Blue light filters don't work – controlling total luminance is a better bet

I found the basic premise of this blog post to be incredibly flawed. The author seems very sure of himself that blue light filters don't work, but making arguments related to cell types and emissions spectra and circadian rhythms is not the way to make a conclusive argument in a topic like this. Science is littered with recommendations about things that "plausibly" made sense, but that turned out to be flawed or just absolutely wrong when actually put to a real, scientific test. One example most people are familiar with: the recommendation against eating eggs in the 90s was based on the fact that eggs have a lot of cholesterol, and we knew high LDL levels in blood were associated with a greater risk of vascular and heart problems. So, "logically", it seemed that limiting dietary cholesterol would reduce heart disease. Except when scientists actually tested those recommendations, they turned out to be largely wrong - when you eat a lot of cholesterol, for most people their body's natural production of cholesterol goes down, so unless you're in the small subset of people who are particularly sensitive to dietary cholesterol, eating eggs is fine.

Making recommendations based solely on a theoretical mechanism of action is bad science. The only way to actually test this is with a study that looks at different types of light restriction and its effect on sleep. Obviously it's kind of impossible to do a blinded study for blue light filters, but you could get close by testing various permutations of light changes (e.g. total luminescence, eliminating only very specific wavelengths, etc.)

As another commenter said, it may be a placebo effect, but if it is, who cares? All I care about is that I get a better night sleep, and as someone (unusual among programmers I know) who really doesn't like dark mode, a screen reddener greatly helps me at bedtime.

>It doesn’t make any sense in 2026 that Gmail doesn’t have a dark mode

I've been using dark mode on gmail for years, not sure what OP is talking about here.

But also, my sleep quality got much better when I turned on f.lux. And it got better still when I added a second light to my bathroom that can do a 1800K super-warm light (that's also very dim).

And as an added pro-tip, I use f.lux during the day to cut my color temp to 5900K (instead of the default 6500K) and it made a huge difference for how long I could work without getting tired eyes.

They absolutely help my eyes not be so strained. If its placebo, its a working placebo.

>Are people actually using Night Shift? >Aggravatingly, yes.

What is the authors problem lol? It feels a lot better on eyeballs to use warm light things. Why does he care?

My overall take (elephant in the room): Blue light filters don't work, it depends on what you do & how you do it.

For example, most people keep watching/scrolling Instagram Reels and TikTok videos. They keep stimulating the brain constantly, not just at electrical level but also in emotional/chemical level too.

I have seen people who are addicted and cannot get rid of the addiction. This is not only the dopamine-boost, it has deeper connections of neuro-chemical stimuli. Just observe around you; people pick up their phone to directly open Insta/TikTok, start scrolling right away every 5-10 seconds. (watching stories included too)

This is to some extent that when you mention even the possibility of such addiction and abnormal behavior, one gets outright resistance and denial of addiction itself. Much like substance abuse...

My point is, majority of the population watches/scrolls these, needing 10g of melatonin to fall asleep.

Obviously if I get engaged in an interesting stuff continuously, the existence of blue light does not matter that much. It matters if/when I am reading a novel which is in a mediocre chapter where nothing that interesting going on. The existence of blue-light or lack thereof may tip the scale at that point.

You can just do things. Not everything needs a study, you don’t have to justify yourself to anyone!

Try things, if you like them, do them!

Try not living a neurotic “study” based life, I am trying it and its pretty great!

the real variable is probably what you're doing before bed, not the wavelength. scrolling social media keeps the brain actively processing new stimuli -- notifications, comparisons, emotional content. reading on a kindle with blue light filter probably sleeps better than watching youtube with it on. the luminance thing the author mentions points in this direction too.

Interesting take for me is that melatonin (over-)usage can be severely harmful for the individuals.

  ... over-the-counter melatonin supplements can contain anywhere between 10 to 30 times as much melatonin as is optimal to maintain circadian hygiene. If you have ever taken melatonin and got immediately knocked out cold, had weird dreams and woke up in the middle of the night sweaty or shivering, you likely took too much—which, to be clear, is not your fault, it’s the default in the US and Canada. The mega-doses in stores serve as hypnotics (punches you to sleep), but wreck sleep architecture. The right dose is ~0.3 mg, which is hard to find in pharmacies but can be found online.

Night shift seems to have a very strong causal effect on my sleep cycles. Up until about ten years ago I was a night owl, rarely falling asleep before midnight and rarely waking up before 8. Then I started getting serious about light hygiene and using night shift and now I'm a serious day person, rarely staying awake after 11 and rarely waking up after 7. But the real clincher is that when I travel I don't change the time zone on my computer (because it screws up my calendar). But my sleep cycle continues to track my home time zone for a very long time. I life in California, but at the moment I'm in Hawaii. I've been here three weeks so far. At home I'd fall asleep around 11 and wake up around 7, but here I'm getting sleepy at 9 and waking up at 5.

My wife, on the other hand, is a hard-core night owl even with night shift. So apparently there is a lot of individual variation.

This article has inspired me to do a control experiment by switching night shift off. Check back here in a week or so for the results.

> Unless your strategy is to create a photo-lab-like screen in pure black and red, or wear deep-red-tinted glasses, it’s unlikely that a pure colorshift strategy will cut out that big of a chunk of the spectrum.

I absolutely think this is the right approach. The glasses which do 'blue light filtering' which barely change your perception are clearly placebo, but a very strong redshift I think is obviously a different creature.

> Unless your strategy is to create a photo-lab-like screen in pure black and red, or wear deep-red-tinted glasses, it’s unlikely that a pure colorshift strategy will cut out that big of a chunk of the spectrum.

The writer is dismissing this out of hand but to me this sounds like a great idea.

I have my phone in monochrome (i.e. greyscale) mode and just subjectively it's much easier to look at especially at night. I have it at the lowest brightness and it's still very readable. Human eyesight is basically monochrome in low light settings anyway.

I bought some amber glasses from blublocker.com[1], because they link to a research paper that actually measured how much of each wavelength their filters allow (as well as other brands). They're pretty dark, so you have to crank up the brightness on your screen, but I'm confident that I'm not getting ANY blue.

1: https://www.blublocker.com/blogs/news/what-blue-light-blocki...

I firmly believe this varies between people significantly.

Blue light filters do not work for me because I fall asleep on command everyday all the time regardless if WW3 is outside.

BUT it also seems the effect of poor sleep seems to be MUCH worse for me than other people. I go from extreme motor coordination to dropping cups in a span of 3 days of poor sleep.

There’s a chemical called adenosine which accumulates over the day that induces sleepiness and there are genetic variations that can affect your susceptibility to it. Receptors notice the accumulation of adenosine and use it as a signal to “scale down.”

I think that I am more sensitive, explaining my ease of sleep but also the effect of it when it accumulates due to poor sleep (sleep flushes it away). Yeah it’s great when I’m in bed but it’s not great when I want to throw a ball and my brain wants to be stingy. It basically means that someone else’s “helpful guide to sleep” is completely different from my “helpful guide to sleep.”

I recall studies showing that reading in poor lighting conditions is a cause of myopia in children. So I'm questioning whether we want to be reducing luminance on our devices at all.

I like my (warm-coloured) lights and screens set to max brightness. I find it's easier to read and lets me work with more distance from the screen.

But what about easier sleep? Could we exercise more? Leave screens out of the bedroom? I have no idea.

Regardless of the sleep effect (or lack of) they absolutely do work for reducing eye strain for migraineurs.

It's noticeable to me all the time, but if I'm borderline migraining, or recovering from a migraine, the difference between shifted and not is something I can feel instantly. Shifting all the way over enables me to eek out some work after a migraine without it flaring back up again.

In summary blue light filters actually do work, through the indirect action of reducing overall light output, but the author has a larger axe to grind about the "technical details" (it's worth reading the article). The warmer color temperature reduces strain on my eyes, which I find both soothing and invaluable.

This is just my own anecdotal experience but I usually get tired around 2130-22 but a few times I've turned off the red filter for various reasons (photo editing etc) and suddenly I'm still there at 0030-01.

I'm not saying it's like this for everyone, but it seems to work very well for me at least.

I really don’t care if they “work” or not. I find it incredibly cozy to have a warmer, calmer screen in the evening.

I replaced all the light switches in my house with smart dimmers and have the lights dim in the evening. It happens in steps so it's noticeable and it's like a clock ticking down. I don't know if there's anything scientific about it but it's pleasant, like the house is going to sleep so maybe I should too.

But have you considered that it feels better?

I actually get head ache after a long session in front of my computer, but putting anti blue light glasses it goes away or never happens.

> That’s all great, but there are websites that still don’t have dark modes. It doesn’t make any sense in 2026 that Gmail doesn’t have a dark mode. If the activity you’re doing most at night is reading email, you might consider an alternative email client.

This reads funny on a website that does not respect your device's dark mode. Guess I'll look for an alternative blog.

It seems pretty clear in the OP that headline is misleading—they do work, just not as well as he would like. I think that a 50% cut in light emission is pretty good—and you can stack that with the other interventions listed, like auto-dark mode and reducing light in your room.

I get frustrated with my dim and red shift app that is the default on my android phone for neither being very red nor very dim...but it's the type of app where every scammy body will put a red shift app whoch sucks up your location, contacts, etc, so I haven't changed.

Best thing to actually do is use as dim a screen as possible closer to sleep. You can do this with external monitors using DDC and actually directly control the physical backlight of multiple monitors.

Also properly color calibrate your monitors

Blue light filters definitely work for me. But it needs to be a strong filter (quite a bit stronger than the strongest setting of Apple's built-in filter).

If you aren't aware, your phone's screen can go much dimmer than the minimum brightness offered by the slider, if it supports HDR. There are apps that use an HDR screen overlay to lower brightness all the way down to the dimmest you can perceive. In my own experience, 'half' the brightness of 'minimum' brightness is plenty dark enough to not disturb sleeping at all if using my phone in bed.

Is this your article OP?

I use blue blocking glasses, like Bono but darker and they do work. I also use UV LEDs to help me wake up, which also works.

I agree with the premise that night shift and other color warmth features are insufficient to have a strong effect, though they do help with eye strain which is still a positive.

> I took a sample of 4 websites/apps (Google, X, Github, and VSCode) with the SpyderX colorimeter + a diffuser to average over a larger area of the screen, and found reductions in luminance ranging from 92% to 98%! That’s huge.

What about TikTok or Youtube?

> Is half a lot?

> No. Human light perception works on a log scale, allowing us to maintain useful vision over 6 orders of magnitude of luminance, from the sun at noon to moonless nights, whereas halving is .3 orders of magnitude. In relative terms, halving light is a tiny blip of the dynamic range of vision.

Kind of missing the point that:

a) a display emits spectacularly less light than the sun, even on very overcast days

b) said "blue light" reduction is presumably intended to happen at night where 1) any comparison with the ability to maintain unsaturated vision in plain sun on a clear day is largely irrelevant and 2) backlight itself is typically lower than in daylight (not for OLED which does PWM)

So given that the amount of artificial light to not screw up with sleep is about equal to "none at all" I'll take a cut in half of what essentially constitutes a flashlight aimed straight at my retinas any day.

> Here are four things that can help. [...] Use dark mode [...] found reductions in luminance ranging from 92% to 98%! That’s huge.

From my anecdotal experience dark mode and other low contrast themes are mostly used by people who set their brightness too high, and conversely people switching to dark mode immediately crank brightness up.

Countless discussions I had:

"my battery holds poorly"

"using dark mode?"

"yes"

"try light mode"

"but my eyes!"

"turn brightness down"

"done. wow I just reclaimed 1-2h of battery"

So the main claim presented here is that reducing blue reduces total "light" (lumens? watts?) by 50% (totally believable), and that reduction in light is all that matters for sleep?

That seems reasonable. The pseudoscience wankery that the fad has brought bothers me a lot too.

... but I'm not sure that's much of an argument against blue light filters, aside from color complaints. That seems to support that it's Useful and Good and is Achieving Its Intended Goal. It's reducing total luminance, because people prefer it over reducing screen brightness overall. I sure as heck do anyway (as night shift modes, they're a more comprehensive option than dark mode), though I think I'll experiment with just reducing brightness a bit.

----

For melatonin in particular, fully agreed. The recent trend of "can't even get <5mg in stores, and >10mg is appearing regularly" in the USA is mind-boggling to me. AFAICT it's exclusively because it's a "supplement" and therefore practically unregulated, and these companies don't give a shit about anyone they harm, just profit.

Start with something like https://a.co/d/0dISg7oa (0.3mg, this is what I personally use) and go up from there, slowly.

Based on my experience, most health benefits are from personal habits over external hardware. But people care health so much, it's a great opportunity for merchants to get revenue.

> That’s all great, but there are websites that still don’t have dark modes.

Such as that very website? ;)

I have a triple-click shortcut in iOS to use the accessibility features to go below the min dim settings.

Otherwise even dark mode is way too bright in a dark room.

I have had success with an extremely aggressive red filter. My unchecked sleep schedule has me going to bed around 4 am, consistent over decades. I don't consume caffeine or any other stimulant. In the last 4 months I switched my lights to LED bulbs to turn red at 6pm and use QRedshift on Linux (Mint) with the temperature set to 1000k at 6pm. I have consistently been falling asleep around midnight. What is remarkable to me is that I am actually feeling tired at night.

I added the “Noir” extension to mobile Safari, now I automatically get dark mode on all websites including Hacker News.

These are the kinds of articles that give science a bad name, and that make people anti-science.

You might as well try to claim hot tea doesn't help you get to sleep, or reading before bed doesn't, or whatever else you do to wind down.

I personally don't care if some narrow hypothesis about blue light and melanopsin is false. I know that low, warm, amber-tinted light in the evening slows me down in a way that low, cold, blue-tinted light does not. That's why I use different, warmer lamps at night with dimmers, and keep my devices on Night Shift and lower brightness. It works for me, and seems to mimic the lighting conditions we evolved with -- strong blue light around noon, weaker warmer light at sunset, weakest warmest light from the fire until we go to sleep. Maybe it doesn't work for everybody. That's fine. But it certainly does for me.

And maybe it's not modulated by melanopsin. Or maybe it's not about blue light, but rather the overall correlated color temperature (CCT), e.g. 2100K instead of 5700K. Who knows.

But this type of article is bad science writing. It shows why one hypothesis as to why a warmer color temperature would result in one other physiological change isn't supported. That doesn't mean "blue light filters don't work" as a universal statement. It's hubris on the part of the author to assume that this one hypothesis is the only potential mechanism by which warmer light might help with sleep.

And it's this kind of science writing that turns people off to science. I know, through lots of trial and error and experimentation, that warm light helps me fall asleep. And here comes some "AI researcher and neurotechnologist" trying to tell me I'm wrong? He says it's "aggravating" that people are "actually using Night Shift". I say it's aggravating when people like him make the elemental mistake that showing one biological mechanism doesn't have an effect, means no other mechanisms can either.

The entire blue light madness is based on a poor study where N was around 8. And the difference in sleep was something like 15 mins. The entire study was based on crap but somehow the entire world has run rampant on the idea that blue light has this profound effect. It just goes to show that bad science is easily propagated, even when there's even more sources of information.

Well they work in that the color temperature of the light in my house is much cooler during the day than at night, and it's nice to match it so it doesn't look jarring.

> Everybody wants better sleep

Bro, as someone who had brutal insomnia for a couple of years and now sleeps "normally" for whatever that means, I can tell you that I don't think about my sleep quality at all. I'm happy to be sleeping.

If you too sleep "ok" for whatever that means, maybe stop worrying about optimizing it and go do something else less insane.

The argument about luminance ranges is wrong. I measure the brightness of monitors regularly as part of my job, and typical maximum luminance values are in the range of 100-500 lux. That puts you right in the steep range of the visual response (especially if you are turning it down and near a max of 100), which is natural — maximizing the slope of the neuronal response to light means that more information will be available to the brain. In fact a good monitor will be tuned according to the just-noticeable difference which aims precisely to maximize the information available according to this characteristic curve. See e.g. the DICOM standard:

https://dicom.nema.org/medical/dicom/current/output/chtml/pa...

The author's basic problem is that he knows too much about the brain and not enough about monitors.

The author goes on to argue that you should be turning your brightness down, but most people already are turning their brightness down; the blue light filter is more comfortable. He does make a reasonable case that you should be reducing green light similarly, but people prefer the incandescent effect of the flux filter to a straightforward color filter — indeed a primary design goal of these filters has been to be pleasant to look at which is why people use them.

It's funny, I'm so comfortable calling this guy an idiot purely based on the fact that I've taken up Bob Ross style painting in like the last 2 years.

Teaches you to pay attention to "objective" colors. And at night, guess what, the colors get more red and less blue. I don't have to pull out as much blue paint for the night scenes.

It would be utterly naive to not thing that there's -- perhaps purely "psychological" (not sure if that's the exact concept but hey) effect by making the "white" on your screen, look like like the "white" you will definitely see in real life, which is going to be orange-r.

I have Night Light perpetually on with all of my devices because I find it softens everything and makes viewing displays less harsh, less garish, less vivid, and less intense. I don't need eye searing HDR constantly cooking my retinas.

my phone has a buried setting for ultra dim, which does help, except outside, where it makes the phone unuseable, and then it's impossible to do the 5 taps and scroll to find it, fuck android going to a linux phone

Why is it that a few people seem to get bent out of shape by redshift and/or dark modes? If you don't like it, don't use it. Whining about scientific evidence is pointless, even if it all only comes down to user preferences with no science behind it, so what? Let people enjoy things.