Become a MacRumors Supporter for $50/year with no ads, ability to filter front page stories, and private forums.

MallardDuck

macrumors 68000
Jul 21, 2014
1,600
2,949
New OS versions have new features (and code bloat). New features take more resources. That slows the machine down - simple as that. You can't run Win 7 on an XP machine, or Win 11 on most Win 10 machines for the same reason. Ditto here - Catalina and Big Sur ran fine with 2GB for the host. Monterey was a big jump and really needs 4GB for the host to run well.

The other reason, especially if you use spinning disks, is the switch from HPFS+ to APFS, which is substantially slower on hard drives.
 

bearinthetown

macrumors 6502
Original poster
May 5, 2018
286
323
This is not scientific evidence. If you want to make a case that macOS slows down existing computers with new releases, you have to provide reproducible benchmarks from dozens of non-battery powered machines (let’s say 30 as a bare minimum to help us be more confident in the results) all configured exactly the same. It’s important not to use battery powered computers because the batteries can degrade over time. Then install the newer version of an OS on all the same machines and run all the same software to benchmark. Then do that again with the next version of the OS. You also likely need to covary for the size of the OS. Newer OS versions will add features, which also needs to be considered. Some of these features are not strictly necessary but others are security-related. These will require resources to manage. The computers should also be free of any other non-Apple software.

Benchmarks could vary but could include start-up times, file copy times, and even benchmark software. What’s best is if any benchmarking software is stable across OS versions and is not affected by low-level algorithmic changes in the OS.

This would be a start to have objective data that Apple slows or does not slow down computers intentionally. Even if there was a “slowing” it would be difficult to argue it was intentional because there are so many factors that could affect performance. Many of these are simply new, more powerful features.
I don't need scientific evidence to notice that something, once snappy, became not snappy.
 

joehohoho

macrumors member
Jan 26, 2016
72
96
Tunbridge Wells
I don’t have any evidence as frankly I can’t be bothered to look into it and see no noticeable differences with macOS.

However I have suspected similar to the OP suspicions regarding iOS.

I’ve had a personal 13 mini for a year. Just got an 11 for work and it feels noticeably faster. I wonder (if there is any intentional slow down) if it’s more related to hardware age as opposed to outright software updates. Would make more sense for apple to do that on iOS devices as these are the main sellers which they want people to upgrade more regularly.

Again not saying this is true but I wouldn’t be surprised if it was.
 

bearinthetown

macrumors 6502
Original poster
May 5, 2018
286
323
I don’t have any evidence as frankly I can’t be bothered to look into it and see no noticeable differences with macOS.

However I have suspected similar to the OP suspicions regarding iOS.

I’ve had a personal 13 mini for a year. Just got an 11 for work and it feels noticeably faster. I wonder (if there is any intentional slow down) if it’s more related to hardware age as opposed to outright software updates. Would make more sense for apple to do that on iOS devices as these are the main sellers which they want people to upgrade more regularly.

Again not saying this is true but I wouldn’t be surprised if it was.
In case of iOS it was 100% true, not sure how it is these days. My iPhone 11 Pro Max is still comfortably fast after almost 4 years, while I used to notice big slowdowns after less that 2 years with some older iPhones. Also, perhaps another controversial claim from me, but I can't resist an impression that these "budget" iPads are designed to literally be annoyingly slow. Like come on, the experience is just awful.
 
  • Like
Reactions: joehohoho

Xand&Roby

macrumors 6502a
Jun 13, 2020
533
486
I'm not sure what you mean.
At first you seem to understand that this is programmed obsolescence, and in fact it is so, for those who want evidence can try to use a device for 10 years (a computer or a Mac, because all other products, Apple and not Apple, hardly get there); then you start saying that it's not like that, you know why you're a programmer, and you try to explain that it's an induced need, and that's true too (and that, I'm sorry, I'm not going to explain it to those who were born and live in the homeland of capitalism).

So what is it that you want to hear you answer? You already know the answers.

And there is no debate, here, because here, but also elsewhere, many pose as if they were shareholders or owners of Apple, and I hope they are because at least they have a gain to hold this position, and with those who have no doubts it is useless to confront.

P.S.: I've been an Apple customer since before many were born here.
I have had and have millions of Apple products, which I have maintained or stored, never thrown away, over time.
I know the haters and have faced them for decades.
I can evaluate the quality of the sw and the hw Apple, and it has decreased considerably in the last decades, despite the fact that the capacities of the sw are basically the same.
This was possible because Apple capitalised on the customer satisfaction index, which was among the highest if not the highest in the world, something like 97~99% of customers, 10 years ago.
And what do you do when you have such a high satisfaction rating? Simple, lower it, go between 90~93%, and with what you save in terms of investment in the product you greatly increase sales.
Et voilà, the numbers, even financial, of Cook.
 

leman

macrumors Core
Oct 14, 2008
19,319
19,336
For example: It was designed to only run on machines that support AVX2 instructions, which means newer CPUs. Yes, there are some folks that can get it to work on non AVX machines, but those machines have a really hard time using it, and it runs very unstable.

AVX2 is available on all Intel processors sold after 2011 and is supported by virtually all currently operated hardware, aside from few folks who insist on using obsolete dinosaurs.
 
  • Like
Reactions: NC12

erikkfi

macrumors 68000
May 19, 2017
1,726
8,087
Those are all very capable machines, with one caveat -- they don't support AVX2 instructions, which is a requirement for Ventura.
A fake requirement, as it turns out. Ventura runs 100% fine on 2013-era hardware with workarounds. Some have had success with even older machines.
 

erikkfi

macrumors 68000
May 19, 2017
1,726
8,087
I'm surprised to see it considered controversial to say that new macOS versions decrease the feeling of speed on older hardware.

My best anecdotes are old, though. My 2009 MacBook Pro went through this: it felt older and older with each new OS X version. One day I got nostalgic for Snow Leopard so I threw that onto a new partition, and was stunned by how fast it was to boot up and do anything in the Finder. Of course, I'm talking about a lot of versions between 10.6 and 10.11, but what a cumulative slowdown! I've got no reason to believe that new macOS versions that add more and more on-device photo recognition tasks and other things don't have the same effect today.
 

diego.caraballo

macrumors 6502a
Oct 18, 2013
621
1,604
I'm surprised that some users claim that they haven't noticed any slowdown in the different Mac OS versions running on the same machine.

There are probably several videos available, but just some samples:

Big Sur vs Sierra: 3 times slower.

Big Sur vs Ventura:

The slowdown is real and is even happening with M2 devices.
My Air now starts around 10 seconds slower after Ventura.
My GF MB Pro 2017 is reaaaally sluggish after Ventura, but was running Big Sur quite fast.

Sure some OS's had added features that maybe would require more things to load at boot time,
but for me, the latest batch of OS's (from Catalina and beyond) are just unnecessary slow.
And I also feel that Intel machines are being much more "penalized" than Mx.
 
  • Like
Reactions: Choco Taco

adib

macrumors 6502a
Jun 11, 2010
711
559
Singapore
Are they intentionally slowing older macOS devices? I doubt it, but they are not putting the effort into keeping them optimal. Is that the same thing as intentionally slowing them down? Perhaps.
This is it. One notable example would be the Music app that engages the discrete GPU if the "Now Playing" section is ever opened — draining the battery for no good reason at all.
 
  • Like
Reactions: LeeW

prefuse07

Suspended
Jan 27, 2020
895
1,066
San Francisco, CA
A fake requirement, as it turns out. Ventura runs 100% fine on 2013-era hardware with workarounds. Some have had success with even older machines.

I agree with you, but I already addressed this on page 1 -- next time, read through the thread before posting, because chances are 99.9% of what you are repeating has already been covered 👍
 
Last edited:
  • Haha
Reactions: erikkfi

leman

macrumors Core
Oct 14, 2008
19,319
19,336
I'm surprised that some users claim that they haven't noticed any slowdown in the different Mac OS versions running on the same machine.

If one wants to cherry pick irrelevant benchmarks, I’m sure one equally find a case where older macOS versions are slower. Macs are optimized for hibernation, not fast boot. Apple designs these systems to be never shut down. How does boot time translate to system performance? It’s like saying that the car got slower because the luggage compartment door opens slowly.
 

one more

macrumors 601
Aug 6, 2015
4,563
5,746
Earth
This is a good question. I am not a specialist, but have used computers for a while.

My guess is that as software becomes more and more sophisticated with various processes running simultaneously, this “eats” into more hardware resources, an obvious example being the RAM - the more there is, the smoother things can run. Similar goes for graphic, sound blocks, etc., etc. We do agree that a game created in 2020, for example, is way more realistic and engaging than the one made in 1995. So people’s consumer expectations of what a “good software” is and how this should run have evolved over time accordingly. So if Apple brought back something along the lines of iOS 5 in 2023, we would all go like “Meh!”. ;)

From my own experience, this “planned obsolesce” pattern was very present with Windows machines, to the point that it was considered a good practice to wipe them clean every 6 months or so.

On Apple side things are actually much better in comparison - I can still use my “old” 2015 MBP with a fresh installation of Monterey quite nicely. At the same time I cannot update a high end 2017 (?) Dell Spectre to Windows 11, as its processor “is not supported”.

Then nobody in my immediate surroundings complains about how their iPhone SE 2020, iPhone 11, iPhone 12 mini, iPad Air 3 and iPad Air 4 can handle iOS 16. Let’s not forget that for a majority of people these are just tools to make their lives smoother.

So, personally, I get the best of all my hardware for as long as I can and then just replace it strategically, not necessarily going for the latest, more expensive or Pro lines. As long as I can use it to run the latest iOS adequately, I am happy. Besides, upgrading your hardware after 3-5 years brings in more tangible improvements vs doing it annually or once every two years.
 
  • Like
Reactions: BigMcGuire

Isamilis

macrumors 68020
Apr 3, 2012
2,092
993
I'd like it to be a kind discussion. I'm expecting this, but saying just in case.

I have an unpopular and perhaps a little controversial opinion that Apple makes old devices obsolete on purpose. Most people I talk about with about this disagree. Most people think that adding features to software makes it slower. I'm a programmer myself and I know this is not the case. Unless this feature works in the background, it has no impact on performance whatsoever.

It's the same with iOS. When we install new versions, our hardware "naturally" becomes more and more obsolete. Or does it not? What is exactly happening in these new versions that use more resources all the time? I remember that before this "battery saving" (yeah, right) scandal blew up, every iOS version made home screen slower on older iPhones. I've been super suspicious about this, because making UI smooth is easy even in resource-hungry environment. It's just the matter of thread priority and Apple clearly didn't want to do this. And even that's assuming that something else used most resources of older iPhones, which I'm quite sure is not the case at all. Luckily, after the scandal they stopped it and now even a few years old iPhones are smooth on home screens.

I opened this topic as macOS thread, as I'm expecting more professionals here than in the iOS thread.

I feel like there are some thinking pattern traps that allow this to happen. One day I realized that it works like that with any software update. Once it's new and branded as "stability improvements", we feel like our software is super stable. Until there's next one - before we install that one, we feel like our previous one is no longer stable and we have bad quality software. But our software never changed! It's not like it got less stable over time. I hope you guys understand my point, because this shows the thinking patterns about hardware too. Many software requirements don't change, because this software remains the same, yet we are always pressured to replace our hardware, because if the new one is released, our becomes "obsolete". This is insane if you think about it. My favorite example are MacBooks, especially the jump from Intel to Apple Silicon. 16" Intel used to be considered a monster, until Apple Silicon came and 16" Intel became instantly "slow", even though the software we use was still the same.

Is anyone with me on that?
It’s not Apple’s intention to make newer system become slower and needs more cpu & ram. But, it’s a natural path of software development where the developer always keep improving and adding new features (innovation) to their product to make them stay on the top.
It’s up to users now, whether to follow this route or choose stay at certain OS version (with some consequences like security and 3rd party app compatibility).

Problem with Apple, probably the cycle is too short.
 
Last edited:

AppliedMicro

macrumors 68020
Aug 17, 2008
2,290
2,644
Getting slower with OS updates was at least very noticeable if you had a non-SSD Mac. SSD / flash storage make a big difference compared to spinning HDD (still very much a thing in 2015), and changes to file system, Spotlight indexing etc. may explain that.
I reinstalled Ventura from recovery screen on the SSD, then 2 hours to transfer old HDD data to SSD, boot from SSD et voila, ten times faster, and the old lady can use her iMac another 5 years.
She absolutely can, if she doesn’t use the internet - or feels comfortable browsing it on a browser and operating system that don‘t receive security updates anymore.
 

melliflu

macrumors regular
Mar 15, 2010
103
383
Getting slower with OS updates was at least very noticeable if you had a non-SSD Mac. SSD / flash storage make a big difference compared to spinning HDD (still very much a thing in 2015), and changes to file system, Spotlight indexing etc. may explain that.

She absolutely can, if she doesn’t use the internet - or feels comfortable browsing it on a browser and operating system that don‘t receive security updates anymore.
As far as I know the Ventura OS on the external SSD will get the standard Ventura updates no?

A cleaner but riskier way is to replace the HDD by a SSD inside the iMac. Apple could definitely propose this for a reasonable price if they really cared about sustainability (most HDD macs after 2013 would be running just fine with a SSD)
 
Last edited:

bearinthetown

macrumors 6502
Original poster
May 5, 2018
286
323
Many people also claim that VS Code text editor is not laggy, while for me it's barely acceptable in terms of snappiness. Some of you make fun of calling it slower perception, but it really is a thing.
 

darkpaw

macrumors 6502a
Sep 13, 2007
700
1,336
London, England
This is not true. Like I said in my post, new features do not slow down the system, unless they are services running in the background.
Actually, they can. How about going from a @2x image to a @3x image? Does it take the same amount of memory or processing power to view these images? What about moving from 1024x768 resolution to a 5K screen or a 4K screen like on the iMac 24"? Same features in the OS, just more pixels to push around meaning more memory is used.
 
  • Like
Reactions: jchap

AppliedMicro

macrumors 68020
Aug 17, 2008
2,290
2,644
As far as I know the Ventura OS on the external SSD will get the standard Ventura updates no?
Yes - but macOS Ventura itself will prospectively become unsupported in two and a half years (autumn of 2025).

And I doubt that macOS Ventura's successor (macOS 14) will be supported on that 2017 iMac - especially with the transition to ARM going on, in which they can't drop support for old machines quickly enough.
 
  • Like
Reactions: melliflu

Smartuser

macrumors regular
Oct 18, 2022
185
348
I'd like it to be a kind discussion. I'm expecting this, but saying just in case.
I'd suggest being kind to others first, the rest will be automatic.
I have an unpopular and perhaps a little controversial opinion that Apple makes old devices obsolete on purpose.
This is by no means an unpopular opinion, it's an extremely popular misconception and comes up all the time in some places.
My evidence is my programming knowledge and my experience with Apple devices.
Sure.
if it's true that you don't notice the slowdown probably means that you have slower perception, many people do.
That's interesting to learn, but unfortunately I can't find anything about that theory.
I just learned the hard way that our perceptions differ. A lot. I notice huge slowdowns.
What you call "perception" is what I call "opinion".
I don't need scientific evidence to notice that something, once snappy, became not snappy.
Sure.
"budget" iPads are designed to literally be annoyingly slow. Like come on, the experience is just awful.
"Come on".


You say you don't need scientific (or any other meaningful) evidence, but you're not going to convince anyone not already on your side without hard evidence.

The best way to get a "kind discussion" is to be polite and convince people by supporting your claims with data, not by condescendingly dismissing all requests for evidence by either doubting people's cognitive abilities , making specious claims about not needing any proof due to your circumstances or simply "come on".

I'm very sorry, but I'm absolutely not convinced of your claims. If everything is as obvious as you claim, then I'm sure you can back them up with solid evidence.
 

jchap

macrumors 6502a
Sep 25, 2009
586
1,061
I don't need scientific evidence to notice that something, once snappy, became not snappy.
I think the point is that even for perceptual issues, which are based on subjective experience, you would need to have benchmarks of some kind to prove what you're saying, at least statistically.

I do get what you are saying, and agree that newer versions of macOS on the same hardware might typically produce more delays or issues that are perceptually-based, not easy to benchmark or measure. It might depend a lot on how you use your machine, what other services and processes you are running and how well those services and processes function on the newer OS versions, how various services interact and so on. It seems incredibly complicated to track down these perceptual issues, so we often label them as the machine "getting slower."
 

Smartuser

macrumors regular
Oct 18, 2022
185
348
[...] transition to ARM going on, in which they can't drop support for old machines quickly enough.
There's no support for that claim. The dropping of support for older machines for Ventura is not significantly faster. There were some articles about that a few months ago, google for it.
 

bearinthetown

macrumors 6502
Original poster
May 5, 2018
286
323
I'd suggest being kind to others first, the rest will be automatic.

This is by no means an unpopular opinion, it's an extremely popular misconception and comes up all the time in some places.

Sure.

That's interesting to learn, but unfortunately I can't find anything about that theory.

What you call "perception" is what I call "opinion".

Sure.

"Come on".


You say you don't need scientific (or any other meaningful) evidence, but you're not going to convince anyone not already on your side without hard evidence.

The best way to get a "kind discussion" is to be polite and convince people by supporting your claims with data, not by condescendingly dismissing all requests for evidence by either doubting people's cognitive abilities , making specious claims about not needing any proof due to your circumstances or simply "come on".

I'm very sorry, but I'm absolutely not convinced of your claims. If everything is as obvious as you claim, then I'm sure you can back them up with solid evidence.
Well, you don't notice the difference, good for you. But please admit you don't find 120 Hz an improvement neither. But that one can at least be "scientifically proven".
 

AppliedMicro

macrumors 68020
Aug 17, 2008
2,290
2,644
There's no support for that claim. The dropping of support for older machines for Ventura is not significantly faster.
Yes there is.

macOS Ventura, released in fall of 2022 doesn't support anything older than "2017"-generation Macs. It even dropped support for some machines that were sold as brand-new by Apple as recently as 2019 (such as the 2017 MacBook Air and the Mac Pro). That gives a period of 5-6 years of supporting the newest macOS release, plus an additional 2 years of security update support.

Mac OS X 10.15 Catalina, their last pre-ARM OS, released in fall of 2019, on the other did include support as far back as 2012 models. Mac OS X 10.13 High Sierra, released in 2017, included support for 2010 and even some popular late 2009 Macs (iMacs, Macbooks).

2022 - 2017 = 5 years (or 4 for the unsupported legacy non-Retina 2017 MacBook Air)
2019 - 2012 = 7 years
2017 - 2010 = 7 years (or 8, if you're counting the supported 2009 models)

EDIT: To expand on this, let's look at the oldest supported model generations among iMacs, MacBooks, MacBook Airs, and MacBook Pros (discounting the Mac mini, iMac Pro and Mac Pro, since those have much lower volumes and have had much longer release/update cycles, often staying in Apple's lineup for 4+ years) by calendar year:

year of macOS release (in fall) - year of release of earliest supported iMac/MacBook models for that OS version
2022 - 2017/2018 = 5/4 years
2021 - 2015/2016 = 6/5 years
2020 - 2013/2014 = 7/6 years
2019 - 2012 = 7 years
2018 - 2012 = 6 years
2017 - 2009/2010 = 8/7 years
2016 - 2009/2010 = 7/6 years
2015 - 2007/2008 = 8/7 years
 
Last edited:

neuropsychguy

macrumors 68020
Sep 29, 2008
2,452
5,908
I don't need scientific evidence to notice that something, once snappy, became not snappy.
Here’s what you said, “I have an unpopular and perhaps a little controversial opinion that Apple makes old devices obsolete on purpose.” (emphasis added)

My comment was about you saying Apple intentionally slows older computers down with new macOS versions (which is the title of this thread you started: "Why does macOS slow down with newer versions").

Yes, you do need scientific evidence to make a claim that Apple intentionally slows down computers, otherwise you're conjecturing. “Feels less snappy” is entirely subjective and cannot support your claim that Apple is intentionally slowing down computers. I understand the "less snappy" experience with computers. That's one reason I try to upgrade my Macs and my computers running Windows every 3 years or so. However, I don't blame Apple, Microsoft, the various companies writing firmware, or the various companies producing software as intentionally making older computers/hardware obsolete. There is some planned obsolescence because at some point companies, including Apple, are not going to put effort into supporting older hardware. Technology moves forward and some things get left behind.

That really is not the point you were making though.

Your responses come across as highly dogmatic and full of confirmation bias because you are dismissive of points I and others are making. You might be correct but to verify what you write will take 1) reproducible, scientific experiments with various Apple computers and 2) evidence that any slowing (if it exists) is a deliberate ("on purpose") practice done by Apple's management and engineers.
 
Last edited:
Register on MacRumors! This sidebar will go away, and you'll see fewer ads.