Building an artificial sun that looks realistic [video] | Hacker News Reader

Building an artificial sun that looks realistic [video]

youtube.com

515 pointsjulvo123 comments

In a sci-fi future where people are making long journeys on spaceships I can imagine rooms decked out with these to mimic a natural day/night cycle and outdoor spaces.

I remember first seeing lights like these a couple years back before from CoeLux (https://www.coelux.com), iirc theirs are very expensive though. Awesome that this guy managed to DIY one!

Spaceships? I can imagine these things in Alaska:

https://en.wikipedia.org/wiki/Whittier,_Alaska

> Whittier is a city at the head of the Passage Canal in the U.S. state of Alaska, about 58 miles (93 km) southeast of Anchorage.[6] The city is within the Chugach Census Area, one of the two entities established in 2019 when the former Valdez–Cordova Census Area was dissolved.[7] At the 2010 census the population was 220, up from 182 in 2000. The 2018 estimate was 205 people, almost all of whom live in a single building, the Begich Towers.

https://en.wikipedia.org/wiki/Begich_Towers

> The Begich Towers Condominium is a building in the small city of Whittier, Alaska. The structure is notable for operating like a small-scale arcology, being the residence for nearly the entire population of the city, as well as containing many of its public facilities. This has earned Whittier the nickname of a "town under one roof".

Riding the coattails of your comment to ask the hive mind: any reason I shouldn't try to build the simpler version from the tutorial using a more powerful (100W) LED from Amazon?

https://www.amazon.com/dp/B07T8RDF8F

Might not be as awesome as a $60k CoeLux, but for 1000x less cost, maybe it's worth a shot?

(Edit: come to think of it, with an 8x10 fresnel lens, it might be way too bright for the size.)

Don't forget that the volume of the device has a cost too. The video shows the construction of a really really nice, but takes up the whole hallway outside that room. Most of the time you'd probably prefer to use your hallway for walking in. On the other hand, maybe your house has an attic above one of your rooms that; there will be a lot less opportunity cost to using up that volume.

Use Fresnel lenses with a shorter focal length and reduce the depth.

It would be useful to start with a standard home-type window from, say, Home Depot, and build a unit about 120mm deep using multiple shorter focal length Fresnel lenses.[1] Then, in a basement or windowless room, mount some of these into a wall. Add curtains. Nice if your home office or workshop is in a basement.

This has commercial applications. Houses have few windowless rooms, but commercial structures have many.

[1] https://www.alibaba.com/product-detail/Fresnel-lens-plastic-...

DIY Perks has previously shown how to do that, and it looked pretty good.

Turning Smashed TVs into realistic daylight panels: https://www.youtube.com/watch?v=8JrqH2oOTK4

Keep watching the video - he builds Fresnel lens versions at the end of the video.

The satellite dish is more dramatic so I don't blame him for leading with it :)

The idea is to build this as a light box behind a standard windowpane frame. Like this one.[1] With a 120mm short focal length, the unit only needs to be about 150 - 200mm deep, if you angled the optics so you get light at a 30 degree angle or so. (Horizontal sun rays would look wrong.) That's a convenient size; you could hang it on a wall, like a rather bulky picture.

[1] https://www.walmart.com/ip/Darice-Wooden-Window-Pane-w-6-Pan...

lenses cause chromatic abberations though, the red/yellow halo around the beam you see in his cheaper version at the end.

I have a chromatic film on my actual windows because I like the effect so there's technically a market for it , especially if it makes it cheaper and smaller

I am really curious about this -- do you have it just for the effect? What does it look like?

It is much easier to take broken lcd screen and remove diffuser.

https://m.youtube.com/watch?v=8JrqH2oOTK4

That’s the same guy as this article link. This one is supposed to be an improvement over that.

the point of the parabolic reflector/fresnel versions are the parallel rays

maybe I should look at things first before commenting ..

I would advise to use some brass chassis for hot elements as fire accident may be more costly.

You'd still have to deal with the glare and chromatic aberration from the mirror sheets. Also the area covered would still be confined to the size of one sheet and you'd have to make a "bank" of them to cover any reasonably large area.

Especially if they emit the UVB light we need for Vit D production

I thought the official advice was to get supplemental vitamin d through oral supplements, rather than through artificial UV lamps?

https://www.nhs.uk/live-well/healthy-body/how-to-get-vitamin...

>Using sunbeds isn't a recommended way of making vitamin D.

There's at least some evidence that LEDs can be more effective than sunlight at synthesizing D. There's a specific wavelength of light that triggers it and I'm not sure if a standard tanning bed emits it.

https://www.nature.com/articles/s41598-017-11362-2

FWIW there's not evidence vitamin D supplements have any health impact (asides from crickets).

I’m confused by this comment. There is evidence that Vit D deficiencies have all sorts of health impacts, and that supplements raise in-blood levels. What other evidence is needed?

https://www.bulletproof.com/supplements/vitamins-minerals/vi... (11 scientific references linked at the bottom)

Most things related to vitamin D are just correlation. If you are weak and frail you may not get out much and won’t have so much vitamin D. That doesn’t mean giving you vitamin D will make you strong. I heard that most studies don’t show benefits from vitamin D a supplements but maybe I’m wrong.

Sounds to me like what you heard could be easily proven/disproven by reading some studies.

I searched google scholar for “vitamin d meta analysis”. I skimmed the abstracts. These are some of the earlier results.

It seems supplements help but randomised controlled trials are needed: https://jamanetwork.com/journals/jamainternalmedicine/fullar...

“Vitamin D supplementation appears to reduce the risk of falls among ambulatory or institutionalized older individuals with stable health by more than 20%” http://citeseerx.ist.psu.edu/viewdoc/download?doi=10.1.1.578...

Oral vitamin D supplementation between 700 to 800 IU/d appears to reduce the risk of hip and any nonvertebral fractures in ambulatory or institutionalized elderly persons. An oral vitamin D dose of 400 IU/d is not sufficient for fracture prevention. https://jamanetwork.com/journals/jama/article-abstract/20087...

High levels of vitamin D among middle-age and elderly populations are associated with a substantial decrease in cardiovascular disease, type 2 diabetes and metabolic syndrome. If the relationship proves to be causal, interventions targeting vitamin D deficiency in adult populations could potentially slow the current epidemics of cardiometabolic disorders. https://www.sciencedirect.com/science/article/abs/pii/S03785...

————————————

At this point I got bored of looking at results (I omitted a few which only looked for correlations).

So while there is some evidence of some benefits for some people, I didn’t find evidence of vitamin d being the magic panacea that it sometimes feels it is claimed to be by people on the internet.

Well, if it may have benefits and costs pretty much nothing, it would make sense to take it...

I think you need to add "among people with sufficient levels of vitamin D" to that statement.

It follows almost from definition that if you are deficient in something, supplementing that something is beneficial. There might be some exceptions, but then we've really just got deficient defined poorly or no effective method of supplementation.

Aside from _in_ crickets experimentally, or aside from crickets as in you hear nothing?

rickets, not crickets

Ohhhhhhh, that makes total sense.

I looked into it for a client's commercial building. $40k each for parts only.

I had a room in an apartment that had no sunlight, and was lusting after CoeLux after it launched. Almost goes without saying, but the first to manufacture an affordable solution at scale would make a ludicrous amount of money, especially with much of the world in lockdown.

Low dose UV light would also be a good addition.

Low dose UV light would also be a good addition.

I'd also like a bit of optional IR so you get a tiny bit of that toasty sun feeling if you want it.

If you use a parabolic reflector and can prescatter the light it's as easy as adding another light source.

A computer chip costs $1 billion to make.

A billion of them cost $1.1 billion to make.

Too right!

If you're too lazy to build something like this out but like the idea I highly recommend bright LED lighting for the day (explained here: https://www.benkuhn.net/lux/). I bought two ~7000 lumens LED corn bulbs and have noticed a huge shift in energy levels from having a brightly-lit room in dark PNW days. I put together a bunch of resources when I was researching buying some here: https://www.notion.so/vimota/LED-World-fce39f79710b42708b92a...

I did something similar to this but was turned away from using random bulbs off of Amazon since most of them lied about their product performance (advertising peak wattage of all the LEDs and then not driving them at max power), safety certifications (fake UL and ETL markings), and in several cases I could not even find valid business licenses where the companies claimed to be incorporated. I need light for the darkness of winter but not at the expense of burning down the house.

Instead, I ended up buying a string light[0] and a box of 1600lm daylight LED bulbs[1]. Installation sucks, but once the whole thing is up, it’s nice that the light is distributed throughout the room instead of coming from a single point source. It was (at least at the time) cheaper than buying corn lights—and as a bonus, you can swap back the bulbs that come with the string light for party lighting.

[0] https://www.costco.com/feit-electric-48'-led-filament-string...

[1] https://www.costco.com/feit-electric-led-100w-replacement-br...

On the higher end, I've been pretty happy with an Eve Light Strip [1]. However that's HomeKit-only. The key option we're looking for is RGBWW (with 2 different "temperature" whites) for a wider range and more natural lighting. We've been considering various more flexible light strip arrangements, but as GP noted, the market is flooded with low-reliability options.

Luckily, we've stumbled upon Sowilo [2], which advertises powerful RGBWW LED strips with appropriate specs (with per-meter max brightness and power consumption listed), provide all the components you need for powering and driving the strips you need, and best of all provide a Hue-integrated controller! (likely from scavenging an actual Hue LED strip controller!). We've ordered a kit last week, and are eagerly anticipating testing out the ~13000lm of purported light.

[1] https://www.evehome.com/en-us/eve-light-strip [2] https://sowilodesign.com/collections/led-strips

Tangentially, we've picked up 4x GE Ultra Bright [3] at Lowes for $10 each to test out the brightness, and those have been great too.

[3] https://www.lowes.com/pd/GE-Ultra-Bright-150-Watt-EQ-A23-Sof...

I actually purchased the 35,000 lumen corn bulb mentioned in the post you linked to. It's actually really big and awkward to place and has kind of a loud fan. Currently I have it propped in a corner since it's too heavy to be held by anything that I have to support it. But it will light up a bedroom just as much as a window of sun will. It doesn't look entirely like sunlight, but for the price, it does an adequate job.

So I'm a little bit confused, the linked 250W corn bulb does indeed claim 35,000 lumens: https://www.amazon.com/gp/product/B07W5DGBJZ/ref=as_li_ss_tl

But when I look at similar 250W corn bulbs on the Home Depot site they only claim 7,384 lumens: https://www.homedepot.com/p/Halco-Lighting-Technologies-250-...

How is that Amazon bulb producing 5X the lumens?

EDIT: To answer my own question, it's because the Amazon bulb actually uses 250W while the Home Depot product listings are "250W equivalent" and only use ~50W. Hence the 5X difference in light output.

Seconding this, although I only bought one giant 14k lumen corn cob, since I didn't have a good spot to put the second. Anecdotally, they've been great at staving off my usual post-lunch slump. Right around the same time as I got it, though, I started waking up much earlier naturally, without an alarm clock (shifting from ~9AM natural wake up to more like ~6AM). I'm careful about turning the corncob off no later than 6PM, and the earlier wake up time might be unrelated, but I was wondering if others had experienced a similar phenomenon with extended, consistent exposure to super bright lights during the day.

Sounds very interesting - i am currently in Europe and can see that the price is about 350 usd, and you can get versions from 3000 (warm) to 6400 kelvin (sunlight?).

What kelvin did you choose, and how large is your room? Did you put something in front of the lamp to diffuse?

Looking forward to get out of the lockdown slump! :)

My diy lamp shade has been wrapping parchment paper around the LED corn bulb (making sure to leave space for airflow so it doesn't overheat!). It's not perfect but it helps diffuse the light and make it less intense to look at.

The one I got is 5k kelvin, which was advertised as sunlight as well, and is about as cold as I can take. If I had to buy it again I probably would get something slightly warmer, tbh. I've got a lampshade on it that diffuses the light at the cost of attenuating it a bunch, but it hasn't lessened the effectiveness as far as I can tell.

I was looking into this last year, but couldn't really find anything similar here in Sweden (or anywhere in the EU). I wonder if there are some regulation here banning ultra high lumen bulbs? The closest match have been on semi-sketchy sites, where they've been ≈ 300 euro for 10k lumen...

How expensive are they in the US?

Also in Sweden - there is this Danish shop :

https://www.ledproff.dk/produkter/11907-kraftige-led-paerer-...

Trustpilot doesn't raise any major red flags, but i don't know the shop, it has "e-maerket" though which is official certification.

13.5K for about 400 usd. So it it must at least be legal in EU? They also have 14.8K if you go up 1 level.

Wait a minute, they also have a swedish shop:

https://www.ledproff.se/led-ljuskallor/11907-kraftfull-led-l...

Ah, nice. I recognise the brand of the lamp from earlier research. I read this post just now https://meaningness.com/metablog/sad-light-led-lux, where he's using ≈ car lights. That got me thinking about construction lights, they are apparently much cheaper. But not sure about the Color Rendering Index. Here [1] we have 8000 for 600 sek, but not in as a nice design maybe...

[1]: https://www.kjell.com/se/produkter/el-verktyg/belysning/utom...

I also looked into this and ended up being successful when searching for photography lights and corn cob LEDs. Corn cob bulbs at 40-60 Watts each are cheap (~20USD in the UK), provide ~4K lumens each, and 2-3 are bright enough to get to daylight levels. Would highly recommend if you want to see whether it makes any difference, and/or which colors temp works best.

I'm curious if the lack of heat in the lightrays will ruin the immersion of sunlight? Perhaps couple this with an IR heatlamp to add the real feel of sunrays hitting your skin?

The main issue is the blue scattering. If you can get away with pre-scattering the light before it hits the dish adding an IR lamp is no issue since you're using a parabolic mirror which has no diffraction. I have no idea if it 'feels' right doing that though.

For full immersion you would also want a number of different light sources since leds are have very narrow spectra.

The method proposed in the video already accounts for the narrow LED spectra by using lights that have excellent color reproduction (the part where he recommends a CR rating of 95%+)

It's not really good enough: https://www.oscars.org/science-technology/sci-tech-projects/...

Mixing violets and greens makes the spectral curve closer to that of sunlight.

Again, without seeing it in person there is no way to know how much impact this has.

Using a parabolic reflector however is the only way to go since you will not get diffraction or blocking of wavelengths.

Without going into the pdf report, the multiemitter doesn't look better than the blended phosphor.

No, you would need to pick better diodes, and wider variety of them. 3 greens and two violets + phosphorus at least.

To answer anyone looking at this now: the Rayleigh scattering needs to produce diffuse light [0]. Which actually makes the project easier to pull off since you just need to place secondary lights all around the roof and walls of the room that emit AM1.5 diffuse light globally and not worry about any sort of filtering on the main light, which you need to match to AM1.5 direct.

[0] Applied Photovoltaics, Earthscan Publications, 2007: http://www.eng.uc.edu/~beaucag/Classes/SolarPowerForAfrica/B...

Yes, patio heaters are very cheap. Such an infrared emitter would have been a very easy addition to this project, at the cost of doubling the power draw.

If you have access to a broken LED TV or monitor, or in some cases CFC, you can just buy these. Search for "led panel light", and you can find them in all sorts of sizes.

I was looking at building one last winter, but on CraigsList people tend to overestimate the worth of a TV with a broken screen. "Awesome 4K TV, paid $1200 3 months ago, just need to replace broken screen, $250." You gotta talk them out of that with a "A new panel costs more than buying the whole TV."

You can also buy just fresnel lenses on Amazon, but then you gotta build the and line it with foil.

He has a video of that project as well. It's not as good as this version but can look very nice.

https://youtu.be/8JrqH2oOTK4

I wonder if you could make a better fresnel version, with less chromatic aberration by using a more expensive lens? The parabolic version is cool, but extremely unwieldily...

Also, are there fresnel lenses with a much shorter focal length? The dream would be to have something like this in a ≈ 10 cm (4 in) thick picture frame hung on the wall...

Check out Edmund Optics. They do have precision fresnels but they're tiny.

The lenses I've seen at EO have focal lengths that aren't much shorter than the side length---20cm for 25cm sheets. I'm pretty sure this is a limitation on the refractive index plus maybe the precision required for steep incidence angles.

I thought there was some people building things similar to this (not as bright) and they were using lenses from the front of LCD TVs. But I may be remembering the details wrong.

This same guy has a video doing just that somewhere in his history. It's not nearly as bright though which is the big difference.

Now when you mention it, I realised that the guy in TFV have actually done something like that: https://www.youtube.com/watch?v=8JrqH2oOTK4

I wonder if you could use a spotlight like a bit like this: https://www.amazon.com/Primelux-8-inch-14400-Lumens-Driving/... to get the intensity and parallel rays.

There is also severe lens flare to deal with for fresnel lenses.

At the risk of making a too simplistic comment, I must say this is an astonishingly great video. Period.

Part of me is like, no way I want that, it isn't natural. And the other part is like, yes way I want that, what I do with indoor lighting isn't natural anyways!

Why do fresnel reflectors seem to be so hard to find? Seems like they’d be a good option here.

Interestingly, you can DIY that too: http://www.dr-iguana.com/prj_FlatPackMirror/index.html

And I think they really could be a good option. Unlike the lenses, there would be no chromatic aberration with a reflector, right? Plus, you can set the focal length to whatever you want (taking practicality into account of course).

I might have to try it out...

If you could find two parabolic mirrors with a focal length difference of less than a couple of inches, I suspect you could take the middle of the short focus one and the outside of the longer focus one and stack them concentrically, to get a flatter dish.

Honestly for this use case, just sliced and stacked a single dish would give you mostly-collimated light, right? If you put this behind a mild diffuser you could probably still trick your brain unto thinking it's sunlight coming through the curtains.

I wonder this as well, perhaps it's a problem related to manufacturing shrinking sized concentric steps?

If my understanding is correct, fresnel lenses effectively operate as a larger lense surface by concentrating light with flat or curved steps getting smaller in size packed towards the core.

Unless I’m mistaken, you should be able to invert a fresnel lens stamp, make the same plastic lens and then silver it, so now it’s a concave mirror instead of a convex lens.

Do you mean just the lens? There's one in each of those old school overhead transparency projectors. Those should be easy to find.

https://www.bhphotovideo.com/c/product/592006-REG/Dry_Lam_50...

No that's a lens.

A Frensel reflector doesn't led light through bit instead reflects the light. Basically the same idea as the parabolic dish, but as a flat sheet.

Sounds like something you could 3D print.

That was a significant fraction of the links I could find, yes.

Interestingly I posted this 2 days ago but it dropped into the ether.

Ever since I seen it I've become convinced something similar will be part of a future garden office I'm planning. With a IR panel heater to add the warming effect. I was also wondering about whether UVB could be incorporated by but frankly I'd be too chicken to mess with that.

I personally hate sunlight for indoor activity. It's just too intense for anything involves vision, reading, writing, looking at screen, etc.

How often do you go out? I'm wondering if this a habit thing.

Seems like this idea could scale very well with more LEDs of lower wattage, like a local dimming array in TVs. That way

1. heat would be more evenly distributed reducing the need for large heatsinks

2. smaller fresnel lenses could be positioned much closer to the light reducing thickness

3. with a lot of LEDs there'd be a less noticeable grid at the edges of each fresnel lens from chromatic aberration.

I'd totally buy a 'sunlight window' product like this.

Do they make "directional" or "angled" Fresnel lenses?

Because you also really want the light to come out of the vertical window at something like a 45° angle downwards.

The physics of it should be fine, I've just never heard of them being manufactured.

And then just add a translucent blue diffusing layer on top.

It seems highly commercializable.

I suppose you could buy a big Fresnel lens and then cut out a square near the side, and only use that; it would focus the light but also change the angle.

There are flat sheets of prisms, that do what you need there. You probably know them from pictures you had as a child that change what they display depending on the angle you are looking under.

Anyone familair with this: https://www.tate.org.uk/whats-on/tate-modern/exhibition/unil...

I saw it at the Tate Gallery. It was one of the most impressive art installations I have ever witnessed.

Also I think those are sodium lights. More or less a single wavelength, hence the monochrome (yellow and black) look. Just about as far opposite to daylight as you can get!

It was a semicircle against a mirrored ceiling which is a neat trick to double the effect. People naturally lay on the floor like being on a beach and saw themselves reflected back way overhead.

Yes that's exactly what I thought of seeing this! It was so miserably cold and sleeting the day I went and inside it felt like an oasis. Everyone lying on the floor basking in front of that giant sun.

Ah, I have fond memories of that, a favourite for me too. I wish it was a permanent exhibit. His exhibition there last year was just great too with lots of light used in different ways.

A fun modification would be to add a mask directly in front of the LED to change its shape, for instance to mimic the weird crescent shadow effect that many of us saw during the last solar eclipse.

Some other fun things to do with it would be to point it through some plants in front of a fan to get the filtered-light-through-leaves effect, or to shine it onto moving water.

It looks amazing already, however another fun thing to do will be to vary the light intensity over time to make it more like the sun.

Then you need to vary the spectra as well.

I set something similar to this (didn't to be blue light scattering) for a friend who suffered from SAD (seasonal affective disorder) but had to stay living in Washington state for job reasons. They already had a basement with two sky lights and we used Cree 95 CRI LED stripss with a "2d parabola" (they were re-purposed from a concentrated solar project). It didn't do the 'sun at a distance' effect either, but the light quality was enough to get them through the winter while staying productive.

How about one for moonlight? My room doesn't have windows, and the nightlights I've tried are too harsh and make it harder to go back to sleep after waking up in the middle of the night.

Coelux has one: https://www.coelux.com/en/news/coelux-moon-at-euroluce2017-2...

Just change the color temperature and brightness.

Leds aren’t a good source of full spectrum light. Metal hallide bulbs are the way to go. 250-400 ways worth has a fantastic warm glow. Bonus points you can grow stuff under them really well

The Yuji LEDs he has are very high CRI (95+) and very expensive. Not sure what exactly metal halide CRI is, but Googling it seems its from 60-90.

I'm looking for a warm glow for my dining room.

Wdym with "250-400 ways" is that a brand?

autocorrect fail: it's "watts"; 250W & 400W are two common size for metal halide lamps.

Unfortunately they are full of mercury and can explode at end of life.

wouldn't that put out a lot of heat? like a lot a lot? that's one of the attractions of LEDs for lighting is they can produce a hell of a lot of light with very little heat coming from the bulb. all of the heat is in the electronics in the back. anything in front of the light will not wilt. of course, i'm thinking in terms of photo/video.

LEDs as you can see in the video aren't as efficient as you might think, since he has a HUGE heat sink / water pump system in place

you're missing the fact that that is the key point. the heat that is generated is not pointed at the person in front of the light. in photo/video, this is huge. talent/models don't sweat from standing in front of the lights. food products don't melt/wilt under the heat either. nobody really cares about the heat pointed away from the subjects.

For photography I agree. But if you're trying to create a "sun at home" because you don't have enough light/no chance to get sunlight, to feel the warm of the light can be nice...

Could indoor or outdoor artificial sun products soon be 'better than sun' (for humans)?

e.g. has the hormone regulation & vit d synthesis benefits, without the skin damage?

Because it seems like we're getting there: https://www.sciencedaily.com/releases/2017/09/170913193101.h...

Just take vitamin d.

The reason that people are deficient is because they don't take enough - the US RDA was shown to be 10x too low, due to an arithmetic error in an ancient paper.

This is an important problem that doesn't get the attention it should.

Quality high density, and dense time based living is the way forward for the poor to the rich. Be it sleeping downtown in NY city to working suburbs in Hong Kong. Good for efficiency and this is good for mental health.

This is AR that matters. It's nice work.

Wow. For some reason I thought it is already product on market with ton of alternatives. Fascinating.

See https://www.coelux.com/ and https://www.youtube.com/watch?v=aJ4TJ4-kkDw .

Other than the CoeLux I haven't seen anything like it

He did another video of something similar using diffusers instead from a busted monitor. I wonder if you could add a very thin layer of blue tape on the acrylic layer to achieve a similar effect?

Couldn't this potentially dangerously amplify ambient RF if you follow the author's advice of re-using an antenna dish & covering it with a non-conductive reflective vinyl coating?

It certainly won’t amplify anything since the only powered component is the LED.

It might concentrate RF at the focus to some degree, but only if you put your head right at the focus, and even then it’s almost certainly orders of magnitude less exposure than using a cell phone.

The focal point of any such amplification would be where the LED is mounted...so I think it would be a non-issue

The 3M ‘chrome’ vinyl wrap is conductive.

Does anyone here have personal experience with CoeLux? I would like to know 1) how realistic does the light look 2) how much did it cost and 3) how easy was it to install.

personally id go with a lot of small LEDs and small reflectors. you could get something pretty thin then.

He has an older video on doing just that: https://www.youtube.com/watch?v=8JrqH2oOTK4

I can't believe it's not the Sun

Agreed if you just took some of the parts where he has the window setup, you'd have a hard time telling the difference.

... that looks realistic to a digital camera that doesn't record with high dynamic range?

Taking it as given he's pre-armed to be enthusiastic, he projects a sense the experiential outcome is good:

Yes, he's biassed. No, that doesn't mean he was lying. The realism is about the subjective experience. If you went into the room, you probably would feel "this isn't bad" before you went to Meh.

This is not new. See this video from 2014:

https://www.youtube.com/watch?v=aJ4TJ4-kkDw

This light is sold by https://www.coelux.com/. They might also hold the patents on this technology.

Yeah, but it show how you can do it by yourself and not spend 30k on manufactured one.

Reply on news.ycombinator.com