f.lux
justgetflux.com
justgetflux.com
Also, if you care about how your eyes and circadian system work, don't forget about all the light from outside your screen too.
Time for you to make us some great sunglasses :) or curtains..
(This is not an endorsement, however.)
With cycling glasses, I had more of relaxing effect due to them keeping moisture on my eyes, instead of it being burned away by the winter heating.
Went with these ones: https://spectra479.com/products/clip-on-lenses-blue-blocking...
Thanks for f.lux.
One should not only maintain screen color and brightness, but also room and overall environment color and brightness.
I bought a "smart" light after listening to that episode.
No clearly visible signs of improvement yet. But it certainly feels good.
All the lights in our house automatically lower their color temperature and brightness at sunset every 15 minutes from 5000k to 2700k and 100% to 10%. SmartThings and the community created Circadian Daylight Coordinator have made this pretty straight forward to accomplish.
I don't have data, but anecdotally I can say it makes an appreciable difference in overall comfort.
Thank you!
Plus, it’s a great example of a really high quality piece of software developed with a lot of care and a huge long term commitment.
f.lux still lets me go to a lower colour temp than either Windows or OSX. I know I’m a niche audience, but it makes a difference to me and I don’t think I’ll get anywhere by shouting at the giant Wall of Apple to expand their range.
It's most likely just that I've conditioned myself, but in practice I don't really care about the cause. I only care that it helps me wind down in the evening.
There's something that feels good about being in sync with the natural day/night cycle. Cool white in the morning, warmer more dim light at night. I also like the idea of having a "sunrise/sunset" wall for any room that doesn't have proper windows but I've never implemented it.
Flux helped noticeably. For true relief FL-41 glasses would extend the time to a few hours.
A few times for late shifts in various businesses I've convinced the execs to shut off the fluorescents and I cannot describe the wave of relaxation that hits almost everyone.
It's so much worse also with the combo of 60Hz monitors and 60Hz AC power.
* is pure OTC saline -- not any of that other stuff people use for hangovers, etc.
I'm sure there would still be plenty of edge cases, but this would be plenty accurate for me.
https://play.google.com/store/apps/details?id=com.urbandroid...
I also used to use f.lux on Linux until i found "Redshift"
https://news.byu.edu/intellect/is-night-shift-really-helping...
I've just donated $25 - https://justgetflux.com/windows/contribute.html
Just said "thanks" to them with the same.
Here's how I do it on Linux with a Dell U2720Q monitor.
sudo apt-get -yy install ddcutil # Install ddcutil on Ubuntu
sudo modprobe i2c_dev # Load required kernel module
# Night - Set brightness to 20% and color temperature to 3000K
sudo ddcutil setvcp 10 20
sudo ddcutil setvcp 1a 43
sudo ddcutil setvcp 18 69
sudo ddcutil setvcp 16 100
# Daytime - Set brightness to 50% and color temperature to 6500K
sudo ddcutil setvcp 10 50
sudo ddcutil setvcp 1a 100
sudo ddcutil setvcp 18 100
sudo ddcutil setvcp 16 100Personally, I just hit the brightness down button on my keyboard a few times around 8pm. I think windows has this as a built in feature now, too? I'm curious to know what f.lux offers over this:
https://support.microsoft.com/en-us/windows/set-your-display...
https://github.com/kelvie/gbmonctl
Along with some systemd scripts to toggle it for sunruse/sunset, with another simple tool:
https://github.com/kelvie/sunrise-monitor
I really wish there was a common (XDG?) interface for monitor brightness so that e.g. Plasma/KDE can hook into this with their sunrise/sunset settings.
More importantly installed the Dark Reader extension (https://darkreader.org). This one is incredible: it's much nicer to browser the Web in dark mode than to make page reddish. I now enable Dark Reader at all time (day and night) and can't stand the screen of my coworkers (so dazzling).
Idem for my Android device.
The only app/webpage I don' apply dark mode/reader to is Google Maps. For now…
Sometimes it leaves black-on-dark for icons and such, but that's negligible, and definitely a good trade-off to my eyes being drained.
Side note: Maps is surprisingly good with Dark Reader!
Edit: I just remembered I've been using an extension called "Hacker News Enhancement Suite" to enhance the layout of HN, and after disabling it, Dark Reader worked so I think that extension was somehow interfering with Dark Reader.
I use "Hacker News Readable Dark" under Stylus and this site looks fantastic.
If anyone knows a similar solution that doesn't require such an extreme, I'd like to hear about it..
Anyway, my vague understanding of what is does is that it inverts all colors on the wrong side of a user-configurable brightness threshold (e.g. backgrounds that are too bright are inverted, and text that is too dark is inverted). This can make website colorschemes quite different, but most of the websites I visit regularly are still still readable, and often still reasonably nice-looking, despite having inverted colors.
The relevant part of my config file looks like this:
# This requests the dark version of the website if it is available. Generally when this
# works, I guess brightness-based the color inversion doesn't happen.
c.colors.webpage.preferred_color_scheme = 'dark'
c.colors.webpage.darkmode.algorithm = 'lightness-cielab'
c.colors.webpage.darkmode.contrast = 0.0
c.colors.webpage.darkmode.enabled = True
c.colors.webpage.darkmode.grayscale.all = False
c.colors.webpage.darkmode.grayscale.images = 0.0
c.colors.webpage.darkmode.policy.images = 'smart'
c.colors.webpage.darkmode.policy.page = 'smart'
c.colors.webpage.darkmode.threshold.background = 128
c.colors.webpage.darkmode.threshold.text = 128
Like, there is simply no way that is not a solvable issue.
Where Dark Reader is full-featured, mine is minimal. It adds a translucent tinted div overlay on every page, with mix-blend-mode:multiply to make the text come out a little sharper. No themes or querying page elements.
You can audit the code yourself if you want, it's only 58 lines of JS across 2 files:
// content.js (this runs on every page):
'use strict';
const C=chrome, d=document.createElement('div'),
update=(o={})=>{
if('tog' in o){d.style.height = (o.tog.newValue | o.tog)+'00%';}
if('color' in o){d.style.backgroundColor = o.color.newValue || o.color;}
};
d.style='mix-blend-mode:multiply; z-index:2147483647; position:fixed; width:100%; top:0; right:0; pointer-events:none;';
document.documentElement.insertBefore(d, document.documentElement.firstChild);
C.storage.local.get(null, update);
C.runtime.onMessage.addListener(update);
C.storage.onChanged.addListener(update);
// popup.js:
'use strict';
let XY=[150,40];
const C=chrome, S=C.storage.local,
send=x=>C.tabs.query({active:true,currentWindow:true}, t=>C.tabs.sendMessage(t[0].id, x)),
torgba=x=>`rgba(${x[0]}, ${x[1]}, ${x[2]}, ${x[3]/255})`,
dropper=(x,y)=>torgba(c.getImageData(x,y,1,1).data),
target=(x,y)=>{c.fillStyle = 'black'; c.beginPath(); c.arc(x,y,30,0,Math.PI*2,true); c.stroke();},
touch_xy=t=>{const r=c.canvas.getBoundingClientRect(), f=(x,ox,r)=>Math.floor(Math.min(Math.max(x,ox+1),r-1)-ox);
return[f(t.pageX, r.left, r.right), f(t.pageY, r.top, r.bottom)];},
init_c=()=>{
c.clearRect(0,0,c.canvas.width,c.canvas.height);
c.canvas.style.backgroundColor='white';
const g1=c.createLinearGradient(0, 0, c.canvas.width, 0), // horiz, hue
g2=c.createLinearGradient(0, 0, 0, c.canvas.height); // vert, lightness
g1.addColorStop(0.2,'rgba(255,120,120,0)');
g1.addColorStop(1,'rgba(255,130,70,0.99)');
g2.addColorStop(0,'rgba(0,0,0,0)');
g2.addColorStop(1,'rgba(0,0,0,0.8)');
c.fillStyle = g1; c.fillRect(0,0,c.canvas.width,c.canvas.height);
c.fillStyle = g2; c.fillRect(0,0,c.canvas.width,c.canvas.height);},
xy_send=(xy)=>{init_c(); target(...xy); send({xy, color:dropper(...xy)});},
xy_save=(xy)=>{init_c(); target(...xy); S.set({xy, color:dropper(...xy)});},
c=document.getElementById('c').getContext('2d'),
tog=document.getElementById('tog');
window.addEventListener('load',()=>{
S.get(null, o=>{
if('tog' in o){tog.checked = o.tog;} else {S.set({tog:tog.checked});}
if('xy' in o){init_c(); target(...o.xy); XY=o.xy;} else {S.set({xy:XY});} // color's xy coords in canvas
if(!('color' in o)){S.set({color:dropper(...XY)});} // hardcoded default xy location
});
tog.addEventListener('change', x=>{S.set({tog:tog.checked});});
c.canvas.addEventListener('mousemove', x=>{if(x.buttons!==1){return;} xy_send([x.offsetX, x.offsetY]);});
c.canvas.addEventListener('mouseup', x=>xy_save([x.offsetX, x.offsetY]));
c.canvas.addEventListener('touchmove', x=>xy_send(touch_xy(x.targetTouches[0])));
c.canvas.addEventListener('touchend', x=>xy_save(touch_xy(x.changedTouches[0])));
});
window.addEventListener('keydown', e=>{
if(e.key==='x'){tog.checked = !tog.checked; S.set({tog:tog.checked});}
});
I feel like during the day most things are bright/cool with darkness as the contrast. Text on paper. Also as much as dark mode has become an option, it's impossible to go an entire day without experiencing a light theme and the switching of light to dark or dark to light for me is worse than whatever problems each theme has. Point being it's easier to only use light than only use dark.
Anyway dark mode without red tint isn't really helping with the blue light problem? Especially if the media you're looking at is mixed with e.g. images of daylight.
I installed Hue bulbs in my home office for this reason, which keeps my lights at a nice energizing color during the day, then warms up during sunset (especially useful during the winter Standard time).
Unfortunately I’m not aware of similar support for MacOS at this time, but maybe others know of options there.
1. When the screen color changes at sunset it's a subtle but powerful cue to know that the work day is coming to a close. And then if I'm on my computer when it changes around bedtime that's a cue to GTF off the computer.
2. All the science I've read suggests removing blue light in the evenings does indeed help with sleep. So why not use it?
I've used f.lux for a long time, but it kicks in in the middle of the afternoon in winter and probably about bedtime in summer!
Unfortunately in northern latitudes, our sunset can be as early as 3PM or as late as 10PM, so it doesn't quite work for that purpose here!
Nevertheless, I find matching the colour temperature with the time of day makes my eyes far more comfortable.
The studies on whether blue light messes with sleep are mixed, but I personally find if I set my lights brighter and bluer in the morning that I feel more awake than I do otherwise.
Yeah, that makes it totally unusable for me, it's ridiculous that they don't let the user set the hours and insist that you must follow the sun, because that's "more natural". Even though that might be true, it's totally impractical, you can't organise your work day after the sun, that's totally incompatible with the rest of society. At least os x night shift let's you set the time like the responsible adult that you are..
I encourage people who are looking to solve sleep problems to not use any backlit screens in the several hours before sleeping. Such evidence as there is on the subject doesn't suggest that redshifting the screen is sufficient.
If you've seen someone using a laptop or phone in a dark room, it's quite clear that they're blasting themselves with a flashlight. Filtering some of the blue out isn't going to repair this basic dynamic.
I don’t recall seeing any studies that directly used f.lux, just studies with blue lightboxes, and that stuck out to me as odd.
Monitors, even modern ones, don’t 100% convert their backlights (which are white) to the displayed color, so you always have an amount of washout, and I was always curious if that spoiled the effect, among other unique quirks of computer monitors.
I haven't had to use f.lux for many years now because it's built in to every platform I use (KDE, Gnome and Android).
[1] https://www.sciencedaily.com/releases/2019/12/191216173654.h...
so it's actually the opposite to the current common belief. it should be blue at night, warm during the day :)
Emphasis on "may".
"Our findings suggest that using dim, cooler, lights in the evening and bright warmer lights in the day may be more beneficial."
Emphasis on "suggests".
Also, the study's title itself.
"Cones Support Alignment to an Inconsistent World by Suppressing Mouse Circadian Responses to the Blue Colors Associated with Twilight."
Emphasis on "mouse".
Science oozes back and forth. You should not strongly interpret individual studies. Debunked is a strong word here.
The current study has has not found links between cool light at night and sleep patterns. That may change in the future, sure, but current science suggests said theory is debunked (and I’m not afraid to use that word).
Maybe we could say the theory is "weakened" by the new evidence or that it is now more "controversial" than it was before. "Debunk" sounds like blue light having an impact or circadian rhythm is some random "fake news" that could be found on Facebook, when it is still backed by some decent evidence.
I have not added my own views on the subject, but merely highlighted existing words that seem to have been overlooked by some who have only read the news briefs and not the full study itself.
From the conclusion: "An especially pertinent question, however, is whether the effects of color described here extend to other mammals, such as humans."
The word "debunk" does not appear on any of the 13 pages of text.
Who said the word appeared in the text? Who said the word has to appear in the text for it to be used in discussions among people?
> I have not added my own views on the subject, but merely highlighted existing words that seem to have been overlooked by some who have only read the news briefs and not the full study itself.
The words were not emphasized in the paper itself. You chose to explicitly point out certain things in order to discredit the comment you were replying to. You know exactly how your comment comes off, and by doing it you, indeed, add your own views to the subject.
=\=
"The effects of color described here extend to other mammals, such as humans."
The first is in the text. The second is an interpretation of the text.
Dismissing the content of a scientific article and replacing the existing words to support your own interpretation of the text is exactly adding your own views on the subject.
This study is but one step in the scientific process towards verifying this fact. Hopefully, the next step is taken soon and the theory is tested on humans.
Should that next paper conclude with "Blue light is not disruptive to human sleep patterns", then I would absolutely agree that debunking has been had.
In fact, since you are pushing to know my own views, I would be thrilled about it. Issues with sleep quality is something that affects me and having a concrete answer about the subject at hand would allow me to make better and more enlightened choices about steps to improve my sleep quality.
But this is not the case here, right now this would be selling the skin before having caught the bear.
Is this a good moment to slow down, analyse and ponder? Absolutely. Do test out if f.lux actually has an effect on you if you are a daily user. Do try to use another color than red in your Philips Hue sleep routine. But now is not the time to take direct actions.
...
Let's step back for a minute and imagine we were talking about a study about medication.
Let's say that an animal study came out that concluded with "An especially pertinent question, however, is whether the toxicity of acetaminophen extend to other mammals, such as humans."
I would without a doubt leave a comment highlighting the content of the paper to show the commenter that no, an animal study should not be strongly interpreted as debunking the safety of acetaminophen in humans.
Would you also write: "You chose to explicitly point out certain things in order to discredit the comment you were replying to. You know exactly how your comment comes off, and by doing it you, indeed, add your own views to the subject" in that situation?
Not really.
The point is this, the whole blue light business popped up because the people who were doing the studies forgot that amber dusks are fairly rare outside of cities/equator.
Those of us nearer the poles have long, blue evenings. Yes, in march and august we tend to have brilliant orange sunsets. However those are the exceptions, not the rule. If you look at the colour temperature from noon till actual dusk, you'll see that it increases as time goes on.
pinning screen colour on the quality of sleep is a fools errand. We do not look at a screen for it's colour we look at for its content. That content has a much great effect on our sleep than the screen brightness and colour temperature.
TLDR: twilight is blue, not orange. Colour temperature and sleep quality is not convincingly linked.
As for the warmer color temperature: I do use it. I prefer it warmer because it burns my eyes much less. Similarly I prefer the brightness of my phone's screen around 40%. I have lenses that "block" blue light (i.e. makes white look more yellow).
If you are a developer doing UI/UX and that sort of thing where color matters, then you probably want to disable f.lux, redshift, or whatever it is that you are using.
We could also get philosophical here and discuss qualia. Is my red your red? :)
So much to discuss!
Stop eating junk food before bed? Probably better sleep. Stop getting poked with a stick before bed? Probably better sleep. Etc.
If you have an e-reader that only emits bluish light like a kindle paperwhite then you can turn down the light intensity and increase the font size to compensate for lost of detail sensitivity.
I use them on my phone (which is the default and I didn't bother to change), but never get used to them on computer screens. I'm pretty sure if I give it more time I can quickly get used to, but just didn't see the appeal enough. I do have a yellow light in my room.
But then again I don't use dark mode either, so maybe I'm just a weirdo.
I also only use daylight bulbs in my house. "Soft white"/yellow bulbs give things a creepy feeling for me.
I wonder if the people that like f.lux are also the same people that claim that light themes are blinding, because I'm of the opinion that if you find a completely white screen is blinding, then either your brightness is too high or you don't have enough ambient light in the room.
In my experience it‘s sufficient to adjust the display brightness to roughly match the ambient lighting; no need for blue-light filtering.
https://github.com/jonls/redshift https://gitlab.com/chinstrap/gammastep
A quick specific complaint (in hopes of learning about workarounds): The implementation on Windows (known as "Night light") seems particularly bad: Sometimes it won't turn on at the expected time, or sometimes it will only apply on 1 monitor but not the other. Has anyone else seen this? Any hints to make it more reliable?
I have my suspicions the reason why it doesn't work consistently has something to do with power savings or hardware profiles or something, but i haven't been able to reproduce it consistently enough to open a bug report.
You should turn it off once late a night, your retinas will hate you for it :D
But if I step away from my computer at night for an hour, and then go back to it, nothing appears unusual or jarring when I return
Redshift can easily just change brightness but f.lux doesn't have an option to just shift brightness last I checked and I haven't found a great way to do this on windows desktop (laptops have more built in brightness options that are at least easy to adjust). It sounds like some monitors have fairly easily adjusted brightness but mine is a pain to adjust anything. I tried redshift on windows but it had some issues that I don't remember. Since I just use windows for games, it isn't as big of a deal since games tend to be less bright (and many have brightness adjustment). As my monitor gets older I have more of the opposite issue of it not being bright enough during the day in the summer.
Also, I tried gammastep with sway, but it seemed to shift at the wrong time like there is a GMT vs local time mixup. I ended up just switching back to X.
I’ve used computers for years without f.lux and haven’t noticed any eye strain symptoms or sleep issues. I also got some blue blocking coating for my eye glasses last time because it was on a big discount and can’t tell the difference (symptom-wise… I can see the included blue laser is mostly stopped and things look slightly yellower) compared to my old glasses.
Just a note to avoid confusion: Lunar serves a completely different purpose than f.lux.
f.lux changes the colour temperature by altering the Gamma tables.
Lunar's main function is controlling the hardware brightness/contrast/volume/input/colors of external monitors using DDC and private macOS methods.
macOS also has a native approach to f.lux called Night Shift. I use it with Shifty (https://shifty.natethompson.io) to get the same featureset as f.lux.
I started to wonder how much of it was the placebo effect, the initial change is certainly noticeable but the benefits are hard to actually quantify for yourself.
The science is not really clear, it’s definitely controversial. Truthfully, I suspect the amount of blue light from our screens is actually much more negligible that we tend to estimate. Unpopular opinion alert, but I don’t think filtering blue light at the screen level, or those glasses they sell really does much of anything. I think it’s more likely the placebo effect is in play here.
They do have an Android app as well as desktop apps. f.lux has been in the space for years and years now.
Flux has been resilient in being featureful through the years and has survived being obviated by operating systems including its basic functionality.
- https://vermaden.wordpress.com/2018/06/28/freebsd-desktop-pa...
Using night mode on PCs greatly reduced the eye strain and sleeping issues I was experiencing due to long days at the computer.
Here's how to enable them on Mac: * Open Preferences * Search for "color filters" * Check "Enable color filters" * Select "Grayscale"
I don't use Windows much anymore, but when I do, I run f.lux there.
I remember there was some widely popular screen dimmer, which didn't even let you manually change setting and relied on location provided by google services which I don't have, not sure if it's not f.lux because it ws one of the most famous options
It's also a visual cue for me. Oh shit my screen is yellowish. Time to get off the computer and wind down.
[1] https://www.imore.com/why-you-should-use-true-tone-your-ipho...
Night Shift was developed much later.
But it’s subjective, some people like the extreme warmth of f.lux more. It depends on how you perceive colors and how well you can adjust to warmer colors.
Shifty is also a very nice addition if you want more control over Night Shift: https://shifty.natethompson.io