A product page that says six hundred seventy nanometres is selling you something more specific than a red bulb. The number does the selling. Daniel's prompt this week is about where that number came from, whether it's doing any actual work in the sleep products that carry it, and how anyone could possibly tell. He's asking four things. One, where does six seventy originate. Two, why it's largely irrelevant to circadian biology. Three, what the dose gap is between the research and the product. And four, how a consumer verifies any of it, including whether an RGB bulb's red mode is even spectrally clean.
The honest answer starts with a confession about what six seventy was never about. It was never about melatonin. It was never about sleep. The number's entire reputation comes from a different field with a different mechanism and a different target tissue.
That's the part that feels almost unfair when you first learn it. The number is real. It's just not real for the thing being sold.
Six seventy comes out of photobiomodulation. The line of research on long wavelength red and near infrared light being absorbed by an enzyme in mitochondria called cytochrome c oxidase. That enzyme sits in the electron transport chain, the machinery that produces ATP. And its absorption spectrum peaks right around six seventy nanometres because of the copper centres in the enzyme complex.
So the number is a property of a copper atom's appetite for photons.
That's not a bad way to put it. The copper centres in cytochrome c oxidase absorb red light, and that absorption changes the enzyme's activity. The claim in the literature is that this improves mitochondrial function, particularly in cells that are under metabolic stress. And the specific application that made six seventy famous is retinal ageing.
The Glen Jeffery lab at UCL.
Right. Jeffery and his group have spent years on the idea that ageing photoreceptors are mitochondrially stressed, and that brief exposure to six seventy nanometer light can improve their function. The headline study was Weinrich and colleagues in twenty twenty. Older adults, brief daily exposure to a six seventy nanometre LED torch, and they measured improved cone photoreceptor function.
And this is where it gets strange, because that's a interesting finding and it has nothing to do with sleep. The claim is about mitochondrial function in retinal cells. Melatonin is a hormone regulated by a completely different photoreceptor system.
Completely different. The circadian system runs through melanopsin in the intrinsically photosensitive retinal ganglion cells. Those cells have peak sensitivity around four hundred eighty nanometres. They're tuned to blue light, not red. That's the system that suppresses melatonin when it sees short wavelengths and lets melatonin rise when those wavelengths are absent.
So six seventy is the signature of a copper centre in an enzyme. Four eighty is the signature of a photopigment in a ganglion cell. Two different proteins, two different jobs, and the product page is quietly borrowing the first number to imply the second outcome.
And the borrowing is the whole problem. The photobiomodulation studies are small. The Weinrich study was, I believe, around twenty four participants. The exposure protocol is very specific. A torch held at a set distance from the eye, delivering a defined irradiance for forty seconds daily. That's not a bulb glowing in the corner of a bedroom.
Forty seconds. That's the entire exposure. And the distance matters because light falls off with the square of distance. Hold the torch twice as far away and you get a quarter of the irradiance.
And the protocols are delivering milliwatts per square centimetre at the cornea. That's the unit that matters. An ambient room light, even a bright one, is delivering orders of magnitude less.
Let's put a number on that. A typical photobiomodulation protocol might deliver, say, forty milliwatts per square centimetre at the cornea for a defined duration. A night light across the room might deliver microwatts per square centimetre. That's a factor of a thousand or ten thousand.
And that's before you account for the fact that the eye is closed when you're asleep. The cornea isn't even exposed.
The wavelength can be genuine. The LED can really be emitting at six seventy. And the product can still be doing nothing with it. That's the distinction Daniel wanted made cleanly.
The number is real, the dose is fictional. That's the whole story in seven words.
So let's go back to the mechanism. You said copper centres. What's actually happening at the molecular level?
Cytochrome c oxidase is complex four in the electron transport chain. It's the final enzyme in the chain, the one that takes electrons and passes them to oxygen to make water. It contains two copper centres, called copper A and copper B, and two heme groups. The copper centres have absorption bands in the red and near infrared. When a photon hits that copper centre, it changes the oxidation state, and that alters the enzyme's catalytic activity.
So you're literally tickling a copper atom with a photon and the enzyme works differently.
That's the hypothesis. The exact mechanism is still debated, but the leading explanation is that red light photodissociates nitric oxide from the enzyme's active site. Nitric oxide binds to cytochrome c oxidase and inhibits it. Red light knocks the nitric oxide off, and the enzyme starts working again.
That's remarkably specific. And it explains why the effect would be largest in stressed tissue. If the enzyme is already working fine, knocking nitric oxide off doesn't do much. If it's inhibited, you get a rescue effect.
That's why the photobiomodulation literature focuses on ageing tissue, damaged tissue, hypoxic tissue. The effect is a restoration of function under stress, not a boost to healthy tissue. Which is another reason the bedroom bulb is a category error. Even if you got the dose right, the target is a stressed retina, not a sleeping brain.
And the melatonin system doesn't work through cytochrome c oxidase at all. It works through melanopsin, which feeds into the suprachiasmatic nucleus, which drives the pineal gland. Different pathway entirely.
The circadian case for red light is almost embarrassingly simple compared to the photobiomodulation story. What protects melatonin is the absence of short wavelengths and low overall intensity. That's it. Once you're out at the long end of the spectrum, six twenty versus six sixty versus six seventy is nearly irrelevant to the outcome. The melanopsin system barely sees any of them.
So the product that says six seventy is borrowing a number from a mechanism it's not using, to imply a benefit it's not delivering, at a dose it's not providing.
And that's before we get to the question of whether the LED is even emitting what it says.
That's the consumer verification question. Daniel wanted the practical toolkit. How would you actually tell?
The first question is whether the manufacturer publishes a spectral power distribution chart at all. That's a graph of intensity versus wavelength. It shows you the peak, the bandwidth, and whether there's any emission at other wavelengths. Most manufacturers of sleep bulbs don't publish these. Which is itself informative.
If you had a clean spectrum, you'd show it.
A spectral power distribution chart takes five minutes to generate with a spectrometer. If a company is making a precise wavelength claim and not showing the chart, the most charitable interpretation is that they don't think it would help the sale.
And the chart would reveal something most people don't realise. Six seventy is a peak, not a line. An LED has a bandwidth of some tens of nanometres. The stated figure describes the centre of a distribution.
A typical red LED might have a full width at half maximum of twenty to thirty nanometres. So a six seventy LED is really emitting from about six fifty five to six eighty five, with the peak at six seventy. That's fine for most purposes, but it matters when you're making a precision claim. The number implies a sharpness the physics doesn't have.
And then there's the question of what kind of red LED it actually is. A true narrowband red emitter is a semiconductor die that emits directly in the red. A phosphor-converted amber is a blue or violet LED with a phosphor coating that converts some of the light to longer wavelengths. The phosphor version has a much broader spectrum and often leaks some of the pump wavelength.
The difference shows up in the spectral chart. A narrowband red emitter has a single clean peak. A phosphor-converted amber has a broad hump and often a secondary peak in the blue or violet where the pump leaks through. If you're buying a red bulb for circadian reasons, that leakage matters. A little blue light at the wrong time is doing more to your melatonin than all the red light in the world.
Which brings us to the metrics that actually answer the sleep question. Melanopic EDI and melanopic ratio.
Melanopic EDI is melanopic equivalent daylight illuminance. It's a measure of how much a light source stimulates melanopsin, weighted by the melanopsin action spectrum. Melanopic ratio is the melanopic EDI divided by the standard photopic illuminance. A low melanopic EDI means the light is not stimulating the circadian system.
So for a sleep light, you want low melanopic EDI. And the colour name tells you nothing about that. A bulb called warm white can have a surprisingly high melanopic content depending on the phosphor mix.
And a bulb called red can have a surprisingly high melanopic content if it's a phosphor-converted amber with a blue pump leaking through. The colour name answers nothing. The melanopic EDI answers the actual question.
What about the cheap verification tools? Daniel mentioned spectrometer apps and diffraction gratings.
A diffraction grating is a piece of plastic with thousands of microscopic lines that splits light into its component wavelengths. You can buy one for a few dollars. Hold it up to a light source and you'll see the spectrum spread out. A narrowband red LED shows a single clean red band. A phosphor-converted amber shows a broader smear, and if the pump leaks you'll see a blue or violet line alongside the red.
You can do this at home with a cereal box and a CD, honestly. The CD's grooves act as a diffraction grating.
The spectrometer apps are less reliable. Phone cameras have their own spectral response curves and software corrections that make absolute wavelength measurements unreliable. But they can still show you the shape of a spectrum. A narrow peak versus a broad hump versus a leaky pump. For the question of is this a true narrowband red emitter or a phosphor amber, that's often enough.
So the consumer toolkit is: look for the spectral power distribution chart, know that its absence is information, understand that the stated wavelength is a peak not a line, ask about melanopic EDI, and if you really want to check, a diffraction grating will show you the shape of the spectrum.
And then the final question Daniel asked, which is the RGB comparison. On spectral grounds specifically.
We've been down the brightness argument before. An RGB bulb in red mode fires one of three dies, and because red dies are the least efficient in the package, you get a fraction of the rated output. An eight hundred lumen bulb might give you a hundred sixty lumens in red. That's a brightness problem. Daniel's new question is different. Is the red mode actually spectrally clean?
And the honest answer is that it depends on the bulb, but the physics is more forgiving than you might expect. A true RGB LED uses three separate semiconductor dies, one red, one green, one blue. When you set it to red mode, the green and blue dies are supposed to be off. The red die emits directly, not through a phosphor. So the red emission is genuine narrowband long wavelength light.
The question is whether the green and blue dies actually go fully dark. And that's a driver question, not a die question.
Right. The dies are controlled by pulse width modulation. The driver switches each channel on and off rapidly to control brightness. In theory, setting green and blue to zero means those channels are off. In practice, some drivers leak a small current even at zero, or the PWM signal doesn't go fully to zero, or there's crosstalk between channels.
And at deep dim levels, the PWM behaviour gets weird. Some drivers don't dim smoothly down to zero, they step down to a minimum level and then jump to off. So there's a range where the green and blue channels are technically on at a very low level.
But here's the thing. At these intensities, does any of it register against the melanopic threshold? The melanopsin system has a threshold. Below a certain melanopic EDI, the light simply doesn't trigger a measurable melatonin suppression response. A tiny leakage current in a green die at one percent of one percent of its rated output is almost certainly below that threshold.
So the honest landing is that spectrally, RGB red mode is probably fine. The red die is a genuine narrowband emitter. The leakage, if it exists, is likely below the melanopic threshold. The real objection remains the brightness one. It's simply too dim to live with for most people.
And that's a different argument from the spectral one. The spectral argument against RGB red mode is weak. The brightness argument is strong. Confusing the two is how you end up buying a six seventy night light that's also too dim to live with, but now you've paid extra for a number.
So if the number is borrowed and the dose is wrong, and the RGB question lands on brightness rather than spectrum, what's the actual takeaway for someone standing in front of a shelf of sleep bulbs?
Three questions. What claim is this number from? Is that the claim being made here? And is the dose anywhere near what the claim requires? That's the toolkit. That's what lets you look at any wavelength number on any lighting product and know whether it's doing work or just doing marketing.
Let's run the six seventy bulb through it. What claim is the number from? Photobiomodulation of retinal mitochondria. Is that the claim being made here? No, the claim being made is better sleep through melatonin protection. Is the dose anywhere near what the photobiomodulation claim requires? Not even close.
Three questions and the product fails all three. The number is real science, but it's real science from a different room of the building, bolted onto a product that doesn't deliver its conditions.
And the precision is doing the work. Six seventy sounds like a specification. It sounds like someone measured something and found an optimum. It sounds like the difference between six seventy and six twenty must matter, because why else would they print it to three significant figures?
The precision is a credential. It's borrowed authority. Six twenty, six sixty, six seventy, all the same to melanopsin. The number is doing marketing work, not biological work.
That's the thing I keep turning over. The number is true and it doesn't matter. Both things at once.
And that's harder to communicate than either a debunk or an endorsement. The photobiomodulation research is genuine. The Weinrich study measured a real effect in a real population with a real protocol. The mechanism has a plausible molecular basis. None of that transfers to a night light.
The transfer is the sleight of hand. It's not that the science is fake. It's that the science is being asked to testify in a case it has nothing to do with.
Let me be careful about the evidence quality, because I don't want to overstate the photobiomodulation side either. These are small studies. The effect sizes were modest. The protocol was very specific and the follow-up was short. This is preliminary, promising work, not settled clinical practice.
So even within its own field, six seventy is a research wavelength, not a therapeutic standard.
Right. The literature is exploring whether this works. It's not at the point where you'd prescribe it. And the idea that a consumer product is delivering a therapeutic dose is, frankly, absurd on its face.
The dose math is the part I want to sit with, because it's the most transferable skill here. Milliwatts per square centimetre. That's power per unit area. The photobiomodulation protocols need tens of milliwatts per square centimetre at the target tissue for a defined duration. An ambient room light delivers, what, single digit microwatts per square centimetre?
Depends on the bulb and the distance, but yes, the gap is three to four orders of magnitude. And it's not a gap you can close by leaving the bulb on longer. The photobiomodulation literature has a biphasic dose response. Too little does nothing, too much does nothing, and the therapeutic window is narrow.
So you can't just run the night light all night and call it a forty second exposure repeated many times. The dose rate matters, not just the cumulative dose.
That's the biphasic response. It's not like vitamin D where more sun equals more synthesis up to a point. Photobiomodulation has a sweet spot. Below it, no effect. Above it, no effect or even inhibition. The forty second protocol is forty seconds for a reason.
And the eye is closed. I keep coming back to that. The retina is the target tissue in the photobiomodulation studies. The light has to reach the retina. A closed eyelid attenuates light significantly, especially at longer wavelengths where tissue absorption is higher.
Red light penetrates tissue better than blue light, which is one of the reasons it's used in photobiomodulation. But an eyelid is still a barrier. And a pillow is a better one.
So the six seventy night light is failing on every axis. Wrong mechanism, wrong dose, wrong target, wrong time.
And yet the number keeps appearing on product pages, because it works. It works as marketing. It converts a vague claim into a specific one. And specificity sells.
I want to talk about the spectral power distribution chart one more time, because I think that's the single most useful consumer tool in this whole discussion.
It's the difference between a claim and a measurement. A claim is a sentence. A spectral power distribution chart is a measurement. It shows you the actual emission of the device, not the marketing intent.
And the absence of the chart is information. If a company is selling a precise wavelength claim and not showing the measurement, the most likely reason is that the measurement would complicate the claim.
The chart would show the bandwidth, which immediately undermines the precision. It would show any secondary peaks, which immediately undermines the purity. It would show the actual peak wavelength, which might be six sixty five or six seventy five, not six seventy.
And the melanopic EDI is the metric that actually answers the sleep question. Not the wavelength, not the colour name, not the marketing copy. The melanopic EDI.
Melanopic EDI is measured in lux, same as regular illuminance, but weighted by the melanopsin action spectrum. A low melanopic EDI means the light is not stimulating the circadian system. For a sleep light, you want melanopic EDI as low as possible while still being able to see.
The melanopic ratio tells you how circadian-active the light is relative to its visual brightness. A warm white bulb with a high melanopic ratio is sneakier than a red bulb with a low one, because it looks dim but it's still poking your circadian system.
The colour name answers nothing. Warm white, soft white, amber, red, sunset, candle. These are marketing categories. Two bulbs with the same colour name can have wildly different melanopic content depending on the phosphor mix.
The consumer who wants to do this properly ignores the wavelength number, ignores the colour name, and looks for two things. A spectral power distribution chart and a melanopic EDI figure. And if the manufacturer doesn't publish either, that's the answer.
If you want to check rather than trust, a diffraction grating is the cheap tool. It won't give you a calibrated measurement, but it will show you the shape of the spectrum. Single clean peak versus broad hump versus leaky pump. For the question of what kind of red this actually is, that's often enough.
Let's talk about the phosphor-converted amber for a moment, because I think it's the sneakiest product in this space.
A phosphor-converted amber uses a blue or violet LED as the pump, with a phosphor coating that converts some of the pump light to longer wavelengths. The result looks amber or red to the eye, but the spectrum is broad and often has a residual blue peak where the pump leaks through.
It looks red but it's actually a blue LED wearing a red coat.
The blue peak is exactly the thing you're trying to avoid. A little blue light at the wrong time does more to your melatonin than all the red light in the world. The melanopsin system is exquisitely sensitive to short wavelengths.
This is why the spectral chart matters. A narrowband red emitter and a phosphor-converted amber can look identical to the eye. The chart shows the difference immediately.
The diffraction grating shows it too. The narrowband red shows a single clean band. The phosphor amber shows a broad smear with a blue line at one end. You can see the lie with a piece of plastic.
The RGB question lands here too. A true RGB LED uses three separate dies. The red die is a genuine narrowband emitter. It's not a phosphor-converted anything. So on spectral grounds, the red mode of an RGB bulb is actually cleaner than a phosphor amber.
The red die in an RGB package is typically an aluminium gallium arsenide or aluminium gallium indium phosphide semiconductor. It emits directly in the red with a bandwidth of maybe twenty nanometres. That's a genuine narrowband source.
The spectral argument against RGB red mode is weak. The red die is clean. The question is whether the green and blue dies go fully dark, and at these intensities, does any leakage register against the melanopic threshold?
The honest answer is that the leakage, if it exists, is almost certainly below the melanopic threshold. A few microamps of leakage current in a green die produces a few nanowatts of optical power. The melanopsin system doesn't register that.
Spectrally, RGB red mode is fine. The real objection remains the brightness one. It's too dim to live with.
That's a different argument. The brightness argument is settled. The spectral argument is new, and it lands in RGB's favour. Which is a rare outcome in these discussions.
It's almost disappointing. You want the RGB bulb to fail on every axis, and it turns out the physics is more forgiving than the marketing.
The physics doesn't care about our desire for a clean debunk. The red die is a red die. The leakage is negligible. The problem is that you're getting a hundred sixty lumens from an eight hundred lumen bulb, and a hundred sixty lumens is not enough to read by.
The person who wants a red light for sleep should buy a dedicated red LED, not an RGB bulb, but for brightness reasons, not spectral ones.
Or a tunable white bulb that can go down to twenty two hundred kelvin. That's the recommendation from the earlier discussion and it still holds. The intensity matters more than the colour temperature, and a dim warm white is often better than a bright red.
Let's pull this together, because I want to make sure the listener finishes with the toolkit.
Three questions. What claim is this number from? Is that the claim being made here? And is the dose anywhere near what the claim requires?
What claim is six seventy from? Photobiomodulation of retinal mitochondria. Is that the claim being made on the sleep bulb? No. Is the dose anywhere near what the photobiomodulation claim requires? Not by three orders of magnitude.
The number fails all three questions. And that's the test. Run any wavelength number on any lighting product through those three questions and you'll know whether it's doing work or doing marketing.
The precision of six seventy is real science. But precision without relevance is just a number.
The relevance is the thing that gets lost. The number is real, the science is real, the mechanism is real. None of it transfers to a bulb in a bedroom. The transfer is the sleight of hand.
I want to go back to the Weinrich study for a moment, because I think the details matter. Forty seconds daily. A specific torch. A specific distance. Older adults. And they measured cone photoreceptor function. That's a visual function measure, not a sleep measure.
Right. The study was about whether red light could improve the function of ageing photoreceptors. The outcome was improved cone function, which is about visual acuity and contrast sensitivity in older eyes. It's a fascinating finding about retinal ageing. It says nothing about melatonin, sleep quality, or circadian rhythms.
Even the headline finding is about eyesight, not sleep. The sleep connection is pure inference on the part of the marketing department.
It's an inference that doesn't survive contact with the biology. The melatonin system doesn't run through the photoreceptors that the photobiomodulation studies target. It runs through the intrinsically photosensitive retinal ganglion cells, which are a different cell type entirely.
The retina has at least three photoreceptor systems. Rods and cones for vision, and the intrinsically photosensitive retinal ganglion cells for circadian entrainment. The photobiomodulation work targets the cones. The sleep work targets the ganglion cells. Different cells, different pigments, different wavelengths.
The dose requirements are different too. The circadian system responds to very low light levels. A few lux of blue light is enough to suppress melatonin. The photobiomodulation system requires milliwatts per square centimetre. The dynamic ranges are completely different.
The six seventy bulb is simultaneously too dim for its borrowed claim and too bright for its actual claim. Too dim to do photobiomodulation, too bright to be a perfect sleep light.
That's a nice way to put it. The number is stranded between two fields, not fulfilling the requirements of either.
Let's talk about the bandwidth issue one more time, because I think it's the most counterintuitive part for a layperson.
Most people think an LED emits at a single wavelength. The number on the datasheet reinforces that. But an LED emits a distribution. The number is the peak of that distribution, and the distribution has width.
A typical red LED has a full width at half maximum of twenty to thirty nanometres. So the six seventy LED is emitting from about six fifty five to six eighty five. The number describes the centre of a bell curve, not a spike.
That's fine for most purposes. The circadian system doesn't care about the difference between six fifty five and six eighty five. But it matters when the marketing copy implies a precision the physics doesn't have.
The precision is a rhetorical device. Six seventy sounds like a specification. It sounds like someone measured something and found an optimum. It sounds like the difference between six seventy and six twenty must matter.
In the circadian context, it doesn't. The melanopsin action spectrum is so far away from six seventy that the exact wavelength in the red is irrelevant. What matters is that there's no short wavelength light, and that the intensity is low.
The sleep bulb that says six seventy is using precision to imply relevance. And the relevance is the thing that's missing.
The number is doing marketing work. That's the whole story.
I think we should talk about what a good sleep light actually looks like, because the debunk is only half the job.
A good sleep light is dim, warm, and free of short wavelengths. That's it. The wavelength number doesn't matter much. What matters is the melanopic EDI, which should be as low as possible while still allowing you to see.
A dim warm white at twenty two hundred kelvin is often better than a bright red, because the intensity matters more than the colour. The melanopsin system responds to intensity as well as wavelength.
The practical recommendation is to use the dimmest light you can function by. If that's a red LED, fine. If it's a warm white at ten percent brightness, also fine. The goal is low melanopic EDI, not a specific wavelength.
The six seventy bulb is solving a problem you don't have with a mechanism it doesn't use at a dose it doesn't deliver.
Charging extra for the number.
Let's go back to the consumer verification question one more time, because I want to make sure the practical toolkit is concrete.
Step one. Look for a spectral power distribution chart. If the manufacturer publishes one, read it. Look for the peak, the bandwidth, and any secondary peaks. If they don't publish one, that's information.
Step two. Look for melanopic EDI or melanopic ratio. These are the metrics that actually answer the sleep question. If the manufacturer doesn't publish them, ask why.
Step three. If you want to check rather than trust, a diffraction grating will show you the shape of the spectrum. Single clean peak versus broad hump versus leaky pump. That tells you whether you're looking at a true narrowband red emitter or a phosphor-converted amber.
Step four. Run the three questions. What claim is this number from? Is that the claim being made here? Is the dose anywhere near what the claim requires?
That's the toolkit. It works for any wavelength number on any lighting product.
I want to make one more point about the dose math, because I think it's the most transferable skill here. The unit is milliwatts per square centimetre. Power per unit area. And the key insight is that area matters as much as power.
Light falls off with the square of distance. A torch held at ten centimetres delivers a hundred times the irradiance of the same torch held at one metre. The photobiomodulation protocols specify the distance because the distance is the dose.
The bedroom bulb is across the room. The distance alone kills the dose, even before you account for the lower power output.
The photobiomodulation protocols are delivering tens of milliwatts per square centimetre at the cornea. A night light across the room is delivering single digit microwatts per square centimetre at the pillow. The gap is three to four orders of magnitude.
You can't close that gap by leaving the bulb on longer. The biphasic dose response means the dose rate matters, not just the cumulative dose.
The forty second protocol is forty seconds for a reason. The therapeutic window is narrow. Too little does nothing, too much does nothing. An ambient bulb is so far below the window that no amount of duration compensates.
The six seventy night light is not a diluted version of the photobiomodulation protocol. It's not even in the same physical regime.
It's the difference between a medical laser and a Christmas light. Both emit red light. Only one is delivering a therapeutic dose.
Hilbert: The number's the thing. I worked for a lighting manufacturer in the late nineties, and we had a box of rejected LED dies with off-spec peak wavelengths. The marketing department's first question was always whether we could relabel them as a special therapeutic wavelength instead of scrapping them.
Of course they did.
That's the whole industry in one sentence.
Hilbert: We sold full spectrum bulbs with precise colour temperature claims years before anyone had measured what full spectrum meant. The marketing team invented a circadian friendly label in nineteen ninety eight. There was no science behind it. They just noticed that precise numbers sold better than vague descriptions.
The colour temperature wars. That's exactly the same dynamic. A precise number implies a precise benefit, whether or not the number has anything to do with the benefit.
Hilbert: The rejected dies were the interesting part. A die that was supposed to peak at six sixty might come out at six forty five. Still red, still worked fine for indicator lights. But the marketing team saw a story. Special therapeutic wavelength. Premium price.
Nobody could check. That's the thing. In nineteen ninety eight, a consumer had no way to verify a spectral claim. The number was pure assertion.
Hilbert: Nobody checked. We sold the relabelled dies as premium products. They were the same dies we'd been selling for indicator lights, just with a different number printed on the box.
That's the part that connects to now. The verification tools exist now. The diffraction grating costs a few dollars. The spectral charts are easy to generate. The information is available. But most consumers still don't check.
Hilbert: The box said six forty five and the customer believed it. Why would they check? The number was specific. Specific sounds true.
Specific sounds true. That's the whole psychology of it.
Hilbert: I want to know one thing. If I take one of those old relabelled dies and run it through the three questions, what happens? What claim is six forty five from? Nothing. It's from a manufacturing tolerance problem. Is that the claim being made? No, the claim was therapeutic. Is the dose anywhere near what the claim requires? There was no dose. It was a number on a box.
It fails all three questions. Same as the six seventy night light.
Hilbert: The test works. That's what I wanted to know.
The test works on a thirty year old relabelled die and on a six seventy night light. That's a decent validation.
Hilbert: The box is still in my flat somewhere. Six forty five nanometres. Special therapeutic wavelength. I kept one.
The wavelength was real. The die really did peak at six forty five. The number was true and it didn't matter.
That's the throughline of this whole episode. The number is real and it doesn't matter.
Hilbert: They sold the whole batch. Every single one.
That's the open question, isn't it. As the photobiomodulation evidence evolves, will the dose problem be solved by better consumer devices, or will the marketing always outrun the physics?
The marketing has a thirty year head start.
Hilbert: The marketing doesn't need to solve the dose problem. The marketing just needs the number.
The number is cheap. That's the thing. The number is free. The dose is expensive. The number will always win.
The one thing I want the listener to keep is the three questions. What claim is this number from? Is that the claim being made here? Is the dose anywhere near what the claim requires? Run any wavelength number on any lighting product through those three questions and you'll know whether it's doing work or doing marketing.
The sharper version. The precision of six seventy is real science. Precision without relevance is just a number.
Thanks to our producer Hilbert Flumingtop for keeping us on schedule, even when the schedule involves a box of relabelled dies.
This has been My Weird Prompts. Email us at show at my weird prompts dot com if you've got a wavelength number you want interrogated.
We'll be back soon.
