Thursday, August 12, 2010

Scart Road

Yesterday we set up our folding stools at the top of Scart Road looking down into Bantry, Ireland. Occasional sprinkles of rain and gusts of wind gave us some trouble: At one point the wind knocked over my water cup and dumped three ounces of water all over the half-finished watercolor.

The thing I found the most challenging was the perspective drawing. Here’s the scene as it looked to the camera. I knew all those buildings have different vanishing points from each other. And the road has its own set of vanishing points below eye level. And I expected I would be tricked by those building fronts, which were extremely foreshortened.

But even knowing all that, I still had to erase and redraw the pencil drawing three times until I was convinced I had it right.

Here’s how the line drawing looked after about three hours of measuring, erasing, and redrawing. I used a regular #2 pencil and I didn’t use straightedges. I tried to pin down some measurements using the central pole as a fixed unit of length.

Just to remind myself how much I need to improve, here’s the line drawing superimposed over the photo, after the fact. I made the yellow building on the right too big, and misplaced the windows and doors on the gray building on the far left. What would have cured a lot of these errors would be to measure across the scene, holding up a level pencil in front of me to judge relative heights—and I’ll do that next time.

Here’s the final painting, after two sittings on location, and about five hours work. I did make a few conscious changes, such as eliminating the wires, and raising the value of the illuminated roof on the red building in the middle distance.

And I regrettably left out of the drawing the wonderful people of Bantry, who passed by with a friendly hello while we were working—especially Brendy, who lives in the house at the far left.

“I like to say hi to me neighbors,” he said to us. “Me neighbors is mankind of every description.”

25 comments:

Vicki Holdwick said...

Wow! This is spectacular!

Thanks for sharing,

xoxo

Sean said...

It's been a real pleasure to keep up with your journeys, especially as you find these visual gems. Thanks for taking the time to keep us updated so well.

Daroo said...

Fantastic job!

Measuring is a muscle that you have to constantly flex in order to stay in shape and use effectively. I'm really impressed with how little your pencil drawing is off. I'm glad to hear it took you two sessions -- when I really try to be accurate, I really slow down -- I'm sort of a measuring couch potato.

That said, the reason to practice measuring is so that you can use it quickly and effectively and then change it if it doesn't serve your artistic intent.

Susan Adsett said...

“Me neighbors is mankind of every description.” is now one of my favorite quotes. Thanks for taking us on your trip - it sounds like you're having a blast.

Walter Wick said...

What a wonderful series of posts! Part travelogue, part art instruction, conducted by a most astute observer.

Question: Could your "error" in making the building on the right too large been a subconscious reflex to balance the architectural masses on both sides of the street? There are other ways in which the painting has suppressed certain details (like the wires) and highlighted others; like the color of the shadows and the stains on the garage. Isn't such interpretation one of goals of painting? If so, why worry about the accuracy of perspective? I much prefer the painting to the photograph.

Bajzek said...

Wonderful job James.
I've keep following your blog through goggle reader.
Interesting to see your careful work on the perspective drawing.

António Araújo said...

Jim: Great work. Loved the quote.

Walter:

>If so, why worry about the accuracy >of perspective?

If nothing else, because it is healthy to test oneself against difficult, pre-determined tasks. If Jim set out at the start to achieve exact measurements (and to play freely with color, for instance) then it is very usueful to check at the end for the exactness of those measurements. A color inexacteness would only be a choice but a measurement error is a real error simply because of the task Jim chose for himself at the time.

I can't speak for Jim, but I bet his ability to do an accurate drawing serves him well when he wants to do one from imagination.

Without clear goals (such as reproducing what is actually there) we have no means of knowing if progress is being made. Setting a clear goal and seeing our mistakes for what they are clears away the human tendency for making up excuses, and helps us improve.

From personal experience, trying to get a drawing like this precisely right is great for training not only measurement but discipline and the old humility muscle. Somehow something always gets on the wrong spot. It is just as instructive as it is frustrating.

And at least to me it is great to see how well someone as proficient as Jim Gurney measures up to the task. It places things in perspective and it's an incentive for the rest of us to improve.

alotter said...

I'm a first-time visitor, and I was blown away by the accuracy of your drawing. I am also impressed by your technical ability to match up and overlay, drawing to photograph.

JG O'Donoghue said...

ah strangely enough my parents have the house about 4 houses on the right, Brendy is a very friendly fella alright, great rendition of the street, seems "the light being bad" didnt affect ye too badly, the bright colors of the houses really suit the lighter values you choose

Walter Wick said...

António,

Even though you "can't speak for Jim", your answer resonates pretty well with me. Thanks.

Given the circumstances, the perspective match is pretty amazing.

digitect said...

James, I've been testing my own measuring skills lately against a photograph. My conclusion is that I could always have solved minor discrepancies by simply taking more time to measure and verify. Do you ever feel that too much time spent measuring is "cheating"? (I'm curious how much time you spent on this drawing vs. painting.)

That's about where I am, and after having just read David Hockney's book, I can't decide if drawing on transparency over glass or some other "imaging" technique isn't just as fair as simply spending hours measuring. I'm curious, how do other artists feel about this?

Steve Hall

SVSART said...

I find myself wondering if some of the perspective differences between your drawing and the photograph could be due to some possible lens distortion? Could that account for some variation between what you see and what the camera captures?

John-Paul Balmet said...

Impressive as always! I noticed some differences in value and color from the photo also. Were those deliberate, or the product of painting on location and two different sittings were you found an "average"?

I always struggle with trying to capture exactly what I see versus pushing the colors and values to make it "feel" more like I want it to or imagine it to.

Kaos said...

Wow... impressive.

There's a question that comes to me while reading the post. How do you deal with things that have vanishing points below of the eyeline (like the road)?

Tyler J said...

Great piece, thanks for breaking it down for us.

I was thinking similarly as Svsart, about the lens distortion vs. your natural eye distortion (including any turning of the head and so forth).

The other question I had was about color. The colors in your painting are more saturated and vivid than the photo; is this due to the camera dropping out colors, the scan of the painting, did you alter the colors for artistic purposes or to compensate for inconsistent lighting, a combination, or none of the above? =)

Thanks again.

Tyler J said...

Steve (digitect),

I am not a professional painter so take my answer for what it's worth.

I think the decision whether or not to use an "imaging technique" ultimately comes down to the artist's guiding values.

If someone were mostly concerned about process, then doing it the hard way, so to speak, would be the way to go; it's the journey and not the destination.

If someone were mostly concerned about the finished product and who cares how you get there, then use any and all tools at your disposal; the ends justify the means.

In this case, I think James was doing a combination, but I think the process is probably taking priority (but as António pointed out, I cannot speak for him).

For me, I would almost certainly do it the hard way, simply because I have a great deal to learn before I start taking shortcuts. That said, I was comissioned to do a pet portrait and I confess to using Photoshop to compose the two images and then trace the major landmarks. Granted I did most of the work "by hand," but I knew that "cheating" would give me a better final product and ultimately that's what the client was paying for.

TLDR: I think it's okay to cheat perspective when the occasion calls for it, but it's generally better to do it the hard way.

António Araújo said...

In order to have cheating you gotta have a rule to break. Meaning you have to know what "game" you are playing.

If you are a professional illustrator or artist, then the game consists in coming up with the most impressive final product, no matter how. Projectors and cameras are perfectly ok. I guess the only rule is that you cannot blatantly copy someone else's work (in some ages, even that was ok).

If you are drawing from life, for your own benefit, then the game consists of learning something. So, if you copy over a photo, or use projectors, you are cheating - namely, you are cheating yourself out of the purpose of the exercise.

But, for instance, if I am interested in doing an exercise in color, I will have no problems in tracing the original drawing before I start. This way I don't get tired and distracted by the hard work of getting a precise drawing, and I'll know for sure that any error comes from the coloring process. In the same way, if I am concentrating in a drawing exercise, I will disregard other considerations.

You just got to know what game you are playing at each moment and what the purpose is.

There is of course a caveat. An artist will be cheating if he claims to have made a drawing by a certain process while he used another (easier) one. But I don't usually observe current artists claiming anything about their process. It's sometimes the fans who feel cheated because they harbored some personal fantasy about how work is done.

The claim of cheating also comes from some artists who notice the unfortunate reality that, tracing and other devices being "allowed", people who have great skills at drawing from life and/or imagination don't get rewarded for it financially (directly), while skills at composing an attractive picture (mostly design skills) are very much rewarded by the market. This may be unfortunate to those of us who are better at the former task than the latter, but that's just life - unlike what Marx thought the value of a work is not the man-hours or the skill employed but whatever the market wants to pay for it.

(It occurs to me that the only market where a "pure" drawing process is de rigueur today is street portrait drawing. The public in that venue would very much consider the use of any device as cheating. Again, it's a particular game being played, with its own rules)

Personally, I too wish that life drawing commanded a bigger market than that but I don't mind that very much since the hard work of drawing without aids is more than payed for by the pleasure it gives me and by the ability to see better the world around me. And I do get the impression that good drawing skills are correlated with a better end result, even where design and composition are involved, so professional artists who don't go through that kind of training have to rely on gimmicks and can only compete in a smaller niche.

Max said...

Do you work mathematically with perspective, such as measuring, calculating degrees, using a ruler, etc? Or are you more visual and intuitive when it comes to perspective?

James Gurney said...

Wow, what a lot of interesting comments! You all expressed things that were sort of in the back of my mind. As you all know, I set all sorts of agendas when I sketch--sometimes I caricature and exaggerate, or combine element or move them around.

In this case I liked the scene pretty much as it was, and wanted to get the drawing as accurate as possible (what the old painters would call a "topographical" sketch or a "veduta essata." I don't think the camera distorts the perspective as long as I'm taking the photo from the exact same position as my sketching position.

I drew the perspective without protractors or rulers, or even a viewfinder (which is a mat board with a rectangular hole proportional to the drawing surface). That last item would have been a great help and I wish I had brought one--probably would have given me more accuracy. I also would have been more accurate if the drawing was up on an easel so that I could do "sight size" comparisons. Since the sketchbook was in my lap I had to do a good bit of head-bobbing.

I agree with the point that any tool is OK depending on what results you want: the only cheating is saying you're doing one thing if you're really doing something else.

Walter, yes, I think I did unconsciously make the yellow building bigger to balance the elements on the left, and that's OK from a compositional point of view, but I was trying to get an exact drawing just to see if I could.

Color distortion of photos is another story, and the photos that I take never capture more than a small fraction of what I see with my eyes. Here I believe the painter can get a much closer match to observational "truth" because we can compensate for extreme variations in light levels and pick up on variations within close values. For close color matching, oils are easier (for me) to control than watercolor.

Of the five hours I spent on this picture, three were spent with the drawing, and two with the color.

And finally, on a down-sloping street, the vanishing point of the street is below the eye level or horizon, while the vanishing points of windows and doors is on the eye level.

cath said...

A photograph will not look exactly the same as that scene seen through your eyes. For a start we have stereoscopic vision, and that creates differences from a camera image unless you were to do the entire drawing with one eye shut! Also the focal length of the lens will change the image, a wider angle lens will make the foreground bigger in relation to the back ground, ie it exaggerates the perspective.
So unless you take these factors into account, using a photo to check whether you have your perspective drawing correct, is not a straightforward matter. The old idea "the camera never lies" was false even before the advent of Photoshop! It doesn't lie, so much as mislead.

António Araújo said...

Cath:

>Also the focal length of the lens >will change the image, a wider angle >lens will make the foreground bigger >in relation to the back ground, ie >it exaggerates the perspective.

my two cents:

What you are describing is perspective distortion. It is a function of the distance and not the lenses you are using. The same field of view will look different if shot with a wide angle lens or a telephoto lens versus a normal (say 50mm focal length for 35mm film) focal distance lens. But this is not really a lens distortion, it is perspective distortion that happens because of the distance at which you are seeing the object. Your eyes do the same thing. It is a property of perspective itself. Try walking closer and closer to an object and observe how the angles change as you approach it. The reason why it becomes so obvious in a photo is that the special lenses allow you to get a field of view that you wouldn't otherwise get at that distance. If you could get it you would see the same effect with any lens.

Now, this is not to say that cameras don't distort images in other ways, but for the purposes of the comparison Gurney is doing they are pretty good. If you shoot with a normal lens from the point and angle from which you are drawing, you shouldn't have any problems.

As you stated, there is also the problem with binocular vision. But again, linear perspective itself assumes monocular vision, and that is how we measure from life (I always find it funny that teachers urge you to draw from life, not photos, using binocular vision as a justification, and then tell you to close one eye in order to draw :)) so the comparison should work well enough (actually I found you can measure pretty well with both eyes open but I can't quantify that so I won't go into it). Also, Jim was standing at a good distance from those houses, far enough that binocular vision is not that important. Try this: one eye closed, align your finger with a building. Now switch eyes. The finger should be completely misaligned. Now align two buildings in your field of view, with one eye. Switch eyes. No real difference, right? That is why monocular is pretty good for landscapes.

Finally, any lens-specific distortion should have symmetries, for instance radial symmetry, and your own drawing errors shouldn't, statistically. So even if there was a lens distortion we could not control (by avoiding edges of the field of view, etc ) we still could use the picture reliably to expose our mistakes. For instance, if one roof on the right side is misaligned with the other roofs on both sides, I'm pretty sure that's not the camera's fault. If angles are consistently wrong near the edges but not in the center then I'd suspect the camera.

I'd say the greatest distortion away from pure perspective probably comes from the way we draw. With our heads still we see only a very small angle, so if we are drawing a big field of view we move our heads to left and right in order to get the rest of the picture, and therefore we really change our projective plane. But again, this should be workable. Just like with the camera, avoid being too close to the target, avoid too wide fields of view, and you should be ok. Jim's drawing seems to be within those parameters.

All in all, the kind of superposition that Jim did is not only reliable but a greatly underrated way of improving our drawings. I always kick myself for not doing it more often.

Walter Wick said...

It's not that complicated to get a good reference photo for perspective. Just lift the camera and shoot from the drawing position as James did. It doesn't matter what lens is used as long as it takes in enough of the scene. If it takes in too much, just crop the photo to match the scene. As long as you get the shot from your drawing position your fine. The only distortion one has to worry about with regards to camera optics is barrel distortion with cheap wide angles, or pin cushion with some telephotos. But it doesn't look like that's the case here. 

I do suspect binocular vision does come in to play with the painter though. Even with the goal of   constructing an accurate perspective framework, I imagine it would be hard not to favor the left eye ( or move one's head slightly to the left) to get a better view to draw, say, a window behind some trees, and likewise favor the "right eye" for other details, such as for the purposes of avoiding awkward tangents of foreground and background elements. You could even say a painting done in such a way is a composite of left eye and right eye perspectives.

romero-leo said...

wow. you are a monster.
by the way I bought your book
Salutes, from Medellin - Colombia

Katherine Tyrrell said...

Oh well done James - you must be pleased with that!

I almost always get one of my dimensions wrong when drawing plein air. However I do the same thing as you of measuring off one standard vertical - it makes measuring altogether simpler.

I enjoyed reading the comments too!

Steven said...

I always enjoy insights into how you work. Thanks for sharing! :)