alt.hn

7/31/2026 at 11:34:24 AM

Increasing the lifespan of a bulb makes it worse in every other way

 https://maurycyz.com/misc/tungsten/

by tonyg

8/1/2026 at 2:00:30 PM

> "Therefore, this was done so that people would always need to buy more light bulbs."

This is wrong, but it's the type of wrong that cites its sources and hides in the part you'd never think to fact check

Just as a general point: I haven't seen a product decision that wasn't sold with some official explanation why it's "really for the good of everyone". Google's Manifest v3 was about reducing performance bottlenecks, FIFA's hydration breaks were about player health and relentless tracking is about "being able to provide better experiences".

The question is if those explanations hold up to scrutiny or turn out to be bullshit, not whether or not they exist in the first place.

by xg15

8/1/2026 at 3:56:18 PM

Well corporations tend to push their worldview on the employees. I remember guy from Google who decided to defend their device attestation, they planned to put into Chrome (likely they still plan). His whole reasoning line was: "obviously advertisers have right to be sure real eyeballs are seeing their ads, so Google doing just that with some regard for the users are actually good guys" (because they could rip us from the privacy entirely, but they didn't)

by dzikimarian

8/1/2026 at 4:03:03 PM

Amazed they actually argued with the interests of advertisers so openly and not something like "well it could help crack down on insecure or hacked devices" or something like that.

And yeah, there seems to be a baseline assumption in big tech companies that it's no problem if they get unlimited control and surveillance abilities, because they're the good guys and would never abuse their privileges...

by xg15

8/3/2026 at 11:28:27 AM

I think it's because it was just engineer responsible for the thing, not someone pr-trained. Hopefully it was sobering experience for him.

by dzikimarian

8/1/2026 at 2:27:06 PM

The macro point is that complex systems are complex and have competing priorities, so it’s facile to say “I don’t agree with the decision, therefore any reasoning that supported the decision must be 100% intentional malicious bullshit.”

I’m old, I’ve had a varied career, I’ve been in small group meetings in startups and VCs and megacorps. I have never once heard the kind of “obviously this is an evil decision, but here’s how we’ll spin it” conspiracy that people so enjoy imagining. It just doesn’t work that way.

by brookst

8/1/2026 at 3:47:01 PM

In every company I have worked so far, for the very large majority of decisions I've seen everyone try to act very ethically and morally. However, every once in a while, a temptation to make a greedy decision comes up, and what I see during the discussions of these kinds of decisions, where everyone already knows they want to make the choice that would be harmful to the customer, is that the usual detailed discussion of pros and cons that normally happens is reduced, and everyone implicitly somehow knows to focus on only talking about just the pros, both for the company and the customer. Sometimes cons for the company are mentioned, and it feels very performative, as if just to give the appearance of balance. At times when I or some other individual speaks up about the obvious harm to the customer, usually ignorance is feigned, people pretend not to understand the harm, or they counter with all the "pro-customer" points that have already been said so far.

It's never an overt thing like how you portrayed it, but it happens every single time when the perceived payoff is too enticing to pass up.

by bananabiscuit

8/2/2026 at 10:20:31 AM

There are a few other paths to these types of decisions that also don’t require overt malice.

The most prevalent, almost universal, is that macro strategy is set without consideration of what it really takes to accomplish the goals. So the discussions are framed by resource constraints and decisions are rationalized by those constraints.

Likewise, group discussions at high levels are almost always abstracted into corporate-speak that removes clarity to the point where nobody can even decipher the original intent. Honestly, even when thinking about decisions or talking 1 on 1, it’s easy to mask real consequences with obtuse language. It’s a coping mechanism for many.

It’s been extremely rare that I’ve ever heard anyone say anything that could be directly considered condoning unethical behavior. Establishing plausible deniability is an involuntary reflex for lots of people.

Generally the decisions made on the ground that seem like greed or malice are more often made as part of some survival strategy.

It’s all quite messy and disheartening, but inevitably human.

by cobzilla

8/3/2026 at 3:36:57 PM

I think every company does what they think will make the execs the most money.

Look at how many companies complied with the 'trump shakedown'.

Ethics are great, until it messes with the bottom line.

It only takes a little thought to come up with a reason why it is really 'the right thing to do, for the customers'. (customers need a good AI. so, in order to support that need, we are going to re-align our staffing to free up some cash. For the customers, of course)

by budman1

8/1/2026 at 10:52:51 PM

People rarely state out loud their nefarious intents when they can easily try and justify it under a completely different framing. Most people aren't dumb enough to state out loud their intent to do what the majority of people find abhorrent even if it is exactly what they intent to do.

by AngryData

8/1/2026 at 10:54:07 PM

I mean if your argument rests on reading the minds of people you haven’t even met, I’m not super convinced.

by brookst

8/2/2026 at 4:38:23 AM

You don't have to read minds, just look at what corporations actually do in the real world. It doesn't matter if the people in the company spend 90% of their time talking or thinking about doing good things if the remaining 10% of shitty things end up being implemented far more often. Defusing the responsibility of decisions away from individuals, both psychologically and legally, enables large companies to do shitty things in the name of profit.

by AngryData

8/1/2026 at 7:18:13 PM

People lie. It's what they do. It's childish to assume decisions that magically align with their interests specifically to the detriment of everyone else aren't actually motivated by their interests.

by DangitBobby

8/1/2026 at 2:29:39 PM

I mean, we have evidence of things that have been said e.g. in Facebook's board meetings. So those things do exist.

by xg15

8/1/2026 at 2:51:33 PM

Facebook is also kind of an outlier IMO.

There will be a scandal where Facebook is caught doing something, and then you get a bunch of people talking about "tech companies", or "Facebook and Google", and then you read the details ... and it was Facebook, with everyone going "well, the rest probably are too".

by saalweachter

8/1/2026 at 3:37:48 PM

Outlier in candor.

by j-bos

8/2/2026 at 5:42:03 AM

That’s why Wikipedia has an Early Life section.

by turgidpost

8/2/2026 at 10:17:11 AM

This seems unrelated?

by AlexeyBelov

8/2/2026 at 10:18:57 PM

[dead]

by sieabahlpark

8/1/2026 at 12:14:38 AM

While the author mentions that they are not trying to defend the Phoebus cartel, I think they may inadvertently be somewhat misleading in their explanation. What they are saying is, from an engineering standpoint, true. But my understanding of the argument against the Phoebus cartel is that it was using legitimate engineering points as the official justification for profiteering business collusion: that the points were real, but were not the actual reasons for their decisions, or reasonable justification for the particular choices of values they enforced.

In doing so, the article perhaps obscures the point that business collusion, I expect, is more often built around legitimate reasoning taken to illegitimate extents and conclusions, than around outright fabrications.

by cge

8/1/2026 at 1:03:26 PM

I more so view it as standardisation of product. A product where there was not that much to change with. Your 60W lightbulb from any maker would produce similar amount light, consume same power and thus have similar lifespan.

Maybe they could have optimised for less light and longer lifespan with lower efficiency. But agreeing to standard in itself is not bad.

by Ekaros

8/1/2026 at 8:50:07 AM

> In doing so, the article perhaps obscures the point that business collusion, I expect, is more often built around legitimate reasoning taken to illegitimate extents and conclusions, than around outright fabrications.

It also doesn't quite come around to the idea of something you might call "convenient" obsolescence. Not planned, not actively colluded to happen (the Phoebus cartel did do this, but are one of few examples), but very convenient for the manufacturer when optimizing for other factors (mostly cost) instead of lifespan.

by Aerolfos

8/1/2026 at 4:50:33 AM

Yeah, moreover, the argument is a particular kind of apologism for one or another kind of profiteering. Basically, "the people criticizing this (profiteering) don't understand the full details of how business (give some business details)". But then, actually, even with all the details, the alleged profiteering is actually happening and naturally people running these business know lots of details but they like money too.

by joe_the_user

8/1/2026 at 2:38:15 PM

I really dislike the rhetorical trick of framing any opposing argument as “apologist”. It’s the new “shill”; an attempt to attack motives rather than argument.

by brookst

8/1/2026 at 3:45:08 PM

What is apologism then?

by inigyou

8/2/2026 at 2:52:06 PM

The point isn't that it sometimes isn't apologism, but that it is used even when it isn't apologism

by GigaDingus

8/1/2026 at 11:46:08 AM

>is more often built around legitimate reasoning taken to illegitimate extents and conclusions, than around outright fabrications.

I don't disagree with you.

Whatever standard we choose I want it applied evenly to all subjects.

A solid 60% of the "policy advocacy" type HN comments fall into this category.

by cucumber3732842

8/1/2026 at 12:35:11 PM

The article is complete and utter nonsense, probably intentionally so because it's the type that will drive engagement. The reason planned obsolescence is insidiously effective is precisely because the typical consumer does not take reasonable account of product longevity. You can easily say 'yeah I like how bright [or whatever else] this bulb is' but lifespans are scarcely taken into account. If they were then massive swathes of the market wouldn't exist - a typical example being the $20 coffee makers which you'll end up paying way more for than a decent one due to the fact they just constantly break.

And all cartels are obviously built around maximizing income, so I don't understand what 'illegitimate reasoning' would be. The same thing is still certainly happening today, but without the paper trail. A clear example of it is in the video game console industry. All the way up until the PS3 era of consoles, both Microsoft and Sony were heavily subsidizing their hardware. The video card in the original XBox was worth substantially more than the retail price of the system itself.

Then in the PS4 era they both stopped, at the same time and started releasing consoles in mid-range retail hardware with minimal subsidization. That makes 0 sense from a competitive point of view. If your main competition stops subsidizing their product, that's the exact time you subsidize harder than ever and claim the market for yourself. But it makes perfect sense if a couple of c-levels happened to have a chat and casually mentioned that ending the subsidization battles would end up being vastly more profitable for both of them. 'Yeah, I agree.'

by somenameforme

8/1/2026 at 12:43:51 PM

Regarding the PS3 specifically, that was the breaking point for subsidizing because it ended up being famously and heavily used for compute clusters. We see the same effects today where any chip useful for high end gaming is also useful for AI.

by mathgeek

8/1/2026 at 1:41:55 PM

While true I expect this likely paid for itself many times over in indirect advertising. People using PS3's as supercomputing clusters was phenomenal advertising for the PS3 that was a generally mediocre system for gaming. It seems that Sony was betting everything on dev's bending over backwards to maximize the system's highly unique hardware and leaving ports to other systems working as second rate citizens, but instead the exact opposite happened. The PS3 did poorly, but I think it would have done much more poorly if not for things like the compute cluster 'marketing.'

Their mistake wasn't the subsidization but launching with extremely esoteric hardware.

by somenameforme

8/1/2026 at 3:55:19 PM

I dunno. At the end of the gen Sony had lost a billion dollars on the PS3 last I read about it.

by mathgeek

8/1/2026 at 4:19:12 PM

They lost money on the console itself, but were already profitable after accounting for software, accessories, and so on by 2008. [1] The console was released in late 2006 and was a beast of subsidization in part because of the heated blu-ray/hd-dvd war which now seems quite quaint. The modern business models are just really taking the consumer for a ride.

If either company maintained the previous business model, they'd have made the competitor look like a clown. It was basically a prisoner's dilemma, but they both chose to collude. It's hard to explain that without assuming that they're colluding, which is basically a tautology.

[1] - https://www.gamespot.com/articles/hirai-claims-ps3-platform-...

by somenameforme

8/1/2026 at 6:27:08 PM

AFAIK the profitable quarters your link refers to were after they made all the hardware changes that rolled back the decisions we’re talking about. Wikipedia has many source about it from the main article and highlights the one you linked.

As an aside, I don’t really care about the collusion claim.

by mathgeek

8/3/2026 at 2:11:27 AM

But the article isn’t saying that no form of planned obsolescence exists. If you think it’s important to make people aware of genuine planned obsolescence, then use examples of genuine planned obsolescence. Using incorrect or highly misleading examples is doing the cause a big disfavour.

You say the article is nonsense but all your points are about entirely different products/markets. So what exactly was nonsense about the article. It’s factually correct, and it’s raising awareness of something important: that we shouldn’t trust narratives just because they’re compelling and useful to make people engaged in a cause (the fight against planned obsolescence).

The kind of standardisation we got with light bulbs is exactly the thing we want: the avoid companies trying to sell you a bad product (one that wastes electricity and dumps heat in your home whether you want it or not), so they can use some less important metric (buy this! It lasts twice as long the competitors!) to sell you stuff. Basically a scam.

We don’t want to throw the baby out of the bathwater. Companies should be allowed to cooperate and set standard (NACS is a recent example) that is mutually beneficial to everyone. Then we create regulations to address egregious cases of planned obsolescence. Like EUs recent regulations for replaceable batteries - though as we now see it’s a double edged sword. Some products were made worse in a mildly significant way. That’s OK though as long as we decided that’s what we want through a democratic process.

by audunw

8/3/2026 at 4:37:38 AM

The article is literally lying for clickbait. You can read a real article with actual research, quotes from the involved parties, and more here. [1] You can also read the government lawsuits with similar investigations. [2] The cartel's sole and only motivation was profit. Getting the bulbs down to 1000 hours was driven solely by a desire to increase repeated purchases - planned obsolescence.

Most criminal organizations give themselves a moralistic argument, probably in part to ease cognitive dissonance as I think most people, including those doing awful things, are inherently moral. For instance many of the brutal drug lords during the hey day of Columbia's cocaine trafficking would spend significant amounts of money supporting education, religion, and other forms of local social services in various poorer areas in the country. The same is true of Mexican drug cartels in modern times.

So it is true that one of the easiest ways to reduce the life of a bulb is to have it burn brighter, but that's largely incidental. It's akin to how Apple actively downgrades performance on older iPhones under the pretext of enabling them to operate longer with the reduced battery life. In reality it makes the phones miserable to use, far moreso than a reduced battery life, and drives users to go purchase new phones. Exact same business. The pretext is just a pretext.

[1] - https://spectrum.ieee.org/the-great-lightbulb-conspiracy

[2] - https://law.justia.com/cases/federal/district-courts/FSupp/8...

by somenameforme

8/1/2026 at 1:13:17 PM

If you put numbers on a package, the consumer will choose the bigger numbers. Taking some statistic out of context like "the typical consumer does not take reasonable account of product longevity" doesn't make sense when the only two things a lightbulb can do better are 'be brighter' and 'last longer', and both are prominently displayed on the package. The typical consumer does not pick coffee makers based on longevity because the coffee makers don't put longevity on the package, they put the number of feature buttons, and then the consumer chooses based on that instead of inferring or researching a longevity, and the actual takeaway is that consumers are bad at inference. And on this you call the article's extensive description of how lightbulbs work and how lifespan trades off with efficacy 'complete and utter nonsense'?

The video game system analysis is junk too. No, Sony and Microsoft are not in a cartel, you can tell by how they're in vicious competition with each other and Sony's winning. Assuming similar moves made at similar times is obvious proof of collusion is assuming that one party's move was just picked out of a hat, rather than market incentives prompting this move and applying equally to both companies causing both to come to the same conclusions for the same reasons. The actual reason the level of subsidization dropped, which you can find by googling for five seconds, is because they switched from bespoke hardware stacks to standard PC components, which were cheaper. No, you do not make the stupid financial move of making your hardware stack even more expensive just because your competitor made theirs more efficient.

by pie_flavor

8/1/2026 at 1:36:37 PM

The level of subsidization has nothing to do with hardware customization. You can have subsidized retail hardware or customized hardware sold at cost. They both went from high end heavily subsidized hardware to mid-range minimally subsidized hardware at the exact same time. Colluding in this way means they both dramatically increase their profit margins, and obviously does not preclude competition in other aspects.

Though they are also overtly colluding in other areas. For instance console exclusives used to be a major selling point, but they greatly empower studios to the consoles manufacturers detriment. The PS1 and PS2 had around 500 exclusive titles each. By the PS3 we're down to 174, 49 for the PS4 and now we're down to 26 in the PS5 era. Microsoft trends similarly and it was a strategic decision, to the point that they made a big announcement recently about returning to exclusives. As if the idea of having a competitive edge is some novel or revolutionary concept.

by somenameforme

8/1/2026 at 6:19:39 PM

It is not proof of collusion that they abandoned bespoke hardware right as consumer hardware was getting good. That is exactly when you want to abandon bespoke hardware. It is not proof of collusion that neither one lowered prices for the cheaper tech. Both were trying to appease shareholders who were sick of the massive up front capital investment that heavily subsidized hardware caused. Again: both had the same incentives, and so independently came to the same conclusions. If they were colluding, Xbox wouldn't've declaratively lost that generation, and with the department on fire today and in danger of being restructured, Sharma would right now be executing one of these strategies you are saying is a smart idea.

The world is not full of conspiracies. You will understand it much better if you look for the actual reason things happen.

by pie_flavor

8/1/2026 at 6:48:56 PM

Since you're ignoring words and just repeating the same fallacies, I'll respond in bullet point format to make it easier for you to read:

- Retail vs customized hardware has nothing whatsoever to do with subsidized or not. You can have subsidized retail hardware, or custom hardware sold for profit.

- Agreeing to collude on one issue that is mutually beneficial does not preclude competition in other ways.

- There have been tens of thousands lawsuits related to price-fixing and other anti-competitive behavior by corporations, and an unknowable of actions that were carried out with no consequences whatsoever, like this one. It happens constantly, because it's highly profitable and the consequences even when caught tend to be negligible.

---

Back to words: Sony and Microsoft were in a prisoner's dilemma with regards to subsidization and high end hardware. If they both collude, then they both see major profit increases and live happily ever after. But if either of them actually competed then they'd end up crushing the other one and take the market, at least in large part, for themselves.

In a competitive environment, this is precisely why you saw heavily subsidized high end hardware be the norm for consoles for decades. Neither could risk the possibility of not subsidizing. Then suddenly Microsoft and Sony stop doing so at the exact same time. It was both extremely overt but also impossible to prove so long as they didn't leave a paper trail.

by somenameforme

8/2/2026 at 1:09:31 AM

"You're ignoring what I said", says the guy who proceeds to repeat himself ignoring what I said.

> You "can" have subsidized retail hardware

But neither corporation's shareholders wanted them to for capital reasons.

> or custom hardware sold for a profit

But that'd be an insane waste of money for a platform that's more inconvenient to develop for.

> Agreeing to collude on one issue that is mutually beneficial does not preclude competition in other ways.

But when you're in danger of going under, it's time to stop leaving cash on the table properly, especially when you will win a bidding war.

> There have been tens of thousands lawsuits related to price-fixing and other anti-competitive behavior by corporations

But not against these corporations in this context.

> you saw heavily subsidized high end hardware be the norm for consoles for decades

Because in that span of time it was a good idea: they had lots of cash and were pushing the envelope of what hardware was capable of.

> Then suddenly Microsoft and Sony stop doing so at the exact same time

Because in that span of time, it stopped being a good idea: PC hardware had started being ahead of what they could offer. And you are calling an eleven year period "exact same time".

> extremely overt but also impossible to prove so long as they didn't leave a paper trail.

In other words, you have literally zero evidence this is true and are just going off of vibes, even though as I've just laid out, everything that happened is exactly what you'd expect to have happened if they weren't.

by pie_flavor

8/2/2026 at 6:17:10 AM

I'm not ignoring what you're saying - you're just repeating plainly falsifiable or illogical claims. By contrast you've laid out no reasonable arguments to what I've said. You even tried to claim companies engaging in anticompetitive was some sort of inconceivable conspiracy, in spite of the fact that Sony and Microsoft have been repeatedly busted for anticompete conspiracies, even within the Playstation and XBox section of their business. You're concocting unsound bad faith arguments, and then repeating them.

And once again, if either company actually competed on hardware from the PS4 gen onward, they would have dominated the market. They chose to collude instead which maximized revenue, but only thanks to the collusion. And the person who paid for this is the consumer who was left paying much more than they normally would to get much less than they normally would.

by somenameforme

8/1/2026 at 12:41:44 PM

> a typical example being the $20 coffee makers which you'll end up paying way more for than a decent one due to the fact they just constantly break.

I chuckled a bit realizing that you probably meant a decent coffee maker to cost far less than what I think a “decent” one does. Took me a moment to shake “you could buy a $20 machine every couple years for decades before you break even.”

by mathgeek

8/1/2026 at 1:46:18 PM

Yeah a genuinely nice machine is going to set you back plenty, but is built to last forever and can be easily repaired to achieve that. On the other hand a French Press costs like $10, lasts forever, and makes an amazing cup of coffee. But the $20 coffee machines are basically just a scam, but one that works because people just don't appropriately weight longevity.

by somenameforme

8/1/2026 at 2:36:19 PM

There are a lot of people for whom the choice is not “spend $20 every two years, or spend $200 today and never again” but instead “spend $20 today or just not have a working coffee maker”

by brookst

8/2/2026 at 8:13:06 AM

Better to spend $20 now and invest that $180 you spend. At 8% that $180 is $210 after 2 years, so buy the next one and invest the $190 etc

by hdgvhicv

8/1/2026 at 2:45:29 PM

There needs to be a term for this argument, because I think it's frequently used but generally fallacious and primarily used to defend the indefensible. It's a sort of strawman, in that you're concocting some demographic as something being reasonable for even though that demographic either doesn't exist, or exists in negligible quantities. The pitifulman?

And on top of all of this - again, French press. Cheaper than coffee maker and makes better coffee than a coffee maker. Are we to now invoke the same pitifulman again and claim that there's this same demographic that can't afford anything, yet works so much that even the extra 2 minutes you'd spend with a French press in impossible for them and their 7 part time jobs?

I mean FFS we live in a world where you can earn $20/hour frying frozen precut fries at McDonalds.

by somenameforme

8/1/2026 at 6:55:06 PM

… but what good is a french press without a $200 gooseneck kettle? :)

by mathgeek

8/1/2026 at 2:51:39 PM

cf Vimes' theory of boots[0] which is basically the same point.

[0] https://en.wikipedia.org/wiki/Boots_theory

by zimpenfish

8/1/2026 at 3:39:13 PM

Again I don't really think many of these concepts really apply any longer. I'm currently wearing $10 Chinese knock-off shoes that are lasting for years comfortably, while there are plenty of low income types running around in Air Jordans because of marketing.

This is part of the reason I have a bit of a personal crusade against marketing and advertising. I think its currently one of the main things that is realistically keeping the poor poor. Endless creation of artificial demand and instilling false hope about what 'luxury' goods would do for a person is just screwing over poor people. Some hip hop and other stuff has started going down this track [1], but it's still outvolumed a few hundred times over by the culture of glorifying being exploited by corporations.

[1] - https://youtu.be/QK8mJJJvaes?t=151

by somenameforme

8/1/2026 at 3:00:09 PM

I don't know if $20 coffee makers are really a scam. I've been using my $20 Mr. Coffee for like 5 years now without an issue, still makes coffee and more than likely will for a long time.

You might debate the quality of the coffee it makes, but I've long since learned I don't have the palate for it to matter.

by somehnguy

8/1/2026 at 3:49:07 PM

Yeah I've never had a $20 coffee maker or toaster die, they're too simple to have anything go wrong. It's always been when I splurged on a slightly fancy one it has broken, since they add a bunch of crap that can now go wrong.

by kalleboo

8/1/2026 at 4:05:49 PM

Toasters I agree with, though I use my toaster very rarely so I'm unsure if that's the nuance. But the coffee makers, cheap coffee makers would have the heating plate fail constantly. And by contrast I'm a 1-2 pot a day person and tended to leave the pot warming, so it's getting used a lot. They still shouldn't fail though.

It's similar to my experience with cheap electric ovens. In that case I'm skilled enough to repair them but there is overt planned obsolescence with the wiring being designed in a way that corrosion and failure is inevitable. It's quite an easy repair-job since it just requires fixing a corroded wire, but the case itself is built in a way that makes it a complete PITA to do. 15 seconds fixing the wire, 15 minutes disassembling/reassembling the casing.

by somenameforme

7/31/2026 at 10:18:59 PM

I've never had LED lights last anywhere near the purported 5-7 year timeframe. The worst ones are the small decorative ones with the tinier screw part that my house seems to have an inordinate number of. Those last maybe 6 months and then they start getting dimmer and then all the lights in the circuit get worse at once. I like LED lights immensely but the idea they last even 2 years is a lie in my opinion.

by freediddy

7/31/2026 at 10:30:02 PM

I bought all new Philip hue bulbs going on 7 or 8 years ago now and not a single one has gone bad. Maybe you're just buying cheap poor quality bulbs?

A 100 out of 100 success rate after 8 years.

by Hilliard_Ohiooo

8/1/2026 at 2:28:41 PM

I think a lot has to do with both luck and power quality as well.

When I got a new apartment ~10 years ago I filled it with LED bulbs. Mostly Philips hue, but also some generic stuff on Lutron controls. Had 0 failures for 7-8 years.

One summer we had weeks of power issues on my block. Transformer blew out 2-3 times and we had no power/low voltage/no power/restoration state multiple times. At this point 5-10% of my hue bulbs were fried. Pretty randomly since a bunch were in the same fixture and only a few of them died.

Its possible the hues died more than the generics since they are constantly "on" rather than behind a switch, so they experienced all of those voltage states directly.

by steveBK123

8/1/2026 at 10:54:52 AM

A recent rewire saw the electrician fit LED bulbs throughout the house. Modern fire safe housings etc, plenty of ventilation.

These were not cheap off the shelf bulbs from the local store.

3 in the kitchen went after less than a year.

I contacted the manufacturer directly. Turns out they had a "bad batch". How many bulbs did I fit?

About 50..

A box of 48 bulbs turned up 2 days later.

Not a single other bulb has gone 4 years later.

Not sure what the lesson there is. You win some you lose some? It's not always down to cheap bulbs?

by _puk

8/1/2026 at 11:41:25 AM

>3 in the kitchen went after less than a year.

Cheap/poor quality LEDs (the only kind I buy) hate high temps. Like really, really really hate it.

I've killed more bulbs near the shower than all other rooms on that floor combined. Meanwhile in the garage/basement the only tube LEDs that I've replaced this decade are the ones in the hot unventilated area away from the door or near hot work.

by cucumber3732842

7/31/2026 at 11:01:28 PM

They have become cheaper and cut costs over time. LED bulbs bought around the l-prize time when the technology was new are heavy, you can feel their metal heatsink. New LEDs from the same reputable brand are comparatively cheap.

by krackers

7/31/2026 at 11:26:21 PM

Yep, I have a 15 year old LED lightbulb that's a "60 watt equivalent" and it works great. It was one of the first LED bulbs I bought, and I only bought one because I remember it costing individually about $25, maybe more. It's got a very heavy metal heatsink where nowadays you have just a plastic "neck" of the bulb. It chugs along wonderfully (though the color is noticable a little bit greener than later cheaper bulbs that have better color accuracy).

by lelandbatey

8/1/2026 at 2:36:57 PM

me too. Those things were built to last - and they do.

by GeorgeTirebiter

8/1/2026 at 6:46:41 PM

> Maybe you're just buying cheap poor quality bulbs?

All of my led bulbs that failed were either cheap poor quality bulbs or expensive poor quality bulbs.

I have a hard time determining the quality of a bulb without running it for 2 years, and then it's a little late.

And then you have the mix and match problem. The same model number from the same manufacturer purchased at the same retailer gets different internals. So if you've got a room with 6 can lights, when one stops working properly, you have to search your inventory to see if you can replace just one, or if you have to replace all six. As long as the color is roughly the same, I don't mind a difference between rooms, but it's unacceptable when one lamp comes on instantly and the other five take a moment, or when they don't dim together, etc. I never had this problem with incandescents, as long as you bought the same color type and wattage, you could mix and match across brands. I did have one fixture that had tight tolerances for the fashion shaped large sphere bulbs it wanted, but that was fixable with a copper shim that cost one cent (and some paint on rubber insulation around the edge)

by toast0

7/31/2026 at 10:32:34 PM

No, I started with Philips Hues but then gave up and migrated to Cree light bulbs with Lutron Casetas because the Hues kept dying. These are over 2 different houses as well. I wonder what is going on if it's just me

by freediddy

7/31/2026 at 10:56:41 PM

The old Cree bulbs that Home Depot used to carry have been great for me. I need to find a new supplier though because the brands they have now are junk that last a year or two if I'm lucky.

by shawn_w

7/31/2026 at 10:40:32 PM

Had a ton of issues with the original Cree bulbs. They underglued the glass bulb and it came off with thermal cycling. They also put the LEDs on a metal core PCB they bent into ~10-gon tower. On the bends, the traces apparently would crack which led to overheating and a burn spot.

Cree is now no more; the bulbs are just a brand for FEIT who imports the same unreliable Chinese white-label bulbs.

by sam1714

7/31/2026 at 10:43:36 PM

Which brand do you suggest?

by freediddy

7/31/2026 at 10:59:09 PM

Philips Hue. I've had issues with color ones, the White + Ambience ones are fine. (Makes sense as they omit the color section)

Hue is basically the only bulbs not being sourced from the same couple of white label manufacturers. Philips/Signify's rapidly shrinking line of dumb bulbs are being sourced from there too.

Never buy filament style bulbs. They are universally unreliable because of poor thermals with the emitters.

by sam1714

7/31/2026 at 10:37:06 PM

Yeah that is very weird. Personally I even have some LED lights that are going on 15 years old, spent a bit more than I'd like on them at the time, but they were on sale!

by Krutonium

7/31/2026 at 11:07:24 PM

Same for me, have Philips Hue bulbs for the past 10 years, some 50 of them and only had one failure unrelated to the LEDs themselves.

by piva00

7/31/2026 at 11:08:12 PM

> Maybe you're just buying cheap poor quality bulbs?

If only that were true. Osram is a well-renowned brand so you'd think their - rather pricey - 'high-wattage' (10W) bulbs meant for outside lights here on the farm would last longer than about a year, all in different armatures, all with plenty of cooling (they're mainly used in autumn/winter/spring when it tends to be rather chilly here in Sweden). You'd think cheap bulbs go bad quickly but the cheap 'Biltema' brand bulbs I bought for some rooms just keep on going without marked loss of output.

All in all the average life time of a LED bulb seems to be around 2-3 years before it either has dimmed so much that it is no longer usable or it has failed outright. There are exceptions like the mentioned cheap bulbs which are still going strong after about 9 years of close to daily use or those Osram bulbs which fail in about a year or less or the (not mentioned) 'fluorescent tube replacement fixtures' which last less than a year before one of the 4 LED segment strips fails due to a failed LED. I repair these by replacing the failed chip, they work for a few months and then the things fail again.

I've used 'efficient' lighting since I bought my first house starting with jam-jar like Philips SL 'bulbs' (compact fluorescent tubes with all the related gizmos packaged in a single replaceable unit) for which I made fixtures out of birch plywood wood, 90° PVC drain pipe bends and E27 sockets because those things were heavy at about 500 grams. None of those SL bulbs had failed when I moved out 6 years later, I still have some somewhere in storage. Later I moved to the lighter PLC version which also tended to last a long time, some of them are still in use here. Then came LED bulbs which were supposed to last for tens of thousands of hours... and I suddenly started to have to think about replacing failed bulbs again.

by hagbard_c

8/1/2026 at 12:38:26 AM

> Osram

Pay attention to the European efficiency rating for LEDs. Anything above B is practically "Dubai LED" class. The common Osram isn't doing good in this category, their LEDs I have seen in retail are rated like E-F.

To reiterate: the article fell short on the point of LED efficiency. LEDs are semiconductors, do you know when the CPU is at peak efficiency? Before it hits the exponential voltage/frequency curve. Low voltage -> cooler LEDs -> more efficient light conversion -> easier to cool -> tens of thousands hours of practical lifespan.

What does it cost? Like less than a dollar in BOM. What does it cost us? Billions in replaced LEDs worldwide and megatonnes of electric waste, because it is not made to last for "objective reasons": lower price tag in storeshelves aka capitalism and oblivious consumers.

by PinkSheep

8/1/2026 at 3:14:44 PM

These are 'A++' marked LED bulbs according to the old energy label standard which would be between C and D in the new label standard. They're supposed to last for '15 years' according to EC 1194/2012 [1] (2.7 hours/day, 15000 hours) and survive 100000 power-on/off cycles. In reality they last for about 1500-2000 hours in well-cooled fixtures.

The lamps I'm referring to have 4 wide (4mm) LED filaments in a normal-sized glass bulb. On the packaging they're marked 'Ledvance' [2]/made in China, i.e. they carry the Osram brand under licence.

In other words LED lamps can be made to last a long time but many are not, brand names and brand pricing notwithstanding. The form factor - E27 socket with glass bulb, fluorescent tube replacement, large heat sink mounted LED strip - does not play as much of a role as you'd think given that the lights with the 'best' cooling (LED strip mounted on a large heat sink) shows the highest failure rate while the 'traditional incandescent lamp' form factor with the electronics in the E27 base shows both the lowest as well as the second-highest failure rate here. It probably comes down to component quality, something which again does not seem to be related to brand name.

[1] https://eur-lex.europa.eu/LexUriServ/LexUriServ.do?uri=OJ:L:...

[2] https://www.ledvance.com/00_Free_To_Use/asset-8344755_THE_AS...

by hagbard_c

8/1/2026 at 1:50:51 PM

We need a powerful non-profit / semi-governmental organisation that regularly tests all these things and tells it like it is. Something like Stiftung Warentest in Germany.

People can make up their own mind about going for cheaper electronics or aiming for longevity, but at least we would have a clearer picture of who is contributing to that pile of electric waste and we could make more informed decisions.

by wvh

8/1/2026 at 2:52:53 PM

There are some who took this view when LEDs were first mandated. Light bulbs were dirt cheap, yes they use more electricity but their waste stream is glass with a bit of metal.

Now we pay more for something that lasts a bit longer with more waste.

It's not "capitalism" if the government mandates a product.

What you are seeing is the typical enshitification that occurs when a product becomes mass produced and low cost to the customers becomes the primary driver.

by themaninthedark

8/1/2026 at 11:02:24 AM

Same here, around 40 Philips Hue lights, 8 years old. All but one are still working fine. In the broken one the electronics went bad after about 4 years, the light was still working but would occasionally turn off or change color for no reason. Got a new one for free.

by t0mas88

7/31/2026 at 11:35:31 PM

Can second this. Did my whole house in Hue and have only had one failure (early on) in 5 years.

by longitudinal93

7/31/2026 at 11:07:32 PM

> I bought all new Philip hue bulbs going on 7 or 8 years ago now and not a single one has gone bad.

I had a bulb go bad once (out of almost 50 bulbs), but I called their support number and they shipped me a new one for free without any hassle.

by gucci-on-fleek

8/1/2026 at 10:11:25 AM

Yeah, same. I have probably around 60 Philips Hue bulbs or lamps, I bought the first one over a decade ago, and so far, one lamp has failed (a Signe that has started flickering).

by InsideOutSanta

7/31/2026 at 10:57:43 PM

2 of the 4 philips hue I bought in 2019 died a few years back, a bulb and a led stripe, I'm not buying expensive ones next time

by toasty228

8/1/2026 at 9:15:41 AM

The bulbs from about 7 to 10 years ago were far more long lasting than the new ones

by pkphilip

7/31/2026 at 10:44:45 PM

I bought a whole bunch of GE LED bulbs figuring big name, probably more reliable. Nope. Diabolically bad. Never again.

by Twirrim

8/1/2026 at 3:38:39 AM

Me too. They were a lovely colour temp and high CRI, really well diffused too. But wow, they died so fast in almost any fixture.

by NathanielK

8/1/2026 at 5:26:51 PM

same on phillips hue but maybe 3 out of 10 have died.

they might not have great QA

by 8note

8/1/2026 at 3:12:54 PM

[dead]

by crypttales

7/31/2026 at 10:32:22 PM

I have never had an LED light bulb fail. I'm still using the same 6-pack I bought over a decade ago, bringing them with me from apartment to apartment. I only have 5 remaining because the glue that holds the glass bulb onto the base eventually failed on one and I couldn't be bothered to figure out what the correct adhesive would be for lightbulbs.

by craftkiller

7/31/2026 at 11:46:18 PM

Fair enough but they don't get hot, nearly any glue would work, Being plastic, I would use CA(superglue).

After failure I like to do a postmortem teardown, it has never been the LED proper, usually it is the capacitor in a capacitance power supply, otherwise the controller IC.

by somat

8/1/2026 at 12:33:58 AM

Good to know, because another one of the bulbs visibly looks on the verge of separation. Thanks!

by craftkiller

8/1/2026 at 12:58:36 PM

I moved into a house with an obscene number of GU10 LED downlighters. The bulbs were all mid market (I think) brands like Energizer etc. I’d guess they’re coming up to 7-8 years old and I have now had to replace most of them, probably 30-40 bulbs from a couple of manufacturers that all just started to fail over about 12 months.

I also have quite a few Philips Hue bulbs, I’ve had 1 fail out of 20-30 and I think that was not the LED itself but some control board. Some are going on 10 years old.

by IrishTechie

7/31/2026 at 10:27:26 PM

Every Philips L-Prize bulb I bought in 2012 is still working. Some have been on about 30% of the time since purchase. A few have been dropped multiple meters onto concrete; oops. Well made LED bulbs last. Cost-optimized ones may not.

by addaon

7/31/2026 at 10:36:15 PM

My OG gray Philips lasted more than a decade constantly on. It was still working at the end but I noticed the yellow remote phosphor plastic panels were disintegrating with a web of cracks. Underneath, the plastic surrounding the LEDs had turned brown: the polymers had photodegraded.

by sam1714

7/31/2026 at 10:30:32 PM

I had a bunch of Philips Hue LED light bulbs until they started dying within a couple of years or less and the replacement costs were ridiculous. That's when I switched to Lutron Casetas.

by freediddy

7/31/2026 at 10:44:10 PM

I run about 50 Philips Hue lights and I’ve replaced 3 in six years, so definitely not as reliable as the L-Prize lights, but not awful. The L-Prize bulbs are built like tanks… when you drop one you check the floor for damage before the bulb.

by addaon

8/1/2026 at 1:45:10 AM

Exact same numbers for me! 3 bulbs stopped shutting off completely. When I would turn them off they would stay around 5% brightness.

by Shadowmist

8/1/2026 at 8:09:19 AM

In the category of private victories without any economic use (any hobby costs you something): I store boxes for LED lights with receipts. Write on the box where you put the light and chug it in the garage. I have not failed to get a new one for free at our local home depot equivalent after 3 to 4 years. It doesn’t help the store that the boxes used to state how many years they would last.

I agree that even Osram doesn’t meet the hours or years, Philips doesn’t by a long way.

by wjnc

8/1/2026 at 9:52:37 AM

Because LED lighting is electronics it hates heat, unlike an incandescent lamp. So the most common cause of shorter lifetimes is that the LED can't discard enough heat to the environment and begins to overheat. Small (or absent) heatsinks, installation which (perhaps for aesthetic reasons) puts the lighting where warm air can't escape and of course just putting the lighting directly in concentrated sunlight will make this worse.

I have a room which is lit for like 10-20 hours per day and so it clocks up hours very quickly and my LED lamps are guaranteed for 2 years so I figured surely the guarantee would trigger but nope, I began keeping receipts and turns out that I only need to replace one every 3-5 years it must just feel frequent because it's annoying, I just dug out the most recent receipt and it's from 2021. I will continue keeping the receipts to feel better about the annoyance.

by tialaramex

7/31/2026 at 10:26:54 PM

Randomly, my Philips Hue E27 bulbs last far longer than any of the China Zigbee clones I've bought, but like all mass produced goods, there are outliers that fail.

The cheap China stuff tend to fail in a few months, while I've only had two Hue bulbs in my collection fail ever, so Philips seems to be doing something right.

by ValentineC

7/31/2026 at 10:30:35 PM

I agree, my Philips hue bulbs have been going strong for about 7 to 8 years now I'm not a single issue. I think people are just buying cheap bulbs then getting upset that they bought cheap bulbs.

by Hilliard_Ohiooo

7/31/2026 at 11:34:34 PM

It's a bit tricky to determine what exactly is "cheap" though. I've got various Philips Hue bulbs/strips with no failures at 20k+ hours. OTOH I've got a few Philips Ultra Definition bulbs over my kitchen island which seem to fail at an average of 10k or so hours.

None of the other bulbs in my house are really on long enough to say much about failures... like I've got dozens of recessed LEDs without any failures over 5 years, but I doubt any of them are over 1-2k hours.

by Marsymars

8/1/2026 at 10:42:46 AM

>> I think people are just buying cheap bulbs then getting upset that they bought cheap bulbs.

What is a 'cheap' bulb though? We paid a few bucks for a bulb for generations. Suddenly we're paying 10x that for a Phillips Hue and anything lesser is 'cheap'.

by basisword

7/31/2026 at 10:32:25 PM

Enclosed fixtures can kill some LED bulbs quickly if they can't dissipate heat.

by xnx

8/1/2026 at 12:17:58 PM

Yes, cooling is important. I don't have a passion for lamps so after 5 years in a new house most of my LED bulbs are still hanging from the ceiling and walls, I can't remember any failure.

I had an IKEA lamp with a pretty tight fitting for the bulb and it killed LEDs after a few months. I reverted putting a fluorescent bulb inside it. No failure.

by pmontra

7/31/2026 at 11:09:57 PM

I would say most LED bulbs. You need ones specifically rated for enclosures, which are way less common, and most people have the wrong idea of what counts as enclosed (the glass hood around your kitchen chandelier bulbs is an enclosure, as are ceiling cans)

by foobarchu

8/1/2026 at 1:56:46 AM

I was killing a lot of LEDs in recessed cans. I replaced them with integrated retrofits, which aren't designed around a bulb shape that fries its own electronics, and aren't made as cheaply as possible.

Here's a nice dim-to-warm recessed: https://www.homedepot.com/p/HALO-RL56-Series-5-6-in-Recessed...

by jerlam

8/1/2026 at 12:42:44 PM

> Those last maybe 6 months and then they start getting dimmer and then all the lights in the circuit get worse at once.

Have you checked the wiring? I may be stating the obvious, but in household electrical wiring every fixture or receptacle is connected to its circuit in parallel.

I can't think of a way for one bulb on a correctly-wired circuit to fail in a way that causes other bulbs to dim, unless the failed bulb is essentially shorting out the circuit but not causing the breaker to trip for some reason.

If you know why it's behaving that way, I'd love to know too. I've never seen anything like that.

by blincoln

8/3/2026 at 2:18:33 AM

Only bulbs I’ve had fail in the last 10 year was some IKEA GU10 bulbs (maybe 2-3 years). I got them replaced for free with no receipt, and the new design had obvious improvements to managing thermals.

It shows that, yeah, bad designs can fail early. But if you buy from a reputable company, it is likely to be fixed.

by audunw

7/31/2026 at 11:11:52 PM

Do you keep your light on constantly?

I used to have a lamp that I kept replacing bulbs for "all the time". I didn't understand why since the light bulbs claimed to last for years and years. One day I read the light bulb package more closely, and it said something along the lines of, "7 years life span at 2 hours per day" in small text (the numbers may be wrong, going by memory). Doing the math, the number of active hours claimed on the box turned out to be roughly true. I don't know if LED bulb packages play similar tricks but it's worth checking out.

by mort96

8/1/2026 at 2:12:49 AM

> I don't know if LED bulb packages play similar tricks but it's worth checking out.

It's perfectly logical that lifetime should be measured in hours of service, not days of existence. People don't know how to translate 7500 hours of service into a real duration, though. Hence the years.

by fn-mote

8/1/2026 at 8:30:03 AM

I would expect 7500 to translate into 0.85 years though, not 7.

by mort96

7/31/2026 at 10:35:54 PM

I had an LED porch light that lasted ~10 years. The fixture had light and motion sensors, so it wasn't on during the day or all night, and often it was not running at full brightness, but still it was on a lot. I was surprised it lasted that long, especially since it was outdoors.

I switched to all LED lighting many years ago. I've had a few that lasted less than a year, but by and large I'd say that nearly all the lights I've had have lasted at least on the order of years.

by usefulcat

8/1/2026 at 10:55:49 AM

Hm... race to the bottom? We bought our current apartment (built in the 1980s) 12 years ago, and I have switched to LED lighting everywhere from the beginning. Main learning: old relays and/or dimmers seem to generate voltage spikes that are fine for old lightbulbs, but kill LED lamps in a matter of months, so that it looks like the lamps themselves might be the cause. After I replaced those, no more problems - most of the LED lamps from the beginning (except for those connected to old relays and dimmers) are still with us! I also tend towards brand name LED lights (Philips & co), which might be a bit more expensive, but have (I hope) better quality checks.

by rob74

8/1/2026 at 11:22:01 AM

The following anecdote doesn’t contradict what you wrote but maybe you’ll find it interesting- I worked for a daughter company of Phillips in the 00s (as an embedded programmer). I was working on mp3 players and set-top boxes but my boss has been with Phillips for decades at the time. He told me how the CD players (the ones you’d connect to speakers at home, not the portable ones) had a variety of brands, from budget to premium. But the only differences were in the external materials & the quality of the disc tray - as those are differences people notice. But everything else (and thus the audio quality) was identical.

by ido

8/1/2026 at 1:17:58 PM

I had that dimming issue in my house... And got the gas boiler serviced cause it had a problem with the ignitor, and all the dimming stopped.

Now the house is ancient, so I'm guessing there was some sort of short happening that was causing this.

by roomey

8/1/2026 at 10:54:09 AM

Yeah my parents had the type with the external driver box from Bunnings, and they died every 6 months or so. Those were DIY with a little pigtail plug.

They were awful. I genuinely believed that LED lights were crap.

In our new house the sparky put in a proper Chinese brand from the commercial sparky store I guess, no external box, dimmable, 4 adjustable hues with a toggle switch. They were like $10 each. Been 7 years no failures.

We have some really expensive ATOM ones above the shower (IP68) and outside around the perimeter. Those were around $75 each, but are the same age as well, no issues.

by aetherspawn

7/31/2026 at 11:41:28 PM

If you are installing LED lamps into enclosed fixtures you need to use LED lamps rated for enclosed fixtures or they will die an early death. A globe around a lamp with an open bottom and a recessed can are both enclosed, even though you can see the lamp. Heat rises.

Similarly, you can’t install recessed cans into a ceiling that is insulated without them being IC rated or they will die an early death.

Spend a few dollars more per lamps for quality ones. Philips makes decent LED lamps, I’m fond of their 2200-2700K dimmable lamps that lower the color temp as they dim. Philips makes an E12 base BA11 shape (candelabra) lamp that should fit your needs.

I buy Keystone lamps LED T8 lamps almost exclusively, but those are pretty uncommon in homes.

Commercial LED fixtures are rated for 50,000 hours typically but they have a separate LED driver and massive heatsinks to provide that sort of lifespan.

by quickthrowman

7/31/2026 at 10:24:17 PM

I've run 30w led modules for 10+ years by - using real power conversion components, making sure they have sufficient cooling and running them at 75% of their rated current. they still seem plenty bright.

by convolvatron

7/31/2026 at 11:01:04 PM

In mid 2016 I replaced around 25-30 incandescent bulbs (7 x 100 W, a couple or so 40 W, and the rest 60 W) with Walmart Great Value non-dimmable LEDs. Basically every incandescent bulb I could find around the house that was not on a dimmer and was A19 shape with an E26 base (except I forgot the attic and crawlspace).

There was some sort of deal between Walmart and Puget Sound Energy that applied an instant rebate automatically so that they simply rang up on checkout for $0.17.

They are all still working fine.

by tzs

7/31/2026 at 11:07:14 PM

We have several hundred smart led lights from I want to say Lutron? In my home going on year 6 soon and we’ve had to replace only 3-4 so far. 3 of those replacements were on the same circuit as well so I question if the circuit isn’t at fault.

Plural of anecdote isn’t data and all that but I think the several hundred quantity adds some weight. Also about 90 of those are in outdoor lamps so there’s a good amount dealing with weather and we’ve not had to replace any of those yet.

by lovich

8/1/2026 at 10:40:11 AM

I owned my last house for 12 years, I replaced 1 bulb?

Half the bulbs were carried over from a previous house, so some are ~14 years old. Admittedly these were early ones, so the original batch had metal bodies etc. the replacement was a Phillips one. This is in the UK though, with pendant fittings, so less heat issues.

by benj111

8/1/2026 at 9:33:52 AM

I guess if people are nerding out on lights in this subthread, maybe someone can recommend me a type or source for low-K (say, 2200K or something thereabout) bulbs with good CRI (yes, I know, but I can still notice a big difference at 2700. I don't know how low it will matter to me, but I want to see)

by duskdozer

8/1/2026 at 10:32:02 AM

I've lived in my current home for 5 years with LED lighting, and we have hardly changed any lightbulbs in the house. I've got no idea how old those bulbs are. I lived in my previous place for 7 years and changed a handful of bulbs.

Can't say my experience matches at all!

by maccard

7/31/2026 at 10:57:26 PM

That is weird. Mine are basically infinite.

by watwut

8/1/2026 at 10:40:52 AM

This sounds like you have electrical issues. I've lived in a couple of apartments before where bulbs would go all the time. Outside of that I've been using the same Phillips Hue bulbs for going on 8 years without any issue.

by basisword

8/1/2026 at 4:40:33 PM

You might have poor quality incoming AC. Capacitors in the power supplies will be under more stress if you regularly have over/under voltages or small spikes and surges.

by Peanuts99

8/1/2026 at 1:36:25 PM

Some of the first LED bulbs we bought are still running after more than 10 years. I don't remember the brand but they were expensive at the time.

by pier25

8/1/2026 at 1:14:47 PM

I bought a bunch of bulps when I moved to my house about 7 years ago. Some are on like 90% of the time since.

I haven't had a single one go bad so far.

Cheap WiZ bulps

by herbst

8/1/2026 at 9:33:44 AM

My n=1 study: I built my living room lights about 8 years ago and have had to swap 1/20 lamps since. Perhaps try a better manufacturer?

by nxpnsv

8/1/2026 at 4:38:54 PM

Counterpoint: I've never had an LED bulb go bad. Ever. I think it's highly dependent on the brand.

by phendrenad2

8/1/2026 at 8:12:50 AM

Ditto. They barely last 12 mos. This goes for both types: LED Light Chips (Strips) and LED String Lights from various manufacturers.

by semerda

8/1/2026 at 10:10:37 AM

I’ve never had a Hue bulb fail in about a decade of using them in every fitting.

by cameronh90

8/2/2026 at 1:48:24 PM

We moved into a new house in November 2022 and bought all lights anew. LEDs all around, some 25 devices taken all together.

44 months later, all of them are still working.

by inglor_cz

7/31/2026 at 11:01:30 PM

BTW, I don't understand why having the driver converting AC to DC inside the LED spots is the default. For a new house, it seems to make sense to me to have at least one external driver for a ceiling of spots, if not one per floor (not one for the whole house because pf the drop in voltage). But 24V led spots seem much less common and I cannot find many reviews. Any idea why it's not more common ? It seems efficiency and thermals would be much easier to optimize without the size constraint, plus economy of scale with components. Any specific 24V led spots / bulbs to recommend ?

by wolfgangK

8/1/2026 at 10:17:57 AM

In the UK having separate drivers is, or used to be, the most common set up. I never had good luck with it. Big problems with compatibility, failing drivers, especially with dimmers, plus weird electrical problems like residual glow.

by cameronh90

8/1/2026 at 1:28:15 PM

The most common setup for installations by lighting designers, sure. I suspect the vast majority of home led lighting is in lightbulb form, from people replacing incandescents.

by psd1

8/1/2026 at 7:28:21 PM

I should have said that I meant common for recessed lighting.

They often use 12V MR16, inherited from the halogen days, and are pretty common in houses renovated from the 80s until the 2010s. The other option was GU10 which ran at mains voltage, but they were less common until LEDs became popular.

Recent new builds, flats especially, often just include hard wired fixtures without interchangeable bulbs at all. They often still have separate drivers above the ceiling, with the wiring to the lights being low voltage, but sold as a single unit. But I think that's just so they can make the light itself thinner...

by cameronh90

8/1/2026 at 9:15:42 AM

> I don't understand why having the driver converting AC to DC inside the LED spots is the default.

because LED bulbs are drop-in replacement for incandescent bulbs. AC is pre-installed

stand-alone LEDs don't have converters, be it singular or be it strips

by NooneAtAll3

8/1/2026 at 12:50:26 PM

They said for new houses.

by guerrilla

8/2/2026 at 9:33:02 PM

Are you going to buy a new house that you can't get lightbulbs for?

by jjk166

8/1/2026 at 1:26:27 PM

he*

by NooneAtAll3

7/31/2026 at 11:09:27 PM

It seems like a pretty normal local maximum situation, no? It's a solution which works "well enough" and means all light bulbs work in all light fixtures and all light fixtures work with all light bulbs. It would be better if we all moved away from LED, got all new standardized 24V light fixtures and stopped putting the power supply in the base of a standard E26 light bulb socket, but those kinds of transitions are costly.

It probably also doesn't help that this would require industry-wide cooperation between companies which would much rather sell their own light fixture + LED combo which people have to replace regularly than a light fixture which people buy once and keep for life.

by mort96

7/31/2026 at 11:32:26 PM

This is also a good solution to the flicker problem. Centralizing the conversion means you can afford better power filtering capacitors and circuit designs.

I notice most BR30/40 bulbs start to flicker after about a year in my fixtures. It's not a happy environment for an electrolytic capacitor. In the summer I guarantee we are out of the operational range of what is in those bulbs.

by bob1029

8/1/2026 at 12:18:57 AM

You can buy recessed LED cans with remote drivers. It’s not super common, but they exist in commercial installations. The largest LED drivers that I am familiar with are 24VDC/96w which is enough for (12) 8w recessed cans.

Individual drivers per fixture allow flexibility with regards to commercial lighting control. The 0-10V dimming signal is sent to the driver, and if you have multiple fixtures per driver, they will all dim together. This behavior can be accomplished with individual LED drivers by wiring the dimming signal in parallel to each driver from the dimming module.

Line voltage circuits can handle a lot of fixtures, particularly at 230V or 277V. A 277V 20A circuit can continuously power 554 8w recessed cans or 138 32w 2x4s on a single pair of #12/2.5mm^2 wire. Two of those circuits can power all of the lights in a 20,000 square foot office.

by quickthrowman

7/31/2026 at 11:25:05 PM

I expected the -48VDC telco standard to become popular as people had more solar, batteries, etc. I guess we will get there the most complicated way possible with usb-c.

by piguin

8/1/2026 at 9:24:32 PM

This might be my bubble showing, but I expect this would have worked better if DC lighting became more popular 20+ years ago, when many many more people had landlines.

(I suppose a lot of people in the US still probably do have landlines, but I know very few, if any, people who do.)

Then again, my house (built in 2009) is still wired for a landline, but the wires (and the voltage on them) remain unused...

by kelnos

7/31/2026 at 11:04:56 PM

It would be nice to have the driver and LED decoupled from each other with a standard socket between driver and LED even if there is one per bulb.

by HappyPanacea

7/31/2026 at 11:20:33 PM

That's what I've always thought as well. The driver could also fit in the standard sockets that bulbs fit in now, with the socket/LED interface being a different standard connector.

by stavros

8/1/2026 at 10:02:42 AM

GU5.3 is 12V standard socket. Then there is G4 but they probably have some issues as well in wrong place. Probably some others, it really is a mess.

by Ekaros

8/1/2026 at 8:57:58 AM

> In order to put the emission peak in the visible spectrum, the filament would need to run at ~5700 °C (...) No metal can survive these conditions: Tungsten melts at "only" 3422 °C.

Hmm, can we possibly find something that does survive at 5700 °C to use as a light source? Perhaps molten metal confined in a magnetic field...

by Aardwolf

8/1/2026 at 9:12:13 AM

mercury vapor

by NooneAtAll3

8/1/2026 at 11:31:28 AM

mercury vapor lamps are not a thermal light source, though

by MathMonkeyMan

7/31/2026 at 11:45:38 AM

Tungsten will burn in an oxygen atmosphere, therefore standard light bulbs actually use an inert gas (like argon). The gas pressure slows down tungsten evaporation and prevents oxidation. The main reason for short living of the bulbs is that the filament gradually evaporating over time until it thins out and snaps.

by dybovalex

8/1/2026 at 1:59:37 PM

I was with the article until this point:

> If long lived light bulbs are worse products, why would they need a cartel to enforce a the thousand hour limit? Well, it's because people think long lasting bulbs are a good product:

Lifespan is something everyone can understand, and has a direct effect on when you will have to go back to the store: if you saw two 60 W bulbs in a store, one claiming to last 400 hours and the other 2,000 hours, you'd probably get the longer lasting one without thinking about it.

This makes no sense to me. The article's whole point was that longer-lived bulbs produce worse lighting: Less brightness, more "orange glow" instead of "daylight", etc.

But light quality is the one thing that a buyer would immediately notice once they actually tried out the bulb. If they bought the amazing 100.000 hours bulb but found out it's a nightlight, they'll probably not buy that bulb again even despite the lifetime. If fact, they'd probably go back to the store really quickly to get a "real" light.

So I don't see why we'd need some industry cartel here to decide what's good for the customers if the customers would perfectly well decide themselves.

by xg15

8/1/2026 at 3:37:57 PM

Seeing how many people buy 6000k LEDs for sunset/night lighting (e.g. a bedroom), no, people don’t give a shit about lighting. The most caring of average consumers would notice their bedroom looks like a hospital, but refuse to buy 2700k or 3000k because "they look like piss" when they go to the store at midday.

And with the stupid habit of talking about lights in input watts (skimmed over in the article), we’d get what’s described in the article, albeit less dramatic: people would buy the bulb that lasts 50% longer and not notice the lower lumen output.

Consumers are not educated at all in lighting, especially when the industry does not help (re: watts, but also temperature, color rendering, flicker, etc)

by tuetuopay

8/1/2026 at 3:49:54 PM

> people would buy the bulb that lasts 50% longer and not notice the lower lumen output.

OK, but if they're fine with the lumen output, where is the problem?

by xg15

7/31/2026 at 11:13:34 PM

There are varying levels of quality for the same bulb. I don't know what the difference is in the bulb but they make a difference in the order of years.

I'm in Switzerland where we can no longer buy halogen bulbs. Every bulb in our apartment is led except 7 bulbs under the cabinets in our kitchen. I used to buy bulbs from a hardware store but they would only last about a year. When they banned halogen bulbs I found that I could buy them from a photography supplier for more money. The product number was different but it had the right specs. They've been going for three years and no burn outs.

by comrade1234

7/31/2026 at 11:16:45 PM

What do you like about halogen bulbs?

by dataflow

8/1/2026 at 9:03:06 AM

They last forever. I moved into my current house over a decade ago and the only bulb I haven't had to change is the single halogen that was already in a bathroom.

With halogen, the lifespan doubles just by reducing the power by a single percent. Having a light fitting far away enough from the DB means that the tiny internal resistance of the wire going to this bulb is enough to make a difference to lifespan.

by lelanthran

8/1/2026 at 8:13:22 AM

The light spectrum they produce.

by shrx

7/31/2026 at 10:35:14 PM

That top paragraph is a masterclass in modern media standards. Tell the truth will hiding facts so your readers can be reassured that they are in fact smart and correct, while you never have to tell a lie.

by WarmWash

7/31/2026 at 10:44:18 PM

You have been misled about spelling misled

by BGizzle

8/1/2026 at 7:48:12 PM

As many have pointed out, I think the cartel is just using the engineering perspective for justification. If avoiding racing to the bottom is the goal, the rule could be that "no one can advertise a lifetime longer than 1000 hrs" as opposed to actually limiting the lifetime.

by SCLeo

7/31/2026 at 10:14:13 PM

Misled

by montag

7/31/2026 at 10:46:32 PM

I've been seeing the misspelling of the past tense of lead (led) incredibly often in the past couple years. It's clearly confusing that lead is an actual word, so it may seem right and likely doesn't get caught by auto-correct, although I assume grammar correction should catch it.

by enobrev

8/1/2026 at 3:28:03 PM

Merriam Webster lists it as a valid spelling, and even uses that spelling in one of the example sentences

https://www.merriam-webster.com/dictionary/misled

by Austizzle

8/1/2026 at 8:48:54 PM

Mislead is the present tense; Misled, the past tense.

I'm not seeing the tenses being used interchangeably with the same spelling on the site, but I may be missing it.

by enobrev

7/31/2026 at 10:22:45 PM

5700 is a terrible temperature for lighting unless that matches the daylight. Soft white and even reddish is much better at night. Presumably, most artificial light it used at night. 2700 is more like it.

My LED lightbulbs change temperature based on the time of day and are dull red at night so you both don't wake up and you don't stub your toe. Show me a non-LED that can do that.

by jsLavaGoat

7/31/2026 at 11:12:11 PM

A dimmed incandescent bulb will shift from 2700k down to maybe 2200k. Maybe not as versatile as your LED but it worked well to somewhat mimic the sun before LEDs.

by francisofascii

7/31/2026 at 10:39:35 PM

> Show me a non-LED that can do that.

Many of use simply don't find LED lights, no matter the color they try to simulate, enjoyable.

I find my LED lights, including fancy decorative ones, bearable only because I always use one real, old-school, warm white (more like yellow really), incandescent (hence non-LED) light in my living room. And, yup, if you know what you're looking for, it's totally possible to still buy incandescent lightbulbs.

LED lights, no matter the shade/color always give such a fake / non-cosy feeling to just about everything.

I'm taking an actual candle + an incandescent lightbulb's warmth feeling any day over anything LED.

Now as I'm also conscious about saving the planet, I'll mix those with mostly LED lights. But having one incandescent lightbulb (and/or one candle) really creates, to many of us, a better atmosphere than pure LED lighting.

It's the old "you're eyes cannot detect more than 24 fps" all over again: "you cannot detect the flickering of a LED light". Except I can instantly tell if it's the cold (no matter how "warm white" the box says it is) light of a LED or the actual warm light of a proper incandescent lightbulb.

by TacticalCoder

7/31/2026 at 11:18:24 PM

Have you tried Philips Ultra Definition bulbs? They have 1/6th the flicker percentage of incandescent bulbs and 1/3 the flicker index.

Here's a review I found of several bulbs for CRI, flicker, dimming performance, and spectral characteristics [1].

[1] https://www.thesmarthomehookup.com/25-soft-white-led-light-b...

by tzs

7/31/2026 at 10:56:06 PM

There's two separate issues where LEDs still have a hard time competing with incandescent bulbs: CRI (really R9 or other deep-red measures, because CRI stats are sort of gamed) and flicker. Most LED bulbs also have poor heatsinks and are more likely to die if kept inverted or in enclosed fixtures.

All of these are individually solved problems, you just can't buy a bulb that's an e26 retrofit that satisfies all of them for whatever reason. If LED bulbs actually lasted 10 years, then there'd be 10 year warranties. Yet even the most expensive boutique LEDs (e.g. from YujiLED) are only 3 years at best.

by krackers

8/1/2026 at 1:03:18 AM

For the vast majority of people the problem was their first exposure to LED bulbs were in harsh temperatures and they wrote them off. The lightbulb wars aren't new. It's an evergreen political football.

If you strongly prefer incandescent that's fine but it doesn't take the 5-7 years the author suggests to recoup your investment. It might, or it might not. Where your curve lands depends, but they do in fact use a lot less energy.

by jsLavaGoat

7/31/2026 at 10:31:37 PM

I do something similar and all my Philips hue bulbs are going strong after 7 to 8 years. I think people just get upset with themselves for being cheap.

by Hilliard_Ohiooo

8/2/2026 at 9:39:59 PM

5700 is the temperature at which the peak is in the visible spectrum. It has nothing to do with lighting quality.

by jjk166

7/31/2026 at 10:44:25 PM

Lighting color is cultural. Go to Asia and you'll find many places, commercial and residential, lit with 5500K, traditionally fluorescents.

I've heard a theory that it has to do with genetics and lower eye sensitivity.

by sam1714

8/1/2026 at 10:32:07 AM

> I've heard a theory that it has to do with genetics and lower eye sensitivity.

Might be.

But I expect that -as you said- it's cultural, rather than biological. One's sense of color is amazingly adaptable [0], but when I do a side-by-side comparison, I find that I strongly prefer the "noontime sun" color to the "incandescent bulb orange" color. As a bonus, the "noontime sun" whitepoint is pretty close to the whitepoint of my monitors, so white things on the screen are very nearly the same color as white things on my desk.

[0] Go play the game "Return of the Obra Dinn" for like an hour straight, and then be amazed at how off your understanding of what black is has become.

by simoncion

8/1/2026 at 9:37:59 AM

Is it by deliberate choice or something like availability? I come across enough people who don't seem to notice the difference at all, or at least until it's directly pointed out.

by duskdozer

7/31/2026 at 10:59:22 PM

It's called a dimmer switch.

by pigpop

7/31/2026 at 11:17:54 PM

Brightness and color temperature are orthogonal. The dimmer switch only controls brightness.

by foobarchu

7/31/2026 at 11:40:22 PM

There are, however, some dumb LED bulbs that have built-in circuitry to warm as they dim. Some of them even adjust the temperature based on the dimmer switch setting without adjusting the brightness for part of the range of the switch. They're quite nice, I use them in various places.

by Marsymars

8/1/2026 at 12:11:44 AM

Incandescents are mostly in the red/warm end of the spectrum anyways so you don't need to change the color temperature at night.

by pigpop

8/2/2026 at 10:22:33 AM

You need to spell and grammar check your writing

by winterbloom

7/31/2026 at 11:44:57 PM 

  Tyrell: The light that burns twice as bright burns half as long,
          and you have burned so very very brightly, Roy
- Bladerunner (1982)

also, I enjoyed the last lines of the article:

Also, to preempt the orange site, I'm not saying that planned obsolescence doesn't exist.

by m463

8/1/2026 at 3:00:49 PM

Unfortunately, Tyrell is incorrect. https://www.replacementlightbulbs.com/ohmslaw.html which gives the equation:

  Life(actual) = Life(rated) x ( V(rated) / V(actual) ) ** 13
NOTE that 13th (!) power. Doubling brightness only requires a small voltage increase, but it causes the lifespan to be roughly 9% due to this law.

Hams have known about this for decades, where is it common to run 14V or 12V lamps on 6.3 volts for dial / panel illumination. I have a Drake 2-B receiver that uses this technique, and I have yet to replace the 2 bulbs that light up the frequency dial.

by GeorgeTirebiter

8/1/2026 at 9:56:56 AM

To the contrary, isn’t the phrase the orange site used as a derogatory sense here? I have seen it used when authors think HN audience just criticize everything without substance.

by ludwigvan

8/1/2026 at 10:45:04 AM

> ...isn’t the phrase the orange site used as a derogatory sense here?

Seems likely, yeah.

> To the contrary...

Why are you -er- proposing the contrary or whatever? From my ~sixteen years of experience with HN commentary, the preemption seems totally fair and reasonable... as does getting some joy from seeing an author publish it.

by simoncion

8/1/2026 at 3:18:03 PM

wOrSE in EvERy WAY

(Is such an outrageous claim true? Click to generate ad revenue and maybe find out!)

by LearnYouALisp

8/1/2026 at 10:26:27 AM

The author critisizes one incorrect claim, but also makes claims that are in turn incorrect.

For instance:

> In truth, increasing the lifespan of a bulb makes it worse in every other way... but people think they want a long lasting lightbulb

Here the author writes "people think they want a long lasting xyz".

Well, people do not merely THINK they want a long lasting xyz. They really WANT it.

Why are projects not infinite in durability?

Of course, creating products with an infinite life time is not possible if only due to entropy alone; but also wear-and-tear damage. Nonetheless, the intrinsic quality and production process SHOULD CREATE PRODUCTS with a maximum life time. Some companies try that. Others deliberately avoid that in order to force customers to spend more on products they need to buy again and again and again. So the author is incorrect in the assumption; people WOULD ideally want a product that lasts as long as possible. Nothing wrong with that.

> Going off modern day prices, electricity costs around 0.10 [$/kW*h], so a 60 W lamp will consume 6$ of electricity over a 1,000 hour lifespan. Considering that such a lamp only costs around 3$, installing one that lasts longer but uses more power would be silly.

Here he is doing another mistake. First, those flat prices are not correct in general; instead, you must take in the opportunity costs. For instance, when I order new LED devices ("lamps"), I do not want to just order one. I want to order several, so that I have backup, and also so I don't have to go to purchase them several times. It is easier to do buy in some quantities, say, 8-10 or so. This lasts quite a long time for my use cases.

Ideally this would last a long time, because if not then I have to do more purchases, which in turn takes away time I could spend doing other things. So the cost comparison is different here. The ideal situation would be to never ever have to replace anything here. Or even capture in solar power and use that for the lamps.

> Instead of "making bulbs last longer", manufacturers spent huge amounts of time and money developing entirely new technology: fluorescent and LED lamps.

Well ... the EU forbade the old technology. While I understand this partially, I don't understand why a dictatorship forces me into having to use new technology. I have nothing against new technology at all, but the prices increased a LOT, so those technocrats ultimately forced me (and force me) to increase expenditure. I absolutely hate that. It is even more annoying with the current inflation. Why do a few people constantly try to abuse the huge majority? Perhaps LED lamps is not the real problem, but this tendency exists throughout society. I feel that both democracies and dictatorships abuse people, just in different amount of quality. We see this with the current trend to force age sniffing onto people. This is even clearer abuse.

by shevy-java

8/1/2026 at 3:49:29 PM

Europe has more cultural norms around the needs of the many outweighing the needs of the few, that's why. Would you add 50% to your city's energy use (at the time often generated locally, from coal - you know what that means - 50% more smoke in your lungs) just for better CRI? Would your neighbor agree to 50% worse air just so you could have better CRI?

by inigyou

8/1/2026 at 3:01:13 PM

[dead]

by crypttales

8/1/2026 at 1:04:30 PM

You ever hear of Dubai lamps? Apparently underdriving LEDs makes them last much longer.

by JTbane

8/1/2026 at 11:41:10 AM

Hey friends, why you do not pull off that second resistor in order to make your LED bulb to emit 30% of power but to have endless lifespawn? Your TV has a matrix of LED with endless lifespawn. Just don't be a blind consumer, also you can glue off the cap which blurs the light winning 10% of light, also you can drill the radiator for helping to cool the debuffed bulb even more.

by eimrine

8/1/2026 at 12:27:17 PM

thanx for downvotes, blind consumers. or check your favorite information sources about the second resistor and about how much lm/w gives 30%-ed lamp related to a 100%-ed one otherways.

by eimrine

8/2/2026 at 7:05:43 PM

Probably the down votes are from people who have no idea what you're going on about, or what it's got to do with an article entirely about the early history of incandescent bulbs.

by quietbritishjim

8/1/2026 at 2:46:49 PM

I don't understand this one. Yes, I want the long-life lightbulb. No, that doesn't somehow make it worse in every other way. The point is so obvious, I couldn't follow the logic in this argument. Nobody is buying or selling 120 Watt short-life bulbs to brag about their tremendous brightness. Long Life (LL) bulbs are not some kind of marketing ploy. We buy them understanding that we're going to minimize the hidden costs of making light, i.e. changing that bulb.

by 1970-01-01

8/1/2026 at 3:40:16 PM

Are you taking about modern long life bulbs? This article is taking about design choices for original format bulbs over 100 years ago.

by quietbritishjim

8/1/2026 at 3:02:59 PM

[dead]

by crypttales