Yeppers. This is a parabola: http://en.wikipedia.org/wiki/ParabolaOriginally Posted by Number_6
Just tell me if I use too much math terms.
Pixeldragoon's parable asked his father for his inheritance early, then went out and spent it all on booze and chicks. It came home asking for forgiveness and Pixeldragoon threw him a party.Originally Posted by Pixeldragoon
LOL, i r teh funnay
Still some good stuff going on here! I'd like to see some people get past spheres and blocks, but the're still things to learn there too. About the parabola: I don't think that is a safe rule. In different lighting situations, and on different materials I think it will fall apart. The white point will change in intensity and brightness with the main (key) light source, its color and brightness, and how shiny (specular) the surface is. The saturated band will become thicker and flatter in most real world situations, as it picks up environmental light and as saturation will vary with light intensity/color. Lastly the bottom of the scale will always be above absolute black, and usualy have some color (saturation). Try running the color-picker over some photos and paintings, how closely do referances follow these guides? I think this is good stuff to think about!
Okay, I'm not a complete slacker... This thing only took a little while to do, and was very informative. This might be a nice way to start a b/w image, or plan values for a paint.
[Always remember that if a topic seems uninteresting, then it's just because you are picturing a solution that lacks vigor.] - William b. Hand
Yeah, still don't get the parabola thing; can it change shape or is it mathematically confined? Anyways don't worry about finding some hidden math secret, hell there might one for every situation (light is physics after all) but it's a lot harder to do it that way if you ask me.
Ok, here is a question for all of you:
What would the path be for a red object in the most powerful sunlight? Just for the sake of simplicity start it at black and end it at white.
@Number_6: Yeah, I got the parable joke. Dork .
@Kitsu: Good, good reasoning against the parabola, and your right. I suppose a lot of other situations might create what looks like a parabola as well though. Careful though, sometimes black is used/necessary. I'd caution against color picking from photo's as well; values and colors are really not as the human eye sees them. Anyways, good job on this last exericise; nice to see that someone is still around. I think you did really well on everything but the red window shutters. Squint at your photo again and see else they closely match in value. Yours seems to be about the second brightest value. The street light might be a little bright as well. You don't need to get it the exact value for the exercise just relative to everything else, however getting it as close as possible will help in the future.
Well, the parabola can be squished and rotated, but the general shape remains the same. Think of an ellipse: there are many different ellipses (almost round ellipses, flat ellipses etc.), but the general shape is still the same in every one.Originally Posted by Idiot Apathy
Mathematically speaking, a parabola contains all the points in (two-dimensional) space that are the same distance from a certain point and a certain line.
Yeah, I didn't mean that it works everywhere, I just noticed it to be handy in that specific exercise.Originally Posted by Idiot Apathy
Finally I get around to posting on this great thread! I usually paint in oils, but here I'll go digital to try to accelerate the process of teaching myself Photoshop (still don't know what two-thirds of the functions do). Please excuse my debut not only for its lateness but also for its lameness - just the first part of the first exercise. But I think I've spotted a few misconceptions about lighting and colour at large on this thread, so I'm hoping it might be useful for me to spell out my process in some detail, and put it up for discussion.
My objective was to draw a shiny red ball under a white main light, with a weaker white ambient light. The first thing I did was to put on my mental polarizing sunglasses and deal with the diffuse reflection (which we see as the red colour), before turning to the specular reflection (which we see as the white highlight).
To represent the diffuse reflection, which gives the modelling of the sphere, I needed to find a series of colours for the main light zones of the sphere: the full-light, the half-light, and the shadow. I also like to have a separate mixture for the centre-light, the most strongly lit area in the middle of the full light (an idea I got from a book on the teachings of Frank Reilly). The trick is to find a series of colours for these zones that will look like a surface of a single colour turning out of a light source - what we can call a "shading series".
Since in this case the main light and the ambient light are the same colour (white), there will be no hue shift between the centre light and the shadow. We want the total amount of light to diminish, but the ratio of wavelengths (which we control by the ratio of R to G to B) to stay the same. For a pure red ball I just had to find (by trial and error) some appropriately spaced numbers for R, keeping G and B at zero:
Centre-light: R 215 G 000 B 000 ( = H 000 S 100 B 084)
Full-light: R 204 G 000 B 000 ( = H 000 S 100 B 080)
Half-light: R 180 G 000 B 000 ( = H 000 S 100 B 071)
Shadow: R 102 G 000 B 000 ( = H 000 S 100 B 040)
Looking at the HSB values for this series you'll notice something very interesting: H and S stay constant and only B changes. This suggests to me that in Photoshop, finding a shading series is much easier than in paint: provided that the main light and the ambient light are the same colour, all you need to do is decide on numbers for the Hue and Saturation of your surface, and then plug in appropriately spaced numbers for Brightness.
(Notice by the way that even though I am painting a pure red sphere, I refrained from using pure red (R 255) even in the centre light. I did this for the same reason that I would have to use a light grey for the centre light on a white sphere: something needs to be left in reserve to show the (brighter) specular highlight. If you use pure red on the sphere you will find that your white highlight persistently refuses to look like a highlight, and just looks like a white spot).
Next, it was just a matter of painting these colours into simple flat shapes whose arrangement was consistent with a definite location of the main light source. I used soft edged brushes and then applied a strong Gaussian blur to get it looking nice and spherical.
Now for the specular reflection (highlight). We see the highlight at the point on the sphere where the surface is at just the correct angle to bounce light from the light source to our eyes - that is, where the angle of incidence equals the angle of reflection. For spheres in general, this in NOT in the middle of the centre-light, but at a point on the line between the middle of the centre-light and the middle of the sphere as we see it. (Light hitting the middle of the centre-light hits the surface at right-angles, and so bounces straight back to the light source, not to our eyes).
For most materials, light does not change colour when it bounces in a specular manner (main exception: coloured metals like copper and gold), so here, where the main light is white, the specular reflection will be white. At the highlight there is actually an additive mixture of this white specular reflection and the red diffuse reflection, but on polished surfaces the specular reflection, which appears as a tiny image of the light source, commonly completely masks the diffuse reflection. Any roughness however makes the highlight more or less fuzzy, as tiny points over a diffuse area will lie at that exact angle that bounces light to our eyes. Over this fuzzy area there will be noticeable additive mixture of the two, which can be mimicked in Photoshop by brushing white over the red with brushes of less than 100% opacity.
Finally I airbrushed a little of the background colour onto the receding planes of the sphere, to suggest the neutralizing effect of specular reflection of light from the background in these areas.
Any comments on my thinking here would be most welcome. I've noticed that quite a few people here seem to have been (1) placing their highlight right in the middle of their full-light area, (2) making their full-light area less saturated (more whitish) than their half-light areas, and (3) giving their full-light area a rather indefinite shape. When all three tendencies occur together they combine to create the impression of a pearly sheen rather than a proper full-light, so that the spheres begin to look more like coloured pearls than solidly coloured spheres.
Last edited by briggsy@ashtons; May 22nd, 2007 at 06:24 AM. Reason: restoring accompanying illustration
@briggsy@ashtons: Your late and lame debut is much appreciated, really good stuff here. Much respect to you and your work. I wish I could visit you at your school, seems really great and exactly the sort of place I've been looking for. You seem to be a really humble and modest guy and I really hope you stick around!
Now, I'm still quite the amatuer so I have some questions if you have the time.
I understand the hue not changing, after all your lightsources were both white. Although I should think the blue background should have quite an effect on hue-shifting but that wasn't really part of the equation I think. Value makes sense as well of course; being directly linked to light after all. However no shift in saturation is where I am troubled. Perhaps I don't quite understand exactly what controls saturation but; is this only the case because the sphere is "pure" red? If it was a weaker red would the saturation not change quite predictably as well? If you have the time and if it would prove the point do you think you could do another one of these; this time with a weaker red sphere or perhaps with different colored ambient lighting?
Second, when can the full-light (not sure if that's the right way to put it) area become more whitish? Is the guidline just that it shouldn't be less intense than the half light? If I understand it correctly after a color becomes as intense/saturated as it can be it can then only increase in value towards white which will also decrease in intensity. Now, this again wouldn't happen on a "Pure" object that absorbed all but one color right? White light is after all, every color. So then assuming all this if your color becomes white the object is then reflecting an amount of all colors interfering with the color it reflects the most? Or is this caused by the specular? (It's a pity that this was the best I could write to explain myself, complicated words for a complicated situation I suppose).
Next question is unrelated to what you've posted but I'm hoping you have the answer, it's something I've been trying to work out.. Complementary colors will produce a gray right? Same with light or not? Now a gray light produced by complements wouldn't act the same as a weak white light either would it? Would it react with only the two hues contained? Or would it only add value since the hues are neutralizing each other?
Anyways, thanks in advance for the help and thanks again for your post; look forward to hearing from you. This stuff... well for lack of a better word excites me.
Edit: What's a jaffa?
Last edited by Idiot Apathy; January 6th, 2006 at 01:52 AM.
Welcome and thanks for the info! I like the way you're not just thinking about painting a sphere, but also about the color, hardness, reflectivity, etc. of the sphere. Two things I noticed about your sphere: First it is too perfect, it looks like a vector graphics illustration rather then a painting. Second the blue fuzz on the shadowed side is way too strong. It is uniform in value and nearly obliterates that shadow edge. To produce that effect either the atmosphere in this image would need to be very thick, or the background very brightly lit. Other than those your conception and execution look good. Your observations at the end are very good/true. Especially the centered specular reflection.
[Always remember that if a topic seems uninteresting, then it's just because you are picturing a solution that lacks vigor.] - William b. Hand
Idiot (mm, that doesn't sound right)
Thanks so much for your quick response. You've raised quite a few issues there. I only have time to respond to a few of them now - hopefully I'll have some more time in a couple of days.
Well, it happens all too often in photographs - when they are overexposed. The full-light areas of one or more elements of the picture becomes too high in tone to stay in their correct relationship to the midtones, and are forced into pale, washed out colours. For example, in a digital photo of a red ball: up to a certain light intensity, the amount of red light in the screen image responds in proportion to the amount of red light coming from the ball - the more red light from the ball, the more the red phosphors on your screen glow, and the more chroma (intensity) of red you see at that point on the screen. When the light increases beyond this range, the red reaches the limit of its response, and only the green and blue phosphors can continue to increase - result: this bit of the image becomes lighter in tone and lower in chroma (more white).Second, when can the full-light (not sure if that's the right way to put it) area become more whitish?
The only reason I can think of at the moment why you would want to make the full-light more whitish in a painting is to suggest the equivalent effect in the eye - that is, to suggest a spot of light that is much brighter than the eye is ADAPTED to. When this happens the light isn't actually white, in the sense of the red, blue and green components being actually equal, but the R, G and B receptors (speaking very loosely here) in our eyes are all fully stimulated so that we can no longer distinguish any colour in the light. When that is the situation that you are trying to depict then yes, the most strongly lit areas should be whitish. But otherwise the full-light colour should be the highest in chroma. The great thing about painting, both digital and traditional, is that you can set the full light colour as the richest one and work down from that to the darker tones, thus ensuring a natural luminosity that is lost in an overexposed photograph.
Not. You can have weaker or stronger light, but not grey light. Greyness is a property of surfaces only. We see something as being grey when we sense that it is reflecting all wavelengths equally, but not reflecting as much light as a white surface would. (It is actually a somewhat subjective property dependent on the surroundings - a surface can look white until you place something that reflects more light beside it, then it will look grey).Complementary colors will produce a gray right? Same with light or not??
Sorry, must be just an Australian thing. Dark chocolate ball about 1.5 cm in diameter, thin hard shell of orange-flavoured candy. Mmm, .... jaffas.What's a jaffa?
Maybe, but are you sure? Try holding up a glossy sphere or cylinder (a shiny coffee cup will do) just in front of a moderately lit wall, and take a close look at the receding planes, especially on the shadow side. The specular reflections of the background here can be surprisingly strong.Originally Posted by Kitsu
My apologies for the excessive perfection.
Please excuse me for my very... VERY late entrance into this thread. I just found it today and have decided to join. What a wonderful idea! Big thanks to Idiot Apathy for creating this.
Ok, I'm going to try and work on every single activity here, so to start off here is my contribution for the very first activity posted:
I started this one off by doing the colored sphere first... just because I felt like it. One major thing that I noticed when I converted the blue ball to grayscale was that on the original gray one, my shading is a bit darker. One thing that I want to work on is learning to add more ambient light to my images... needless to say that I didn't exactly do so in this excercise. I saw one experiment that dealt with that in here and I really think that that will help me once I get to it.
Any constructive criticism will be greatly appreciated. Thank you.
briggsy@ashtons: Thanks for taking the time to reply, I'm sure you must keep pretty busy. Any insight is greatly appreciated, I'm full of questions and am so eager I could probably sap your strength with a barrage of them; *ahem* which I am about to do... please answer only at your leisure!
Now, are you saying that an object's full light should quit when it can no longer become more intense? Once it actually has to loose intensity to gain in value? (re-reading that it seems a little off; rather the full light is as intense as the object should become? ... no not even that works properly I think) And then the colors above this should be reserved for say highlights? This being said, an objects full-light could of course exist in that area it all just depends on what it reflects - an off-white color if you will, right?
Good way to describe gray; made a lot of sense to me, thank you. But what about a hypothetical enviroment where two lights of complementary hues combine, what would this light look like? The object that would reflect it could only take on the properties of these hues however would it appear gray in some situations?
Lastly, Jaffa's sound tasty. Apparently the name comes from Israel; so perhaps they aren't just Australian... hmm, sweet tooth.
@Diabolic: Yeah, you should have been here earlier! Quit slacking; but seriously exercises are always open; there isn't really an order and you can pick and choose which you would like to do. I think that there are some you should do before others though, did I make note of these... can't remember. Anyways, I think you got the point of the exercise and I can see the effort you put into these; good work mate. Look forward to seeing what you do next .
Thank you very much, Idiot Apathy. I'm very glad to be here. I might skip around a bit then, if that's the case.
I decided to do part A of Prject #4 because it looked interesting and because I thought it would go well with the project that I just worked on. I'll do part B of this project when I get the chance.
My main issue with this was getting the reflection to actually look real. I don't even think that my reflections look real with the image that I have posted just because thet colors I chose were a bit too saturated. The ball looks as if it could be made of metal because of the reflection colors that I used. I wanted to go for a more matte look, but eh... maybe I'll try this out again sometime and try to achieve that.
Critiques always appreciated.