Hacker Newsnew | past | comments | ask | show | jobs | submitlogin

I tried doing something like this last year, I purchased lots of 1000 lumen LED bulbs.

The light was blindingly bright to look at and it looks very strange in a room.

My takeaway was:

* that you need some type of diffuser.

* The light leak out of the apartment was crazy. My apartment looked completely insane from the street.

* The light warmth is very important. And proper sun coloured lamps don't exist or are crazy expensive.



I used "rails lights" (the kind they often use in boutiques) and pointed them up at the ceiling (rails are mounted on walls, not the ceiling). There is almost nowhere in the apartment where you can look directly at a light source. The result: bright but pleasantly diffused light.


Very cool. I had considered something like this: install "crown moulding" a few inches shy of the ceiling and running rope lighting along where it will not be seen — but can scatter across the ceiling.


This is a super common design technique - you can buy LED ready moulding:

https://duckduckgo.com/?t=ffab&q=crown+moulding+led&ia=web


My brother has that in his apartment and it's not very bright. It's fine in a bedroom if you just want to relax, but not as a main light source.


"Warmth" and "naturalness" are produced by high CRI index (90+).

I have 8 lamps with CRI 90+ in my living room, at 5000K color temperature, 1500 lumens each. They produce nice light that seamlessly mixes with daytime sunlight from the window: they seem to have the same color.

I also have a few LED strip installation over desks and tables, very bright at max dimmer setting. A diffuser is a must, even though it lowers the brightness noticeably. Without it, anything moderately shiny will reflect painfully bright dots of light.


Quick guide to CRI

<90: Crap

90-95: Can be ok, or not. Depends. Different lights will likely not mix at the same color temperature.

98: Good

99-100 full spectrum cyberlight: As good as Halogen and only ~30x as expensive.


R9 is also important in the higher CRI ranges. CRI is based on pastels - it's an artifact of measuring traditional florescent bob performance - and R9 captures the vivid reds that are important to truly accurate color rendition.


Just from the scale I think it's clear that CRI is not that good a measure. As you say, CRI uses rendition of a dozen or so colors as a measure. The scale is heavily compressed - orange sodium vapor streetlights which are almost monochromatic "achieve" a CRI of 30, which is nowhere near "colors are 30 % good". Similarly, mercury vapor lamps have a CRI of around 50, and yet are complete garbage. All cheap LED fixtures are CRI 80 or 85 which is also garbage. 95 and 98 sounds very close, but aren't.

SSI - spectrum similarity index - is probably a better measure, because it doesn't look at a few colors, but at the entire spectrum, and so heavily punishes peaky and discontinuous spectra as produced by many LEDs. High SSI values also result in much better mix-and-match behavior of light sources. Even CRI 95 LEDs often don't mix well, usually on the pink-barf-scale (tint).


How can one find, or measure, the SSI of a given light bulb?


Funny quip I heard a few days ago from a guy designing LED systems for parking garages: "in a public lot your CRI only needs to be good enough to see the difference between blood and motor oil"


Can you share a link / the name of the product?


I'm a big fan of Seoul Semiconductor's SunLike LEDs.

http://www.seoulsemicon.com/en/technology/sunlike/casestudy/


Thanks! Looks like they make the GE Sunfilled bulb, which is available for $2.57/bulb right now at Lowe's: https://www.lowes.com/pd/GE-GE-SUNFILLED-60W-A21-5000K-1CT/1...

They claim to have a CRI of 97. They're dimmable and claim to be circadian-rhythm-friendly by not having a "blue spike", but don't claim to dim to a warmer tone.

Bought 21 of them for use with the 7-bulb splitter that Ben Kuhn now recommends, linked elsewhere in this thread: https://www.amazon.com/gp/product/B01FKIE6M4/ref=as_li_ss_tl...


I found that:

1) Most modern LEDs are surprisingly good for being sun-colored. It's luck-of-the-draw, though. I once bought some very expensive CFL bulbs from B&H, rated as 90+CRI, and they were literally tritone. I bought some bottom-of-the-barrel LED bulbs at Walmart, and they were great (until they started failing a few months later).

2) You can test this yourself with a $1 pair of diffraction glasses. They're sold as "3d" or "rainbow" or similar kids toys at dollar stores, or with a shipping markup Fleabay.

3) Diffuser is important, but it's also possible to just install many bulbs all around.

4) Consistency is important. If your hallway feels dimmer or brighter than your room, it's not great.

5) Changing color temperature is nice.

I don't think there's a business model here, though. This stuff isn't expensive, but most of the cost is educating the customer. Once a customer is educated, they can do the same thing themselves without your help (even with cheap Walmart lights).


The market is flooded with fakes and shittily QA'd products (often under previously good brand names) - having a turnkey solution with guaranteed quality would have a lot of buyers, and there is a lot of potential to grow a new solid brand.

If you do, please don't sell to private equity right away. Let us all get a couple years of a real solid product at least.

I'm still pissed at the (well known brand) USB stick I bought (marked high speed!) due to shitty engineering would overheat on any transfer longer than about 10 seconds, then throttle to KB/s read/write speed. Average speed was 1.5MB/s.


> You can test this yourself with a $1 pair of diffraction glasses.

I'm reading between the lines here: I think you want to see as continuous a spectrum as is possible with LEDs, not three discrete bands (R, G, B)?

I know LED are not spectrally continuous like incandescent (black body) radiation. Maybe the phosphor coating excites in a more continuous spectrum?

> I don't think there's a business model here, though.

I disagree. I would pay a premium if a company could offer a line that hit the mark on all points: temperature, brightness, spectrum, non-flickering.

My alternatives today appear to be a kind of roulette. And a lot of money is wasted at roulette....


Correct. You can see a continuous spectrum on some bulbs, and bulbs split into three (R, G, and B) for others. It's very obvious. You can't tell a CRI of 80 from 85, but in my experience, it doesn't matter very much.

You're correct that the alternatives now are a kind of roulette. A new company would add one more bullet (or empty chamber) into the barrel. There is no way for a customer to know "Company X solves this problem correctly." If you run LightBulbX and do everything right, and I run LumiNosiTechNo and have the same marketing pages as you, people won't know better.

If reputation builds up somehow, LightBulbX is just as liable to get bought up by an investor who milks that reputation by selling $2 products for $1000.

A value proposition isn't the same as a viable business model. I think there's a clear value proposition which many people would in abstract pay for in an abstract world of perfect transparency and information, but we're not in that world. I don't think there's a viable business model.


You should try with a short arc xenon bulb like the Osram XBO series. They are among other things used architecturally for emulating daylight inside buildings.

Downside is you need a 70A 28V power supply and a semi-massive cooling system. And bulb lifetime is approx. 1 year of normal use.




Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: