programmer but I do tend to run a lot of Docker containers but otherwise I am sure the computer is not under pressure.
Hi. I wanted to ask your advice, your opinion if possible: I work in the AI field particularly on deepfakes and I am used to work with Nvidia Rtx graphics cards so with CUDA architecture and with tensorflow. I'm intrigued by the idea of being able to work with this machine for all the reasons you listed in the wonderful review you did, but I'm afraid I won't get the same working speed and results with Metal and I wouldn't know which libraries are best optimized for this machine, if any exist that at least manage to approach tensorflow with Nvidia's Cuda architecture.
oh.. wth... That's how they binned the chip. I always wonder how can Apple can make CPU profitable without binning. They might bin the chip by allowing higher variant to have high performance power
10:50 the mac series M4 no longer have the same frequency the more cores the chip has the less frequency it will have the M4 max base model and the M4 pro have almost the same performance but the M4 max has more bandwidth which makes it slightly faster the 16 cores c
Alex fam here😂. Dude is similar to a large age gap older brother that encourages you to be a software developer, all the while being a comedic nerd.
I think it would be fair to perform battery test like:
- low power mode on all M4, then
- automatic on M4 Pro/Max and low power off (default) on M4
- high power on M4 Pro/Max
I might be confused, but it seems like you ran your m4 pro and max battery tests on balanced mode, but the performance tests on low power mode. Outside of running the fans more aggressively I don’t think high performance mode changes much from balance mode. So as long as you’re not thermal throttling you should see very similar performance and battery life between those two modes. What we want to see is what the battery life is lot on low power mode. It should be substantially longer.
@@arowden Yeah doing the performance test but not battery test in low power mode is odd.
did both battery test and perf test in high and low. Sorry if it's not clear in the video, but all the results and charts are there.
I think that the problem is that for the m4 pro and m4 max there are 3 modes: high power, automatic and low power.
@@AZisk yeah, but where is the battery drain graph for low power mode?
@@AZiskSorry but where exactly? At 5:15 is the high power mode for the Pro and Max, and at 6:20 is the Automatic mode for Pro and Max. I combed through the video multiple times but can’t see a battery life chart for Pro and Max low power mode.
hey there Alex! really appreciate you doing all these battery tests on so many different models of laptops. I think you are one of the only people I've seen that don't just use a lazy video playback test and call it a day. you have made a really sophisticated test that represents daily use. just wanted to ask if it's possible for you to publish your results in some sort of blog post or something similar? I would really like to compare the battery life across a wide range of products and that would make it easier to compare all of the ones that are not included in one video together.
cheers
Thanks! I'm considering setting up something like that.
I love the diagrams, but as pretty color blind, I can’t tell the difference between the lines - so would it be possible to indicate which machine is which line? Using arrows, text, or symbols?
at least there is some realistic battery test now.
But I am a bit confused about low power mode; where is the battery drain graph for that? (did you forget to put that)
Great video Alex.
Now, get M1 Max vs M2 Max vs M3 Max vs M4 Max base models for ultimate showdown!
The screen is one a substantial power draw and creates a notable difference in battery endurance (in my rudimentary tests my Mx series laptops). While I agree with you that I love a bright screen, it is shocking how many more hours I’ll get on battery when I drop the screen brightness.
I use low power some times, but I’ve become a ProMotion snob, and now it feels like the computer is lagging once it’s in 60fps. If it wasn’t for that I probably wouldn’t notice it’s in low power.
M4 Max: 12P + 4E = 156 seconds
M4 Pro: 10P + 4E = 163 seconds
M4: 4P + 6E = 247 seconds
This one's missing but I reckon...
M4 Pro: 8P + 4E ~ 176 seconds
Having a great time @Alex thanks!
Just put out the mac mini farm video already I’m restless 😅
it's coming soon.
Same, I'm really curious! Local LLM cluster 👀??
Love that you are doing this massive test. But how is it possible that the m4 pro lasts longer in high power mode than in automatic mode? And did the m4 achieve double battery life by switching to low power mode? Seems like a crazy difference.
I wish creators made a comparison between same models different ram configurations, this is one of the most sought questions. m4 pro macbook pro 24 vs 48gb
From where does the 22 hours come from on all Macbook ads ?
I've got a M4 Max and keep mine in low power mode most of the time. everything's still faster than my old M1 Pro! only if I need to do anything more demanding do I crank it up to high power.
Babe wake up, this year alex z's ultimate battery test has dropped!
And where are the Macbooks Pro 14 and 16 low powered test? You checked them with high powered and automatic, but not low powered (which uses more the e-cores) - in the first part of your video and the "coding simulation" ?
no low power battery life times, that would be the only reason to set low power ?????
I got the base M4 model in part because battery life was a concern. I'm not compiling code or encoding large videos; I am, however, determined to keep using my laptop during a long flight or meeting. And the performance is still good enough that it's walloping high-end MacBooks from just a year or two ago.
Hi Alex. How long did the M4 Max 16 inch last in low power mode? Did I miss that? You tested high power mode and automatic but not low power mode, right?
Did I miss it or did you forget to run the battery test on all on low power mode? I was really interested in that result.
Hi Alex! Thank you!
Alex we love you, thanks for taking care of all of us. Greetings from Argentina !!
Gracias! I appreciate it.
When are we going to see that 5xM4 mac mini exo cluster benchmark results? I can't wait!
Do you recommend the 12-core pro chip or the 14-core pro chip in the m4 14inch?
The battery would last longer on the 12-core but the performance is 10%?
We need m3 max vs m4 max ultimate stress test
Thanks for creating this video
Thanks for your testing 👌
Correct me if I’m wrong, but having an M1 Pro, I really wish they’d just give me a high power mode. I paid launch prices, so just give it to me! Not complaining about the current performance, but put that choice in my hands, Apple!
exact the video i was looking for in the last days, your´e a genius
Thank you for this video.
I like low power mode, but I hate that it disables promotion and there is no way to change it. It just makes everything feel slower.
What exactly does low power mode do?
Vdry interesting video. Perhaps it' z s bit out of topic but which one go choose between zM4 Pro 12 Core CPU and 14 Core CPU? I'm using Adobe Crestive Suite programs in particular Lightroom and Photoshop but not for a living. I als do some encoding using Handbrake but again not in a very intensive mode. I can afford the 14/20 but I think for my needs It would be a waste of money.
My 14 inch M4 max lasts around 6 hours with video editing, 85% brightness and 8 hours with camera tethering to capture one (auto apply presets while shooting).
That Tim Cook impression spooked me
We all know developers don’t get to pick Notion. It’s all Jira and downhill from there /s
Coffee ☕ I'm going to go with 33% :) - What's the BTP, Battery:Time to Performance ratio? :| How would you measure that? I have a Max.... Do a "CRUNCH test" Crunch numbers and see how many numbers you can get before it dies.... lol 😂 Find pi numbers or something. You can do more stuff in less time!
nice vid, would love some graph showing the battery life at the end of the first test in hours. Instead of showing the clock that shows 12 houra and then additional 2 hours or what was that?
the graph shows minutes.
Nice video as always. Could you also compare the battery life for the m4 pro and m4 max in low power mode?
yes. done. watch the video again. 😊
the only channel you need! 👍 thanks so much for this test, I still not sure about keeping my m4 max, I see a drop of 20%+ of battery drain running cinebench 24 in about 10m, which is quite ... impressive^^ ... sure, no real life test, but nevertheless ...
and not amazed by the keyboard of the 16" m4 max, do you also get different sound and haptic on keys in the middle of the keyboard or near the trackpad, compared to ones at the edges? really annoying imho 🙄
No macs were harmed, Really 😅😂😂😂😂
maybe a little
@@AZisk 🤣
I like the tilted youtube plaque but it is too distracting for me that it is not in the middle between door and 'whiteboard' or that it is :)
You could give the "bokeh" data visualization library a shot. Really nice stuff. (It's Primarily in Python) :)
I would like to see m4 pro 16inch MBP
Great video! Its just s shame you didnt test how long the pro and max chips last in low power mode.
In my limited experience (the few days I had the m4 pro) the automatic mode is very similar to high power mode. Maybe apple can tweak this in a future update but it doesn’t work as I would expect it. I did some light coding but the battery was draining much faster than it does in low power mode so I just keep it in low power mode all the time.
2:28 I thought the video was over 😂
M4 MacBooks on sale for $1,399 already for Black Friday!!!
Talking about building taking time makes me miss incremental compilers like with the Lisp language that avoid linking.
I thought this was how long they last til they become outdated when the next gen of NVIDIA gpus drop.
Guys & Gals, I always use low power or automatic mode and include the switch to and from high power mode in my build, export, run etc. scripts. That can be achieved with "sudo pmset -Y lowpowermode X", where Y is "b" for battery and "a" for all powersources and X is 0 for automatic, 1 for low power and 2 for high power. Note: This will result in a warning about disksleep, which I would ignore when using SSDs. Otherwise the pmset manual has options to deal with that. None of the Apple Silicon Macs need more than lowpower mode when editing, surfing, email watching etc. And even a bunch of VMs don't matter much, unless they always load the CPUs.
Hey Alex, did you see that MS released an official ISO of Windows for ARM? Would love to see a video of you testing that against Parallels or VMWare.
The new name for the device that cannot currently be mentioned should obviously be: The Finger Banger
please do M4 32 GB/M4 Pro performance test with coding, docker, database, Unity, illustrator, browsing, youtube all at once and check the performance.
Alex, you're doing these machines a disservice by always leaving them in high performance mode. Apple's E-cores are so good that the Low Power Mode could literally double your battery life whilst doing general tasks.
Alex can you do a video with a more in-depth explanation for Glossy vs Nano texture. Because opinions online seems to be very mixed. You briefly mentioned it in a previous video.
my opinion is i don’t need it. also, i want to wait a generation or two to see how the screens degrade after a couple years, if at all
@@AZisk Okay makes sense. Quick question how big are the performance difference between 16 inch M4 Max MBP with 64 GB of ram vs 16 inch M4 Max MBP with 128 GB? In terms of work flow and performance? because I bought the 64 GB and am having buyers remorse lol
I‘m a bit confused, does the M4 Pro in low power mode come close to the M4 in terms of battery life?
Dude please use bar graphs, I can't reas those lines.
what kind of graph will show you how the battery behaves over time?
this year's chips are great for Apple's mini pc categories.
is 24gb ram enough for most developer multi task? what will be the case its not enough?
you have some lighting there and they can get to 1000 nits SDR. Am I wrong !
Why didn't you do the test with all computers in low power mode to compare apples to apples? it wouldn't make sense to compare the default modes in all computers, except on the m4.
Very nice video can't wait for 6 core SD chip
i have a weird question. which macbook would have their fans spin up less for undemanding work? the m4 or m4 pro. i live in a dusty environment and want to limit fans spinning
I think the real test you should do is xcodebench on high power vs automatic to see how much extra perf you’re getting for that faster battery drain. Also did you factor in battery size into that efficiency graph??
Alex I’m torn between the m4 pro binned version but 48gb ram or m4 pro 20core 16 core version with 24gb ram. What do you recommend? I need one for school work, 3D modeling, editing and I game sometimes.
I don't know about those other things, but 16GB of RAM is sufficient for 99% of video games.
I have not heard the fan of a mac since intel. My Max can get loud when I push it. Also the battery can be drained in an hour with large ai models
Great video, thanks for your work! Would you mind puting your results data on a public google sheet ?
Nice tests, but the second test is kinda confusing. Why is M4 Pro/Max not in Low Power Mode but only M4? Of course M4 will lasts the longest… 🤔
You owe us the M4 family test on local generative AI Vs Windows PCs
Good numbers to use excel 😅, I can’t wait 😍
infact the pro and the max lasted less in automatic compared to high power?
Pretty weird result.
Can we add to the test a Next.js project build? That's the only reason why I want to upgrade 😆
1 minute ago is crazy
nice!
It is strange that the Elite X and Lunar Lake seem to be much more efficient however in real world use my MacBook M4 Max 36gb 14 core lasts way longer for me for my workloads and is much faster. I am a programmer but I do tend to run a lot of Docker containers but otherwise I am sure the computer is not under pressure.
The line is not linear 🎉
Hi. I wanted to ask your advice, your opinion if possible: I work in the AI field particularly on deepfakes and I am used to work with Nvidia Rtx graphics cards so with CUDA architecture and with tensorflow. I'm intrigued by the idea of being able to work with this machine for all the reasons you listed in the wonderful review you did, but I'm afraid I won't get the same working speed and results with Metal and I wouldn't know which libraries are best optimized for this machine, if any exist that at least manage to approach tensorflow with Nvidia's Cuda architecture.
compare x Elite surface 7 with macbook pro m4 base model
Thanks!
This video helped alot was thinking of what to buy
Wow! Thanks!
Use plotly or any BI software - I prefer Tableau
Was that the Razer Orochi v2 mouse
Yes
Can you please just tell the hours?
oh.. wth... That's how they binned the chip.
I always wonder how can Apple can make CPU profitable without binning.
They might bin the chip by allowing higher variant to have high performance power
probably 16 pro in low performance would be the longest lasting
We need charging test
That statement is for the mac book pro not the max . Max actually has less battery power compare at the apple website m4 max vs m3 max
I want to buy a MacBook, plz help what is better for ML and AI:
- m4 pro 12/16, 24 unif.M
- m4 10/10, 32 unif.M
- wait m4 air + colab
Commented on last video
Not it's here😊❤
Why can’t we get m4 max in a desktop
6:18 M4 Pro and Max are terrible. They're nowhere near 3x the power of M4, yet only last 5 hours.
If its not linear it could be the BMS was not calibrated.
plotly is nice
10:50 the mac series M4 no longer have the same frequency the more cores the chip has the less frequency it will have the M4 max base model and the M4 pro have almost the same performance but the M4 max has more bandwidth which makes it slightly faster the 16 cores c
9:11 Why you can't?
4%
We need a conclusion the whole 11 min video is just so confusing
they're good.