Colors Glow

Hi...
Jesse In Utah. I Have Trouble With Radiosity Glow Effect…Deeper Blues,Purples,Reds And Oranges Don’t
Glow At All Much. When You Turn Up The Intensity Lever's Higher… Too Say 5 Or 6…Still Doesn’t Have Any Effect. The Rest Of The Other Colors Have Good Glow. Is There A Way To Fix It…Sending You An Attachment...A 3D Model. You Can See It…


Thank You…Very Much

mailjesse@icoud.com
 

Attachments

  • Jesse 3D.zip
    23.5 KB · Views: 26
Could be a monitor/display setting or different perception. I find it quite believable. I used Falcon/Solid shader intensity 2,0, no extra light source:
GlowingSolidShader1.jpg
GlowingSolidShader2.jpg
 
I Have Trouble With Radiosity Glow Effect…Deeper Blues,Purples,Reds And Oranges Don’t
Glow At All Much. When You Turn Up The Intensity Lever's Higher… Too Say 5 Or 6…Still Doesn’t Have Any Effect. The Rest Of The Other Colors Have Good Glow. Is There A Way To Fix It…Sending You An Attachment...A 3D Model. You Can See It…
It's partly about the renderers, but it's mostly about the spectrum. It appears you're expecting all the colors to be equally visually bright, but you can tell by looking at them that they aren't. 3D renderers know that.

Frank is very correct that Falcon renders radiosity colors across the spectrum more consistently. Also, your monitor produces colors using RGB LEDs - primary and secondary light colors will just be stronger. All the more trial and error needed to explore it.

In the light spectrum, the primary colors are additive: adding Red to Green gives you Yellow which is a brighter (more energetic) color than either of them alone. Adding Red to Blue gives you Magenta which is brighter than either of those two colors, but still less energetic than Yellow because neither Red nor Blue starts with as much energy as Green. Blue and Green gives you Cyan, which is closer to being as bright as Yellow, and combining all three gives you White, the brightest color available.

Ditch starting with purple when you initially think about color strategy - start with the primary and secondary colors (RGB and CMY), then tweak them to taste, because anything off of those primaries will have 'costs' of brightness and/or vividness that add up. It's harder to regain those losses than it is to purposely discover those that are acceptable.

I.E.: the value of the purple you chose - it's very close to the hue value of the Blue test color. Your test needs a Magenta object with a similar intensity setting to the one you used with Red - then you'll see a similar radiosity halo effect (under the Cheetah renderer). Choosing your initial test values using RGB settings instead of selecting values visually will yield more consistent results.

Consider this: a few years ago, I tried to locate real-world LED lights that were Maroon, to decorate my house (lots of Texas Aggies in my family). Do you know what the professional lighting Trade Definition for LED-based Maroon lighting is? Red LED lighting at 50% power - no blue at all. Laundry detergent cleans your clothes, but making them "whiter than white" is not misleading - it leaves behind brighteners that reflect at the extreme upper end of the visual register. Those reflections overwhelm stain colors - even natural white itself - to appear to be WHITE. That's why ultraviolet lights make white clothes glow in the dark. The human interpretations of light can be funny, so start by relying on the math when you can.

Also, remember what color objects and ambient light are doing the reflecting. Low energy reflected light numerically approaches neutral lighting - which means they appear to be more and more neutral in a render than the reflective source (Purple light starts near the numerical definition of gray, dim lighting is close to gray, etc.). Gray objects that catch radiosity reflections may brighten but still be much less vividly colored, and colored objects will weakly reflect colors they don't normally reflect well. (I own a translucent green water bottle that I keep on my bedside table - the green LED of my clock makes it appear orange. It freaks me out.)
 
In addition to it possibly being a monitor issue, the properties of LIGHT may be an impact. The spectrum of light begins with infra-red and ends with ultra-violet; both of these are NOT visible to the human eye. As Infra-red, being the lowest vibration of light, begins to increase it's frequency, we begin to see red light, and just before the highest frequency of visible light moves into invisible ultra-violet, we see the color violet. Since we are not working with single frequencies of light (which are like laser-light), any color covers a bandwidth of color. In the case of reds and violets, some of that bandwidth may be in the invisible parts of the spectrum. As such, the same "power" of light in reds and violets may look more dim than other colors, because some of that "power" is in the invisible part of the spectrum. In my experience, rendering red objects in a dark scene is tricky and can take some trial and error since red is right before it goes black. weird but true.
 
One of the things I found out while researching yard lights was that some hi-power multicolor LED components are actually made of a Green LED, a Blue LED, and TWO Red LEDs. Also, Blue LEDs are a relatively recent thing, as it was hard to reproduce large quantities of them reliably.
 
Another interesting LED anecdote I just saw this morning. I'm not sure it would play into pure 3D renders, but since renderers are designed to emulate cameras (but with RGB channels deeply entwined with that process) some may actually reproduce a similar issue. Who's to say whether, at that level, they are emulating film cameras or digital cameras?

 
Back
Top