The Art of the Shadow: How Painters Have Gotten It Wrong for Centuries (mit.edu) 29
An anonymous reader shares a report: Shadows can do some adventurous, sometimes malignant, poetic things: They move, rebel, hide, refuse to be identified, vanish. All these visual aspects provide fertile ground for complex metaphors and narrations. Shadows are so visually telling that it takes little to move into emotionally tinged narratives. But it is the visual aspects that we primarily deal with here, with a special focus on several types of misrepresentations of shadows -- shadows doing impossible things -- that nevertheless reap a payoff for scene layout and do not look particularly shocking.
Painters have long struggled with the difficulties of depicting shadows, so much so that shadows -- after a brief, spectacular showcase in ancient Roman paintings and mosaics -- are almost absent from pictorial art up to the Renaissance and then are hardly present outside traditional Western art. Here, we embark on a journey that takes us through a number of extraordinary pictorial experiments -- some successful, some less so, but all interesting. We have singled out some broad categories of solutions to pictorial problems: depicted shadows having trouble negotiating obstacles in their path; shadow shapes and colors that stretch credibility; inconsistent illumination in the scene; and shadow character getting lost. We also find some taboos, that is, self-inflicted limitations on where or what to depict of a shadow. [...]
Painters have long struggled with the difficulties of depicting shadows, so much so that shadows -- after a brief, spectacular showcase in ancient Roman paintings and mosaics -- are almost absent from pictorial art up to the Renaissance and then are hardly present outside traditional Western art. Here, we embark on a journey that takes us through a number of extraordinary pictorial experiments -- some successful, some less so, but all interesting. We have singled out some broad categories of solutions to pictorial problems: depicted shadows having trouble negotiating obstacles in their path; shadow shapes and colors that stretch credibility; inconsistent illumination in the scene; and shadow character getting lost. We also find some taboos, that is, self-inflicted limitations on where or what to depict of a shadow. [...]
Shadows (Score:1)
Re: (Score:2)
It hasn't always been so easy, though. I remember one game I made with an engine which didn't support textured polygons where I was subdividing triangles in real time to render some of them with a darker colour.
Re: (Score:2)
Re: (Score:2)
It was the in-house renderer, and the company was a firm believer in using only in-house libraries. It finally adopted OpenGL after I'd left.
Re: (Score:2)
Re: (Score:2)
Gradients, yes. Textures, no.
Re: (Score:2)
Re: (Score:2)
No, the triangles were too large relative to the detail required in the shadow.
Technology changes (Score:3)
Shadows are an artifact of the type of lighting in a scene. Sharp, point source lighting is a relatively recent development. So sharp, well defined shadows are rare. Window light tends to be more diffuse, so that's what painters had to deal with. Take a look at some of Vermeer's work for examples.
Re: (Score:2)
I saw a documentary on Vermeer's work that I understood to be arguing that his method was practically a 'manually generated mechanical photograph'.
Re: (Score:2)
"Tim's Vermeer", by Penn Jillette.
It's only speculation though.
Monstly nonsense (Score:5, Insightful)
The article mostly looks at paintings where the artist has "done enough" to suggest lighting without confusing the composition or placing figures into darkness. There are a few that *are* just mistakes (like the cherub and the steps), but mostly these are artistic choices exactly analogous to a photographer placing lights to "fill in" areas or emphasise certain things. Nobody* is writing articles about how photographers have been getting it wrong for centuries.
*Of course somebody is. Probably on a blog.
Re: (Score:3)
it's news for shadowy nerds, but it's still interesting. i haven't read the full text so i don't know if the author has some spin, i just skim read and recognized most of the issues from the images. it is a nerdy inventory of failures, but it also illustrates .... how to put this ... an intergenerational learning process
in early times painting was purely figurative, where artists just produce pictorial representations of concepts they had in their mind in a way that was recognizable by peers, but not necess
Re: (Score:2)
analogous to a photographer placing lights to "fill in" areas
Or now-a-days for people who think they are photographers because they can afford an expensive camera, but have no clue about lighting or anything else about optics that affects the photographic image, to use lighroom, darkroom, or capture one to lighten shadows. And often making the photos weird looking/obviously manipulated.
Why are they incorrect? Two reasons.. (Score:2, Insightful)
-- The artist didn't pay attention to what they were painting.
-- "Our Amps go up to 11!"- They did it that way because that's the way they wanted to do it.
The only way to find out for sure - ask the artist. Unless they left detailed notes on the painting, any other thought as to why is they did it that way nothing better more than a guess.
So many problems... (Score:3)
Did they try updating their graphics drivers?
Anybody else noticing this trend? (Score:3, Insightful)
I've been seeing this trend, and not just here, in tech circles for a while now that anything done by humans is wrong, broken, imperfect, and usually the message includes a strong hint that we should probably let machines take it over because humans are fallible and weak. This story feeds right into that. It's not artistic choices, it's "wrong."
I keep hoping I get a natural check-out before we completely give in to the urge to have the machines run the world for us, but it seems we're speeding right along towards that future faster every day. In twenty-two years we went from, "protect us government, no matter the cost," to "the machines will get it better!" And they probably will, when they pull back our population to a naturally sustainable number and use us mostly as tools of scientific observation and study.
Re: (Score:2)
Re: (Score:2)
TFA discusses artistic choices, how some of the physically wrong shadows go unnoticed, how some of them don't, how some correct shadows would interfere with the composition, and how depictions of shadows in paintings have evolved. So, the headline might "feed right into it", but the story does
The Art of Fudging (Score:5, Interesting)
I'm an amateur artist, I quite often fudge shadows and shading away from realism to emphasize or de-emphasize something. Knowing how and where to rig realty is a big part of skillful art and illustration. (I can't claim I always fudge properly, if there is such a thing.)
While it is true that early artists had less published training material, their critics did also. Artists often have to please both customers and art critics to further their career, and art critics are more likely to know the state of the art of shading, so there is a bit of cat-and-mouse chase over history as art knowledge grew.
Sometimes I'll make token shadows/shading so I can say "see, it's there" to any detail-oriented critic, but otherwise mostly fog it up out to make the customer or intent (message) happy at the expense of accuracy.
Movie directors often admit they fudge the laws of physics and lighting. When fudging cliches become too obvious or common over time after over-use, directors move on the next fad or gimmick. Entertainment and accuracy are often not friends. Shares much in common with politics.
Re: (Score:3)
I'm an amateur artist, I quite often fudge shadows and shading away from realism to emphasize or de-emphasize something. Knowing how and where to rig realty is a big part of skillful art and illustration.
This is true. I just wanted to add a lil to what you said about adding details at the expense of accuracy:
The main reason this happens is is you're converting a 4-dimensional space into a a 2-dimensional medium. When you do that you throw a lot of information out, so what is left has to complete the story. If you have a character in your piece, for example, and their shirt is the same color as the wall in the background, in person you might see them just fine but through a viewfinder (or paintbrush) you
Re:The Art of Fudging (Score:4, Insightful)
I think it was PIcasso who said, "painting is a lie that tells the truth". What we see with our eyes is a dynamic scene, mitigated by our brain processing light, color, and form in real time and interpreting it in various ways. We like to think vision is real, but it's really just as much a processed image as any false-color infrared picture you get from a space telescope.
Your paintings are static, and then they're going to be re-processed by human eyes and brains, and lit in various ways in your house or a gallery.
It's fudge all the way down.
I dabbled in all that in my 20s. The interesting thing about shadows in particular is to think about what color they are. I asked a guy one time, "what's the color of my shadow on that white stucco wall" and he confidently answered that it was gray because it's a darker white. The reality? It's blue because when you took away the Sun it was mostly reflecting the sky. It might have also reflected some trees; but in that location it was sky. I blew his mind when I pointed it out to him. You'd have to use blue from your palette to paint it, I imagine--but maybe not since I've never actually done that. Certainly a good artist is going to build a knowledge base of how to portray such things, and it would probably be counter-intuitive.
Now add on the fact that if you're outdoors painting, the lighting outdoors is nothing like where the painting is going to be viewed, and the paint is *wet*, and a good artist can account for that.
It's pretty cool what humans can do sometimes.
Excellent explanation (Score:2)
I have thought along those lines but was never quite able to express this quite that thoroughly and concisely.
I just have to ask. (Score:2)
Pity the poor AI image interpreters. (Score:2)
It just struck me how this (and similar artistic technique vs. real-world physics factors) would make things difficult for software attempting to infer 3-D models of 2-D images when it is confronted with artwork.
The artistic techniques and conventions would end up tuned to how the visual processing of the human brain makes sense of what the eyes detect, as the artistic communities develop techniques to create the desired effects in the mind of the viewers. An AI trying to interpret both photographs and non