[r4indeer]

> Now an old fashioned light bulb shouldn't be expected to last a decade, but an oven?

Funnily enough, there actually was the Phoebus cartel [1] which sought to reduce the lifespan of incandescent light bulbs to around 1,000 hours and raise prices.

[1] https://en.wikipedia.org/wiki/Phoebus_cartel

[gnicholas]

I recall reading a quasi-debunking of this. I didn't go down the rabbit hole far enough to understand all the details, but it seems the situation was more complicated than just corporate greed. [1]

The topic has been discussed here in the past a few times, including [2] and [3]

1: https://readmedium.com/en/the-phoebus-cartel-was-never-reall...

2: https://news.ycombinator.com/item?id=21596792

3: https://news.ycombinator.com/item?id=17606748

[0xEF]

I would not call any of that a debunking, even quasi. Just a different dance around the same hard-to-swallow pill.

Company X makes a great product that everyone only needs one of and lasts a long time. Over time, the market starts to dwindle and. Company X is going broke. Now, Company X must either invest in innovation or reduce the lifespan of its current offering.

There's nothing inherently evil about this concept, but we tend to want to chalk it up to greed when Company X really just wants to survive and make a profit, which I suppose is the point.

The problem is the concept is ripe for abuse. If Company X makes their product worse, but starts charging more while laying off employees, posting record profits during recessions, adopts unnecessary subscription models cosplaying as continued service and development, etc...now we get to the greed part. There seems to be a line between designing a product to secure the longevity of Company X and straight up using your customers as micro-transaction ATMs with planned obsolescence. Some companies conspire to cross it.

[mmkhd]

Nice video from the well known channel Technology Connections: https://www.youtube.com/watch?v=zb7Bs98KmnY => It's complicated. Yes, there was a cartel, but it was not all bad. There were legitimate reasons to go for 1000h light bulbs.

[dennis_jeeves2]

>but it seems the situation was more complicated than just corporate greed

Then their strategy worked - if you really believe it's more complicated. Haven't investigated this particular subject, but many others subjects are _made_ complicated to achieve a particular outcome. Along lines of: 'let's protect the children' argument.

[stronglikedan]

I'm convinced this happened recently with LED bulbs as well, even though I've found no definitive proof. The LEDs I installed in my house 10-12 years ago are still going strong, but every newer one I've purchased gives up the ghost within a couple of years. And I only purchase brands with a good reputation, like Feit and the like.

[surge]

The rule of thumb I've found with light bulbs is similar to the Boots Theory (https://en.wikipedia.org/wiki/Boots_theory), which is you need to spend at least $8-$10 on a bulb to get something that will actually last. Feit is good but its hit or miss on life span, especially when I get them close to the same price and incandescent, often times its the little A/C to DC converter that dies (really need DC light circuits or dedicated converter in the light fixture). I feel its worth spending the extra to not replace them.

[TonyTrapp]

There's a variety of reasons:

- Lower-quality components (especially capacitors) being used to meet the lower price point. This is by far the most common failure mode I have experienced, it's never the LEDs dying but the power supply.

- Higher-quality LED light is usually result of driving the LEDs harder, causing them to fail earlier.

- Probably some other reasons too.

[404mm]

I recommend watching this:

https://hackaday.com/2021/01/17/leds-from-dubai-the-royal-li...

Only shows you bulbs can be made well and last long. But those are not for you. (Assuming most readers here are not Saudi)

[CapitalistCartr]

The LEDs themselves are made in a handful of factories around the World and are usually robust. The power supplies are the weakness. Each bulb manufacturer makes their own, and it's a race to the bottom.

[Scoundreller]

At least you have a fighting chance of fixing your LED bulb, unlike an incandescent.

Usually they’re over-driven and you can jump a burned out LED and scrape off a bit of a resistor to reduce the amount of current going through to (over-)account for the reduced current need.

https://youtu.be/JBKF7rKB3zc

[seventyone]

There is definitive proof. They over-drive the LEDs which is why they die so quickly. If they were under-driven they last much much longer. It's the heat that kills them IIRC.

[Youden]

It wasn't as simple as them wanting to make more money: https://www.youtube.com/watch?v=zb7Bs98KmnY

Key points from an AI summary:

- Incandescent bulbs had to balance factors like light output, efficiency, and lifespan - hotter filaments produced brighter, whiter light but reduced bulb lifespan.

- Longer-lasting bulbs were less efficient and produced dimmer, yellower light, so they were not simply "better" products being suppressed.

- The 1,000 hour target was a reasonable compromise that balanced these competing priorities, not necessarily a sinister plot.

- Even after the Phoebus cartel dissolved, the 1,000 hour lifespan remained the industry standard for general-purpose incandescent bulbs.

[jimmydorry]

You're missing the fact that the Phoebus cartel fined members that sold lightbulbs lasting longer than 1,000 hours. Aftr reaching a stable equilibrium, it's not surprising that 1,000 hours remained the industry standard. It drove sales!

[1] https://www.bloomberg.com/opinion/articles/2022-05-04/cheeri...

[AnthonyMouse]

This sounds like classic corporate bamboozlery. Find some real trade off that actually exists and then exaggerate its importance or pretend that no other solutions can be found when in fact they don't want solutions because the problem is profitable.

Undoubtedly there are some alternate materials you could make a light bulb out of that present a trade off between longevity and efficiency. But there will also be materials that last a long time and have high efficiency. Moreover, even if they want to use the filament material that emits whiter light and then burns up faster, they could then use more of it so it still doesn't burn out quickly. But they don't want to do that, because it would cost marginally more and more importantly then you wouldn't have to buy as many light bulbs.

It's no good to pretend this isn't possible. There isn't an inherent trade off between brightness and efficiency, because inefficiency is just the percentage of the electricity that goes to producing heat rather than light. At the same power consumption, a more efficient bulb is brighter. LEDs are rated as "100W equivalent" even though they consume ~20W. And the LEDs themselves last far longer than the equivalent incandescent light, but then they purposely combine them with a power converter that burns out much sooner. It's marketing, not physics.

[macNchz]

Well, without the cartel there could have presumably been bright white, 1000 hour bulbs on the shelf next to dim yellow 2500 hour bulbs, and people could have chosen accordingly.

Additionally, the companies set up a whole compliance regime with bulb testing and fines, not for bulbs being too dim, but for bulbs that lasted too long, which I think clarifies the intent more than anything else.

[HPsquared]

It's a question of temperature.

Hotter filament gives more efficient and whiter light (the black body radiation has more visible and less infrared), but the hotter filament doesn't last as long (faster evaporation rate).

It's perfectly possible for end users to use a dimmer switch to make incandescent lamps last much, much longer at the expense of less light and a "warmer" colour.

Lifespan is very, very sensitive to the temperature.

[aitchnyu]

Which cartel is making consumer bulbs and streetlights with advertised 50000 hour led but with 5000 hour drivers? Indian market BTW.

[notoverthere]

There's also the Centennial Light [1], a light bulb made in the late 1890s. It was first lit in 1901 and it's still alight today.

[1]: https://en.wikipedia.org/wiki/Centennial_Light

[kibwen]

The centennial bulb is less a lightbulb than it is a toaster oven. Planned obsolescence is real, but the centennial bulb is not evidence of it.

[Moldoteck]

i guess a lot of lights will work a lot longer if powered at such low voltage and not switched on/off like most ppl do, but this would reduce a lot nr of cases where such a light can be used

[afiori]

Which to be honest has the power efficiency of a dim campfire

[hoseja]

It barely glows. The "lightbulb cartel" was basically a consumer protection because barring major inventions, any deviation from the thousand hour lightbulb would have severe drawbacks in terms of power efficiency or light output.

[Dylan16807]

There was, but also the hotter 1000-hour bulbs are more efficient, and the alternative of 2500 hours still gets you nowhere near a decade of use.

[afiori]

Planned obsolescence is very real, but the reality of incandescent light bulbs means that lifespan, efficiency, and luminosity are not independent.

The 1000 hours limit is in practice a lower bound to a combination of luminosity and efficiency

[Onavo]

LED diodes can theoretically last decades given the correct drivers (current and heat needs to be significantly limited), unfortunately they are the very definition of planned obsolescence.

https://www.newyorker.com/business/currency/the-l-e-d-quanda...

A stable DC current and temperature limited LED can easily last decades.

[jajko]

Literally in that linked wiki article:

> "A longer life bulb of a given wattage puts out less light (and proportionally more heat) than a shorter life bulb of the same wattage"

As long as we can recycle (or at least safely get rid of) the burned out ones I'd say its a win from ecology perspective, and at least in some cases also for end users. But this wasnt the main driver of the change, it was the good ol' corporate greed as per the same wiki page.

[mratsim]

Reminded me of Bastiat's candlestick maker's petition:

http://bastiat.org/en/petition.html

[tiberious726]

The candle that burns twice as bright lasts half as long. The converse is just as true of tungsten filaments as it is of candles.

I know this is a common pop-history thing to cite on the Internet, but I would think hackernews would understand the benefits of standardization.

If every brand's lightbulb has different luminousities how on earth would architects decide how to space fixtures?

This "cartel" is how we avoided a dimness war, like the loudness war we had in digital music a decade or so back