I purchased a five-head gooseneck lamp So I can point the light all over the place and still have tons of it.
Those Costco bulbs mentioned above are really good and check all your boxes. Here’s the 100W- equivalent (actually uses only 17.5W) version: https://www.costco.com/feit-electric-100w-replacement-5-cct-led-a21-bulbs-12-pack.product.4000185974.html
They are also CRI 90+, which is really important for good color reproduction. They let you select between three color temperatures so you don’t have to guess which one you prefer.
A few years ago, the government started requiring “lighting facts” labels on bulbs too. They look like nutrition facts and tell you what you need to know about the bulb: https://www.bulbs.com/learning/lightingfacts.aspx
It should show you:
- Brightness in lumens
- Power usage in watts
- Color temperature in Kelvin, along with how warm or cool it is (yellowish or blueish)
- More recent bulbs of higher quality should include the CRI too
Really helps to look at those instead of the marketing on the box
Those are what I use too. The first package was such a revelation! I can actually see in the rooms where I have replaced incandescent with them.
Actually, 5 color temperatures.
Whoops, you’re totally right. Any shade of off-white you want!
I have a bunch of these around the house, but set them all to 2700K. They work great.
That is an excellent illustration of color temperature. Thanks for posting it.
I’m big into photography and flashlights and I’m going to load that on my phone so I can explain to folks.
It makes sense to me. The ones at the coolest end make me think of a January day in Saskatchewan with clear blue skies and blowing snow with a high windchill. The ones at the warmer end make me think of sitting by a bonfire.
Back when CFLs were the rage, I bought a case of 100W incandescents because the CFLs were so poor. I still have…um…all of the case. If you really want, contact me off-list and we can negotiate a hostage exchange.
Awesome - I will give it some thought!
[I didn’t read the thread]
I’ve got no end of eye problems.
Willing to spend some money? This may be just what you’re talking about:
You may also find/decide that they are only really important in one or two areas of your house, and you can economize in the other areas.
Geeze, those are really expensive at $29/bulb, vs the $3.60/bulb for the Costco Feit ones.
For a difference of 5 CRI (95 vs 90) is it even worth it? Can anyone actually tell the difference without lab equipment?
I went with the first resource I saw. They’re available cheaper elsewhere, eg:
When I was woodworking, the few extra CRI – coupled with my eyes – made a big difference in my shop, particularly when it came time to apply finish.
YMMV, of course.
ETA: an Amazon search for full spectrum LED bulbs
I bought these light fixtures at Home Depot for my cellar. They were the right size and price, but when I got them home I realized that the 4000K color was just too bluish and unpleasant.
I didn’t want to return or replace them, so I bought color gels, like I used for theatrical lighting, to lower the color temp to about 3000. Much more pleasant.
Characterizing bulbs by their color temperatures seriously pisses me off.
Color temperature is a concept that makes sense if the spectrum of your source closely resembles that of a classic blackbody (an idealized opaque material that radiates equally in all wavelength regimes). Once you give the temperature of a blackbody, you can draw its spectrum - it’s completely determined.
Using blackbody temperature as a metric made sense when sources closely resembled ideal blackbodies in their behavior. Get anything hot enough and it starts to behave like a blackbody. Hot tungsten filaments act a lot like blackbodies. The sun, as already noted, acts a lot like a blackbody. surfaces coated with carbon black act like blackbodies. Nernst glowers act like blackbodies. People used these kinds of things for lighting, so it all made sense.
But it makes no sense o try to characterize, say, a laser beam by color temperature – all the light is coming out at a single wavelength, which doesn’t model well as a blackbody curve. It didn’t stop people from doing it, though.
Now look at LEDs. Th way they make white LEDs is not to combine output from several different color LEDs at the same time (although people HAVE done that). It’s to take a short wavelength LED emitting in the blue or purple and add a phosphor that absorbs that short wavelength light and re-emits the energy across a broad swath of longer wavelengths. As a result, the spectrum of a white LED is a spike at short wavelength followed by a wide peak centered around the yellow:
This isn’t a good match to the spectrum of a blackbody.
What they do, in a case like this, is determine the Chromaticity coordinates of the light in a CIE diagram or the like and look for the blackbody temperature with the CIE coordinates that most nearly matches
Just for clarity, are you talking about the difference between a 95 CRI “full spectrum” bulb and another high-CRI bulb, like the 90+ Costco/Feit ones?
Or are you comparing the 95 CRI bulb to some generic cheap low-CRI bulb (which may or may not even have a CRI rating?)
CRI as a metric is generally useful for LEDs, but I would be very surprised if 5 CRI could make that big of a difference. Maybe it can (I have bad eyes), but just wanted to make sure that’s what you meant.
My wood shop bulbs were 4’ fluorescent tubes. In the area directly over my workbench, I had two two-tube fixtures, separately switched, so that I could have some light or a lot of light on my bench based on the task at hand.
Which gave me the opportunity to experiment with a couple different CRI’s in a ‘controlled’ fashion. For me, the higher CRI made a meaningful difference, particularly – as I said – when the differences in what I was working on/with were very subtle.
I’m sort of extrapolating that, in LED, CRI is CRI is CRI (as compared to fluorescents). Is that valid? I can’t be sure.
Not quite sure what you mean here — and @CalMeacham can probably explain this better — but CRI is a number from 0 to 100 that can be applied to any bulb type, and generally higher is better (more able to accurately show colors).
A regular cheap florescent (or a cheap LED, for that matter) might be in the 50s or 60s, which would be a huge difference vs the 90+ CRI “premium” lights. You can get high-CRI lights in both florescent and LED. An incandescent would be 100 or very close to it (it’s almost a blackbody).
It is measuring the similarity of a light’s output spectrum vs that of daylight: Lumistrips The Intricacies of Color Rendering Index: A Comprehensive Guide to CRI and Lighting Technologies
For example:
A typical florescent has a bunch of spikes in odd frequency ranges, giving them that unearthly glow many of us will remember from the old days. There are higher-quality ones used in the film industry.
LEDs were similarly bad at first, especially a decade or two ago. It wasn’t until recently (last 6-7 years, I think?) that high-CRI LEDs became both common and cheap enough to be the "default’ option at Costco. You can still find crappier bulbs at other retailers like Ace Hardware, Home Depot, Safeway, etc.
There may also be non-CRI tests that show even more colors, and some “full-spectrum” bulbs can do better there:
It is entirely possible that even among two 90+ CRI bulbs, one of them has a better “strong red” performance, like your bulbs for example, and that your eyes can detect that difference even if the CRI test doesn’t cover it. It’s just harder to quantify.
Some “full spectrum” bulbs are just marketing fluff that don’t do anything special beyond having an above-average-ish CRI (if even that). It’s not a legally protected term, so anyone can call anything full-spectrum.
But in the specific case of Waveform Lighting, they actually do perform and publish lab test results, so it’s very possible that their bulbs are above and beyond what a high-CRI bulb would get you — I just don’t know if it’s noticeable for an average person, or at least noticeable enough to justify a more than sixfold price difference.
Certainly if you compare their bulb to a shitty shop LED with no indicated CRI, it should appear much better. I don’t know how it compares to a 90+ CRI Feit with no special boosting of “strong red” . You’d have to look at them side-by-side with eyes better than mine, or use some lab equipment.
[bolding mine]
That was the part that I was assuming: that CRI meant the same thing, and was measured the same way, across all bulb types, because my personal experience with CRI was in fluorescent tubes.
Gotcha. And you’re saying the Waveform full-spectrum LEDs compared favorably to high-CRI fluorescents? (I entirely believe you, btw, just making sure I understood you… I am always on the lookout for “better” lights and would consider one of these for the main office light).
Edit: But it still is possible for two bulbs of similar CRI to have a different output spectrum, like:
Those are all 80+ CRI bulbs, with the Philips and Dicuno being 90+.
Compare that to the Waveform one and you can better see the differences:
(From Ultra High CRI LED Lighting | Waveform Lighting)
(A spike in the blue approaching UV, no red spike, other visible colors are very well balanced)
I don’t have any personal experience with Waveform Lighting’s products. They were the first result that I saw in a Google search for full-spectrum LEDs (which was my search criteria informed by my experience with high-CRI fluorescent tubes).
ETA: And the living areas in my last house … I converted the OEM CFL’s to “full spectrum” CFL’s. I liked the difference a lot, though the application was just ambient lighting, nowhere as visually demanding for me as the wood shop was.