What's the wavelength of pure green?

[NightExplorer96]

Crap! I voted 515nm

Oh well, still love 532nm

 

[zyxwv99]

:wave:It looks like the HEX convertor may have been right about green after all: 510nm.

I had seen that number before in various studies, but couldn’t wrap my brain around it. It conflicted with my own experience of 532nm lasers, and with what others here were seeing. However, I forgot to take into account how color vision studies are usually done: with monochrometers.

Photopic (color vision) covers about 9 orders of magnitude of brightness. The bottom 2 overlap with scotopic (night) vision. The area of overlap is called mesopic (twilight) vision. In that region our color vision is comprmised. The 2 top levels have problems with glare, flashblindness, and the M-channel (green) getting overloaded. So that leaves about 5 good levels in between. In typical color-vision studies, before people look into the monochrometer their eyes are dark adapted for 10 minutes. Then they get something like 27 trolands (with a 2 mm artificial pupil). In other words, at the bottom of our 5 good levels of color perception.

To find out which wavelength is the most green, they show people a bunch of wavelengths at 10nm intervals, presented in random order, and with brightness automatically adjusted for each wavelength so they look equally bright. Then they have them answer questions, starting with “How red was it?” They have to answer on a scale of 0% to 100% in increments of 10%. Then the same thing for green, blue, and yellow. The test subjects have to make sure the percentages add up to 100%. Then one last question: how saturated was it? The results for each wavelength are then weighted according to the perceived saturation level.

When the results are plotted on a graph, the blue and yellow curves hit zero at the same point where the green curve is highest. That wavelength is then called “unique green” because green is the only color reported as being seen at that wavelength.

So then I looked at the data on Bezold-Brucke shift. It turns out that Bezold-Brucke only apples to colors seen simultaneously (or near-simultaneously). Otherwise there’s a similar phenomenon called “no-contrast hue shift.” I think that’s what going on here. In either case, 532nm, if it were bright enough, could easily shift to the equivalent of 510nm.

Now I’m trying to figure out how bright our laser dots are. If you have the specs on the laser (including wavelength, beam diameter and spread in mrad) it’s not too hard to calculate the brightness of the dot at a particular distance. However, I’m not sure how to convert that to perceived brightness in terms of lux or trolands.

 

[IsaacT]

My guess is 510.

 

[trencheel303]

This entirely depends on how you define green. Similar to how I describe what a lot of people see as yellow to be orange (i.e., SOX lighting to me looks positively orange), I suspect a lot of what people call "true green" is actually what I see as turquoise/teal/cyan.

I admit I've never seen a 520/515/510nm laser in person, but judging by the cacophony of photos and people's sigs, I'd always describe 532 as "about as green as it gets" and anything lower to be trending towards the blues. Such a colour has actually put me off buying diode greens, as I like blue, and I like green, but not the two of them mixed.

 

[IsaacT]

My 520 is extremely green. My 532 looks yellowish in comparison. Here is the best comparison pic I have. It is just ever so slightly more pure green than 532. No blue to be found yet. I am hoping to make a 510 someday and then I will make a comparison of all three at equal powers.

 

[trencheel303]

that is a great pic, but it mightn't be accurate as both the reds look red-orange.

I'll see one eventually, but being a diode laser (I much prefer DPSS) I have no real hankering for one.

 

[IsaacT]

Actually that pic of the reds is the most accurate I have. That is exactly what they look like to me. Bear in mind monitor settings can affect things as well, altering white balance and whatnot.

 

[BurningLaserGuy]

I voted 525nm even if my seeing balls aren't so acurate, as DrMario said...

Lol, another pure wavelenght thread... at least we will stop sometime because not a lot colors remains:eg:. We should make a pool called "What is the pure wavelenght of IR":crackup:

 

[Atomicrox]

:wave:It looks like the HEX convertor may have been right about green after all: 510nm.

I had seen that number before in various studies, but couldn’t wrap my brain around it. It conflicted with my own experience of 532nm lasers, and with what others here were seeing. However, I forgot to take into account how color vision studies are usually done: with monochrometers.

Photopic (color vision) covers about 9 orders of magnitude of brightness. The bottom 2 overlap with scotopic (night) vision. The area of overlap is called mesopic (twilight) vision. In that region our color vision is comprmised. The 2 top levels have problems with glare, flashblindness, and the M-channel (green) getting overloaded. So that leaves about 5 good levels in between. In typical color-vision studies, before people look into the monochrometer their eyes are dark adapted for 10 minutes. Then they get something like 27 trolands (with a 2 mm artificial pupil). In other words, at the bottom of our 5 good levels of color perception.

To find out which wavelength is the most green, they show people a bunch of wavelengths at 10nm intervals, presented in random order, and with brightness automatically adjusted for each wavelength so they look equally bright. Then they have them answer questions, starting with “How red was it?” They have to answer on a scale of 0% to 100% in increments of 10%. Then the same thing for green, blue, and yellow. The test subjects have to make sure the percentages add up to 100%. Then one last question: how saturated was it? The results for each wavelength are then weighted according to the perceived saturation level.

When the results are plotted on a graph, the blue and yellow curves hit zero at the same point where the green curve is highest. That wavelength is then called “unique green” because green is the only color reported as being seen at that wavelength.

So then I looked at the data on Bezold-Brucke shift. It turns out that Bezold-Brucke only apples to colors seen simultaneously (or near-simultaneously). Otherwise there’s a similar phenomenon called “no-contrast hue shift.” I think that’s what going on here. In either case, 532nm, if it were bright enough, could easily shift to the equivalent of 510nm.

Now I’m trying to figure out how bright our laser dots are. If you have the specs on the laser (including wavelength, beam diameter and spread in mrad) it’s not too hard to calculate the brightness of the dot at a particular distance. However, I’m not sure how to convert that to perceived brightness in terms of lux or trolands.

Still not convinced...

I've seen 532nm on a white wall at noon on a sunny day and it still looks far more yellowish than blueish.

And IMHO the brighter the dot the more it shifts to yellow, not to blue.

This entirely depends on how you define green. Similar to how I describe what a lot of people see as yellow to be orange (i.e., SOX lighting to me looks positively orange), I suspect a lot of what people call "true green" is actually what I see as turquoise/teal/cyan.

I admit I've never seen a 520/515/510nm laser in person, but judging by the cacophony of photos and people's sigs, I'd always describe 532 as "about as green as it gets" and anything lower to be trending towards the blues. Such a colour has actually put me off buying diode greens, as I like blue, and I like green, but not the two of them mixed.

I don't see any blue at all in 520nm, just a "softer" mint green. And I do see a very slight yellow in 532nm (or a lot of yellow, if they're side by side), but I'm still convinced the purest green is between 520nm and 532nm.

I do have some LEDs labeled as "cyan" with WL around 505-510nm and while they're obviously not cyan they do have a slight blue tint. Maybe teal would be the better name.

 

[zyxwv99]

I think you're right. 532 is not the most perfect green at any level of intensity, and I was getting confused on the hue shift direction. Brighter green should be more yellow. Also, some of these studies are going in 10nm increments, so I'm still looking into this.

 

[Atomicrox]

In the specific case of green I think we have an additional issue:
"Computer" 0x00FF00 looks yellowish, desaturated and very far from "pure green".

This isn't as evident on the other colors. "Computer" yellow, blue and cyan look fine to me. Red could be slightly less orange but it's fine.

 

[zyxwv99]

I agree with your assessment of the color computer dots, although I'm not sure if you and I have the same thing. I have cheap generic sRGB. I am determined to have more than one color space in this HEX converter. Maybe we should start with two: sRGB and Adobe RGB. If we can figure out how to do that, adding more should be easy.

I've also been looking at the other options. I think the brightness of the wavelength is probably more important than everything else put together. Ideally people should have the option of entering a number from a light meter, but otherwise there could brightness options like "clear sky at twilight," "white paper under reading light," that sort of thing, with numbers in trolands, lux, candelas per square meter, etc. Other options could go under "Advanced."

 

[IsaacT]

What about a bar that you can drag along that would linearly alter the color along the line of 400nm to 700nm. That way people could post they see. Especially since your 520nm is probably not exactly 520. None of this is precise(unless you have a spectro), so it doesn't make the most sense to be super precise about inputs. Just have a scroll so once you see what you beam looks like it will give you that HEX code.

 

[NightExplorer96]

*sorry for the bump-up*

Glad I found this thread. Im trying to go for a complete rainbow right now, and trying to found the "purest" looking green out there. I was originally going to go for 532, but as many people have said 525 is around the sweet spot it seems like 520 would be closer to true green!

JetLasers has some 70mW 520 units so I'll probably skip that instead of the 532's. Anyone here who owns a 520. What's the duty cycle like? I had a PGL-III-A awhile back at <5mW of 520 and it could only be run for like 1min, which sucks! If im not mistaken 520 diodes give off a lot of heat, no?

Thanks!

-Alex

 

[Richie89]

Lmao i just guessed and it ended being the most picked one! 525nm

 

[BurningLaserGuy]

*sorry for the bump-up*

Glad I found this thread. Im trying to go for a complete rainbow right now, and trying to found the "purest" looking green out there. I was originally going to go for 532, but as many people have said 525 is around the sweet spot it seems like 520 would be closer to true green!

JetLasers has some 70mW 520 units so I'll probably skip that instead of the 532's. Anyone here who owns a 520. What's the duty cycle like? I had a PGL-III-A awhile back at <5mW of 520 and it could only be run for like 1min, which sucks! If im not mistaken 520 diodes give off a lot of heat, no?

Thanks!

-Alex

Well, you said you want to go for a complete rainbow... doesn't that mean you need as many different wavelength lasers as possible? Sorry, I cannot answer your question concerning the 520nm laser for I am out of the laser hobby since 2015 or so... I just wanted to tell you that you awakened a thread from a really hated series of threads if I remember correctly. At least so it was back in 2014.