Posted on 06/19/2019 8:23:57 AM PDT by fireman15
Planned obsolescence, or built-in obsolescence, in industrial design and economics is a policy of planning or designing a product with an artificially limited useful life, so it will become obsolete (that is, unfashionable or no longer functional) after a certain period of time.
No company will admit they have this policy, but we can observe this from a products’ average failure rate and the company’s sincerity in fixing it by making available parts required to fix it. This is why I am a huge fan of Japanese products and also Apple. Products are not merely hardware, Apple on the other hand is known for updating their 4 or 5 years old iPhone with the latest OS. As for parts, it is easy to find spare parts for most Japanese brands.
(Excerpt) Read more at geckoandfly.com ...
Someone in the comments linked to a completely outdated a SquareTrade laptop repair frequency report PDF from 2009. Others noted the date and started conflating that link with the data in the article.
re: “What part of my owning and operating a cross platform computer support business for 40 years “
So, this is what accounts for your obtuseness and lack of knowing what goes on in the REST of the world. All this makes a whole lot more sense now. You don’t have widespread experience upon which to make a knowledgeable judgement.
As to points about “security” and the like, revere engineering by knowledgeable folk (those with access to JTAG, chip removal and disassembly) and other in-plant tools used for development renders some of that moot; I have not looked into what hackers are able to do to the iPhones these days.
I think you are still unaware of a LOT of what used to be standard practice with regard to consumer gear ‘back in the day’; buying specialty ICs from Apple doesn’t compromise the security either, because, those with interest in that area have access to **whole phones** which can be reverse engineered.
I don’t think you know the design side of things either; that’s where I’ve spend some of my time in industry, the last stint with Cisco and the development of new infrastructure equipment.
That is why it is always so fun to give you a bad time. I started out as an enthusiast a little more than 40 years ago. I worked for a while maintaining computers in a professional capacity. After that I helped others keep their computers and related hardware and software running... sometimes at work, sometimes as a hobby, and as a friend. From my perspective it seems that the types of failures have shifted a bit from heat related hardware failures to build quality related mechanical failures. With your business experience, you probably know more than I about that.
When we were paying an arm and a leg for those now ancient laptops it seems like the hinges didn't fail and the keyboards didn't fall apart after a year or two. HP, Lenovo, Dell, Apple and all the other manufactures know almost exactly how many times a lid can be opened and closed before a hinge will start to pull out a nut secured in cheap plastic. And they are hoping that you will blame yourself for that type of failure or if you have an extended warranty that the repairs will be more expensive to complete for the insurer than just sending the insured a new or refurbished computer made by them.
They know that a certain percentage won't make it through their initial warranty period and they will lose a little on that. But most will go a few months to a year or two past their manufacturer's warranty and then the consumer will find that it is a better choice to buy a new improved computer than try to patch together the old one. Mechanical failures are likely much easier to predict than heat related component failures and the last couple laptops that we have purchased run very cool anyway. The fan on the one that I acquired a week ago barely comes on except during very processor intensive tasks. So the manufacture’s best bet to force you to purchase a new laptop is to install hinges that they know will bind up within a year or two and pull the nuts out of the plastic. Simple, effective and something that can blamed on mishandling or abuse by the customer. It also can be blamed on pressure to produce devices as cheaply as possible.
If I didn't have a pile of fairly recent laptops with broken hinge assemblies purchased by me or given to me by friends and acquaintances I wouldn't have thought of this strategy. But this is something that seems to have started mostly in the last ten years or so. Before that if a laptop hinge assembly came apart it didn't seem to be such a pain in the rear to fix. As a retired professional, I am curious if you have had similar observations.
Thank you! I didn't go through the comments closely enough in the original article to figure out what all the posters here were referring to. I believe that since we are discussing long term failure rates, charts derived data from warranty service providers from a year or two ago are still appropriate. It does not surprise me that Apple has the best record, but I would tend to dispute the second chart in the article from 2016 showing that HP, Toshiba and Apple were tied in annualized failure rates. Toshiba maybe, but my recent experiences with HP would indicate that they have produced some real dogs in the last couple of years and the next chart of annualized failure rates will likely show something very different. If it doesn't I would think that the data has been skewed by some unknown factor.
Once upon a time HP built computers and printers like Mack and White Freightliner built semi tractors. . . Metal parts. . . Metal and Nylon gears, the only plastic was outside on the case. That was when you needed a Mack truck to move a printer amd many computers. Remember ten pound laptops? Now, everything is plastic except for a few rods and springs and even some rods and springs are getting plasticized.
Warranty repairs are a cost included in the bill of materials of every product made, and the length of time it will be covered is part of that calculation. Increase the length of the warranty, as did the EU, and the price of the product has to go up. TANSTAAFL! Someone has to pay for the parts and labor of warranty repairs. That cost is amortized over all units of each model sold and reduces the profit margin on the product when sold or increases the retail price. They calculate it to the penny and are usually quite accurate using historical experience. . . But it is always their best professional guess. They also include the guesstimated cost of consumer lawsuits. They have to.
Frankly, I saw more laptops with multiple missing key tops (why should so many keys just spontaneously pop off, especially keys that were not keys frequently used?) that made them unusable and lower cases splitting (users swear they did not drop them) than broken hinges, but also some with just floppy hinges where the friction no longer kept the screen in place. The other likely failure point was power jack ports, usually applied with only two extra mechanical solder dollops plus the actual power input solder points instead of screw connection mechanical strengtheners.
Commoditized laptops have to fail at a known rate. The business model of the industry demands it. The manufacturing/selling company requires more sales to remain in business because their margins are far too narrow to survive without repeat sales every so often, at least every two to three years. So, either the keyboards have to fail, the hinges, the logic board, or power supply has to die an economically justifiable, non-repairable death, requiring a replacement. . . Preferably with one of their own brand, but they really don’t care if it is.
Being a commodity, if their customer doesn’t buy the same brand again, some one else will jump ship from another brand to their brand just as easily when his old one died an equally justifiable, non-repairable death. It’s a game in commodity computers where brand matters less than OS, just as in commodity Android phones.
There comes a time when hardware can’t keep up with what the software wants to do.
Kinda like your mind making promises your body can’t keep. h/t Little Feat
Thank you for the very good explanation. All of us our very fortunate to have you participating in this forum. And yes every repair you mentioned has afflicted laptops that I have owned or worked on. Some of them I have refused to let die long past the point where I should have given up on them.
My recollection of heat related failures is skewed because I tended to try to get a little too much performance out of of my laptops by using them for photo and video editing at a time when it was really better to do that type of project on a desktop machine. When processors were still socketed on laptops I sometimes put in a more powerful one when the prices dropped, which could increase the demand on the cooling system. I had one Compaq laptop that I replaced the motherboard several times on. Some surface mounted components would heat up to the point where they became loose. I was sometimes able to solve this type of problem with a heat gun, but usually this was just a temporary fix.
Of course power supplies would die frequently from operating at too high a temperature for too long. And the power supply jacks still get broken frequently by people who accidentally get tangled up with the cord.
But as you say the major brands all sell so many commodity type laptops at low enough margins that they have to be very aware of the likely failure points to help calculate how much that they have to sell them for to make a profit.
One of the most maddening issue to me is when the pcb board on a nearly new hard drive fails and the actual platters are perfectly healthy, but none of the manufacturers will sell a matching board or supply the software necessary to rewrite the ROM or even just send the command to reset a corrupt NAND on a hybrid drive. If you want your data back you generally have to pay a lot of money to recover it even if it takes a technician just 5 minutes to send a reset command, or replace a couple surface mounted diodes in a protection circuit. It is almost as bad as ransomware in certain situations.
It wasn’t a throttling to force upgrades - it was a step to protect the device and battery. Personally, I would prefer a slightly slowed device that helps the battery last a bit longer before it HAS to be replaced.
The “scandal” wasn’t the throttling, the scandal was that they didn’t’ tell consumers about it.
Wrong - a quote used in the article was from 2009. Data from 2016-2017.
Yep. My Crucial 960GB SSD cost $110 on Amazon. Great value and you’re right, the price has gone down considerably. I get mine in the mail tomorrow and am psyched to get this cooking Friday evening!
I tend to get five or more years per Win laptop (Lenovo T’s and W’s), and more than that on a desktop. And even when I decide I want something else, it’s not typically because the old one quit. And I’m typically buying them used off-lease besides.
IF...you’ve never had an SSD, you’re going to like it. The 1st one I ever bought runs the OS on this computer. It’s a Samsung I purchased back in 2013. According to Crystal Disk, it has 53,447 power on hours behind it with a health status still at 100%.
Take care brother.
I have looked into it, extensively, my dismissive FRiend, and am quite expert on those fields you only think you know. My knowledge is quite broad and just having access to the whole device to “reverse” engineer it does them little good on modern SoC ICs these days. They also have to reverse engineer the code inside those locked multilayer chips. It ain’t as easy as you think or the Chinese would have been turning out knock off iPhones long ago. . . And no one has cracked the Secure Enclave yet, now going on five years.
Do you think it would be easy to reverse engineer an Apple A12X Bionic 7nm SoC with its 7 billion transistors just because you have access to one of the Apple devices with one installed and you’re a whiz at desoldering components from itty-bitty circuit boards? Do you even know how many multiple layers of circuits there are in it? The A10 had six.
Do you remember when computer chips on desktops were all socketed? An easy fix back then for a non=operative computer was the drop repair, which you didn’t want the customer to see. Pick up the computer about six inches and drop it. This had the effect of seating any loose chips in their sockets and often gave you a quick, down and dirty fix. LOL! It was still a good idea to open it and press them all down by hand, but you knew what was wrong.
I don't believe the issue that he raises has anything to do with the capacitance value of the part. He states that when the issue is seen, the voltage across the capacitor has dropped to 0.3 volts rather than the normal 1.052 volts. This is a clear indication that the capacitor in question has failed, and is displaying excessive leakage current. He mentions using a replacement that is physically larger. Though he doesn't so state, it sounds like he is using a replacement with a higher voltage rating.
It's hardly a "minor part" if its failure results in a totally inoperative unit.
Capacitor manufacturers generally recommend operating tantalum capacitors at no more that 50% of the rated voltage for optimum reliability and longevity. Apple's choice of a 2 volt rated part would be marginal from this standpoint. A 2.5 or 3 volt part would have been a better choice, albeit at a slightly higher cost.
It sounds like he is adding in a safety factor which is prudent as long as the larger part will fit.
“I have a huge pile of obsolete and broken hardware... I am going to have to get rid of because we are currently moving and downsizing. But it is a challenge to do.”
I took a bunch of older computers to a trusted shop. They wiped the hard drives and put a couple of them in their inventory.
I’m writing this on my 6+ year old Compaq laptop, which originally had Win7. It now has an Ubuntu OS and runs like a top.
I remember buying motherboards, memory chips and the CPU and having to press them all in by hand. Then you had to fill the ISA slots with the boards to control the drives, hook up the monitor, add serial ports and add a printer, if you had enough slots on the motherboard or found multifunction I/O boards you might have enough room for a sound card. And then you had to figure out how to set all the dip switches to get it all to work together.
This morning I am struggling with my new HP laptop. I purchased a cable and rubber boot to add an additional hard drive, but it seems that HP has the latest BIOS programmed to reject any hard drive that is not on the list of original HP approved parts... reminds me of their efforts to keep people from refilling their ink cartridges. I have been looking for a work around.
I recall trying to figure out why a terminating resistor on a daisy chain of SCSI drives, which was supposed to be on the last device in the daisy chain, but often did not allow the last three devices to work, and instead, the terminating resistor only allowed the full chain to work if it was placed on the fifth or fourth or third device in a six chain series. . . or it would work correctly with no terminating resister pack at all! Sometimes, rearranging the order of the SCSI devices would solve the problem, but sometimes it made it worse. Often you had to open them up and found DIP switches with built in termination enabled with a switch setting and NO DOCUMENTATION, but that was rare. Some SCSI devices had auto-termination, but they were supposed to sense additional down line SCSI devices. I never did figure that one out. It was why IT work was often an art rather than a science.
Another one was trying to figure out how to get the switches set on a video card to get a multi sync monitors switches to work with any particular computer at the correct resolution. . . With the correct connecting cable which ALSO had a pack of switches in it (all with between four and eight DIP switches). . . And all three of them had documentation that was written by techs who had never, ever taken a course in how to write and none of them agreed on terminology. AAARRGGGH!
Then there was the placement of a chip, but one leg had to be NOT inserted, bent up so it didn’t get put in the socket. . . With the instructions requiring a jumper wire being added from the leg to another socket because the designer of the motherboard couldn’t figure out how to run a trace, to fix the problem, very carefully soldering onto the leg with a heat sink attached to prevent the internal solder from disconnecting the leg from the IC inside.
I’ve had that issue. The platters were good, the circuitry was blown by a power surge that got through a metal-oxide varistor power-surge protection power strip (they are actually good to stop only one surge, then they act as only a power-strip). The data was extremely valuable for a client, and of course he had not bothered to back it up.
I bought a working hard drive that matched exactly his old one(which took some doing to find), and carefully replaced the circuit board. When I re-inserted it in the computer it booted perfectly. It resurrected all the data and I proceeded to make several backups. I still did not trust it with my rebuild with new electronics for everyday use, so we put in a brand-new, larger HD and re-installed the OS and restored everything from one of the backups. It was expensive to do, but worth it due to the value of the data.
My point is that the small number of failures of those capacitors indicates that Apple’s engineering decision is the right one. Just because he has run into a few, less than 1/100 of 1% of the total production run, with failed capacitors which he can repair at the board level (which is uneconomical for Apple to do) is meaningless.
Disclaimer: Opinions posted on Free Republic are those of the individual posters and do not necessarily represent the opinion of Free Republic or its management. All materials posted herein are protected by copyright law and the exemption for fair use of copyrighted works.