Vermeer’s Secret Tool: Testing Whether The Artist Used Mirrors and Lenses | Hacker News Reader

Vermeer’s Secret Tool: Testing Whether The Artist Used Mirrors and Lenses

vanityfair.com

177 pointsshawndumas44 comments

There's also this recent Kickstarter, the NeoLucida, that was all about remaking similar lens / optic devices artists and painters used in the past: http://www.kickstarter.com/projects/neolucida/neolucida-a-po...

Less kickstarter-y webpage: http://neolucida.com/

I have one of these. It arrived recently; I haven't really tried it yet, but will soon.

Wikipedia: http://en.wikipedia.org/wiki/Camera_lucida -- it notes the basic optics were described in 1611 by Kepler.

I have an antique camera lucida, and it is startlingly effective.

You use it to draw/paint? Can you please share something you've done? I'd very curious in seeing anything you've painted with it.

Interesting article (if a bit clumsily written) but I don't understand why it's so controversial. It should be pretty easy to tell if a photorealistic painting was made with mechanical or optical aids. There will be small positional errors in the image either way -- places where the perspective is almost but not quite perfect. The magnitude and direction of the errors (and how they accumulate) will be very different compared to those in a freehand painting, and shouldn't be too hard to analyze.

It's not that easy, for a few of reasons. One is the establishment (no better word for the mixture of subjectivity, entitlement, money, and fashion in the art world). Rewriting art history is frowned upon unless it makes the genius artist look like a greater genius artist. Applying a scientific method doesn't work there.

Another reason proving this is hard is that some artists do have a great ability to reproduce reality (this was really the most common way to measure an artist's ability before photography came along), but the economy of an artist's technique may reproduce some aspects of reality exactly, while other aspects are just suggested or completely ignored, so you'd need complete consistency.

One more reason is that simply seeing a camera obscura image or a similar 2d image may change an artist's technique, they may become aware of things they hadn't considered before (this was the case with linear perspective, look up Brunelleschi's experiment - though nothing as elaborate as a camera obscura). For a camera obscura projection, lighting conditions are very specific, that alone might shift an artist's (mental) focus from being 100% descriptive to being very vague about contours. Basically, you can't "unsee it".

Dutch artists from Vermeer's era are perfect candidates to have used lenses, because they had the technology. If you can find undeniable proof anywhere, it's probably there. Really the more proof adds up, the more ridiculous the naysayers appear. Some natural selection would do this field some good.

Thanks, in four paragraphs you gave me a whole new perspective on Hockney's book on this.

Hockney does not argue that every painter of realistic paintings used the techniques he describes. He in fact goes through a series of painters and discusses who probably did and didn't.

Basic clue: if the painter seems more interested in showing off his command of perspective details (e.g. the lettering on someone's desktop globe, or intricate floor patterns) than, say, people, then he's quite probably using the technique.

Interesting supporting factoid about the lenses not mentioned in the article: Antoni van Leeuwenhoek, inventor of the microscope, born in the same year (1632) lived two streets away from Vermeer in Delft.

It was a tiny city back then, the men very likely knew each other, so Vermeer would have had access to the best lenses and the world's leading expert of that time.

I've read Hockney's book, and it is completely convincing to anyone who (to paraphrase P. G. Wodehouse) isn't being paid to be unconvinced by it. In his book he finds geometric errors (mismatches in the textures and folds of tablecloths and floors, for example) and depth of field artifacts exactly where you'd expect to find them if the painter were using optics to focus images onto a drawing canvas, and had to change focus as he moved out of focus planes.

He also points out that Caravaggio's greatest paintings (among many examples) are a combination of astonishingly deft lighting and detail combined with gross errors that you'd get if (essentially) copying and pasting bits of photographs onto a canvas. (And, hey, I love Caravaggio...) The critics would have you believe that Caravaggio was a godlike draftsman who cold realistically paint the veins on people's eyelids but couldn't put their arms in the right position with respect to their bodies.

These guys invented (analog) Photoshop in the 17th century and it was one of their trade secrets. Apparently art critics find this insufficiently awesome, or the wrong kind of awesome. (Think: science awesome vs. romantic artiste awesome.)

Hockney also traces the passing of the secret technique by putting the works of various artists on an up-down (geographic) and left-right (time) chart and can show how two artists (one of whom had godlike drafting abilities and the other of whom was merely competent) crossed paths and then suddenly the lesser artist became godlike.

The argument that Hockney is just jealous of the talent of the old masters is equally ridiculous when he shows his own work with and without the techniques. One could level the charges that the art "authorities" that disbelieve him are simply jealous of their own utter lack of talent.

tl;dr -- most "art experts" are self-deluded idiots. News at 11.

> I’ve got a file of counterarguments to my own theory

Thank you for having scientific integrity. Falsifiability for the win.

>"But no one understood exactly how such a device might actually have been used to paint masterpieces." //

Seems a stretch.

I vaguely recall seeing a documentary - perhaps many years ago, probably on the BBC - on Vermeer's "The Music Room" and how it was understood to be created using an optical system [camera obscura?]. They showed a replica being used to trace an image.

Related http://www.math.nus.edu.sg/aslaksen/projects/MC-CL-Vermeer-a... uses the assumption of the accuracy of the dimensions to establish two of Vermeer's painting were likely done in the same room.

Thats close, but not exact. The difference is in the technique of using partial images of the subject overlayed over the painting image, to get the colours and overall colour and lighting gradients matched exactly, at a fidelity the unaided eye wouldn't be capable of. The trick of matching the patch so closely that the edge of the mirror disappears is the real secret. A straightforward over-painting projection of the whole subject wouldn't help with that.

> The brightness of any surface becomes exponentially less bright the farther it is from a light source, [...]

Is this true? The brightness decays quadratically with the distance, so I guess that the reflections decay also quadratically not exponentially.

Sadly I think the answer is, "according to the article it is true." I doubt many HNers have access to high resolution scans of Vermeer interiors. Until someone else repeats Jenison's analysis of the paintings we will have to take his word for it that this is an accurate description of Vermeer's use of light.

But this isn't a question about Vermeer's use of light, it's a question about the physics of light and the psychophysics of visual perception.

Brightness decays quadratically, but I'm not so sure about luminance (which is supposedly what the author is referring to.)

Edit: actually, luminance depends only on angles. But since the perceived brightness of a reflected source is not a point-light source, I still don't know how it all ties together.

Visual perception of luminance decays exponentially, according to the strongest distinction.

I think it may be more accurate to say perceptual description of luminance may be exponential (that humans report the difference from 1 to 1/2 as being the same as 1/2 to 1/4). But I think this is all irrelevant if you are trying to reproduce tones correctly (other than it says the error-model is relative error on tones). So I would say it takes a very labored reading to think of the article statement as correct (as light falls of 1/R^2 not exponentially). But that also makes the article's point: if you are painting what is there you are more immune to wrong assumptions.

Not necessarily. If you are making a painting, you don't have the ability to make some sections literally brighter (higher luminescence) because you don't really have control over the lighting conditions, and oil paints only give you so much ability to adjust reflectance. So what you do is subtly adjust things like tone, texture, detail or technique to 'trick' the human perceptual system into thinking the difference is there. These tricks are what imparts that life-like quality into the old master's works.

The term “luminance” refers to a physical quantity, the intensity of the light. It can be measured in candelas/square meter. The relevant technical terms for the perceptual attributes of a color are “brightness” (which is the perception of how bright something is on an absolute scale given current adaptation of the eye/surroundings), and “lightness” (which is the perception of how bright something is relatively, taking into account knowledge of the scene lighting).

Luminance (light intensity) is going to fall off with the square of distance.

Brightness/lightness do not have an exponential relationship with luminance. The curve is (very) roughly like a square root.

In any event, the article was clearly mistaken about this point. It’s a fairly trivial mistake though. Vanity Fair’s fact checkers probably didn’t bother to ask an optics / color expert about the statement.

Brightness decays exponentially for light traveling through smoke: https://en.wikipedia.org/wiki/Beer–Lambert_law. But I think the article author probably just messed up.

No. But when journalism, art, and science intersect, expect casualties.

That was fascinating. I would have preferred some more details on the evidence found during this that a camera obscura actually was used. I suppose not having it makes this article a pretty good teaser for the documentary.

There's a wonderful book by David Hockney called "Secret knowledge: recovering the lost techniques of the Old masters" where he lays out some pretty compelling evidence for this idea. It's also beautifully designed, as you might expect from Hockney. I keep meaning to try out some of the techniques mentioned to see how well they work but haven't got round to it yet.

The book is great and I'd say the case he makes is pretty water tight (and I say that as someone who wasn't much of a Hockney fan going in, and was very resistant to the idea that some of my favorite artists "cheated" ... but of course artists always "cheat" -- their goal is produce their artwork, not follow some arbitrary set of rules on how to get there).

I have a hard time understanding the dilemma here. Could someone please break it down? What difference would it have made to use mirrors and lenses, and how would that have improved his paintings?

I think they are claiming his view would be split between a zoomed in portion of the screen and the canvas. He could then mix the color until it matched up so closely with the mirror that the line between the two seemed to fade out. I still don't really know what the setup would actually look like and how far zoomed in it was, etc.. You would think the article would have a diagram.

Tim Jenison was interviewed by Leo Laporte on Triangulation (http://twit.tv/show/triangulation/118). They start talking about it between minutes 35 and 36, and he describes the process in a little more detail, including magnification, distance from the canvas and so forth

Artist here.

Okay. Working out perspective by hand is hard. It's a lot of work, that takes a lot of arcane knowledge.

So's doing a photorealistic painting of a thing. You have to look at the thing, consider a bunch of relationships of the particular feature you're interested in at this very moment, and transfer those relationships to your canvas. That's a pretty complicated set of habits and brain circuitry you have to build up over a period of years.

Tracing an image is immensely easier. There's all the perspective, there's all the proportion relationships. There's all the colors. Right there in front of you. You can spend an hour or two essentially doing manual edge-detection, and then start filling in areas.

There's still stuff to learn - how to mix the paint to get the colors you want, how to apply it to the canvas, stuff like that. And if you trace organic stuff too closely, it starts to look stiff and dead. But compared to the amount of skill you have to learn to draw this kind of stuff out of your head? It's nothing.

I mean, hell. Take a photo, print it out. Tape it to a sunny window, tape another piece of paper over it, and trace it. Then take that same photo and try to draw a copy of it, without tracing. There will be a huge difference between the two drawings you just made. The one made without tracing will probably have taken a lot longer, and have many more bits erased and redrawn.

A lot of modern artists have this weird thing going on with respect to tracing. It's the worst thing ever, they'll say. Tracing another person's work to learn from, or tracing off of a photo, they're different kinds of art crimes, but they're still art crimes. Leaping around learning to break a scene down in your head and reconstruct it from scratch - essentially building a wetware 3D modeller in your brain, really - is considered CHEATING by some people.

So Hockney's book on Vermeer's work is essentially saying "dude Vermeer was totally a tracer". Which in many artist's eyes is like calling him Not An Artist. Artists get weird about what is Art and what is Not Art, it's this whole thing that's about as inexplicable to a layperson as a programmer's choice of text editor.

By using mirrors and lenses to project an image onto his camera to trace, Vermeer only had to learn the basic painting technique stuff. If he did, if he traced all this stuff, then some people will start saying that his work is no longer Art. It's just Illustration. Which, in the eyes of people who care about this, means that it is a lower, less respectable thing. Now it's just a filter applied to an image, albiet manually instead of via Photoshop or whatever.

The truth of the matter (IMHO) is that a lot of the people who will scream bloody murder when someone is tracing are amateurs. Real pros can and will project images and work over them, build reference models in the real world or in 3D programs, and do any damn thing they can to make the hard parts of their work easier, to get all those little details they'd never make up out of whole cloth.

I am pretty sure I've repeated myself a couple times here, thanks to the tiny window HN allows to type in. But hopefully this helps.

tl;dr: If Vermeer used these techniques then he's a dirty cheating TRACER and is thus not a REAL ARTIST in some people's eyes.

It doesn't make Vermeer any less of an artistic genius. He had to invent new or improve existing devices, set up the scene, not to mention execute the painting itself.

No one says photo-realistic painters are fakers or cheaters when they recreate photographs on the canvas. Working from a photo does a lot of the same work as a camera obscura; it flattens the perspective, defines the colors, etc.

egypturnash isn't endorsing what you're opposing here.

"Modern" in the art context has a time period attached to it (20th century), you're looking for "contemporary" most likely.

IMHO it's time to stop using "modern" as the tag for the art of the late twentieth century and find a new name - but yes. n.n

>I am pretty sure I've repeated myself a couple times here, thanks to the tiny window HN allows to type in.

Insightful comment... sorry to only directly reference this line, but you can expand the text field by using the drag-handle in the lower-right.

For some weird reason it wasn't working for me when I wrote that comment.

Occasional amateur painter here.

Yes, tracing images is immensely easier. Problem is it yields lousy paintings. Photographs have different perspective and lighting than what you see with your own eyes and paintings that have been made by tracing or even merely copying photographs almost always look, well, like someone smeared paint all over a photograph.

If you look at a Vermeer painting, you'll find it does not look photographic at all. The objects in the picture are almost tangible and seem to jump off the canvas. This is an effect that is only achievable by having very strong 3D skills: you basically have to be able to paint even more realistically than mere realism. So it might very well be that Vermeer used some tools to get a feel for what it was he was going to paint and maybe even to study the fleeting effects of lighting, but the final paintings themselves are without a doubt made by someone with exceptional skill.

TL;DR;Vermeer did not use these techniques because his paintings are way better than what can be achieved using only these techniques.

Another artist chiming in.

I disagree with you. I believe that Vermeer used photographic techniques to aid him. It is very obvious to me (imho) that Vermeer's works is of a wholly different character than any other Dutch masters working at the time. His paintings are incredibly flat and tonal, and very, very photographic.

I also think that his compositions are his true genius. His compositions are beyond masterful. They are brilliant. This is hard to describe to a non-artist, but the way that he uses tight negative spacing along with his use of large spacious gradations is magic. He had the eye of a master photographer. This is where the question comes in for me. Did he set up his scenes and then limit himself to what he was able to manage there, just tracing what he saw? Or did he allow himself the freedom to change and modify the spaces and proportions to what looked right to him. I personally believe the latter.

To me, Vermeer is one of the greatest photographers thats ever lived. Every 2D design class should devote long study to his compositions.

Something else to consider for this post and parent. Using photographic techniques to make a painting does not guarantee a photographic "look." If you look at Warhol's traced drawings, the only clue is an even, utilitarian line, he was clearly not interested in copying visual reality (or he gave up on it when he realized it wasn't that easy to do from a projection). As counterpoint to that, you can look at Ingres' drawings, which are a world apart (it's likely he used a camera lucida, judging from the small size of the drawings and type of line). There is ample opportunity for the artist to bridge the "technological" gap with their own effort and ability, and it's likely this would happen to different degrees.

If using unconventional tools is part of your assigning value to artwork/artists or not, is another question.

Oh, I'm not disputing Vermeer may have used photographic techniques to aid him. There is definitely a very photographic look to his work that is unlike that of most of his contemporaries. What I am disputing, however, is that they were merely used as a shortcut and that the use of these aids is the be all and end all of Vermeer's talent. Even without his alleged use of photographic techniques Vermeer was a masterful painter: his handling of paint is otherworldly and that is something no photographic aid can help you with.

It's also true that his work is much flatter than that of his contemporaries, but to my eyes there's still a lot of 3D construction lurking behind those dazzling light effects that is completely missing in paintings that we know for sure have been made as copies of photographs. Like I said, to me Vermeer is a much better painter than the grandparent gives him credit for.

The 2006 monograph by the british artist David Hockney (for anyone interested).

Secret Knowledge: Rediscovering the Lost Techniques of the Old Masters[1]

[1] ISBN-10: 0142005126> http://www.amazon.com/Secret-Knowledge-Expanded-Edition-Redi...

This work is referenced many times in the article.

On Caravaggio: http://www.webexhibits.org/hockneyoptics/post/grundy7.html

Hockney's book is great, but for a quick and entertaining look at the theory, there was a (BBC I think) documentary with the same title. Probably on iPlayer and YouTube by now, covers more than just Vermeer (no links since I'm not sure about the copyright status.)

Reply on news.ycombinator.com