I understood this as a “the next generation will cost more”. By now I am sure apple can fairly accurately predict device sales for each season and so they likely have a decent backlog for the current generation, procured at reasonable cost from hardware manufacturers.
It’s when they had to negotiate for the next generation where the price would be hiked.
Like the author I wouldn’t bet more than a beverage on this though.
On some level the Neo is just a new chapter in the story of the original iBook. The difference between 1999 and now is the much stronger Apple brand, so people on the fence were quite ready to throw down with a $600 Mac.
Classic MacOS never had an entirely reliable network stack for browsing the web but hey, they had a GUI running in 128k of RAM in 1984. Not to say that macs didn't have their charms during the .com bubble but it wasn't until Mac OS X came out in 2001 that a mac was a purchase that made sense.
Were you actually a Mac user? The initial release of OS X was quite inferior to OS 9--it was an incredible proof of concept and perfect developer target, but for an end user, not so much.
I've been in computer labs full of Mac OS 8/9 machines happily browsing the internet so I'm not sure what your claim about "unreliable network stack" is referencing. Unless you mean "a crash requires a reboot" which was true, but also often true for Windows 95/98 as well!
If you had good enough hardware to run OS X, running OS 9 stuff in Classic meant that at least you didn't have to restart your whole computer when something froze, just all your OS 9 software. OS X was definitely slow and a bit buggy early on, but IIRC it got good enough for daily use around when Jaguar came out (late 2002?).
OS 9 running Microsoft Internet Explorer! I can remember the puck mice and the noise of the hard drive grinding away while pulling up Ebay and the campus webmail (Horde/Roundcube debates going on and on...)
When I was in grad school I worked in a corner which had a mac on one side and an IBM POWER system with a huge screen on the other side.
Early on the mac was crashing all the time when I was browsing the net, at some point in the 8-9 era they added a bunch of locks to stop the crashes and then it was beachball city all the time.
My understanding was that classic was built on the assumption that events come in from the keyboard and mouse and once you added more events from the network it exposed race conditions. It probably didn't help that we were on "Internet 2" had were early to get 100 Mbps ethernet. If you were using dialup it was probably not so bad.
People think "you have race conditions or you don't" and that is true on some level but if your utilization factors are low they might cause problems once a month but increase the load and those problems are happening once an hour. I saw all the symptoms that I expected such as a brand new and faster mac crashing less often than the old mac because it had more capacity to process events and less overlap between them.
Yeah '95 was bad, NT not much better despite the fanbois saying otherwise (had terrible fights w/ a prof who was a roommate of Bill Gates who got a grant from Wintel to get new x86 machines.) Once we got those machines there were two of us you'd always find in front of the NT machines: the student who liked NT and myself who would use VNC to log into one of the few Linux machines. One day that prof came around and said "you win Paul!" and announced that most of those NT machines would be switched to Linux.
The rainbow beachball cursor was added in OS X. I think your memory of the Mac OS 9 machines might be mixed up.
Classic Mac OS used a cooperative threading model which meant that one program locking up would lock up the whole computer, but Windows 95 was only slightly better in this regard. It didn't have any classical race conditions with the network as you can't "race" in a world with no preemptive threads--control is only ever handed off when a program explicitly yielded.
Win 95 had real processes and threads, memory protection and all of that but it wasn't terribly effective. The whole machine would go down frequently (as did Win NT despite Microsoft’s insistence otherwise; Microsoft kept repeating "It's the hardware stupid", Win2K was a great improvement but WDDM in Vista meant you could crash your display system and not crash the rest of the machine and I think got us in "rock solid" territory except for the the power management crashes that still dog us today) and it was absolutely endemic that applications would crash, I think because of memory safety problems. The industry went through a transition of "those crashes are annoying" to "those crashes prove your system is vulnerable to hacking..
I have had to argue with people so many times over that "you can't have race conditions in systems that have cooperative multithreading" in teams that couldn't fix those race conditions because they didn't believe they can exist.
Like yeah, thread A can't stomp on data that thread B depends on but you can definitely have situations where different things happen because of different orders of events. Like I was working on a knowledge graph editor in GWT around 2005 or so that had problems because it would try to cache results of XHRs and very different things could happen if a function called a callback synchronously with data from the cache or if later that callback got called by a callback that was called by the XHR. If you are systematic about things you can do it right, everybody is aware of these problems in threaded systems and threaded systems give you tools for protecting your data but in cooperative systems people will tell you don't exist even when the systems are crashing and corrupting data right before their eyes.
(Look at how Python has synchronization primitives for async that are roughly parallel to the ones in threading. In old school cooperative multithreading you are taking code that could be threaded and cutting it up into pieces manually and that gets super-difficult as complexity. Even if you don’t want to face up with concurrency means concurrency doesn’t want to face up to you.)
Only if the application explicitly set the cursor. There was no default indication that an application had stopped processing events - your only clue was that everything stopped responding.
> My understanding was that classic was built on the assumption that events come in from the keyboard and mouse and once you added more events from the network it exposed race conditions. It probably didn't help that we were on "Internet 2" had were early to get 100 Mbps ethernet. If you were using dialup it was probably not so bad.
They added preemptive multitasking around System 7.5, but anything that used the Toolbox still had to run in the main thread and be cooperatively multitasked so it wasn't much of an improvement (hence the lockups).
For that matter when they added SMP support to Linux they put most of the kernel behind a giant lock so, effectively, applications could run multiple processors even though the OS could only handle one. I guess you could have done the same thing with a single-process OS.
The list of classic Mac OS programs which actually spawned preemptive threads was probably in the single digits. Very few of the machines were multi-core, so it didn't matter much.
I'm not sure exactly how many programs used them, but support was fairly widespread in a lot of popular programs because it was the only way for Classic Mac OS to make use of the second processor in multiprocessor machines. Some examples of software that used preemptive threads are Photoshop, Final Cut Pro, After Effects, Premiere, Lightwave, and Cinema 4D (see https://www.macworld.com/article/159778/mp_mac.html ).
The network stack was fine (they licensed the same STREAMS implementation also used in AIX, UnixWare, VxWorks, etc) - it was everything else around it (cooperative multitasking, no virtual memory, no memory protection) that was a house of sand.
At least the iBook would come out of sleep when you opened the lid, I don't think Windows laptops could manage that until 2007.
You still have to turn off USB power management on a windows machine to avoid serious problems just as you have to turn off Bluetooth power management if you don't want to be connecting and reconnecting your headphones several times a day.
> At least the iBook would come out of sleep when you opened the lid, I don't think Windows laptops could manage that until 2007.
I think that was very hardware-dependent. I don't recall what hardware I had, but it had a > 90% success rate of resuming from suspend and/or hibernation under Windows XP.
Apple's advantage there being pretty obvious: they control the entire range of hardware
> You still have to turn off USB power management on a windows machine to avoid serious problems just as you have to turn off Bluetooth power management if you don't want to be connecting and reconnecting your headphones several times a day.
The irony being that I still have to turn wifi and bluetooth off and on on my macbook air regularly today, as airplay stops working every ~10-20 transfers / every other week.
I think it would be uncharacteristic of Apple to raise prices anytime from now till new products are announced in September. It doesn’t match Apple’s brand image (like the author says). As pointed in another post by John Gruber, Apple kept selling the trash can Mac Pro for a very high price for years without any updates. So it can certainly afford to bear this pain for a couple of more months and bundle all the price hikes together.
The threat of a price hike may increase sales in the near term (especially the back to school sale) and could tamper down the drop in profits a bit. After all, the hardware bill of materials is not the only thing deciding the product price.
A bigger hike now could have a snowballing effect on “switchers” and the potential services revenues they could bring.
I’m guessing that Apple will increase the prices of all products with the iPhone and Apple Watch launches in September. The increase in prices for currently selling products will be a store update, without any press release or news or tweet or any notification. That’s the (quiet) Apple way of doing things.
It's also a financial optimization: Even with the very expensive RAM upgrade prices, it's way better to use the same memory chips to manufacture and sell four Mac Minis or two Mac Studios, rather than a single high-RAM unit.
Demand for 512+gb hasn't gone away though. I suspect we'll see M5/M6 Ultra later this year with 512gb-1tb of ram going for the biggest premium we've ever seen from the company.
> I suspect we'll see M5/M6 Ultra later this year with 512gb-1tb of ram
I suspect the opposite. 512gb of LPDDR5X is enough memory to make sixty four Macbook Neos. Apple's HPC/server audience is not very large or demanding, it seems likely that they will avoid memory upgrades during the shortage to focus on delivering value to low-end consumers and driving service sales.
It was probably two birds one stone: they could use that RAM for other more likely to be purchased models, and the loss from selling 512GB of "pre-RAM-AI-tax" costs would have been too high. They would have certainly sold those models though.
It'll really depend on apple's contracts with suppliers and the types of ram used in certain products. I could see them having medium term contracts, with first right of refusal to purchase on the longer term.
Memory has been a boom/bust industry since the 1970s, so I imagine people are careful with long-term agreements, but that's just me spitballing.
I suspect you’re right, but if they don’t do it until the new stuff is announced I suspect the price hike would drown out the news of all the new stuff. And I really don’t think Apple wants that.
If they announce it on, to pick a date, July 1 then by the time the iPhone is announced it will be somewhat old news. It will still get mentioned but it won’t be the main feature of every story at announcement time.
Like so many other companies Apple is between a rock and a hard place with this stuff.
> I suspect the price hike would drown out the news of all the new stuff. And I really don’t think Apple wants that.
I think they could get away with it if they do something like drop the price of the lowest device tier (or minimally hold price steady) and only increase prices on their more premium line up, where the premium buyers are less price sensitive.
The problem is the things going up are core. It’s not like the cameras are getting more expensive, it’s the RAM and processors and storage. Which all the phones have to have.
If anything the percentage increase is probably higher on the low end phones relative to total bill cost.
> As pointed in another post by John Gruber, Apple kept selling the trash can Mac Pro for a very high price for years without any updates. So it can certainly afford to bear this pain for a couple of more months and bundle all the price hikes together.
It seems unlikely that Apple created a rainy day fund from offering an legacy product at niche prices almost a decade ago. Equally unlikely that Apple will continue to sell at a loss today out of a traditional disregard for decreasing component costs.
This sure seems like the Apple way to do it. "The old stuff is now more expensive" is a worse marketing announcement than "the new stuff is awesome, and with it comes a higher price just like most years".
They might add it as a surcharge the way Ubiquiti is doing it. I recently bought a bunch of stuff and it all comes with a memory or tariff surcharge that is annoying but it's hard to argue it.
Also we know that it's coming soon, that's why Cook is running cover for the new CEO. They don't want the new CEO to be the one taking the fall on higher prices, so before September 1 will be my guess.
Thank God I made the right decision and I bought a max'ed out Macbook Pro 5 Max with 128 GB of memory a couple of months ago. I think prices will continue to keep going up.
You can avoid the Ubitiquiti tariff surcharge and shipping fees if you buy via B&H. They don't stock all Ubiquiti products but they have most of the common stuff.
Unfortunately, a lot of B&H's Ubiquity stock has the memory surcharge built into the base price, so it can't avoided it just isn't labeled as obviously. Shipping, usually can be avoided. But also note that (annoyingly) UI curtails the warranty period on all gear purchased third party.
They already did, to an extent. The base Mac Mini used to be $600 with a 256 GB SSD, but they got rid of that and now the cheapest option is the $800 512 GB model. The typical Apple strategy is to make the base model of a given computer the price competitive version, and price gouge on additional storage and RAM to build up their margins. I guess the RAM price increase made the margins on the $600 model too low for Apple to want to sell.
> Gabe Newell worked at Microsoft for 15 years - people need to start considering Valve an extension of Microsoft. Their unofficial App Store for games.
This is a really weird take, considering Valve's push into Linux.
It’s really a factually incorrect take. The only reason Microsoft puts its games on Steam is because they have to for their game studios to be commercially viable. They have already tried getting a foothold with the Microsoft/Xbox store and the truth of the matter is that Steam holds all the cards in their overwhelming user base.
Valve is not an extension of Microsoft, they are a company that out-competed them.
I actually find it somewhat surprising that Microsoft is as warm as they are toward Valve. If there is a personal relationship factor, maybe that’s part of it.
But I think another part of it is that Microsoft ultimately makes more money via Steam than if Steam wasn’t a part of their ecosystem.
Well Apple has no excuse for this. They literally had a quarter trillion dollars sitting, they could had bought an entire ram manufacturing company along with its entire supply chain (sand included).
The argument “the market cornered us” is ridiculous for a behemoth like Apple. They had the opportunity to end, android, pc, consoles overnight.
Apple’s services make more revenue than all Mac and iPad revenue combined, at a higher profit margin.
I could absolutely imagine Apple buying or creating a new DRAM plant, with a strategy of Apple being the primary customer and being able to undercut the rest of the consumer electronics industry.
If they had extra capacity they could sell chips at market rate as a bonus, but they’d be able to prioritize themselves as a customer.
Seems like a strategic slam dunk to me. Apple has only been getting more vertically integrated over time. With the way Apple subscription services work they have much more of an incentive to prioritize market share over hardware margins than in the past.
I have been buying an Asus laptop with Intel for our employees in the last 1.5 years, an ASUS ExpertBook P5 90NX0861-M007X0 to be precise, and it did not see even a cent of price increase.
I suspect this is because of its CPU, Intel Core Ultra 7 258V with 32GB RAM directly soldered on the CPU package, being overstock or something, or Intel secured the RAM supply upfront. I don't know but it's wild that this laptop is still early 2025 prices.
I was planning to wait until the M6 generation to replace my 16" MBP M1 Pro. But after Cook said they will have to increase prices soon, I went out and bought an M5 15" Air with 24GB/1TB.
I think the M6 Pro/Max generation will have a substantial price increase given the new design, OLED, N2 TSMC node, and memory prices not coming down.
It’s pretty incredible to get a Mac Studio that performant for only two weeks’ pay… Well, two weeks’ pay for doing what you’d need to be doing to justify that Mac Studio, of course.
Money aside since it’s probably getting to a used car territory pricing, but completely agree how amazing the Mac Studio is if you need it and know what you are doing
Can't imagine increases to existing products. Would seem almost certain it would be for new product/refreshes and they may simply continue to remove the entry level configs in the meantime. That is all still significant for Apple for whom pricing is part of the brand. Been waiting for a maxed out new MBP coming later this year/early next, which already is about 7K, so with the prices increases, scared to think what the 8GB SSD and max ram will come in at.
>I also do not think they’re going to raise the prices of existing products mid-cycle
This surprised me too. I'd accepted that price hikes were coming for the new range...that's expected. But hiking prices on the existing range felt like a step too far!
Might have been a marketing stunt to nudge people into upgrading.
Well, if that was the plan, it worked. I just caved and bought an M5 to replace my older one. Boo.
Why is that a step too far? e.g. the PS5 price has increased like 25% and it's a 6 year old product now. Similar for other consoles. It feels much more acceptable to me on something like a Mac that's less than a year old (and going to last a long time + have good resale value).
> Why is that a step too far? e.g. the PS5 price has increased like 25% and it's a 6 year old product now
Precisely because it's a 6 year old product. The $499 the PS5 cost at launch in 2020 is equivalent to ~$650 in 2026 according to the inflation calculator [1]. Within a year it's harder to justify that. Nobody believes Apple is paying the price of the day for components instead of having them negotiated at least for the whole run of a model.
> It feels much more acceptable to me on something like a Mac that's less than a year old (and going to last a long time + have good resale value).
That sounds like it should be exactly the other way around? A PS5 from 2020 is substantially identical with a PS5 from 2026 except maybe for some minor HW optimizations. They are completely fungible. A Mac from this year will compete with a faster model next year, and another even faster model the year after that.
PS5's increased to compensative for the many Sony gaming flops, like the $100m Concord (closed down within a couple of weeks) and the $200m Marathon (still on life support but they had to cancel the much more popular Destiny 2 to prioritise it).
Macs have much shorter lifecycles than consoles, and Apple doesn’t have a history of reducing prices over that lifecycle. Consoles have much longer lifecycles and typically drop in price over time (as components get cheaper).
It seems fair to expect that behaviour to work in both directions.
Yes, exactly. Price hikes mid-cycle is not something I have seen Apple do historically.
If true, one reason could be that: Cook is retiring, while Ternus is on his way in. Ternus wouldn't wanna start off with his first announcement being a ginormous price hike. Makes him look bad.
So Cook becomes the fall guy. i.e., he increases the price by 15% in the existing M5 range.
When Ternus comes in, he keeps prices stable on the M6 range.
Makes Ternus look like an awesome guy...and gets him off on a flying start in his CEO tenure.
Sure, and what I'm saying is that I'm ok with them doing that. It's a matter of consistency.
The PS1, PS2, PS3, PS4, Xbox, Xbox 360, Xbox One all followed the pattern of dropping in price over the course of their lifetimes, because components got cheaper over time. The fact that the PS5 and the Xbox Series X|S have gone up in price is consistent with the general price elasticity.
I'm also OK with Apple having a rigid pricing structure and never really doing any sales or discounts, but then I expect them to not raise prices on the current M5 hardware, and leave those price hikes for the M6 generation that I assume is just around the corner.
I haven't seen other people call it but I think the ultimate force behind Moore's law is not transistors getting smaller but transistors getting cheaper. Like you can make them smaller all you want but if you can't make them cheaper you can't afford them.
In the 2020s we've blamed extrinsic causes such as the pandemic and the AI bubble but I am suspecting the intrinsic cause that new processes are more expensive to develop and more expensive to operate.
The tragedy of the PS5 is that even though it has sold a lot of units it doesn't represent the kind of generational change like the PS3 was. There are roughly 15 exclusive games and closest I come to one that I want is a remake of one of my favorite games from almost 20 years ago.
Perhaps I wasn't clear. I'm talking about the product lifecycle, not the device lifecycle.
The M1 Mac mini, the PS5 and the Xbox Series X|S were all released on November 2020.
Since then, Apple has released the M2 mini (Jan 2023) and the M4 mini (Nov 2024), and I'd be willing to bet we'll see an M6 mini later this year. The PS5 has had the habitual mid-generation "Slim" refresh and the Pro (a smallish specs bump), and Xbox Series X|S has seen barely any changes.
My M1 Max MBP is still a beast of a laptop, but it hasn't been the current model in aeons. My Xbox Series X is still current.
This is understandable given the market but part of me really wishes this would wait a year even though it won't.
Mac hardware is so close to being really useful for local LLMs and it's shared memory architecture could be a direct shot across the bow of NVidia's aggressive VRAM Market segmentation but it just can't compete with the raw FLOPS and memory bandwidth of NVidia. You can buy a Macbook Pro with an M5 Max with 128GB of RAM for $6k currently. I expect that will go up by 20-50% in the next generation.
It's safe to say that no current Apple product will get a RAM bump for the next 1-2 cycles at least.
I think this is going to impact NVidia too but in a different way. Normally in NVidia's product cycle we'd expect 50x0 Super mid-cycle refreshes. It's clear that's not happening this time around. We might expect the 6000 series late next year. I think there's zero incentive for NVidia to do that so that'll likely get delayed into 2028 or possibly 2029. 5090 prices keep going up even though it's 1.5 years old.
Anyway, as for Apple I'm keenly watching for the anticipated refresh of the Mac Studio lineup. The previous gen (M3 Ultra, M4 Max) just don't have the raw horsepower even though they had configs up to 512GB (512GB and 256GB now discontinued). It'll be interesting to see what the max config is and when these come up. Q3 2026 is widely expected but I wouldn't be surprised if it slips into 2027.
I think we’re likely to see some kind of tiered storage (maybe still 8GB of GDDR5, but some additional memory running at slower speeds that’s volatile and then segmenting the SSD into a bigger, slower part and a smaller really fast part.
The technology all exists to do this and it’s ideal for the kind of local inference Apple wants to push.
Aren't their storage prices so inflated that they could just eat the difference?
For years with every other OEM I have bought the laptop with the minimum amount of memory and saved $800 or so buying the largest memory sticks that work with the machine from Crucial (R.I.P.) Doesn't work if the memory is soldered to the board though!
Hah, yes, with my cheesegrater Mac Pro 2019, Apple wanted $3,000 for 160GB of memory (to go from 32 to 192GB).
I bought it with 32 and OWC sold me the exact same sticks of memory (manufacturer, timings) for $1,050.
Same. $3,000 for 8TB of SSD. $1,200 for 4 x 2TB Samsung 990 Pros and a 4xM.2 NVMe PCIe enclosure. Which actually ran about 500MB/s faster than the Apple SSD.
This was inevitable. The better question is if AI related hardware costs drop after the AI bubble implodes, will Apple drop the prices? My answer is negative.
I think we might end up in a weirder situation: Apple _does_ drop their prices back down to current levels for the same quantity of ram, but ASP goes much higher, at least for the Pro tier buyers. My reasoning is that depending on how the benchmarks look, many of us may try to go big on ram on our next hardware purchases to run models locally as a way of hedging either model costs, or to ensure access.
I don't have apple-issued devices, so no money flows from me
into apple, but it should be pointed out that with an increase
in RAM prices, many of us already paid more in general, and
apple may benefit here indirectly when they can add their own
extra-cost onto devices on TOP of that increased RAM prices.
So this should be considered rather than merely focus on
(only) "Apple is increasing device prices". They are all milking
us.
I also think now, with RAM prices increased, ALL hardware manufacturers
should be considered an illegal mafia aka cartel. It can not be that
they steal money that way. That is not how capitalism and free market
work. This is a de-facto monopoly. States need to do something; the
USA under Trump is just a corporate disguise right now. They are doing
nothing about it. The EU is not much better, slow and like a behemoth
focusing on "data privacy" (but then handing over all of our data to
the USA anyway and on top of that mandating age sniffing soon). They
don't protect consumers from exploding RAM prices.
I understood this as a “the next generation will cost more”. By now I am sure apple can fairly accurately predict device sales for each season and so they likely have a decent backlog for the current generation, procured at reasonable cost from hardware manufacturers.
It’s when they had to negotiate for the next generation where the price would be hiked.
Like the author I wouldn’t bet more than a beverage on this though.
> By now I am sure apple can fairly accurately predict device sales
The Neo being an exception I beleive.
That’s a new product line. Should have clarified but I meant mostly their existing product lines and specifically iPhone
On some level the Neo is just a new chapter in the story of the original iBook. The difference between 1999 and now is the much stronger Apple brand, so people on the fence were quite ready to throw down with a $600 Mac.
True, but it's effectively a 'new' market segment when it was introduced and thus demand was less predictable.
Apple was on the ropes in 1999!
Classic MacOS never had an entirely reliable network stack for browsing the web but hey, they had a GUI running in 128k of RAM in 1984. Not to say that macs didn't have their charms during the .com bubble but it wasn't until Mac OS X came out in 2001 that a mac was a purchase that made sense.
Were you actually a Mac user? The initial release of OS X was quite inferior to OS 9--it was an incredible proof of concept and perfect developer target, but for an end user, not so much.
I've been in computer labs full of Mac OS 8/9 machines happily browsing the internet so I'm not sure what your claim about "unreliable network stack" is referencing. Unless you mean "a crash requires a reboot" which was true, but also often true for Windows 95/98 as well!
If you had good enough hardware to run OS X, running OS 9 stuff in Classic meant that at least you didn't have to restart your whole computer when something froze, just all your OS 9 software. OS X was definitely slow and a bit buggy early on, but IIRC it got good enough for daily use around when Jaguar came out (late 2002?).
Oh sure, by the time 10.2 rolled around, the system was quite usable. All the major apps had been Carbonized.
OS 9 running Microsoft Internet Explorer! I can remember the puck mice and the noise of the hard drive grinding away while pulling up Ebay and the campus webmail (Horde/Roundcube debates going on and on...)
I remember browsing using 5 slightly offset IE windows stacked diagonally, since this was somehow before tabs in browsers.
When I was in grad school I worked in a corner which had a mac on one side and an IBM POWER system with a huge screen on the other side.
Early on the mac was crashing all the time when I was browsing the net, at some point in the 8-9 era they added a bunch of locks to stop the crashes and then it was beachball city all the time.
My understanding was that classic was built on the assumption that events come in from the keyboard and mouse and once you added more events from the network it exposed race conditions. It probably didn't help that we were on "Internet 2" had were early to get 100 Mbps ethernet. If you were using dialup it was probably not so bad.
People think "you have race conditions or you don't" and that is true on some level but if your utilization factors are low they might cause problems once a month but increase the load and those problems are happening once an hour. I saw all the symptoms that I expected such as a brand new and faster mac crashing less often than the old mac because it had more capacity to process events and less overlap between them.
Yeah '95 was bad, NT not much better despite the fanbois saying otherwise (had terrible fights w/ a prof who was a roommate of Bill Gates who got a grant from Wintel to get new x86 machines.) Once we got those machines there were two of us you'd always find in front of the NT machines: the student who liked NT and myself who would use VNC to log into one of the few Linux machines. One day that prof came around and said "you win Paul!" and announced that most of those NT machines would be switched to Linux.
The rainbow beachball cursor was added in OS X. I think your memory of the Mac OS 9 machines might be mixed up.
Classic Mac OS used a cooperative threading model which meant that one program locking up would lock up the whole computer, but Windows 95 was only slightly better in this regard. It didn't have any classical race conditions with the network as you can't "race" in a world with no preemptive threads--control is only ever handed off when a program explicitly yielded.
It was a wristwatch back then, right?
Win 95 had real processes and threads, memory protection and all of that but it wasn't terribly effective. The whole machine would go down frequently (as did Win NT despite Microsoft’s insistence otherwise; Microsoft kept repeating "It's the hardware stupid", Win2K was a great improvement but WDDM in Vista meant you could crash your display system and not crash the rest of the machine and I think got us in "rock solid" territory except for the the power management crashes that still dog us today) and it was absolutely endemic that applications would crash, I think because of memory safety problems. The industry went through a transition of "those crashes are annoying" to "those crashes prove your system is vulnerable to hacking..
I have had to argue with people so many times over that "you can't have race conditions in systems that have cooperative multithreading" in teams that couldn't fix those race conditions because they didn't believe they can exist.
Like yeah, thread A can't stomp on data that thread B depends on but you can definitely have situations where different things happen because of different orders of events. Like I was working on a knowledge graph editor in GWT around 2005 or so that had problems because it would try to cache results of XHRs and very different things could happen if a function called a callback synchronously with data from the cache or if later that callback got called by a callback that was called by the XHR. If you are systematic about things you can do it right, everybody is aware of these problems in threaded systems and threaded systems give you tools for protecting your data but in cooperative systems people will tell you don't exist even when the systems are crashing and corrupting data right before their eyes.
(Look at how Python has synchronization primitives for async that are roughly parallel to the ones in threading. In old school cooperative multithreading you are taking code that could be threaded and cutting it up into pieces manually and that gets super-difficult as complexity. Even if you don’t want to face up with concurrency means concurrency doesn’t want to face up to you.)
> It was a wristwatch back then, right?
Only if the application explicitly set the cursor. There was no default indication that an application had stopped processing events - your only clue was that everything stopped responding.
> My understanding was that classic was built on the assumption that events come in from the keyboard and mouse and once you added more events from the network it exposed race conditions. It probably didn't help that we were on "Internet 2" had were early to get 100 Mbps ethernet. If you were using dialup it was probably not so bad.
They added preemptive multitasking around System 7.5, but anything that used the Toolbox still had to run in the main thread and be cooperatively multitasked so it wasn't much of an improvement (hence the lockups).
For that matter when they added SMP support to Linux they put most of the kernel behind a giant lock so, effectively, applications could run multiple processors even though the OS could only handle one. I guess you could have done the same thing with a single-process OS.
The list of classic Mac OS programs which actually spawned preemptive threads was probably in the single digits. Very few of the machines were multi-core, so it didn't matter much.
I'm not sure exactly how many programs used them, but support was fairly widespread in a lot of popular programs because it was the only way for Classic Mac OS to make use of the second processor in multiprocessor machines. Some examples of software that used preemptive threads are Photoshop, Final Cut Pro, After Effects, Premiere, Lightwave, and Cinema 4D (see https://www.macworld.com/article/159778/mp_mac.html ).
The network stack was fine (they licensed the same STREAMS implementation also used in AIX, UnixWare, VxWorks, etc) - it was everything else around it (cooperative multitasking, no virtual memory, no memory protection) that was a house of sand.
The iBook also started at over $3,000 adjusted for inflation ($1,599 in 1999).
That was a pretty average price for "value" laptops at the time, much like the pricing of the Neo today.
At least the iBook would come out of sleep when you opened the lid, I don't think Windows laptops could manage that until 2007.
You still have to turn off USB power management on a windows machine to avoid serious problems just as you have to turn off Bluetooth power management if you don't want to be connecting and reconnecting your headphones several times a day.
> At least the iBook would come out of sleep when you opened the lid, I don't think Windows laptops could manage that until 2007.
I think that was very hardware-dependent. I don't recall what hardware I had, but it had a > 90% success rate of resuming from suspend and/or hibernation under Windows XP.
Apple's advantage there being pretty obvious: they control the entire range of hardware
> You still have to turn off USB power management on a windows machine to avoid serious problems just as you have to turn off Bluetooth power management if you don't want to be connecting and reconnecting your headphones several times a day.
The irony being that I still have to turn wifi and bluetooth off and on on my macbook air regularly today, as airplay stops working every ~10-20 transfers / every other week.
I think it would be uncharacteristic of Apple to raise prices anytime from now till new products are announced in September. It doesn’t match Apple’s brand image (like the author says). As pointed in another post by John Gruber, Apple kept selling the trash can Mac Pro for a very high price for years without any updates. So it can certainly afford to bear this pain for a couple of more months and bundle all the price hikes together.
The threat of a price hike may increase sales in the near term (especially the back to school sale) and could tamper down the drop in profits a bit. After all, the hardware bill of materials is not the only thing deciding the product price.
A bigger hike now could have a snowballing effect on “switchers” and the potential services revenues they could bring.
I’m guessing that Apple will increase the prices of all products with the iPhone and Apple Watch launches in September. The increase in prices for currently selling products will be a store update, without any press release or news or tweet or any notification. That’s the (quiet) Apple way of doing things.
Maybe that's why the 512GB RAM models are gone: they opt to pull a model without further announcement rather than increase its prices.
It's also a financial optimization: Even with the very expensive RAM upgrade prices, it's way better to use the same memory chips to manufacture and sell four Mac Minis or two Mac Studios, rather than a single high-RAM unit.
Demand for 512+gb hasn't gone away though. I suspect we'll see M5/M6 Ultra later this year with 512gb-1tb of ram going for the biggest premium we've ever seen from the company.
> I suspect we'll see M5/M6 Ultra later this year with 512gb-1tb of ram
I suspect the opposite. 512gb of LPDDR5X is enough memory to make sixty four Macbook Neos. Apple's HPC/server audience is not very large or demanding, it seems likely that they will avoid memory upgrades during the shortage to focus on delivering value to low-end consumers and driving service sales.
Probably the first >20k$ macbook ever ...
It was probably two birds one stone: they could use that RAM for other more likely to be purchased models, and the loss from selling 512GB of "pre-RAM-AI-tax" costs would have been too high. They would have certainly sold those models though.
It'll really depend on apple's contracts with suppliers and the types of ram used in certain products. I could see them having medium term contracts, with first right of refusal to purchase on the longer term.
Memory has been a boom/bust industry since the 1970s, so I imagine people are careful with long-term agreements, but that's just me spitballing.
I suspect you’re right, but if they don’t do it until the new stuff is announced I suspect the price hike would drown out the news of all the new stuff. And I really don’t think Apple wants that.
If they announce it on, to pick a date, July 1 then by the time the iPhone is announced it will be somewhat old news. It will still get mentioned but it won’t be the main feature of every story at announcement time.
Like so many other companies Apple is between a rock and a hard place with this stuff.
> I suspect the price hike would drown out the news of all the new stuff. And I really don’t think Apple wants that.
I think they could get away with it if they do something like drop the price of the lowest device tier (or minimally hold price steady) and only increase prices on their more premium line up, where the premium buyers are less price sensitive.
The problem is the things going up are core. It’s not like the cameras are getting more expensive, it’s the RAM and processors and storage. Which all the phones have to have.
If anything the percentage increase is probably higher on the low end phones relative to total bill cost.
Which is where you accept the loss. Lower cost products are often required to be loss leaders to get people through the door.
> As pointed in another post by John Gruber, Apple kept selling the trash can Mac Pro for a very high price for years without any updates. So it can certainly afford to bear this pain for a couple of more months and bundle all the price hikes together.
It seems unlikely that Apple created a rainy day fund from offering an legacy product at niche prices almost a decade ago. Equally unlikely that Apple will continue to sell at a loss today out of a traditional disregard for decreasing component costs.
The best path and most likely path will be Apple stop the 256GB price point.
Customers will be forced to buy the higher capacity iPhone. So iPhone 17 512GB start at $999.
Effectively Apple hasn't really up the price, customer just don't have the lower storage option.
This sure seems like the Apple way to do it. "The old stuff is now more expensive" is a worse marketing announcement than "the new stuff is awesome, and with it comes a higher price just like most years".
Apple doesn’t increase prices most years, so that doesn’t really make sense. The iPhone Pros this year also look to be a minor update.
came to say this, there's almost zero history for this so unless their supply chain got meaningfully distorted I can't see why this would happen now
Yes, there's been a meaningful distortion in the global RAM supply chain.
They might add it as a surcharge the way Ubiquiti is doing it. I recently bought a bunch of stuff and it all comes with a memory or tariff surcharge that is annoying but it's hard to argue it.
Also we know that it's coming soon, that's why Cook is running cover for the new CEO. They don't want the new CEO to be the one taking the fall on higher prices, so before September 1 will be my guess.
Thank God I made the right decision and I bought a max'ed out Macbook Pro 5 Max with 128 GB of memory a couple of months ago. I think prices will continue to keep going up.
You can avoid the Ubitiquiti tariff surcharge and shipping fees if you buy via B&H. They don't stock all Ubiquiti products but they have most of the common stuff.
Unfortunately, a lot of B&H's Ubiquity stock has the memory surcharge built into the base price, so it can't avoided it just isn't labeled as obviously. Shipping, usually can be avoided. But also note that (annoyingly) UI curtails the warranty period on all gear purchased third party.
They already did, to an extent. The base Mac Mini used to be $600 with a 256 GB SSD, but they got rid of that and now the cheapest option is the $800 512 GB model. The typical Apple strategy is to make the base model of a given computer the price competitive version, and price gouge on additional storage and RAM to build up their margins. I guess the RAM price increase made the margins on the $600 model too low for Apple to want to sell.
I said “$600” in a comment about the Neo. You said “$600” twice here about other things.
One thing is for sure, people consider that Apple hardware starts at $600.
They are not raising prices. They are keeping them low (in marketing) to outcompete others like Microsoft (aka Valve).
Gabe Newell worked at Microsoft for 15 years - people need to start considering Valve an extension of Microsoft. Their unofficial App Store for games.
> Gabe Newell worked at Microsoft for 15 years - people need to start considering Valve an extension of Microsoft. Their unofficial App Store for games.
This is a really weird take, considering Valve's push into Linux.
It’s really a factually incorrect take. The only reason Microsoft puts its games on Steam is because they have to for their game studios to be commercially viable. They have already tried getting a foothold with the Microsoft/Xbox store and the truth of the matter is that Steam holds all the cards in their overwhelming user base.
Valve is not an extension of Microsoft, they are a company that out-competed them.
I actually find it somewhat surprising that Microsoft is as warm as they are toward Valve. If there is a personal relationship factor, maybe that’s part of it.
But I think another part of it is that Microsoft ultimately makes more money via Steam than if Steam wasn’t a part of their ecosystem.
Well Apple has no excuse for this. They literally had a quarter trillion dollars sitting, they could had bought an entire ram manufacturing company along with its entire supply chain (sand included).
The argument “the market cornered us” is ridiculous for a behemoth like Apple. They had the opportunity to end, android, pc, consoles overnight.
Tim blew this.
> they could had bought an entire ram manufacturing company along with its entire supply chain
And then what, sell it for cheap? Right, Apple is well-known for subsidising their hardware.
Apple’s services make more revenue than all Mac and iPad revenue combined, at a higher profit margin.
I could absolutely imagine Apple buying or creating a new DRAM plant, with a strategy of Apple being the primary customer and being able to undercut the rest of the consumer electronics industry.
If they had extra capacity they could sell chips at market rate as a bonus, but they’d be able to prioritize themselves as a customer.
Seems like a strategic slam dunk to me. Apple has only been getting more vertically integrated over time. With the way Apple subscription services work they have much more of an incentive to prioritize market share over hardware margins than in the past.
Just keep selling when everyone else will not be able to find a single dimm.
I have been buying an Asus laptop with Intel for our employees in the last 1.5 years, an ASUS ExpertBook P5 90NX0861-M007X0 to be precise, and it did not see even a cent of price increase.
I suspect this is because of its CPU, Intel Core Ultra 7 258V with 32GB RAM directly soldered on the CPU package, being overstock or something, or Intel secured the RAM supply upfront. I don't know but it's wild that this laptop is still early 2025 prices.
I was planning to wait until the M6 generation to replace my 16" MBP M1 Pro. But after Cook said they will have to increase prices soon, I went out and bought an M5 15" Air with 24GB/1TB.
I think the M6 Pro/Max generation will have a substantial price increase given the new design, OLED, N2 TSMC node, and memory prices not coming down.
Tim justifying selling a computer for more than 10k when that Mac Studio comes out
It’s pretty incredible to get a Mac Studio that performant for only two weeks’ pay… Well, two weeks’ pay for doing what you’d need to be doing to justify that Mac Studio, of course.
Money aside since it’s probably getting to a used car territory pricing, but completely agree how amazing the Mac Studio is if you need it and know what you are doing
Can't imagine increases to existing products. Would seem almost certain it would be for new product/refreshes and they may simply continue to remove the entry level configs in the meantime. That is all still significant for Apple for whom pricing is part of the brand. Been waiting for a maxed out new MBP coming later this year/early next, which already is about 7K, so with the prices increases, scared to think what the 8GB SSD and max ram will come in at.
>I also do not think they’re going to raise the prices of existing products mid-cycle
This surprised me too. I'd accepted that price hikes were coming for the new range...that's expected. But hiking prices on the existing range felt like a step too far!
Might have been a marketing stunt to nudge people into upgrading. Well, if that was the plan, it worked. I just caved and bought an M5 to replace my older one. Boo.
Why is that a step too far? e.g. the PS5 price has increased like 25% and it's a 6 year old product now. Similar for other consoles. It feels much more acceptable to me on something like a Mac that's less than a year old (and going to last a long time + have good resale value).
> Why is that a step too far? e.g. the PS5 price has increased like 25% and it's a 6 year old product now
Precisely because it's a 6 year old product. The $499 the PS5 cost at launch in 2020 is equivalent to ~$650 in 2026 according to the inflation calculator [1]. Within a year it's harder to justify that. Nobody believes Apple is paying the price of the day for components instead of having them negotiated at least for the whole run of a model.
> It feels much more acceptable to me on something like a Mac that's less than a year old (and going to last a long time + have good resale value).
That sounds like it should be exactly the other way around? A PS5 from 2020 is substantially identical with a PS5 from 2026 except maybe for some minor HW optimizations. They are completely fungible. A Mac from this year will compete with a faster model next year, and another even faster model the year after that.
[1] https://www.usinflationcalculator.com/
PS5's increased to compensative for the many Sony gaming flops, like the $100m Concord (closed down within a couple of weeks) and the $200m Marathon (still on life support but they had to cancel the much more popular Destiny 2 to prioritise it).
Macs have much shorter lifecycles than consoles, and Apple doesn’t have a history of reducing prices over that lifecycle. Consoles have much longer lifecycles and typically drop in price over time (as components get cheaper).
It seems fair to expect that behaviour to work in both directions.
Yes, exactly. Price hikes mid-cycle is not something I have seen Apple do historically.
If true, one reason could be that: Cook is retiring, while Ternus is on his way in. Ternus wouldn't wanna start off with his first announcement being a ginormous price hike. Makes him look bad.
So Cook becomes the fall guy. i.e., he increases the price by 15% in the existing M5 range.
When Ternus comes in, he keeps prices stable on the M6 range. Makes Ternus look like an awesome guy...and gets him off on a flying start in his CEO tenure.
They’re not dropping in price though. PS5 is 25% more expensive than at launch six years ago.
Sure, and what I'm saying is that I'm ok with them doing that. It's a matter of consistency.
The PS1, PS2, PS3, PS4, Xbox, Xbox 360, Xbox One all followed the pattern of dropping in price over the course of their lifetimes, because components got cheaper over time. The fact that the PS5 and the Xbox Series X|S have gone up in price is consistent with the general price elasticity.
I'm also OK with Apple having a rigid pricing structure and never really doing any sales or discounts, but then I expect them to not raise prices on the current M5 hardware, and leave those price hikes for the M6 generation that I assume is just around the corner.
I haven't seen other people call it but I think the ultimate force behind Moore's law is not transistors getting smaller but transistors getting cheaper. Like you can make them smaller all you want but if you can't make them cheaper you can't afford them.
In the 2020s we've blamed extrinsic causes such as the pandemic and the AI bubble but I am suspecting the intrinsic cause that new processes are more expensive to develop and more expensive to operate.
The tragedy of the PS5 is that even though it has sold a lot of units it doesn't represent the kind of generational change like the PS3 was. There are roughly 15 exclusive games and closest I come to one that I want is a remake of one of my favorite games from almost 20 years ago.
My experiences have been that Macs last far longer than consoles.
Perhaps I wasn't clear. I'm talking about the product lifecycle, not the device lifecycle.
The M1 Mac mini, the PS5 and the Xbox Series X|S were all released on November 2020.
Since then, Apple has released the M2 mini (Jan 2023) and the M4 mini (Nov 2024), and I'd be willing to bet we'll see an M6 mini later this year. The PS5 has had the habitual mid-generation "Slim" refresh and the Pro (a smallish specs bump), and Xbox Series X|S has seen barely any changes.
My M1 Max MBP is still a beast of a laptop, but it hasn't been the current model in aeons. My Xbox Series X is still current.
This is understandable given the market but part of me really wishes this would wait a year even though it won't.
Mac hardware is so close to being really useful for local LLMs and it's shared memory architecture could be a direct shot across the bow of NVidia's aggressive VRAM Market segmentation but it just can't compete with the raw FLOPS and memory bandwidth of NVidia. You can buy a Macbook Pro with an M5 Max with 128GB of RAM for $6k currently. I expect that will go up by 20-50% in the next generation.
It's safe to say that no current Apple product will get a RAM bump for the next 1-2 cycles at least.
I think this is going to impact NVidia too but in a different way. Normally in NVidia's product cycle we'd expect 50x0 Super mid-cycle refreshes. It's clear that's not happening this time around. We might expect the 6000 series late next year. I think there's zero incentive for NVidia to do that so that'll likely get delayed into 2028 or possibly 2029. 5090 prices keep going up even though it's 1.5 years old.
Anyway, as for Apple I'm keenly watching for the anticipated refresh of the Mac Studio lineup. The previous gen (M3 Ultra, M4 Max) just don't have the raw horsepower even though they had configs up to 512GB (512GB and 256GB now discontinued). It'll be interesting to see what the max config is and when these come up. Q3 2026 is widely expected but I wouldn't be surprised if it slips into 2027.
> You can buy a Macbook Pro with an M5 Max with 128GB of RAM for $6k currently. I expect that will go up by 20-50% in the next generation.
That config can be had for $5100 already: https://www.apple.com/shop/buy-mac/macbook-pro/14-inch-space...
> It's safe to say that no current Apple product will get a RAM bump for the next 1-2 cycles at least.
The Neo seems likely to.
I think we’re likely to see some kind of tiered storage (maybe still 8GB of GDDR5, but some additional memory running at slower speeds that’s volatile and then segmenting the SSD into a bigger, slower part and a smaller really fast part.
The technology all exists to do this and it’s ideal for the kind of local inference Apple wants to push.
Aren't their storage prices so inflated that they could just eat the difference?
For years with every other OEM I have bought the laptop with the minimum amount of memory and saved $800 or so buying the largest memory sticks that work with the machine from Crucial (R.I.P.) Doesn't work if the memory is soldered to the board though!
> Aren't their storage prices so inflated that they could just eat the difference?
Yes but to a point. Like Microsoft raised their Surface series prices. There's internal threshold of what they can tolerate before raising prices.
You can see the price and trend here: https://www.pcpartpicker.com/trends/price/memory/
Yet, remember that costs are more complex with currencies and shipping prices.
Hah, yes, with my cheesegrater Mac Pro 2019, Apple wanted $3,000 for 160GB of memory (to go from 32 to 192GB).
I bought it with 32 and OWC sold me the exact same sticks of memory (manufacturer, timings) for $1,050.
Same. $3,000 for 8TB of SSD. $1,200 for 4 x 2TB Samsung 990 Pros and a 4xM.2 NVMe PCIe enclosure. Which actually ran about 500MB/s faster than the Apple SSD.
[dupe] Discussion: https://news.ycombinator.com/item?id=48580466
https://news.ycombinator.com/item?id=48576901
All of these comments aged well.
Um did something happen? I'm pondering an M5 make to replace my perfectly-working-but-won't-be-able-to-get-software-for i9 MBP.
Never mind, I saw the story on the front page. wah-wah sad trombone for me.
Still getting a MacBook Neo for my teen now rather than wait.
Assuming a refreshed Macbook Neo, the second generation of Macbook Neo should have 12GB of RAM
I'm assuming they'll stick with 8GB RAM to keep from increasing the price much
if they are using the next gen iphone soc it'll have 12gb soldered on.
This was inevitable. The better question is if AI related hardware costs drop after the AI bubble implodes, will Apple drop the prices? My answer is negative.
I think we might end up in a weirder situation: Apple _does_ drop their prices back down to current levels for the same quantity of ram, but ASP goes much higher, at least for the Pro tier buyers. My reasoning is that depending on how the benchmarks look, many of us may try to go big on ram on our next hardware purchases to run models locally as a way of hedging either model costs, or to ensure access.
Yep, this tracks. Open source models seem to be getting more and more attention.
September.
[dead]
[dead]
[flagged]
[flagged]
I can't believe that anyone would actually think that Steam is vaporware.
I don't have apple-issued devices, so no money flows from me into apple, but it should be pointed out that with an increase in RAM prices, many of us already paid more in general, and apple may benefit here indirectly when they can add their own extra-cost onto devices on TOP of that increased RAM prices. So this should be considered rather than merely focus on (only) "Apple is increasing device prices". They are all milking us.
I also think now, with RAM prices increased, ALL hardware manufacturers should be considered an illegal mafia aka cartel. It can not be that they steal money that way. That is not how capitalism and free market work. This is a de-facto monopoly. States need to do something; the USA under Trump is just a corporate disguise right now. They are doing nothing about it. The EU is not much better, slow and like a behemoth focusing on "data privacy" (but then handing over all of our data to the USA anyway and on top of that mandating age sniffing soon). They don't protect consumers from exploding RAM prices.
The EU should have figured out how to open a RAM factory and make sure it stays cutting edge and relevant.
my father in law bought a samsung with gemini for 350€. at what moment shall we pay more for what has become simply a portal to the almighty AI?