Slashdot Mirror
Spectrophotometer Analysis of Crayons
Volhav writes "Like many as a child, the photographer Mark Meyer wondered what the difference between Yellow-Green and Green-Yellow was in that Crayola box of crayons. Using a monitor calibration tool and the Argyll 3rd party software he evaluated a box of 24 Crayola crayons, and plotted them out with sRGB values. He even included a nice printable poster size version of the chart in his blog post. For the geek or curious this was a pretty interesting plot."
24, not 64.
He's also on twitter: @markmeyerphoto
Just in case you want to say hi.
(Always pimpin' my fellow Alaskans)
Support FSF: Stop thinking with your wallet, and think with your imagination. (cc/non-commercial)
Note that chemists have real recording spectrophotometers, not just monitor calibration gadgets. I spent way too long in my youth, if I recall correctly, classifying iron ore samples using one. Its a fairly elegant technique because accurate ultra wide range light sensors have been old stuff for decades. It seems like I took an entire 200 level quantitative chemical analysis chemistry class where all we did was F around with a spectrophotometer in different ways.
If I recall correctly, the infrared ones were the cats meow before NMR got "cheap" for classifying organic compounds.
"Science flies us to the moon. Religion flies us into buildings." - Victor Stenger
Has anyone done a similar plot evaluating the taste of the crayons?
And who pinned these mittens to my jacket?
One step closer to understanding some of the recent legislative changes...
I "work" with color science. We have an X-Rite spectrophotometer just sitting around. Takes an artist's thinking, I suppose.
I ran into a paper a while back where the author captured spectrum of 100s of "natural" objects. Rocks, leaves, skin, etc etc. Made for an interesting chromaticity diagram.
"But there's way too much information to decode the Crayola. You get used to it. I...I don't even see the spectral signature. All I see is blonde, brunette, red-head..."
Set your phasers on "funky"!
Disclaimer -- I'm a flashaholic (see candlepowerforums.com for a sample of the geeks who adopt that label), your interest may vary...
The spectra are where it's at.
See, LEDs, CFLs, MH, and basically every other light source that isn't a glowing wire all suffer from the same deficiency -- their spectral distribution is "wrong", causing them to light up some objects well, and others poorly. Matching color of the light isn't enough -- mixing monochromatic light of two wavelengths (say, appropriately selected blue and yellow) are enough to make a "white" light, but almost any object will look wrong. An object reflecting only, say, red above 650nm, would appear black, since there's no red light for it to reflect; a broader spectrum red might appear yellow, but nothing will appear true red.
White LEDs typically have a narrow blue peak from the LED proper, and a broad yellow peak from the phospor coating, which combine for "white", but have serious red and minor green/cyan deficiencies. High-CRI LEDs have multiple phospors to broaden the yellow peak more, reducing (but not eliminating) these weaknesses. Other light sources have their own characteristic spectral deficiencies (including incandescents, even halogens, which while often touted as standard (thanks to CRI), are rather weak in blue due to their low CCT).
The most common measure for "properness" of a spectrum is CRI, which is based on accurate rendition of a series of standardized dyes compared to a blackbody source of equivalent temperature (roughly, color) -- there are numerous reasons it's a bad system, but it's in use, and a whole lot better than nothing. However, anything properly selected for high reflectance in the LED's weak points and weak reflectance in the LEDs strong points will allow similar comparison of LEDs to halogens. For those of us who experiment with using red and/or cyan LEDs to "fill" the white LEDs' spectrum, but can't afford a spectro, this info should be pretty damn helpful -- observe the difference between a box of crayons under halogen and LED lighting, see which crayons are most different, look at the SPDs, and you can see where your wavelength deficiencies are.
full-color PostScript plotter uses crayons as drawing filiments...
does it exist?
The Admin and the Engineer
"Using a monitor calibration tool and the Argyll 3rd party software he evaluated a box of 24 Crayola crayons..."
... that's our secret recipe, just like Kentucky Fried Chicken (excuse me, KFC) or Coke (excuse me, Coca-Cola (Excuse Me, Coca-Cola Classic)). Now kids can just use it for free on the intertubes?!? And *no* royalties or renewables to parents to renew!!??!!??
But...but
Release the lawyers!! Let the DCMA slaughter begin -- Leave No Chemist Standing!!
(...and someone get me Jackson about that D-M-R + microwavable chip thing he was blathering about again.)
If the universe is someone's simulation -- does that mean the stars are just stuck pixels?
A monitor calibration tool is not a spectrophotometer, its a spectroradiometer.
Only the bottom end of the calibration gadgets for monitors use characterised colourimeters. Many of the medium and higher end monitor calibration gadgets are proper spectrophotometers. Typically many of the devices where one device is capable of doing the entire colour workflow from scanner to monitor to printer and compare them to pantone charts are the real deal. The i1 PhotoPro is also roughly 10x the cost of your run of the mill "monitor calibration gadgets" like the i1 Display.
In this case his results even include a lovely spectrum as a result. Sure they don't cost $10k, they don't have a wide compliance for lux, they probably aren't 100% accurate and don't read outside the visible band, but that doesn't not make them real spectrophotometers.
For me a spectrophotometer is an AAS (Atomic Absorption Spectrophotometer) or an ICP-OES (Inductively Coupled Plasma - Optical Emission Spectrophotometer) or, if you want really low levels, an ICP-MS (Mass Spec)... Looks like working 15 years in an analytical lab kind of skews your view on more common equipment :-)
Has anyone done the conversion to sRGB values via CIELAB?
GCHQ Quantum Insert installed. If only our tongues were made of glass, how much more careful we would be when we speak