is it possible that one color still be the same under 2 different light sources? it should from my standpoint, if both colors matches the maximum color of the material, lets say we've got this dark gray R:100 G:100 B:100, and 2 light-sources that got much more color, an orange light R:255 G:155 B:100 and a cyan R:100 G:155 B:255. but it seems to be so wrong, and its hard to test this in reality cause of little speculars involved on almost every material.
€: okey, my RGB calclations like "a material reflect what it gets" does not work out somehow, ugh, ... :/
Last edited by Mister Janchichan; December 31st, 2013 at 01:42 PM.
So just looked through the thread and it looks like most my questions have been answered. so thanks!
Mister Janichan: actually i think there is an explanation on this very forum, have a look for LINRAN color tutorial he goes into how to work out the correct values, hue etc using just the rgb slider.
Im really gutted about the whole band of saturation before the terminator not being the case, are there any specific circumstances when this would happen? (apart from the large light source)
still like to know if its possible that one object looks the same under 2 light-sources.
@siphonophores: somehow i dont get what you are trying to ask, what do you mean?
MJ: no problem the question was pretty vague. Basically there seems to be two schools of thought when it comes to shading.
1.) principle of uniform saturation
2.) highest contrast at the terminator.
However to my knowledge the first option is actually the realistic 1, while the second confuddles the specular and diffuse highlights. I just was hoping there might be a case where this whole band of saturation existed in reality, although i havent seen it as of yet.
also have a look at this thread, this has been discussed a few times.
but i tried to illustrate an example of one situation with speculars involved where this band of saturation occures, hope ill get corrected if someones knows better:
so, this is how i create such situation, on the red sphere, where its lightest (on the upper edges) its more desaturated because of the specular from the white surrounding. the intensity of the specular decreases as the angle changes, so its weaker where your eyes are straight on (shall be around 0-10%). so where the specular is weaker you see more of the balls local color, hence more saturation in this case.
the local color is a very pure red, so the color wont be much affected by the reflecting blue light of the pillow.
because of that i created the orange ball, which contains more blue and green in its local color, so it will be affected by the pillows reflected light. now you've got the following:
low saturatuion on the upper edges, higher saturation in the half-tone and lower saturation in the shadow
my thoughts on this: break it down into steps - first local color, shadow and light, then add speculars, think about surrounding light that bounces into you spheres shadow that could change your color/saturation in the shadows.
if i misunderstood someting and told you stuff you already know - i am sorry, but probably it helps!
but i would like to here someones thought on this!
i got something here:
it looks really plain, but i guess that it would look plain if it is lit 360° equally. i just lit the edges a bit cause of specular light from the surrounding.
is this something a white tower looks like on the shadow-site? i just took the blue from the sky and painted the tower with it...from my understanding -> if white gets lit by blue light it reflects blue..so in photos the shadows are always very greyed out and darker, but i dont trust the camera, a darker/grey would result from cloudy skies and/or 2nd-light reflect into the shadow.
Last edited by Mister Janchichan; January 2nd, 2014 at 10:57 AM.
somehow my attachments in the post above are broken:
MJ: I think those spheres are nice u r really making progress in this area. however I think you are having trouble with specular reflections. basically specular reflections are mirror reflections of the surrounding environment (the clarity of this reflection depends on many factors. the material is a big factor). The white dot u see usually on spheres is the light source (i.e the sun or a lamp etc) However i think i understood fro your answer why there is this band of saturation. (the specular reflection interacting with the lambertian/diffuse ref)
so i wanted to paint my spheres with very low specular-power,see the white desaturation on the edges which is weaker in the middle of the ball
i found a quote from david in another thread that explains exactly what i've done:
by the way, have a look @skin-tones, usually you will find this band of saturation in the half-tones as well, for the same reasons as explained OR because of subsurface-scattering of the blutvessels
Last edited by Mister Janchichan; January 2nd, 2014 at 04:35 PM.
mJ: thanks ill check that out.
Your light source seems to be coming from above but your specular seems to be like the ball is back lit. i thought it was the fresnel effect. Yea it def does depend on the mat to change the strength of the specular, i think the description u posted describes this
edit: also, i tried looking, im pretty sure this is floating around the net i just thought id throw it out here. Color vibrancy? does it only effect saturation or can it effect hue and value as well? thanks again
Hi Briggs (or anyone who has the answer) ALso just wondering what do you think is better the colour triangle or color square for hsv? Or are they the same.
SO i did some color balls recently and was wondering what people thought about them, they are meant to be in direct sunlight.
hey siph, these look great, but you could challenge yourself a bit more and tease a background and a groundplane so its better to tell if your effects are right. also you could vary the power of the specularity, i find this very challenging.
thanks MJ! I think thats a good idea i might try that soon. I think i missed how strong direct sunlight is.. so i did an edit.
here are some BALLZ! some are more glossy, some less
dood, all my posts disappear and stuff, wtf... next try...
and one question that REALLY REALLY gets me into trouble, i mentioned it before, but describe a bit simpler:
if a color reached its maximum of reflectance under skylight, how will it look under sunlight? the same right? i dont know..since i am working in RGB instead of HSB i stumble upon this alot. and i just mean diffuse reflection!
Last edited by Mister Janchichan; January 4th, 2014 at 03:59 PM.
sry i dont know that one, nice balls
Hi guys, my post from yesterday seems to have vanished permanently, but anyway I seem to have managed to get my email notifications working again. It's great to see so much activity here, but by far the most important thing I can say is to repeat what I said to Mr J on the last page:
If you are trying to paint simple objects from imagination make sure you actually understand the theory properly, for example with regard to the position of the highlight on a sphere (it has to be on a line from the light source through the middle of the sphere, not off to one side) and the importance of shading series.
Siphonophores: I hope you can see that due to your decisions regarding the colouring, you've made your spheres look highly translucent, which I don't think was your intention. Also, to show strong sunlight on an opaque object requires a strong contrast between the light and shadow sides. Did you examine a coloured sphere in sunlight when doing the exercise?
For the exercise of painting spheres into a photograph, I'd like to draw attention to this post, also on the last page:
In my attempt at one of Mr J's photos below, I first painted a white sphere under the lighting I inferred on the left, and then a coloured version on the right, using the procedure set out in that post. To follow this procedure, it is essential that you closely examine your photograph for clues that tell you about the relative strength and colour of the different light sources. I've marked in some of the clues I found useful for this into the photo, but I want you to see if you can follow for yourselves how I used this information.
you've noted the rgb distribution on my screenshot, it has a lot of color, very bright, the left ball, even in your version has less RGB color in sunlight than the sky has to offer, so why should it get darker in the shadows? i solved alot with just observation without asking anybody, but i am running at a wall with this.
is it just because sunlight penetrates the material more than the skylight or something?
if there is no answer to this i probably should rely more on the values than RGB-Distributions and create my shadows first of all on this information (sky value compared to white/sunlight) and alter the local-value at this factor.
Edit: Probably i got a Solution, Needs some validation, but i am optimistic
Last edited by Mister Janchichan; January 5th, 2014 at 06:54 PM.
Briggs: Thanks! I did not see that at all. I really struggled with trying to increase the contrast. I couldnt find many refs for direct sunlight in fact one i saw was a tennis ball which might explain the translucency thing. I will keep experimenting and seeing how to do this
Edit: i think i wanted a little translucency but not as much as i had there.
MJ. Im not sure i understand completely what u mean but i will try answer. Also Great job on those balls the yellow one to me really looks like direct sunlight.
Ok so I think you are on the right track. There is less light hitting those areas of the sphere. Also the reflected light due to the ANGLE, is less powerful (45% of the original intensity) keeping in mind that it is already reflected and therefore much weaker than the original lightsource that i think might explain it.
Also you can figure out the values in the rgb slider using linrans method tho personally i prefer the triangle thing.
Its been a long time since ive seen the sun.... It forgot these lands, went south for the winter
i slightly changed how i establish my colors based on the last talk. if i want to paint something under very colored light i first have to paint the locals compared on white to better think of how it would change after colored light, but i am sure after some time i am able get more confidence.
here are 2 of my latest little results
even if those sphere-studies are fun im looking forward to paint something more vauleable finally
edit: oops, didnt realise how small they are
as mentioned i compare the colors to whire first, thought it would be a nice excercise like i did here, no multiplymode or such:
Last edited by Mister Janchichan; January 6th, 2014 at 06:51 AM.
brigggs: Thanks! changed my approach on this now heh. time to go back to it
MJ: 1st one looks excellent! very representative
Ok so I tried to take what was said on board and came up with these. Now the first bunch were done before the crits were given so they are off. hopefully i got closer with the second batch. Then the third batch was a little light adjust to simulate that burning retina feel.
Edit: they were posted in the wrong order. No 1 is the burning retina.
So i was wondering, i was reading about the eye and specifically in regards to its ability to compress brightness in order to adjust the image.
Would you know how much they adjust the brightness for different times of the day like sunrise and sunset etc?
Heres what I found out so far.
There is a Logarithmic relationship between luminance and subjective brightness perception in the human eye.
Human eye uses a low pass filter and the band pass filter to compress the illuminance coming into the eye.
The fall off for light compression is halved every time the frequency doubles.
it can also divides 1/4 every time between the frequency doubles.
But how does the frequency apply to the amount of light in the eye? Is the frequency the amount of light in the eye?
From what i understood i thought there may be a significant decrease in the amount of compression that happens in low light circumstances.
Edit: FOund the info i was looking for.
Last edited by Siphonophores; January 12th, 2014 at 08:18 PM.
How does the principle of uniform saturation account for the brezold brucke shift? Does it account for the abney effect? and if it doesnt how could we incorporate this?
The striped sphere in white light looks really good, but your method only seems to be letting you paint setups where the main light and the shadow illumination are the same colour. The way I showed lets you show multiple light sources of different colours.
Great, those last blocks look a lot more convincing than the spheres. One thing to at least be aware of though is that the gamut of digital colours is quite differently shaped to the range of commonly occurring natural colours, and extends well beyond the latter for many hues, especially in the greens and in the red-magenta-deep blue range (not so much in the yellows and greenish blues). So unless you tone down some colours below the maximum brightness and saturation possible, they'll tend to look a bit unnatural or even fluorescent (as in your green and magenta blocks).
thanks! yea this is definetly something to consider, I was thinking about it but i wanted to keep the high saturations in to simulate artificial colors and surfaces.
Now i read somewhere that a way to compensate for it could be to select your shading series in a kind of curve rather than a linear selection. The one question i had after reading your post was,
If the shift happens naturally would it only happen if your computer monitor was in Direct sunlight or is it a property of the Intensity of the light itself?
So say we select a color with 100 brightness (which i think equals 1000 Lux) Will this because of its intensity automatically create the brezold brucke effect, or is this only effect of the high (v high) intensity of Direct sunlight?
Sorry, I haven't really gone into the BB effect in great detail myself because to be honest it seems a rather arcane issue in the context of practical painting. There's even some debate over whether the effect exists at all in object colours, as opposed to lights. I just meant that within the range of brightness of RGB colours, if there is any BB effect then I assume you would get that automatically as colours of a particular "Hue" and "Saturation" got brighter. If what you are concerned about is how to depict the BB shift of a light that gets brighter than that (?), then yes presumably you would have to adjust the hue appropriately. Any BB effect would of course be overprinted and possibly swamped by a difference in colour of the illumination between the shadow and light areas.