If I were to guess, it's likely that sales projections are down right now, and they're hoping by keeping the existing line a bit longer, new buyer numbers will be larger in the spring. Most people don't upgrade every generation and a lot of people are still running M1/M2 devices.
I would also speculate that there may be some growing pains for the n2 production from TSMC, and/or a desire to get there in the AZ fab production before launch to avoid tariffs hitting their bottom line. They'd rather pay 12-20% more for just the CPU than eat large tariffs on the full cost. I don't think they'd be able to significantly raise prices further based on tariffs, like some other companies with smaller margins are forced to do, on order to be competitive.
I upgraded from an M1 to an M4 MBP recently and although the performance gains were mostly incremental, the matte screen (fucking finally Apple) is really nice and a good reason to upgrade if you ever plan on using it in a brightly-lit area. It’s a must-have.
Don't forget SSD degradation since it's not a user replaceable part and you can't do true external boots (it copies to the internal storage). That's my biggest worry.
Due to the limitations of MacOS, my M1 MacBook pro is pretty much exclusively a thin client (software dev, gaming) and I see no reason to upgrade it - unless battery degradation warrants it.
It's fantastic as a thin client - though it's a bit annoying carrying around a mini PC with me when I travel.
I mostly use my M1 Air as a browser, sometimes I'll ftp or ssh out. VS Code + ssh remoting to my personal desktop when I actually need to be productive.
Regardless of the CPU, the time to upgrade every year is long gone, even without taking M chip series into consideration.
Most of my home stuff is from around 2010, my work laptop is from 2021, and neither at home, nor at work, do I need more programming power.
I have enough cores, RAM, disk space, and graphic cards, to still code whatever I feel like locally, with exception of ray tracing or latest AI models, neither of which are that relevant to my computing activities.
So I will keep using them until they die, or there is no hope left of the battery and need to acquire a new one to take outside the house.
Really, they could probably get away with selling M4 devices for another couple cycles. As far as performance is concerned, even going back to the M1, the only group feeling inadequacies large enough to make upgrading a “must” are those whose needs sit within spitting distance of cusp of consumer/prosumer computing. Battery life is still industry leading with only a handful of competitors just recently being able to claim similar real world numbers (thanks mostly to Lunar Lake, which isn’t available for many models and comes at the cost of some performance).
I’ve used both M1 Max and M4 Max machines extensively and while the latter is a good deal faster, it’s only really noticeable with longer sustained tasks and particularly large projects. The high-RAM variants of M1 models in particular should continue to be quite servicable for some time to come.
You should also consider liking Linux, or wanting to support the right to repair movement. Or disliking Apple for whatever reason, such as their generally very closed nature. My biggest criterion for buying consumer electronics is being able to replace the software easily, followed by being able to replace the hardware easily. Nothing else is particularly important.
OS is still pretty terrible. From janky kid focused UX, to trying to hide errors and information from you (e.g. you try extending your screen to an iPad, and it just says it failed. No information whatsoever about why or what you could do, just fuck you), to basic features missing (did you know you can't have a different scroll direction between scroll button and touchpad? There are two toggles in two menus, but they just change each other).
The OS is one of the important part of the experience of using a computer. So it makes sense to filter first on what kind of OS matches what you want. I run a M2 MBP and I wouldn't use it if I couldn't run linux. But in terms of hardware it's a truly great machine. The first laptop where I'm satisfied by the power. M4 doesn't have linux support(yet).
Linux support is not industry leading, nor presence of Ethernet port. Sadly others are following in lack of ports as I can't find a new Thinkpad that has both an Ethernet port and an SD slot.
Apple seems to overlook how much timing matters for Mac sales in academia. Macs—especially MacBooks—absolutely dominate among professors (I’d guess ~90% in my department).
The academic fiscal year often ends in Aug/Sep, and new faculty usually get a “technology fund” for buying their first computer. A lot of us use that to get the latest Mac. Historically, Apple’s October refresh was just late enough to miss that budget window, but people would still wait a month or two for the new models.
If they push announcements even further (as the article suggests—early 2026), it’s a different story. New hires can’t wait half a year with no laptop, so they’ll just buy whatever’s top-of-the-line right now. For research folks who need GPU power, that could easily mean a 5090-based laptop instead of a Mac.
What about university students? They also start their classes in September, and while they have no "budgeting deadline", they still need to buy some computer around that time.
Apple has pretty sophisticated modeling for their sales cycle so I assume that it they aren't simply looking at the academic lifecycle.
Professors are a very small % of the education market, most of their sales come from high school and college students and during back to school season.
I think it might just be the other way around, if they front-loaded a lot of inventory shipments before tariffs were due to hit they might be loaded with unusually high inventory levels that needs shifting and will be hard to do so at price if a new model is out.
Add to this the recent economic uncertainty and prospective buyers might just have been holding up purchases (thus further adding to inventory if they already front-loaded before tariffs).
As for people buying powerful machines that could be worth going to a 5090 based machine instead, they're probably a fairly small part of the Mac purchaser market in the big picture.
Like I said, most people want something that's the "latest"; M4 is already one year old.
wrt the second part of your comment: Academics care about speed, RAM, battery life, the ability to run the latest AI models at a decent speed (M4 is still relatively slow).
The advantage is all of that legacy software that some process relies on and hasn't been meaningfully updated in 10+ years and won't be ported over to the ARM processors that you damned kids are running on because back in my day we paid for one copy of x86 software and that got us through 10 winters, dammit.
I do. Wanted a discrete GPU and ability to run all the games I love on the go, including those that may want a little bit of GPU performance and don't have a macOS port. Can't realistically do this on non-x86.
What I’d most like in future MacBooks is the continuation of increases in memory size and bandwidth.
Apple has carved out a niche for itself in the local LLM space, yet it continues to overcharge for RAM and under-deliver in terms of bandwidth.
I have no hopes that Apple will decrease its prices, particularly on top-of-the-line models like those with 128 GB of memory and above.
Yet I certainly believe that it can deliver even more RAM and, in particular, memory bandwidth. Apple clearly offers much more VRAM than consumer NVIDIA GPUs, but Macs are still behind in terms of memory bandwidth and, relatedly, overall performance.
It would be silly of Apple not to jump at the opportunity to eat even more of NVIDIA’s market share among the general public.
I love working in MacOS but I refuse to buy another new Macbook that has the storage soldered to the motherboard. Non-volatile memory has a limited amount of write cycles and let's be honest, NAND can fail long before its forecasted lifespan is up, which would render a $5k+ computer inoperable. I understand the benefits that come with SOC architecture but I draw the line at storage. This design choice wasn't about "saving space" it was about exploiting the Mac-tax.
I got M1 Max with 64GB and 32 core GPU fro $1500 refurbished with zero cycle battery on 100%. As most companies doing refresh/write-off after 3 years 2025 is really a year where you can get a beast of machine for the money. I also have M4 Max for work and differences are only on really heavy tasks but for 3X less money I guess M1 Max is still good deal. This delay also means that M2/M3 ones will be good buy next year as well.
That's what I have too. I upgraded about 9 months ago from an M1 Max 32 GiB + 1 TiB SSD to 64 GiB + 4 TiB SSD, and donated the old one to a FOSS developer in Europe who can now support macOS. The "new" one is used, but had about 15 cycles on it and 99% battery health... like it was in someone's closet for 3 years.
It's a total waste of money paying zillions for tiny improvements.
Would be nice to do that for OS releases too. I feel like the yearly macOS releases are too ambitious at this point, and Apple's software quality is suffering. 2-3 year cycles would be much more sustainable. Hardware is good enough now to not need a new release ever year.
Refresh cadence has little to do with sustainability.
Consider cars: manufactures come out with new ones each year with marginally differences. Is that somehow unsustainable and they should instead keep manufacturing an old design for years? Does that mean once the 2026 model starts manufactuing you go dump your 2025?
It makes more sense to iteratively improve your design and stop manufacturing old things if you can manufacture something better.
The real sustainability argument is about support length (which apple does well), and repairability (which apples does ). Changing to a 3 hear cycle is orthogonal to both of these.
Who's forcing you to get every upgrade every year?
The yearly releases make a lot of sense for everybody, because then you can upgrade on your own schedule, instead of delaying the upgrade because the product was released a full 2 years ago, at a time your older one is on its last breath.
In fact, yearly releases are then also more sustainable, too, since the purchasing would be spread out to each year (on an as-needed basis), instead of having a month-long cycle every 3 years, necessitating the extra infrastructure all along the way (from the stores to manufacturing to shipping).
For Apple computers, the last breath comes almost a decade after they were built.
If you wait until it dies, then you will want to get what's available at that time, but, if you plan from the start, you'll have a lot of flexibility with these machines.
My MBP is the same age as my Thinkpad, and looks much nicer.
5+ years for home users and systems administrators. I have a 2020 M1 MacBook Air, it's fine for everything I need. The only issue is, as other points out, the external monitor issue. Only one monitor, and certainly no daisy chaining of displays.
I have a 16" M2 Max with 32 GB of RAM and I do pretty much everything on it, and I've yet to have a moment where I thought "ugh I wish this was faster"
I used to upgrade computers every 1-1.5 years but I think I could easily roll with this for another 2-3 maybe even 4 or 5 years more.
Heck, even if I was given free money I'd be too lazy to switch to an M4 Max just because of the hassle of transferring data that isn't on iCloud/Time Machine.
I was waiting for the M5 to have some other substantial changes like in the display or sound etc.
Readit News
I would also speculate that there may be some growing pains for the n2 production from TSMC, and/or a desire to get there in the AZ fab production before launch to avoid tariffs hitting their bottom line. They'd rather pay 12-20% more for just the CPU than eat large tariffs on the full cost. I don't think they'd be able to significantly raise prices further based on tariffs, like some other companies with smaller margins are forced to do, on order to be competitive.
The only motivation to upgrade is battery degradation or getting more RAM to run larger LLM models locally.
It's fantastic as a thin client - though it's a bit annoying carrying around a mini PC with me when I travel.
Most of my home stuff is from around 2010, my work laptop is from 2021, and neither at home, nor at work, do I need more programming power.
I have enough cores, RAM, disk space, and graphic cards, to still code whatever I feel like locally, with exception of ray tracing or latest AI models, neither of which are that relevant to my computing activities.
So I will keep using them until they die, or there is no hope left of the battery and need to acquire a new one to take outside the house.
Dead Comment
I’ve used both M1 Max and M4 Max machines extensively and while the latter is a good deal faster, it’s only really noticeable with longer sustained tasks and particularly large projects. The high-RAM variants of M1 models in particular should continue to be quite servicable for some time to come.
What excuse besides liking windows or wanting an even cheaper machine do you have not to get one?
Market share, right to repair, upgradability, no open source
The academic fiscal year often ends in Aug/Sep, and new faculty usually get a “technology fund” for buying their first computer. A lot of us use that to get the latest Mac. Historically, Apple’s October refresh was just late enough to miss that budget window, but people would still wait a month or two for the new models.
If they push announcements even further (as the article suggests—early 2026), it’s a different story. New hires can’t wait half a year with no laptop, so they’ll just buy whatever’s top-of-the-line right now. For research folks who need GPU power, that could easily mean a 5090-based laptop instead of a Mac.
Professors are a very small % of the education market, most of their sales come from high school and college students and during back to school season.
Add to this the recent economic uncertainty and prospective buyers might just have been holding up purchases (thus further adding to inventory if they already front-loaded before tariffs).
As for people buying powerful machines that could be worth going to a 5090 based machine instead, they're probably a fairly small part of the Mac purchaser market in the big picture.
Deleted Comment
The average user, even the "power user", does not care/know the difference between an M4 and whatever the M5 will be.
wrt the second part of your comment: Academics care about speed, RAM, battery life, the ability to run the latest AI models at a decent speed (M4 is still relatively slow).
The advantage is all of that legacy software that some process relies on and hasn't been meaningfully updated in 10+ years and won't be ported over to the ARM processors that you damned kids are running on because back in my day we paid for one copy of x86 software and that got us through 10 winters, dammit.
I'm enjoying my framework AMD laptop although the battery life with suspend is miserable.
Apple has carved out a niche for itself in the local LLM space, yet it continues to overcharge for RAM and under-deliver in terms of bandwidth.
I have no hopes that Apple will decrease its prices, particularly on top-of-the-line models like those with 128 GB of memory and above.
Yet I certainly believe that it can deliver even more RAM and, in particular, memory bandwidth. Apple clearly offers much more VRAM than consumer NVIDIA GPUs, but Macs are still behind in terms of memory bandwidth and, relatedly, overall performance.
It would be silly of Apple not to jump at the opportunity to eat even more of NVIDIA’s market share among the general public.
It's a total waste of money paying zillions for tiny improvements.
Leopard followed by Snow Leopard.
Every other year is just a major bugfix/rewrite release.
Consider cars: manufactures come out with new ones each year with marginally differences. Is that somehow unsustainable and they should instead keep manufacturing an old design for years? Does that mean once the 2026 model starts manufactuing you go dump your 2025?
It makes more sense to iteratively improve your design and stop manufacturing old things if you can manufacture something better.
The real sustainability argument is about support length (which apple does well), and repairability (which apples does ). Changing to a 3 hear cycle is orthogonal to both of these.
Who's forcing you to get every upgrade every year?
The yearly releases make a lot of sense for everybody, because then you can upgrade on your own schedule, instead of delaying the upgrade because the product was released a full 2 years ago, at a time your older one is on its last breath.
In fact, yearly releases are then also more sustainable, too, since the purchasing would be spread out to each year (on an as-needed basis), instead of having a month-long cycle every 3 years, necessitating the extra infrastructure all along the way (from the stores to manufacturing to shipping).
For Apple computers, the last breath comes almost a decade after they were built.
If you wait until it dies, then you will want to get what's available at that time, but, if you plan from the start, you'll have a lot of flexibility with these machines.
My MBP is the same age as my Thinkpad, and looks much nicer.
I used to upgrade computers every 1-1.5 years but I think I could easily roll with this for another 2-3 maybe even 4 or 5 years more.
Heck, even if I was given free money I'd be too lazy to switch to an M4 Max just because of the hassle of transferring data that isn't on iCloud/Time Machine.
I was waiting for the M5 to have some other substantial changes like in the display or sound etc.