Not talking about M1 Ultra bandwidth 800 GB/s compared to M2 100....i mean physics cannot be bend, some users cannot understand whats raw performance and whats PR marketing naming
Got a tip for us?
Let us know
Become a MacRumors Supporter for $50/year with no ads, ability to filter front page stories, and private forums.
Apple M2 SoC Analysis - Worse CPU efficiency compared to the M1
- Thread starter Retskrad
- Start date
- Sort by reaction score
Lol...my point exactly. It's ridiculous for these people to be touting m2s when some m1 variant is still better.
Any m2 should be better than any m1, period.
If not, it's just marketing bs and they should keep naming the m1 variants until they actually have something better to offer.
Is marketing and M1 Ultra should be better than M2 period
Or M1 Pro should be better than M2, period
What do you thought ? that M1 M2 etc are not marketing names but what some material used into the SoC ?!
What I gather is that the M2 is 18% faster at peak power draw, but has a lower start-up/idle power resulting in less usable time on equivalent batteries.
MacWorld Review of M2 MBP
Based on benchmarks quoted in the MacWorld article, you can expect an 18% boost in multicore performance about 12% in single core, and about 89% of the battery life. Therefore, in the most optimistic circumstances, the work done is power*time -> 1.18*0.89 = 1.05 for the M2.
For single core 1.12*0.89 = 0.997 for the M2, essentially breaking even.
From my point of view, I would call the machines equal when it comes to the amount of computation they can complete on a single charge. With the M2 being a little faster, it will be more effective when plugged in for long periods of time.
As someone who comes from a field where analyses are done by writing single core, single thread macros that are then run over large local data sets or distributed over a large cluster for final results, I am ecstatic that Apple is providing processor solutions with high end single core performance on the lowest priced machines.
Where did you get that from? I get the same battery from the M2 MacBook Air as I did with the M1 MacBook Air (within a reasonable margin.)
From the linked article comparing the M1 and M2 13” MBP models which very likely use batteries with the same capacity. They do not note the methodology, but based on Apple’s marketing graph, it makes reasonable sense.
Edit: The M2 Air has a 52.6 Wh battery compared to the M1 Air’s 49.9 Wh battery. With 5% more battery and fewer things to power, I’m not surprised the two compare favorably in battery life.
The battery life section from the article:
13-inch MacBook Pro M1 vs M2: Battery life
Battery life is just the same as the M1 model, at 20 hours. This is despite power-efficient improvements in the M2 processor that we thought could lead to battery improvements, but it seems the M2 uses more power than the M1.In our battery tests the 2022 MacBook Pro fell short of Apple’s claimed 20-hours battery life at 16-hours. In our tests the M1 MacBook Pro managed 18 hours. It’s still a lot more than the Intel MacBook Pro models offered though.
In Macworld's M2 MacBook Air review they said that the M2 MBA had longer battery life than both the M1 MacBook Air and the M2 MacBook Pro by an hour and half. I don't know what they do for battery life tests but given the larger battery in the M2 MacBook Pro this doesn't make much sense. I would take their undisclosed battery tests with a grain of salt.
M2 MacBook Air vs M1 Design Performance Price
They don't post any methodology so it's kind of difficult to substantiate their claims. Notebookcheck (https://www.notebookcheck.net/Apple...view-Debut-for-the-new-Apple-M2.631003.0.html) found same idle power usage and slightly higher power usage under load (as expected), with battery life for video playback being practically identical and the M2 scoring 7% worse in their WiFi browsing tests. I don't think there is much here to suggest a noticeable reduction in battery life in common workloads. Maybe by a bit — especially if your work features frequent bursts of CPU activity, but hardly something that can be substantiated. Not trying to be polemic here , just interested in a realistic perspective.
Just look at the one I posted earlier in the thread. They even include the testing methodology in the video as well as a timelapse of the notebooks running until they run of out juice.
The M2 MacBook Pro lasted 1 hour and 50 minutes longer than the M1 MacBook Pro. The difference between these two models is down to the M2 chip as the other components are otherwise identical. This should give us the cleanest comparison compared to the total redesigned MacBook Air and the older MacBook Air.
Pressure, thanks, I'm not sure how I missed that back on the first page of the thread. With that type of battery life, the M2 is more efficient. If we give credit purely to the CPU, then it would be about 30% (27%-34%) more efficient based on my analysis above, although I'm sure a chunk of it is due to GPU and accelerator performance increases as well. I consider that a pretty significant improvement for a refinement step.
Edit: In reality, the truth likely lies somewhere in between the break-even pessimistic analysis, and the optimistic analysis here.
Edit: In reality, the truth likely lies somewhere in between the break-even pessimistic analysis, and the optimistic analysis here.
It will depend on your specific workload, obviously.
In the link I posted they had it cycle 1 hour of the following with a script and screen brightness was locked at 160 nits:
- Constant music streaming
- 10 minutes of Microsoft Excel
- 10 minutes of Microsoft Word
- 10 minutes of Microsoft Publisher
- 10 minutes of web browsing
- 10 minutes of 4K video streaming
- 10 minutes of instant messaging
