Old Macs are so underrated. I gave a 2010 MacBook Pro (maxed out at the time, upgraded to an SSD eventually) I had to my friend (he didn't have a laptop except a school Chromebook with 2gb ram), and he uses it a ton. (The best part is the MacBook has more processing power than the 2023 school Chromebooks!)
With the old Macs, you can hack them, upgrade them, use them to death, because they're old and you don't care if you break them. You don't have to baby it. They're truly tools, not just devices.
Because almost nobody takes advantage of that. Even on machines they were upgradable.
Of the 4000 odd machines we’ve run through our fleet in the past few years all have had m.2 slots or sata ports and maybe a handful have had drive replacements. Under warranty.
Upgrades aren’t a thing because they were purchased with the proper spec drives in the first place.
Increases failure rate, size, power consumption, less space for battery and reduces performance potential due to electrical signal degradation through longer traces and sockets.
Increases failure rate, size, power consumption, less space for battery and reduces performance potential due to electrical signal degradation through longer traces and sockets.
Non professional consumers of Apple products would be more likely to properly spec their devices were Apple not charging ~1,000% markups on upgrades. Were the base devices better specced with these cheap commodity parts, or upgrades available at a price related to their value, it would be less of an issue than it is today.
Soft pass. Not worth engaging someone making gross generalizations during the middle of his mid-morning java regular, sending parallel output through both of his exit ports (without an available input port, as apparently his manufacturer removed it for “improved stability”).
Increases failure rate, size, power consumption, less space for battery and reduces performance potential due to electrical signal degradation through longer traces and sockets.
People built online stores selling parts for Mac and repair services. Many of them exist to this day, and Apple decided to win a competition with them. It's more profitable to do it in this way. It's a logical decision.
I hope they don't invent a power cable that you can't replace without their approval.
It has nothing to do with the ARM limitations. It's a change from buying a computer to renting it under some conditions from Apple.
In my view, Apple sells services, and the M Mac series is a small part of them.
Maybe I should create a separate topic: what is the difference between owning and renting a computer? I see many users point out problems, but I think we don't understand the source of these legal issues.
The old Power Mac / Mac Pro towers were fantastic. If you could afford a new one, you'd have not just a cool toy but also a dependable computer for many years to come. If you could only afford a used one, simply upgrade it with components like graphics card, more RAM, better HDD, etc. and it'd be much better than some brand new systems still. Apple killed an entire enthusiast community which had formed around these towers.
They weren't even horribly expensive, especially in their lower spec form. We would always buy Apple Refurbished for work and the lower specs and upgrade ourselves as needs arose.
Heck, even my old Mac Mini G4 I upgraded the RAM & Hard Drive.
Apple is in a seriously sad state of affairs in this area now.
Maybe I should create a separate topic: what is the difference between owning and renting a computer? I see many users point out problems, but I think we don't understand the source of these legal issues.
I’m taking note here. I’m noting the contrast between an extreme edge case you present (data science within the academy is on that edge) against the low-brow mockery you kept (t)rolling out earlier during this discussion.
It was not constructive to watch you double-down on mocking thread participants who raised (and re-iterated) everyday/quotidian use-cases of ongoing utility for extant hardware (of which there’s plenty still in circulation and plenty of uses which don’t require excessive, quote-unquote, “horsepower”) — to extend product lifespans beyond turnstiles of synthetic obsolescence set in place by a manufacturer.
You revealed a bit about what your background. In collegial spirit, I’ll do the same.
My (inter)disciplinary background is, in great part, focussed on ecological restoration, urban impact analysis, and environmental life cycle analysis. How this relates to consumers is in the seven-stage environmental life cycle of all materials humans use for extracting value (and how we have a famously long history of poor stewardship around satisfying the latter stages in this analysis). Neglect toward the less lucrative, less thrilling stages — like our waste — mushrooms into a mess which we, the self-aware inhabitants of the same shared rock, can no longer avoid and must deal with today and now.
So whereas you may argue for a certain edge case (which, yes, some of Apple’s current products meet your very specific, but quantitatively uncommon needs), these bear a limited relationship to the mean (or averaged) case of consumers who, as part of their consumer buying sway, are imploring hardware companies like Apple to acknowledge and repond to how proprietary components; tamping down components with needless cryptography; and downgrading priority of parts interchangeability are immanently hostile to consumers and, with growing awareness, immanently hostile to what we are bequeathing as a species — one self-aware enough to know better than how we’re behaving.
It’s not about a bunch of nerdy people in a basement doing the kind of stuff you may have done as a skint enthusiast in some basement some twenty years ago. (That was your deal.)
It’s that everyday users — with work to do, jobs to fulfil, deadlines to meet — are not pliantly aboard with the direction which one manufacturer, Apple, are taking. Theirs is one of demanding consumers, including longtime customers, to forgo the ways they get their work done because it doesn’t comport with what the company expect (and want) now from their consumers. It’s ignoring how the mean consumer wants to keep their product(s) working smoothly for a very long time; to be able to have them grow alongside their needs; and who has an emergent awareness that we can’t go on with a culture of disposability and expect, simultaneously, to leave something viable for any who survive us.
Parts interchangeability and not tamping down components with needless, proprietary cryptography are two asks by these consumers. Many are repeat consumers over decades and have the buying power to go with it, as well as the leverage to order buying decisions for workplaces they oversee — workplaces which may not be in tech fields.
These asks are not exceptional, and technology doesn’t come to a halt because it happens to be modular or based on industry standards. USB didn’t grind the industry to a halt; PCIe didn’t, either. And where would industry be without the IEEE 802.11 working group or standardized voltages for power rails?
This ask may mean selling two lines of products — a professional, premium line able to handle high-demand edge tasks in fields as yours, and a general line whose specs don’t necessitate TB/s rates for GPU calculations. Apple once did this. They haven’t for a very long time. Apple’s post-2005 marketing game of slapping “Pro” onto products is no more meaningful or substantive than a automaker slapping “Special”, “Limited”, or “Sport” onto a line of vehicles.
Absent fulfilling that, longtime consumers who’ve made valid (and I’d add, strong) cases for maintaining older hardware in continued use are, in their own ways, doing the total environmental life cycle analysis which the manufacturer won’t (though manufacturers, via their marketing communications arm, know how to talk good game: we call what they do “greenwashing”).
Humanity has long been expert around the first four stages — extraction, processing, manufacturing and distribution — and I’ll skip them.
The fifth stage is using the materials in the finished product. We’ve actually gotten worse with this over the last half-century. Extending this stage, something we should be doing anyhow, reduces pressure on the sixth stage, disposal, which too often ends up being shunted to a waste field far afield, out of sight and mind of all who generated it. We, as a species, confront this stage as we would an all-you-can-eat buffet (and it’s why we find ourselves in this planetary crisis). Easing pressure on that is needed now because the seventh stage, reclamation, is our most undeveloped, but the key to interspecies survival — including our own — on this finite, ball-shaped rock.
That’s my angle in this discussion. That’s why I look for ways to keep synthetically obsoleted equipment in purposeful usage — whether for work, research, play, or dork around on Netflix — to prolong that stage, and also to find people who vow to do the same with that which I come upon but make use of myself.
Getting co-operation from (or, in absence of co-operation, mandating through public regulaton) a hardware company to act progressively on the last stages of the environmental life cycle with the same verve they do for the first four? This is not a big ask.
My fear is Apple’s principal shareholders, the ones with the power to make nominations to the board, won’t shake themselves alert to figure this out in time. We can’t go on as we have.
You see, and now you are getting dangerously close to suggesting that we don't need high performance computing at all, and all that matters is a notion of a "computer growing with your needs" (presents by you in a matter which I consider rather limiting). Which kind of loops back to my earlier mention of people who want to make computing objectively worse because of their marginal view/needs. Yes, I do acknowledge that you propose maintaining two lines of products, where modularity is emphasized at the baseline, and I do think this is wishful thinking that won't solve anything. I will get back to this later.
I think the ecological concerns you voice are fundamentally valid and extremely important. I also think that your presentation conflates multiple separate (and often difficult to reconcile!) concerns and that your conclusions are flawed because of a partial misrepresentation of the industry, it's possibilities, as well as consumer needs. Currently I do not have the presence of the mind or the time to write a detailed, careful essay, so apologies in advance that my thoughts will be presented in a crude and incomplete way. I will try to mention a few general points, which I consider the most important.
First regards upgradeability and repairability, and how they relate to ecology, environmentalism, and consumerism. It seems very popular to present and discuss upgradeability and reparability as a single concern. I believe this view is misrepresenting the reality. The notion of modular computer is most important to the DYI crowd, and its demographic has changed a lot over the decades (we saw it with other markets historically as well, for example radio). In the early day of computing, DYI was pretty much the only way to have a computer you wanted. Now it's mostly gamers seeking to upgrade components and improve experience on a limited budget (costs are important, I will come to this later!). And this market is fundamentally toxic and anti-environmental. It is dominated by components that consume tremendous amounts of power and waste resources on a gargantuan scale, simply because it's a cheap way for the companies to make money and satisfy the demand driven not by actual need but by gamer psychology. At the same time, upgradeability offers only very limited utility to the "casual" user with moderate computing needs, simply because the resource demands have massively slowed down in the last decade. A 1GB RAM computer bought in 2008. years ago would become obsolete within 1-2 years. A 8GB RAM computer bough today will still be usable and useful over its entire projected lifespan. A popular argument is that of environmental damage caused by disposable appliances. This argument is undoubtedly valid, but what about the environmental cost of modularity with its resource usage overhead and the opportunity cost for economy and innovation? There are better ways to deal with appliances than forcing them to be modular in some narrow sense. Again, I will get back to this later.
Second, your notion of the computer that works smoothly for many years, and that can adapt with users needs over its lifespan. I believe this notion to be fundamentally misguided and mischaracterizes both the consumer needs and the industry's possibilities. You essentially advocate a stagnation model for consumer hardware. This would inevitably lead to much higher costs across the board (because the companies need to maintain profits and because they cannot efficiently amortize their R&D costs anymore). It would also have only a questionable benefit to the customer because things won't change much from the current status quo. Again, if your computing needs are very low, going from 8GB to 32GB on a modern computer won't make a noticeable difference over its entire projected lifespan. Especially under a stagnation model where the software will inevitably adapt to the stagnating hardware. There is at least one benefit for the stagnation model though — software will be able to take advantage of the fact that the hardware doesn't change to become more efficient (just as we see with gaming consoles historically). But again, what will be the opportunity cost?
Which brings me to the final point: what most customers care about is neither upgradeability nor repairability, but the total cost of ownership. They have moderate computing needs and want some assurance that they can keep using the device without the risk of flooring hefty repair bill or having to buy a new device where the old one would still be sufficient to their needs. This is often sold as "repairability". But it has nothing to do with repairability. It is about warranty and hardware support. Customers generally don't care if they get back a new computer or their old computer with parts replaced or repaired from the shop, as long as they don't have to pay extra. This is a very strong customer motivation, and there are multiple ways to satisfy it.
Sow with all this in mind, here is my vision for addressing the issues you mention:
1) Introduce strong environmental protection laws that enforce recycling, material reusability, and component level repair; give the manufacturers permission to sell refurbished or recycled components as new products
2) Hold manufacturers to stricter accountability standards (warranty) and implement universal hardware extended warranty insurance that will cover repairs of fairly hardware for a meaningful timespan (say, 10 years); this insurance would cost 3-5% of the original device price per year, with premiums being paid out to the device manufacturer; ideally this insurance would be mandatory for each consumer who keeps the device beyond the initial warranty period maintained by the law
3) Introduce a strict legal framework that regulates the matters of part replacements and authorization; authorization and decision power should always be with the device owner and not the manufacturer; however, it is the manufacturer that mandates which parts can be replaced and how the replacements should be certified (the legal framework would also regulate the certification); maybe mandate that the customer have the right to have their storage medium upgraded to any capacity offered with their model, although it would be very difficult to find satisfactory wording
4) Introduce a "heat" tax that penalizes computing devices that use excessive amount of energy to outclass the competition (to deter companies from releasing consumer hardware that consumes more power than a stove to get that 10% performance lead), introduce clear regulations how products are labeled and advertised (to disallow companies from selling processors that draw 200 watts as "energy-efficient" 60 watt products); I wouldn't mind a maximal power consumption to volume/weight regulation
Points 1-2 address the basic environmental and financial concerns associated with long-term device ownership. It would allow the device manufacturers to innovate and design freely, while rewarding designs and industrial practices that maximize repairability and efficient resource use. It would reward reliable designs — if a company makes a computer that never fails they can cash in the insurance premiums as an extra source of income. At the same time, a company who makes unreliable hardware will have to spend extra cash for repairs. It would also give the consumers financial protection from the stochastic nature of hardware failures. Point 3 would regulate the question of part replacement. Point 4 would address the environmental impact of the toxic PC intrudes that uses manipulation and customer emotions to deliver products that devastate the environment.
But of course, this is very unlikely to happen. Gamers still want their 5% faster CPUs, and people who want cheap computers don't want to pay "for other people's repairs". So nothing will change much. The industry will come up with new ways to save money (like UCIe), maybe we will have a couple of half-assed laws that regulate repair practices to a limited degree, hopefully a law that mandates that consumer PC towers use under 800 watts of power (if they don't inevitably disappear into oblivion until then), and that's it. There will be still people grumbling about how they could upgrade computers in the past and how the device as appliance mentality is ruining the world, but nobody will really listen. This whole thing is a shame, if you ask me. We could have a nice, sustainable, innovation-friendly path forward. But both the industry and the average customer is unfortunately too cheap.
You are asking them to either make Apple Silicon modular (which would compromise their their laptop lineup) or develop a special modular chip for some selected Macs (which would significantly raise costs).
In this sense, “taxing” heat or power consumption of computing hardware (best applied to the sticker price at time of purchase) hews about as logical or as effective as that U.S.-oriented tax for the automotive industry known as the “gas guzzler tax”. It’s basically a one-time pittance fee (maybe a few thousand dollars, max, atop the six, seven, or even eight-figure price of that vehicle), tacked atop the purchase price of a new vehicle whose fuel economy falls below a federally regulated baseline.
Classically, “guzzler taxes” have included the expected, high-horsepower sports cars, super cars, and top-shalf luxury cars (this tax also omits all “light trucks”, which is a gloriously ugly — and uniquely North American — loophole).
It will be worthwhile to observe over time, as even more efficient electric-only power storage systems come online for vehicles, whether a power-guzzler tax may one day apply to purpose-designed battery systems whose designed function is to dump high yields of energy for maximum horsepower and not, say, for efficiency. This would be, if so, less impactful than taxing the first-cycle-use of specialized raw materials, like cobalt, to encourage battery innovation using more conventional materials for power storage. From that example, cobalt extraction is not only incredibly damaging ecologically, but the labour ethics around that resource extraction industry are, with limited exception (i.e., what gets mined from, say, northern Ontario, in Canada), abysmal, even horrific. Think sub-Saharan children, pulling out ore from hand-dug deep pits, using their bare hands, with no labour laws or workplace safety in place to speak of. And that’s just cobalt.
Thing is, consumers of those kinds of energy-intensive products tend not to be not deterred from buying said products. There’s no reason to deduce a top-end gamer or specialist application (say, large-language modelling, modelling macro-scale processes like post-Big Bang evolution of matter into now-familiar galaxies, and so on) would be deterred, either. A “heat tax”, as a usage/consumption tax in this circumstance, is counter-intuitive. Consumption taxes work best when there is a finite resource — say, land — on which too many want to use it at the same time, and whose usage negatively contributes to conditions. This is why a congestion tax in major cities is quite intuitive (and the data continue to present their efficacy in early-adopter cities as London).
What would be something worth exploring, long-term, particularly for stationary computing gear (server racks, desktop servers, and the like), is integrating a heat pump system, whose inclusion can be used to capture and re-use that heat energy in other meaningful ways, rather than losing it to dissipation into the room. The technology around heat pumps still has a lot of tremendous innovation ahead of it.
How much are you willing to pay for that? Intel can sell CPUs for "cheap" because the R&D is shared with their server lineup. And this incidentally one of the main reasons why their CPU are large and hot.
Appel could build a really modular Mac Pro will all the bels and whistles. And it would cost over $60k to break even.
Because we see manufacturers increasing power consumption by 2-3x just to squeeze out 5-10% performance improvement in order to claim superiority over their competitors. Less than 10 years ago CPUs that drew 60 watts were classified as "desktop enthusiast performance". Today an "Ultrabook" Intel CPU draws over 120 watts in bursts.
You don't see the problem with this? I do. This promotes crappy engineering, dishonest marketing, and puts extra pressure on the power grids. It is also an obvious environmental target. Most of all, the excessive power consumption is entirely unnecessary. We should reward innovation, not bad engineering.
Nah. Even that stuff starts to top about around 1500. Anything more than that and you’d be looking to buy server psu and they aren’t gonna fit in those rainbow gaming cases anymore
I vaguely remember with one MBP I still have, a 2006 MBP purchased in 2015 I think. My first Intel Mac - but not my primary Mac. A 2008 MBP came in 2018 and performed better, but again not my main Mac.
My primary Mac was a PowerMac G5 Quad from 2017 to 2020. Before that it was a 2.3DC PowerMac G5 and before that it was a PowerMac G4 Quicksilver. Before that it was a 2001 TiBook 400 and before that it was homebuilt tower PCs from 1990 to 2001. Before that it was a Commodore 128 and a Commodore 64 and before that a TRS-80 back in 1980-1982. So you can see the length of my time with Intel Macs versus other computers.
Since May 2020 my primary Mac has been a 2009 MacPro. You see, you probably remember all this because you use MBPs as your primary Mac.
I use towers. Desktops. MacPros and PowerMac G4/G5s.
So, I'm not really sharing your experience here. And just because the fans on MBPs of this era happen to annoy you, you cannot compare that to my desktops. Not the same experience. Not everyone uses (or likes) laptops.
PS. This post typed in on a 2023 M2 13" MBP - used because it's assigned to me by my company for work. I don't use it outside of work (outside of work I use my 2009 MacPro). But I mention it because it's connected to two Aluminum 30" Apple Cinema Displays with a wired keyboard and wired Mighty Mouse for a 'desktop' experience.
If I have to use a laptop, I'm going to use it like a desktop.
I’ve owned them all, we have a similar computing history starting with C64’s, Amigas, went to PC for a while but started using G5 Macs, and purchased as soon as the Intel transition happened.
Owned Mac Pros, Mac Minis, iMacs, MBAs, MacBooks and started using MBPs mainly towards the end of the Intel era. I loved the 2013 MBP, it was great. I’ve probably owned 20 Mac’s or more for work or personal.
But as Jobs died and Ive kept going thinner and it intel not being able to get a grip on the thermals for the MacBook chips, I found a lot of issues with the intel units I owned, fans running high in Mac minis, MBAs and MBPs. Even the iMac sometimes. Last Mac Pro I had was the 2013 model and it ran well, but it did use desktop class chips and had at least a better enclosure.
I appreciated the intel era for what it allowed my Mac to do, but the units I owned with mobile processors definitely had issues with heat and fan running hard.
I’ve owned them all, we have a similar computing history starting with C64’s, Amigas, went to PC for a while but started using G5 Macs, and purchased as soon as the Intel transition happened.
Owned Mac Pros, Mac Minis, iMacs, MBAs, MacBooks and started using MBPs mainly towards the end of the Intel era. I loved the 2013 MBP, it was great. I’ve probably owned 20 Mac’s or more for work or personal.
But as Jobs died and Ive kept going thinner and it intel not being able to get a grip on the thermals for the MacBook chips, I found a lot of issues with the intel units I owned, fans running high in Mac minis, MBAs and MBPs. Even the iMac sometimes. Last Mac Pro I had was the 2013 model and it ran well, but it did use desktop class chips and had at least a better enclosure.
I appreciated the intel era for what it allowed my Mac to do, but the units I owned with mobile processors definitely had issues with heat and fan running hard.
I estimate that I am 15 years behind other Mac owners. When new Macs were Intel, I was solidly PowerPC. Now that Silicon is the new Mac, I am solidly Intel (and only four years into it).
I can get this 2009 MP of mine to run Sonoma. And I can upgrade it with more ram beyond the 32GB it already has. Since I sit in front of it for most of my day (I work from home), laptops really aren't relevant to me much anymore.
I do have a 2015 MBP and as I mentioned a M2 MBP. But those are work issued and have always been run in clamshell mode since I started WFH. I have never used either one as a personal laptop. They are the only Macs I shut off each day - because I don't want to deal with them beyond using them for work. I have my own Macs.
Right now, Apple is still allowing iMessage on versions of MacOS as low as El Cap. I'm primarily on Mojave. So, moving to Sonoma is going to give me at least ten more years with this Mac I think. And once doing modern stuff on Sonoma becomes more trouble then it's worth, that'll be my next upgrade point.
A lot of things can change in 10+ years. Right now my thought is a Linux box, but that's only because I predict Apple to be locking things down even more. But who knows where Apple will be then and who knows whatever else might be happening in the tech world then. So we'll see.
But right now I have no issues. Possibly, that is because the youngest Macs I own are 2009 and not the later years.
PS. I have been told in the past that the 2010 MacPro was one of the better and most expandable MacPros of the entire lot. I was told this because that was the Mac I was using at a job I had before the one I have now. So, at some point, despite my 2009 MP being upgraded to a 5,1 (from 4,1) I will actually be adding a 2010 MP into my mix.
PPS. There are a lot of upgrades CPU wise I can do to both. I have just not done any of that yet because what I have at the moment has been more than enough to do what I want/need. That's one thing that limited me with PowerPC that I no longer have to deal with.
A hair dryer that uses 3x more power to dry your hair 2 seconds faster? Sure, I'd ban that as well. Again, I am talking about energy consumption relative to a meaningful baseline on the same hardware. I have no problem with powerful hardware. I am talking about anti-environmental and anti-customer practices of shipping massively overclocked hardware and presetting it as a technological achievement. To make it clear: I'm fine with companies shipping a 1000 watt CPU. I am not fine if a company configures that CPU to draw 3000 watts to produce 5% more performance.
I didn’t say that regulation got Apple out of the desktop business, but focusing on power efficiency may have.
Of course it’s a business decision. It’s a slightly different conversation but it feels like apple isn’t really interested in computers any more. They are a services company that builds disposable appliances to consume their services.
