Retina Macbook 2015 Teardown
ifixit.com
ifixit.com
>The MacBook 2015 Repairability Score: 1 out of 10 (10 is easiest to repair)
There are several reasons why this happens; the optimist says that to save space Apple had to glue and solder so much, the pessimist says it's so that Apple can charge more for repairs/replacements.
But I think both can agree that this generates more e-waste than necessary, and I wonder whether this may be an incentive for the EU to step in. They've already standardized mobile chargers - starting 2017, all mobile chargers in the EU have to be of the same standard to reduce electronic waste.
I wonder whether a similar law for PC/laptop components could work and how it would look like...
I'm not so sure. Apple have free recycling and even the most repairable equipment will end up being chucked eventually.
A similar thing happened in the auto industry. http://en.wikipedia.org/wiki/Vehicle_recycling#Policies
Data leakage via computer disks is a real and potentially very serious problem. It can happen if you send a failed drive back for replacement. It can happen if you take a computer into a PC store for repair and the guy on the help desk starts browsing through your files. It can happen if you take a retired PC for recycling without proper precautions, which you can't always take if you retired it because something failed. This is not a hypothetical risk: all of these things have happened and been publicly reported plenty of times.
If you give someone else your drive after you've stored anything sensitive on it, you're asking for trouble. If you do it as a business that works with sensitive information such as financial or health data, you are almost certainly breaching regulations in a lot of jurisdictions.
If you're saying more easily repairable laptops would stick around longer, I'm not sure that's how most consumers treat computers that break on them, especially laptops. Many times it would be "I need a new computer".
I would think all of the produced-as-cheaply-as-possible laptops would be far worse in their environmental impact, but I can't back that up.
Apple introduced non user replaceable batteries because it made sense: capacity and longevity was at a point where the average customer never replaced their battery.
(for the rest of the laptop, it's barely repairable because there's barely anything in it. a logic board and a battery.)
Not to mention that it wasn't too long ago -- still isn't, in many devices -- that the "logic board" was not a discrete piece of hardware, but a place where you could plug in discrete components to repair and/or upgrade your computer.
Yes, you can hand your device back to Apple to get it repaired. But two or three years down the road, that's gonna be more expensive than just chucking it, at best recycling a few of the materials. If it was more maintainable, your nerdy teenage relative might have a fair chance of recovering the device instead (free laptop!).
And yes, there's a tradeoff between repairability and a host of other desired attributes -- cost, compactness, etc. Presumably Apple is judging that the market demands the latter while it doesn't care about the former -- at all. It's that totality that I take issue with.
What's more, Apple, more than any other tech company, continuously shapes what the market wants: if they brought out a series a smartphones that are user-servicable (say as a feature of an entry-level Apple Phone for emerging markets), the other manufacturers would fall in line quickly. It's rare that it works the other way round, although they've done it twice now for mobile device size, first for tablets then for phones, sort of a blow to the Apple mystique.
And of course only a fool would recognise that the manufacturers interests are not at all aligned with consumer interests regarding maintainability. Planned obsolescense is what makes this market -- a perfect storm of rapid technological progress and lack of repairability. When consumers want ever sleeker and faster devices, that's something Apple engineers can deliver, again and again. But a device that will just last them a decade is a device you only sell once in a decade.
Finally, Apple is hardly unique or particularly noteable in any of this. As usual, they're just good at what they're doing. Unless there's a surprising change in market demands, I fully expect most other laptops to look like this in a couple of years, just like most phones now come with glued in batteries and most laptop displays are fused with the front glass, turning a 30 USD repair into a 600 USD one. If anything, the fact that Apple devices are well-built and only start at the middle-end going into high-end means that people will be disinclined to just chuck them and get a new one instead.
Technically, iFixit has given an orange a repairability score of 0 (https://www.ifixit.com/Teardown/Orange+Teardown/13470) so apple products don't get the worst repairability score.
That's an O, not 0:
> Orange Repairability Score: O out of 10 (10 is easiest to repair).
It's a 0 on the left-hand side of the meter, the meter is bottomed out whereas it's at the first mark for the 2015 macbook.
If your equipment are not Apple, they will still pick it up for free via Sims Recycling Solutions. [2]
[1] - https://www.apple.com/recycling/ [2] - http://oem.srsapp.com/applepoweredbysims/
Technically nothing on that page days they don't just throw all the returned iPods into a lake. Well of course they don't, but neither do we know what they actually do.
Actually, if you click through on the GPs link Apple provides information in two links with further links to PDFs and FAQs.
Trade-in would imply that you are also purchasing from Apple as well. I didn't read anything condition requiring any purchases to take part in Apple's recycling program.
You can't critique them on building a product that is hard to repair (and wow, this one looks next to impossible for your average non-technician) - but the ewaste/standard charger concerns don't appear to be valid.
Even Samsung, with the Galaxy 6, no longer includes a replaceable battery.
My guess, is that in order to increase reliability, and reduce cost, companies have come to the realization that it's most straightforward to do that by abandoning multiple "plug in" components, and just soldering the memory, cpu, flash onboard.
For better or worse, I think the Macbook is the way of the future - and I suspect that in 2-3 years, the vast majority of the laptops being sold will be just a small stick of logic-board+flash+memory in the same way the Macbook is today.
It makes me wonder how much of the "we do this because it's what our users want" is what users genuinely want, and how much of it is actually the company using marketing to tell users what they want.
Of course, users lie, and although they say they want a replaceable battery, almost nobody actually replaces the battery. You wind up making a huge design sacrifice to implement a user-swappable battery that nobody swaps.
Here is a video featuring Jony Ive about the unibody design: https://www.youtube.com/watch?v=lJx6cF-H__I
Specific explanation regarding the unremovable battery: https://www.youtube.com/watch?v=PD6sh9purb0
Original page about the unibody design: https://web.archive.org/web/20100414070055/http://www.apple....
You'd wonder if it's really that difficult to add a strap, instead of glue to secure a battery.
For instance, if the MacBook had a standard SSD, it would need the following:
- A SATA controller chipset on the motherboard
- The physical connector on the motherboard
- The physical connector on the SSD
- A SSD controller designed for SATA
- Maybe even a plastic enclosure on some laptops
Instead, Apple can use a lower-level interface, embed the flash controller and directly put the NAND cells on the motherboard. That's two chips.
But that is not the essence of the term. To avoid ewaste, you should avoid creating any in the first place. Creating products that will be useful for 10+ years instead of until the next model cycle.
Well that's just a matter of opinion, isn't it? Nothing's going to stop you from running this laptop 10 years from now just like nothing's going to stop you from running a PC laptop 10 years from now. The question is _are you going to want to_ continue using a 10 year old laptop? That's an eternity in technological terms.
Sure you can upgrade a PC laptop to squeeze a longer life out of it, but there are practical limits to that to (i.e., it's unlikely that a 10 year old chipset is going to support the amount of RAM that we would consider "bare minimum" in 10 years, likewise for the CPU).
I agree with your argument in principle but "10+ years lifetime" is not really a good point to argue.
Except for the parts that fail and can't be easily repaired...?
Incidentally, I know many people who are still using machines in the 6-8-year-old range, and it fits their use cases perfectly well. We're long past the days of multiple-times improvements in efficiency with each new generation of CPU, the improvements are much smaller now.
If the original PC from 1981 could work after 30 years with only a bit of repair...
http://www.pcworld.com/article/237878/can_you_do_real_work_w...
...do you think you could do the same with this Macbook in 2045? For what tasks would it still be useful if you could?
An interesting related article on this: http://www.bunniestudios.com/blog/?page_id=1927
And there aren't parts like that on a more open PC? I mean, if your USB port or display stops working, it's going to be a bit of an undertaking fixing it, glue or no glue.
> Incidentally, I know many people who are still using machines in the 6-8-year-old range, and it fits their use cases perfectly well. We're long past the days of multiple-times improvements in efficiency with each new generation of CPU, the improvements are much smaller now.
Indeed, I still have a very old laptop kicking around. But it's not my daily machine or even free from dust for that matter. I think there's a bit of masochism involved in using a machine WAAAY past its expiration date (6 years isn't, but 10 certainly is) when it's likely that a $299 Walmart special will smoke it.
> ...do you think you could do the same with this Macbook in 2045? For what tasks would it still be useful if you could?
The repairing part would be much more difficult, yes. But unless you're interested in antique computing, are you seriously going to make a go at using this thing on a day to day basis in 30 years? I really don't think that's the standard by which you should judge this thing.
And for the record, this thing would be much more equipped to handle computing 30 years from now than a machine from 30 years ago is at handling today's computing. But that's not saying much.
Depends how you define "more open".
With a home-built desktop, if the USB port goes, you can either get a splitter for one of your other 8 USB ports, or a PCI card with USB ports on it. You can get a new graphics card or monitor.
An extensible laptop will be easy enough to fix for most failures excepting, as you say, display (as long as not ALL your USB ports go out, you can still get a splitter).
And then there are apple-wannabes who glue everything together and charge a kidney for repairs. Such is life...
> But unless you're interested in antique computing, are you seriously going to make a go at using this thing on a day to day basis in 30 years?
Who knows? George R.R. Martin keeps an ancient DOS warhorse around so he can use Wordstar.[0] People have different needs.
I've changed the display on my Dell XPS 15 laptop and it was very easy, since there's no glue. Open the case (~8 screws), pull out keyboard, unscrew the entire "monitor" part (1 screw), pull out the cable between monitor and motherboard, pry open the "monitor" with a guitar plectrum or a screwdriver to get to the glass, push in the new glass, perform above steps in reverse.
I've never done that before and with YouTube's help to find out which screws to leave in and which to take out the whole operation took about 30 min. No glue!
You present this as a normative argument, that things should be a certain way. But do you have a cogent argument for why we should be retarding the growth of memory and speed with our personal computing devices?
That might have been true 10 years ago, when typical speeds and storage capacities were increasing dramatically from one generation to the next and those increases were actually useful.
But even entry-level PCs have been able to run typical home or office applications -- things like reading e-mail, browsing the web, writing a letter or putting together a spreadsheet -- for a very long time. You don't (or at least shouldn't) need an i7 hexacore processor overclocked to 4+GHz and 16GB of RAM to write a letter. The only sharp improvements in recent years have been high-resolution displays (much more important on smaller screens, but generally a noticeable benefit) and SSDs (particularly in much faster boot/load times on otherwise modestly specified machines), and for applications that need them, GPUs.
Even for more demanding applications that really do benefit from greater RAM and processing power, the real world specs for both processors and typical system RAM have barely moved the needle for probably 5 years now, with the emphasis more on low power consumption and more cores within a single CPU. The cost/benefit for upgrading even a high-end workstation from a few years ago to the state of the art today is rarely favourable.
The biggest problems with longevity today, in my experience, are monolithic hardware where components can't be swapped out on failure and, more than anything, software. The reason my third-generation iPad isn't as useful any more has nothing to do with the hardware spec, it's because Apple chose to write more recent versions of iOS in a certain way and then forced the app ecosystem to upgrade because they have absolute control via the App Store. Now we have a choice between not using a lot of newer apps or upgrading to a new version of iOS (which is widely reported not to work reliably on our equipment, and with no way to downgrade again if those reports prove to be accurate in our case). All of this is entirely due to artificial problems caused by a throwaway software culture, and nothing to do with the hardware.
The companies involved have researched the market and the needs of the market and come out with what we have now, trying to argue the end result is wrong would need data to show otherwise.
From an engineering stand point, integrated/monolithic hardware fails less, there's a whole class of failure scenarios removed (ie connectors and interconnects, better sealing, etc).
Hardware has reached the point where it's on the same level with every-day things people throw away and buy new versions of without even thinking, like kettles or toasters, it's fine to argue that's wasteful but that doesn't stop most of the world from doing so for practical reasons.
No, technology companies are researching how to make more money. That frequently means making changes that are not in consumers' interests, in particular when it comes to either closed ecosystems and lock-in or when it comes to built-in obsolescence.
See also: "Smart" TVs, the history of pricing in the printer industry, Anything-as-a-Service, multi-year contracts in the mobile phone industry that obscure the true cost of these devices, DRM, and countless other trends in technology.
From an engineering stand point, integrated/monolithic hardware fails less
In some cases, yes. But when it does fail, you lose the entire device. It is far from clear that any benefit in reliability is a net win relative to any loss from making all failures catastrophic when they do happen.
Hardware has reached the point where it's on the same level with every-day things people throw away and buy new versions of without even thinking, like kettles or toasters
Not even close. That's just what people who make a lot of money selling new devices that don't actually offer much benefit over "old" devices want to convince everyone to believe with their marketing.
The current wasteful attitude is not sustainable, either economically or environmentally. We've only had this "two years and you throw away your $500 iThing and get a new one" kind of mentality quite recently, and most people have been through these kinds of change no more than once or perhaps twice so far.
Just like no-one doing corporate IT assumes you'll throw away a PC every three years any more, the market will learn, and probably very quickly. I already know plenty of people who have had exactly one generation of Apple gear and then become frustrated with the short lifecycle. I know almost no-one going the other way. That's just anecdotal of course, but looking at market share trends and general commentary over time, it looks like the experience and sentiment within my network is not unusual.
I've had my current Apple laptop for three years, and see no need to replace it any time soon. The Apple laptop I had before that lasted me five years, and a relative is still using it (so, eight years). The HP laptop I had before that lasted about a year and a half before literally beginning to fall apart.
If anything, I would have said that Apple, along with Lenovo, are about the _best_ for longevity in this space.
Other than that, 8GB RAM and a 120GB SSD gave it the much needed boost (5 years ago...).
Which actually proves the GP point that making devices not easily repairable and upgradable does generate more e-waste
My iPad is also 3 years old and I will probably not update it until 5 years or so. Even then I will just give it to a family member so it will still have a useful life after me.
>Creating products that will be useful for 10+ years instead of until the next model cycle.
Which products does apple sell that only last this long? Most of my old phones have gone to either family members, craigslist/ebay or resellers like Gazelle.
> Which products does apple sell that only last this long?
My own experience with iPhones has been that as new software comes out, they get slower and less stable. My MBP, as mentioned above doesn't give me great hope for its future (for reference I have a Thinkpad from 2008 that's still trucking along nicely with Linux, and when I do boot into Windows it performs nicely). I don't think that Apple is any worse than competitors in the same price range, but that doesn't mean they all shouldn't strive to be better.
You could simply sell batteries already glued to the cover, the rest is just screws and connectors. Might be a bit more expensive, but it shouldn't be too hard to do the replacement then.
How so?
https://www.apple.com/recycling/
The components are separated for recycling.
Where is that stated? I read these pages and all I could find was Apple saying that Apple requires its external recycling companies to respect local laws, i.e. they are shipping 20ft containers to East Asia and are done with it.
I think it's a combination of optimising for space and cost - every connector that needs to be aligned and connected, and every screw that needs to be installed, requires time and labour. It's far cheaper and faster to glue and solder everything together. This is known as DFM in the industry: http://en.wikipedia.org/wiki/Design_for_manufacturability
One of the more obvious examples of this in laptops is the move from keyboards with shaped keys to flat "island-style" ones; the latter are cheaper to produce.
And yet the keyboard alone uses both glue and about 90 screws (https://d3nevzfk7ii3be.cloudfront.net/igi/YYpaGaNumhQuR5M6)
I don't think there was a similar class about product life cycle available.
(Which doesn't mean that it will be ignored, but I think we are sort of in a place right now where cost control sees too much emphasis while value is overlooked. And I think there are lots of decisions that could be made to favor value without impacting costs, which would be a good reason to include value maximization in design practice.)
I doubt micro USB will cut it in 2017, but I also doubt everyone will be using Lightning. So I wonder if USB Type C will end up being that single standard - it'd explain why Apple's keen to try it out in a new product line already.
And yet this is the first criticism of the macbook 2015 by iFixit. In their opinion, moving to USB-C is creating waste since you cannot reuse your magsafe adaptor with this new macbook. Then they state moving to USB-C will cause more accidental damage than using magsafe, thus creating more waste.
It just seems dumb to me not to be cognizant of the fact that batteries tend to kick it after 3-5 years. Same thing happened to my ThinkPad, but the battery was easily replaceable.
Typical for Apple? The iPhone 6 gets a score of 7; the Galaxy S6 and HTC One M9 get a score of 3 and 2 respectively. The HTC One M7 got a score of 1 (they considered it impossible to open without damaging the case).
https://www.ifixit.com/smartphone-repairability
Their tablets are bad, looks like all of them got a 2: https://www.ifixit.com/tablet-repairability
There is no simple overview for laptops/PCs so I randomly picked some:
>iMac 27" Retina 5K >5/10
https://www.ifixit.com/Teardown/iMac+Intel+27-Inch+Retina+5K...
(a slightly different version got the same score: https://www.ifixit.com/Teardown/iMac+Intel+27-Inch+EMC+2639+... )
>MacBook Retina 2015 15" (not OP's 13") >1/10
https://www.ifixit.com/Teardown/Retina+Macbook+2015+Teardown...
>iMac 21.5" EMC 2638 >2/10
https://www.ifixit.com/Teardown/iMac+Intel+21.5-Inch+EMC+263...
>MacBook Pro 13" Retina late 2012 >2/10
https://www.ifixit.com/Teardown/MacBook+Pro+13-Inch+Retina+D...
>MacBook Air 13" Mid 2011 >4/10
https://www.ifixit.com/Teardown/MacBook+Air+13-Inch+Mid+2011...
So I change my above statement: Their current mobile phones and their desktop machines are alright to good in repairability scores, tablets and laptops are bad
We don't see people complaining about the repairability of their R-Pis, Chromecasts, SoC systems etc. That's the way hardware is going - because that's what we've asked for.
A big part of the annoyance with Apple is their tendency to explicitly take steps against repairs: insane amounts of glue, multiple "weird" screws (the macbook here uses pentalobe, torx, philips and tri-wing in a single machine), non-standard formats & connectors (I don't expect them to ever swap their non-soldered SSDs for M.2, even though they've altered the connector and format multiple times).
Having the machine be "hard to repair" because it's highly integrated is one thing, taking decisions which deliberately hinder repairability and have basically no other advantage is quite another.
> We don't see people complaining about the repairability of their R-Pis, Chromecasts, SoC systems etc.
Nobody complains about the repairability of an RPi because no specific step to make it harder to repair for no value were undertaken.
This thing is underpowered and overpriced, right now, just as the Macbook Air was when it launched. And it's being ridiculed to some extent for those reasons, just as the Macbook Air was when it launched. I think it will be pretty nice in another couple cycles.
[1] http://www.anandtech.com/show/9173/more-pictures-of-2015-mac...
(Usually bigger silicon has more pins, and vice versa, but not necessarily)
But this is not the case if they stack the flash chips with the controller chips. Then only a few external pins are needed for communication and power. But the flash chips actually need the silicon area for all those tiny flash cells.
Read more here: http://newyork.digitalrighttorepair.org/
Does anyone have an idea of what this is used for?
[1]: https://www.ifixit.com/Teardown/Retina+Macbook+2015+Teardown...
https://news.ycombinator.com/item?id=9384791
Something is badly broken here, and it's not just the fact that you can't repair the new MacBook. :-(
We changed the URL on this submission from http://arstechnica.com/apple/2015/04/surprise-ifixit-says-yo... to the article it points to, as we normally do.
Here's a constructive suggestion on how to fix the problem:
1. As long as the article is not on the front page, give the Nth upvoter K/N karma points, where K is the number of karma points the article eventually gets.
2. When an article not on the front page gets an upvote, put it back at the top of the New page (i.e. sort the New page based on most recent upvote, not submission time).
This would provide an incentive for people to browse the new page and be early upvoters of quality articles. In order to prevent people from simply upvoting everything, you would need to:
3. Charge one karma point to upvote an article
and in order to start everyone on a level playing field
4. Give everyone 100 free karma points (or allow karma to go down to -100 before users lose their upvoting privileges, at which point they can only redeem themselves by submissions)
and to prevent submission spam:
5. Limit users with low karma to 1 submission per day or something like that
I'll even offer to implement this for you if you think you might make this change.
That's progress. Now, Apple, please replace each LED with an infra-red or a visible-light sensor, so that someone can write software to put a representation of the keyboard onto the screen that shows which keys have a finger hovering over them, so that the user never has to look at the keyboard except when the keyboard needs cleaning.
The user should learn to touch-type.
You just reach over and press the key and you never press an adjacent key by mistake?
And, as another commenter also mentioned, I don't use the layout that's printed on the keys anyway, so looking at the keyboard would not help.
None of this is particularly uncommon for anyone who can touch-type.
You can fix it, it just takes a few weeks, just cover your hands with a cloth or other barrier so looking down won't help you reposition, and fix any typos using touch rather than looking. If you're in a real rush do the same thing, but now spin up a copy of Typing Of the Dead or similar typing tutor.
PS - Also, yes, your speed will decrease initially. It slowly returns and will eventually overtake your looking speed.
Once you learn to touch type you no longer care about what is printed on the keys or that there is a light to light up the keyboard because it is no longer important.