All electronics (including GPUs) have a legally required 5 year warranty in Norway. Would this reduce e-waste?

[MrBubbIes]

This law has existed for a long time in Norway and the used market is thriving there. Could this reduce e-waste if it where implemented in NA?
 

Section 26.The consumer's rights in the event of defects

If a defect exists that is not due to the consumer or circumstances related thereto, the consumer is entitled to claim compensation.

 

Any complaint must be submitted no later than two years after the consumer took possession of the item. If, when in normal use, the item or any part thereof is meant to last substantially longer (electronics), the deadline for submitting a complaint is five years.

The consumer may choose whether to demand that the seller cause the defect to be rectified or deliver a corresponding item (redelivery).

 

Legal code

[jaslion]

Oh absolutly there are a ton of devices that would be redesigned asap so as to not cost the manufacturer money whwn they break as planned

[johnno23]

Agreed 100% and I do not want to sound negative but Norway is a small market for the large corporations who will and do fight tooth an nail to prevent any form of legislation that will impact their bottom line. Just the simple right to repair has for so many years been battled in courts for as long as I can remember.

The built in die at end of warranty is the holy grail for the stockholders.

[Alex Atkin UK]

The EU has a similar law, unfortunately it doesn't really help as you'd have to sue the company to get them to do anything about it.

 

These laws are useless unless the government are willing to actually enforce them rather than leaving it up to the legal system that is too expensive or too much hassle to deal with.

[OhioYJ]

I don't think it will change much.

 

Say I'm a manufacturer of cheap POS $15 USB hub. Customer calls says it died, I live in Norway it's covered under 5 year warranty. No problem, ship it to return center in this country, and pay return shipping of X amount, we'll fix it. Customer hangs up, USB hub goes in trash.

 

Not to mention how much of everyone's personal Ewaste has little to do with warranty, and more to do with just buying the next new shiny thing? **

 

(**Note: I'm not throwing stones, I have a new computer sitting on my desk too. I do try and pass along my old hardware, to others, so it at least gets used)

[Kilrah]

Doubt so. When something fails and gets returned it'll still cost too much for the manufacturer to repair it so it'll be replaced and trashed the same way.

If replacements cost too much in the long run prices will just be increased to compensate.

[DarkSwordsman]

Agreed with Kilrah. It will reduce waste on behalf of the consumer, but the company can choose what they want to do with it.

The only way to reduce e-waste would be to have this warranty law, but also have laws about e-waste recycling that would heavily fine or tax companies, forcing them to recycle.

[johnno23]

 

16 minutes ago, Kilrah said:

When something fails and gets returned it'll still cost too much for the manufacturer to repair it so it'll be replaced and trashed the same way.

 

Had not thought of that aspect...but i do remember years ago in when living in Eindhoven Holland I had a philips TV that died on me. I went to the repair center where it was plugged into the wall the guy told me please wait and came back with a new in box replacement and my dead tv was simply dumped in a huge metal bin which I could see through the doorway....I imagine today is no different which is rather sad.

[Mark Kaine]

Of course it would, dont kid yourself most (<-- thats an important distinction) electronics are specifically designed to last 2 years, or sometimes 5 or 8... but they definitely aren't designed to last much longer  - which from manufacturers point of view makes sense - but i think 10 years should be the minimum (yeah, not a fan of throwaway electronics at all)

 

 

28 minutes ago, DarkSwordsman said:

Agreed with Kilrah. It will reduce waste on behalf of the consumer, but the company can choose what they want to do with it.

Thing is if manufacturers need to offer replacements or repairs for longer they will maybe start making stuff of higher quality, because all those replacements will cost them, i think that's the idea behind this law.

 

 

1 hour ago, Alex Atkin UK said:

The EU has a similar law, unfortunately it doesn't really help as you'd have to sue the company to get them to do anything about it

which law would that be? here its like 2 years, but already after 1 year you have to prove its a manufacturing or design fault as a consumer ‐ its super anti consumer and pro e-waste actually. 

 

I think most will honor the warranty for 2 years, but legally they dont have to, they can just claim "user error" after just 1 year, no matter how ridiculous it is - because almost no consumer will sue. 

 

 

26 minutes ago, johnno23 said:

I had a philips TV that died on me. I went to the repair center where it was plugged into the wall the guy told me please wait and came back with a new in box replacement and my dead tv was simply dumped in a huge metal bin

me on the other hand bought a cheapo Funai TV 15 years ago (cost 100 euros) which broke after a year,  went to the supermarket where i bought it and they actually had it fixed (i checked with serial number)  and guess what the TV still works! 

So from my point of view this is more a case of a shitty seller rather than the norm.

 

Sure cheaper stuff like cables wont get fixed, but honestly if its a bit more expensive and actually repairable item it usually *does* get fixed in my experience. 

 

 

[johnno23]

I suppose it really depends upon availability...philips is based in Eindoven so grabbing one off the shelf was simply quick and easy maybe ??? either way the real issue is that in todays world so much of what is in fact thrown away is due to the ever increasing demand for the newest iteration or in many cases a situation where items do appear to fail shortly after a warranty period.......not always bit to often to be coincidental but i stop now otherwsie it gets into conspiracy theory about bad evil mega corporations.

[LogicalDrm]

It would make consumers life bit easier. But there are still problems. First, it still wouldn't be global and you could still buy cheaper electronics from international stores (classical example is AliExpress or Wish). As the law only applies those stores and companies that are directly operating on the soil where laws are effective. Because of that, EU implemented another law that counters cheap electronics, but mainly effects other cheaper stuff. So previously products that cost under €22 could be imported without extra costs. After mid-2021, all items, regardless of price, would be under import regulations.

 

Second, it would raise prices. And we all know how touche Americans are with any extra to their costs, even if the end result would be for the better.

 

Third. Warranty doesn't equal quality doesn't equal less production. E-waste comes from high production numbers first and foremost. Yes, parts of it come from products being low quality and break easily or manufacturers refusing easy repairs. But still, production is the biggest reason of waste.

[Lurking]

Cheap crap usually fails within regular return period. Good stuff lasts longer than 5 years 99% of time.

 

I suspect some items now won't be sold in Norway anymore, or will be more expensive to make up for warranty cost. So the consumer doesn't really win.

 

And after some years you don't want the same thing. Would I now want to get a 4 year old smartphone? And the manufacturers would have to store old electronics to give you your 4 year old MB. That alone will create huge cost and e-waste.

 

More bureaucracy isn't the solution. What if the company goes bust? And who guarantees they provide speedy and cheap RMA shipping. Sure, you get your replacement item. But it takes 4 months and you have to have kept all original packaging and ship it in at high expense. And then we find some BS reason to not honor warranty. 

 

Edit: If you want to avoid e-waste, tell MS to allow W11 installation on PCs without TPM. There will be a huge wave of perfectly fine PCs being obsolete in 2025 (if you want to use the internet securely). I'm all for using old hardware, but most of those ideas fail due to lack of proper software support. Same with smartphones. 

 

[Rex Hite]

E-waste originates at manufacture. It is integral and necessary  to the process of consumption  It's a continuum in that utility, useful life, value, and obsolescence can be arbitrarily assigned. E-waste can't be an objectively negative idea since it is a necessary part of bringing new items to market. If we want new stuff and we want people like LTT to promote it on videos then we must not merely tolerate or excuse e-waste, we must embrace it.

 

 

[Chris_UK]

The uk and eu have similar laws, and it helps to some degree however for most European countries UKs SOGA overrides the eu legislation as it gives greater rights. 
 

basically though if an item fails before 6 years you can get it replaced by the retailer you purchased from. But only if the item is expected to have lasted longer. Ie a £10 toaster from Argos probably doesn’t qualify. 
 

but the kicker is after 6months the onus is on the individual to show it failed due to a manufacturing fault and not through general use. 
 

 

the law in Norway also has a expected to last longer clause in it.

 

 

question what would the expected life of a gpu be with fair usage and if over clocking and or over volting turn this from fair to consumer caused issue?
 

 

 

personally laws are needed to reduce e waste and even the uks and no ways rules being better than average isn’t enough as there is enough wiggle room for manufacturers and the burden gets passed onto the retailers not the manufacturers directly 

[ImorallySourcedElectrons]

On 2/5/2023 at 3:27 PM, Mark Kaine said:

Of course it would, dont kid yourself most (<-- thats an important distinction) electronics are specifically designed to last 2 years, or sometimes 5 or 8... but they definitely aren't designed to last much longer  - which from manufacturers point of view makes sense - but i think 10 years should be the minimum (yeah, not a fan of throwaway electronics at all)

Other than malicious firmware, no one sane designs electronics to last a certain amount of time. You do reliability testing to see if it'll last the expected lifetime, but that's often an unrealistic scenario, so not very representative. Basically, it's hard enough to get get circuits to work in the first place.

And other than going full blown redundancy, there aren't many ways to design things to last longer beyond not trying to scrape the last cent from the bottom of the barrel.

[Mark Kaine]

1 hour ago, ImorallySourcedElectrons said:

Other than malicious firmware, no one sane designs electronics to last a certain amount of time. You do reliability testing to see if it'll last the expected lifetime, but that's often an unrealistic scenario, so not very representative. Basically, it's hard enough to get get circuits to work in the first place.

And other than going full blown redundancy, there aren't many ways to design things to last longer beyond not trying to scrape the last cent from the bottom of the barrel.

BS

 

https://en.m.wikipedia.org/wiki/Planned_obsolescence

[mr moose]

49 minutes ago, Mark Kaine said:

BS

 

https://en.m.wikipedia.org/wiki/Planned_obsolescence

Although I do feel planned obsolescence is a thing,  its a very hard thing to do with electronic circuitry, for electronic goods it has to be done with some other part of the product,  maybe insufficient heat sinks so the GPU is stressed faster?  

 

The first thing I learnt when I studied EE (because valves where still in the text book then), was that barring a physical beak down, solid state electronics should work forever. 

[Mark Kaine]

48 minutes ago, mr moose said:

Although I do feel planned obsolescence is a thing,  its a very hard thing to do with electronic circuitry, for electronic goods it has to be done with some other part of the product,  maybe insufficient heat sinks so the GPU is stressed faster?  

i feel its the opposite tbh, like you just need not so good capacitors for example,  saves money and is basically guaranteed to just last a certain amount of time. also solder quality etc, it's very easy to estimate how long things will last based on the quality of components. 

 

 

That's why many PSUs have 7+ year warranty because they use high quality capacitors. 

with GPUs it'll be a middle thing, not the best but also not the worst , as warranty is typically 2-3 years, if it lasts longer fine,  but manufacturers wont cry if it doesn't...

 

that's besides some companies were already caught having so called "kill switches" ... basically components that are pretty much guaranteed to last only a bit longer than the warranty time.

 

48 minutes ago, mr moose said:

was that barring a physical beak down, solid state electronics should work forever. 

well, yeah? as per above example, just use cheap low quality solder there's your physical breakdown = )

 

 

btw i also learned this in school,  manufactures *will* design their products to literally have a limited lifespan,  they would put themselves out of business otherwise.

[mr moose]

4 minutes ago, Mark Kaine said:

i feel its the opposite tbh, like you just need not so good capacitors for example,  saves money and is basically guaranteed to just last a certain amount of time. also solder quality etc, it's very easy to estimate how long things will last based on the quality of components. 

 

 

That's why many PSUs have 7+ year warranty because they use high quality capacitors. 

with GPUs it'll be a middle thing, not the best but also not the worst , as warranty is typically 2-3 years, if it lasts longer fine,  but manufacturers wont cry if it doesn't...

 

that's besides some companies were already caught having so called "death switches" ... basically components that are pretty much guaranteed to last only a bit longer than the warranty time.

 

 

Like any product though, the higher quality the parts the more expensive it gets, so when you are talking about companies using lower quality parts they are also selling at a lower price bracket.  In order to make profit you have to be in  general better than your competitors but also price competitive.  Having worked in manufacturing I can honestly say that the product s we made were not designed to fail, They failed more from human error during production.

 

When it comes to soldering,  that's something they can't fuck around with, if they try to make it a bit dodgy then they introduce to many faults and end up with wasted production which costs more.  That is a best they can do or a not at all process.

 

Remember the big capacitor issue in the late 90's and early 00's,  There was a sudden spike in electronics failing due to a mass amount of dodgy caps that came out of china.  They were all from the same manufacturer,  product fail rates were all outside warranty periods and were across many different industries, most makers honored the fix,  fail rates prior and after were much lower.  This is further evidence of what I have experienced where most manufacturers make more money from producing the best quality they can because it reduces returns, complaints and bad PR.   

 

I know some companies have gone out of their way to create fail points, I really don't doubt it, but I don't think that being able to point to a handful of examples is evidence that it is a wide spread issue.  Especially in electronics were, as I said,  trying to design in a point of failure and place it after X hours is way harder and more costly than just making something good enough or the best you can for the price point.

 

 

 

[ImorallySourcedElectrons]

1 hour ago, Mark Kaine said:

BS

 

https://en.m.wikipedia.org/wiki/Planned_obsolescence

Ok, so you don't understand the difference between designing to fail after two years and planned obsolescence.

 

Planned obsolescence is cutting corners and saving costs left and right, and under its worst form intentionally preventing repairs. But I must say, this rarely crosses anyone's mind when actually building/designing products. Usually you have pretty aggressive cost targets that must be met. In practice, this means that you go for things like adhesive instead of an exquisite machined part with a nice release mechanism. In practice, electronics are already quite difficult to get working, between some nice sketched out circuit diagram and simulations, and a functional circuit, there's quite a bit of ground to cover - something managers rarely understand, and neither do you apparently. Most engineers are happy if we get a circuit to behave well over the entire temperature range and if it passes tests like 85/85 (85% relative humidity at 85°C - accelerated aging). Obsolescence is then usually achieved by management bumfuckery, like failing to budget properly for long term support, ordering some consultant in India to strip "non-necessary components" (like additional ESD protection on connector pins), or other garbage like that.

 

As to why planned failure is incredibly difficult to implement, it's quite simple: you're sourcing components from all over the world (hundreds in many instances) while often second-sourcing from unknown suppliers, have tolerances of 10% to 20% to deal with, purchasing is then going to send it off to some PCBA plant in who-the-hell-knows rural China where electricity was only introduced ten years ago, it'll then sit in a warehouse without climate control until it ends up getting shrink wrapped with zero precautions for static electricity and loaded onto a not entirely sealed shipping container, experiencing the nice sea breeze and salt spray for several weeks, after which it'll arrive in Europe or the US, where some idiots will then throw with the boxes to unload them from the shipping pallet. (This most definitely ain't based on practical experience /s) Which is to say, the fact that it already manages to work reliably for two years in the first place is already a miracle, and you then wish to intentionally implement obscure features to make the product die after two years? The only way you're going to be doing that reliably is by sticking a clock in there under some form, and you're only going to be doing that digitally if you want it to be reliable. At which point you're very much leaving a trail, which will probably get you sued out of existence.

 

So when you show up here saying that we design things to fail after two years, I have no fucking clue how to even start doing it without causing a ridiculous amount of crib deaths with the associated financial mayhem that would cause.

 

1 hour ago, mr moose said:

Although I do feel planned obsolescence is a thing,  its a very hard thing to do with electronic circuitry, for electronic goods it has to be done with some other part of the product,  maybe insufficient heat sinks so the GPU is stressed faster?  

 

The first thing I learnt when I studied EE (because valves where still in the text book then), was that barring a physical beak down, solid state electronics should work forever. 

Running things hotter on modern CMOS will mostly increase the power bill, should still last for a good couple of years if you ain't pushing it beyond 120-140°C junction temperatures. Usually bad designs will cause things like dielectric breakdown of the gate over time through various effects. But such effects are very unpredictable.

[Mark Kaine]

1 hour ago, ImorallySourcedElectrons said:

But I must say, this rarely crosses anyone's mind when actually building/designing products.

what do you think they make laws like this for then?

 

you should also inform yourself better, its very much a thing how companies operate  - the times will vary obviously, you don't want to have a car become unusable after 2 years, but a GPU or a fridge? who cares, just buy a new one, right!

[ImorallySourcedElectrons]

4 minutes ago, Mark Kaine said:

what do you think they make laws like this for then?

Because there were douchebags doing the software timer that killed devices after 2 years and 1 day, and some manufacturers do no reliability testing at all, but that's just scraping the bottom of the barrel and cutting corners once more. Not intentionally malicious.

[Kilrah]

7 minutes ago, ImorallySourcedElectrons said:

Because there were douchebags doing the software timer that killed devices after 2 years and 1 day.

And that's why things are more subtle now, can always say you chose a 1000h 85°C rated cap instead of a 5000h 105°C one "because it was needed to meet the price target" when the expected drop in lifetime was an intended "benefit" all along but is now impossible to prove.

Said douchebags won't just have gone "oh well we tried, let's do good things now", they'll simply be doing a bit more to find their ways around getting slapped again.

[Mark Kaine]

30 minutes ago, ImorallySourcedElectrons said:

Because there were douchebags doing the software timer that killed devices after 2 years and 1 day, and some manufacturers do no reliability testing at all, but that's just scraping the bottom of the barrel and cutting corners once more. Not intentionally malicious.

Thats where we disagree i guess, it can be done by "cutting corners" but it can and is done by "malicious intend" as well.

 

thats why i said you should inform yourself better, i can surely find some examples - but it's pointless because you'll dismiss them anyways i feel (look up Phillips light bulbs for example)  

 

this is a known fact that several industries do this systematically - idk how common this is in computer industry,  but using cheap solder would be a very easy way to do this for example (same as with capacitors) and saying "yeah, well its a cheap product so what do you expect" isnt the best of excuses either when we're possibly talking about several hundred bucks per item...

[ImorallySourcedElectrons]

5 hours ago, Kilrah said:

And that's why things are more subtle now, can always say you chose a 1000h 85°C rated cap instead of a 5000h 105°C one "because it was needed to meet the price target" when the expected drop in lifetime was an intended "benefit" all along but is now impossible to prove.

Said douchebags won't just have gone "oh well we tried, let's do good things now", they'll simply be doing a bit more to find their ways around getting slapped again.

I hate to break it to most folks here, but except for industrial, medical and aerospace gear no one actually does thermal simulations or looks at capacitor lifetime ratings. Count yourself lucky if they bothered to point a thermal camera at the board at any point during the development process. This is how you end up with things like thermostats that generate so much heat that they throw their own measurement off. And that's before purchasing has its way with the bill of materials, they're just going to grab the cheapest 100 µF 16 V capacitor that they can find unless if management forces them to listen to engineering. I've ran into issues like this on aerospace projects, so imagine what it's like on consumer products. And that's without getting into the fact that most of the time these ratings are complete garbage, especially on the electrolytic capacitors you're most likely referring to. Most manufacturers do not properly derate components, so when you buy a 16 V rated capacitor from them you probably shouldn't be using it above 9.6 V, and other guidelines like that. 

 

Basically, expect to pay double the BoM price if you don't like purchasing cutting corners, but it'll probably last until it goes to the trash pile due to old age.

 

5 hours ago, Mark Kaine said:

Thats where we disagree i guess, it can be done by "cutting corners" but it can and is done by "malicious intend" as well.

No, you're not getting the part where I'm saying that no one is intentionally going to make their design worse to cut the lifetime of the product, hoping that they're going to buy a new one when it fails. Management will cut corners to increase their profit margin today, they're too short sighted to look beyond half a year into the future in most instances. They're not thinking about someone rebuying a device in two years time.

 

5 hours ago, Mark Kaine said:

but using cheap solder would be a very easy way to do this for example (same as with capacitors) and saying "yeah, well its a cheap product so what do you expect" isnt the best of excuses either when we're possible talking about several hundred bucks per item...

Regarding the soldering, there's no real difference between SAC305 and SAC105 if the solder profile is adjusted accordingly (aka offset by about 6°C). Bad quality PCBA is usually due to outsourcing to questionable parties for cost cutting.

 

Basically, the gist of what I'm saying is, it ain't intentional, it's just them focussing on short term profits most of the times. Now the anti-repair stuff, that's garbage ghoul-like behaviour, no argument there.

 

Edit: fixed typo