Drawing from photo reference- Bad habits

[Romans6]

Hey everyone,

Along time ago my teacher explained why drawing from photo references is a bad idea, something about when the photo is taken the camera clips or distorts the image. However I cannot remember exactly what he said, could anyone explain in more detail how the camera distorts the image? I also realize that drawing from life in general is more effective and gives you a stronger understanding. It's just bugging me trying to figure out what his exact explanation was.

thanks

[Flake]

"Camera Distortion", Page 193 "Creative Illustration" (265 on the pdf).
http://acid.noobgrinder.com/Loomis/

There are other reasons as to why photo ref should be used with caution.

-Cameras have one eye, we (hopefully) have two
-Cameras have a very limited dynamic range compared to human eyes- we can distinguish many more shades of light through dark than cameras can.
-Your eyes can "white balance" on the fly, focus and aperture settings are kinda automatic, cameras generally can't do this well.
-The colours are always wrong.
-Unless you shot it yourself, it's probably been Photoshopped to death.

[Romans6]

thanks alot this helps!

[Romans6]

it is however still slightly confusing to me even after reading the link you sent me, I just dont understand the interior of the camera enough I guess, it's hard to comprehend how much the figure is being compressed and distorted when we take a photo at a closer range. I understand stand that it does now, I just do not fully comprehend the underlying meanings.

[DavePalumbo]

Everything is relative though. Our eyes are far from perfect. The thing about working from photos (regarding distortion) is just being aware of what lens distortion is and why it occurs. There are all kinds of situations where it may or may not be appropriate to distort or correct your subject in a drawing or painting. If you're doing a dynamic action scene with limbs and such flying at the viewer, you'll approach the figure (in terms of foreshortening and distortion) very differently from an image of somebody seen from a great distance.

As far as compression of color and value ranges, these are things that you can learn to correct by doing alot of study from life. Once again, you can better control it once you've had a fair amount of experience working both with photos and with direct observation.

so to better understand the underlying meanings, you really just need to roll up your sleeves and start experimenting. Shoot photos up close, shoot photos far away, shoot under different lighting, shoot with wide lenses and with telephoto lenses, shoot from eye level, shoot from knee level. You'll understand better with hands on experience. Then, compare what you see from the camera to what you see in reality. The more you play around with these things, the more you learn how things can be seen differently and how to make this work for you. Even if you then choose to never use a direct photo reference, I personally feel that knowing how and why a camera and lens operate will strengthen your ability to create believable space.

[Elwell]

Read this.

[Zirngibism]

This may have been included in the links already, but I think a big proportion disparity is that the lens shape causes the image to be fish-eyed (slightly convex). Things closer to the middle of the image are represented as being slightly larger than they are in life

While I haven't experimented with this myself, I believe some of the photo distortion could possibly be corrected by adding a very slight reverse fish-eye effect (concave distortion) on an image in Photoshop.

(Can any of you prove/disprove this?)

Also, I was told that some digital cameras actually squeeze the sides of images slightly, in order to "improve" the image by making the subjects look thinner...

[DavePalumbo]

Lenses come in a wiiiiiiiiide variety and some fisheye by design (for super wide angle, popular for landscape photography) while others distort due to simply being inferior quality. Longer lenses should give minimal distortion in this respect, but they also require you to be very far from your subject and tend to give everything a very flat look.

I've done some distortion correction in photoshop (really only when shooting paintings though, where the subject is flat and it's easy to see if the square of the painting surface is wonky), though that was with my old camera which had a kinda crappy lens. I use a Sigma lens at 35mm now and it has barely any perceptible distortion for shooting paintings, so I haven't had to bother about it.

As far as shooting people, I don't know if it's practical to compensate in photoshop. It seems a much better method to me to just study and have a good understanding of both the human figure and the equipment you're using. You, in the end, need to decide what looks right and what looks wrong.

[TASmith]

Also, photos flatten three dimensional objects. If you don't understand the forms first, then your studies/copies will feel just as flat if not more so.

[timpaatkins]

Interesting point is though, that Sargent, Schmidt, Speed, Leffel, right side of the brain and countless others say the less you know about your subject the easier it is to accurately depict it, In fact, I think he said somewhere he'd rather you know nothing at all about you subject ( So as to paint what you SEE, not what you think it should look like)

We are usually also taught to paint patches of light, and focus on abstract elements, NOT to try to depict form (at least not until the final stages of a painting) for a more accurate representation. Kinda goes againt the whole " photographys bad because it flattens shapes and simplifies values". Thats what one does with a Lorrain mirror too, AND squinting for that matter.

Then its up to you where you take the piece after that. Most artist would like their paintings to have a little something else than what looks like a straight up photo anyways right?

They all had extraordinary abilities of course and knew forms inside out, but still, Its an interesting contradiction, don't you think?

Discuss!

[Dadaist]

As a fan of comics, in which photoreferencing is taking over, I dislike this method because the action ends up looking static and expressionless. It's referecing an object or person stuck in time. If the artist is just doing a poster or a cover it's not so bad; but in sequential storytelling, where movement must be captured, it renders a lot of actions unintelligible.

[DavePalumbo]

the less you know about your subject the easier it is to accurately depict it, In fact, I think he said somewhere he'd rather you know nothing at all about you subject ( So as to paint what you SEE, not what you think it should look like)

it's not a bad argument, and it certainly makes sense that even if you have no idea how, say, a hand is constructed, it will look like a hand of you exactly replicate the shapes of light and dark. The problem with this is that it's very very difficult to do this by eye and takes a tremendous amount of practice. Meanwhile, you're probably going to start understanding those structures anyhow. Supposing that you don't though, you're training yourself to be 100% dependent on your reference and unable to make choices which will sacrifice "reality" for message, aesthetic improvement, or any number of other reasons why one may choose to deviate from, or flat out ignore, some aspect of their refs.

Knowing why light reacts the way it does on different surfaces, how perspective works, the structure of the human body, etc... knowing these things doesn't stop you from seeing them in front of you. I think it helps you to see them because you know what to look for and how to process it. Otherwise, there's just too much information to absorb and transform into a 2 dimensional image. You need to make choices. Lots and lots and LOTS of choices. Wouldn't you rather make informed choices?

Sargent and Schmidt can talk in abstract theory about this and break it down as to what we really are looking at or creating in pushing pigments around on a piece of cloth or wood, but it's just theory. They both have a terrific understanding not only of the human form (and particularly the head), but of light and form in general.

[J Wilson]

Interesting point is though, that Sargent, Schmidt, Speed, Leffel, right side of the brain and countless others say the less you know about your subject the easier it is to accurately depict it, In fact, I think he said somewhere he'd rather you know nothing at all about you subject ( So as to paint what you SEE, not what you think it should look like)

I see where you are coming from with that, but I'd disagree. Knowing your subject matter helps to clarify what it is you are seeing when your observation or other conditions fail. Having knowledge beyond what you can see gives you confidence and allows you to make choices. For example the situation where what you see is actually a confusing mess. Knowing what is actually there allows you to make decisions so that your painting isn't the same confusing mess that reality may be.

If you had absolute beginners, and your goal was to get the best image out of them, rather than to teach, I could see giving them something they weren't familiar with to force them to actually study what was in front of them. Maybe you've forced them into observation because they have nothing else to fall back on, and gotten good results, but I'm not sure you've impressed upon them the importance of observation in the same way you would if they had painted something they thought they knew and showed them how wrong they were. It's like the trick where you paint something upside down. It's a good exercise for learning to copy what you see, but it isn't a good learning tool for understanding WHY what you see looks like it does. I think those tricks are only of limited value, and understanding is more valuable than just forcing a limited copy of what you see.

[Elwell]

Interesting point is though, that Sargent, Schmidt, Speed, Leffel, right side of the brain and countless others say the less you know about your subject the easier it is to accurately depict it, In fact, I think he said somewhere he'd rather you know nothing at all about you subject ( So as to paint what you SEE, not what you think it should look like)

There are different kinds of knowing. The counterargument is that you can only see what you know what to look for. Drawing is a complex process, which requires the artist to continually switch between several different modes of thought, depending on which one is useful at that particular moment. If I'm drawing or painting a figure, for instance, sometimes I'm looking at it as an abstract collection of flat tones and shapes, sometimes as a collection of three-dimensional forms in space being effected by light, sometimes I'm thinking about the names of the bones and muscles and what they are doing, and sometimes I'm thinking about what my artistic intention is with the piece and how I have to interpret and modify what I see in order to fulfill it. Training that leaves out any of those modes of thought is incomplete. Making art takes your whole brain.

[Romans6]

dave palumbo: your posts have been both interesting to read and educational, you're right I need to get out there and do the work to understand it all better, I can read all I want but until I go through trial and error myself my works will continue to have flaws

elwell: great link, I've always been a big fan of Norman Rockwell, nice to see he used reference but understood when to make corrections to the image

tasmith: that makes sense, if I dont observe from life first my art work will never have withhold a true sense of realism, it will look flat and boring

everyone else and timpaatkins thanks for the interesting discussion, I personally feel like knowing my subject and surroundings better will give me a stronger understanding of what I'm looking at when I'm trying to reproduce it in an artistic manner that captures a true underlying feeling of the particular subject

[Flake]

This board is so much more interesting when it has discussions like this, that are actually about art..

[dbclemons]

As a fan of comics, in which photoreferencing is taking over, I dislike this method because the action ends up looking static and expressionless...

Blame the artist for that. When they use photos, of those preliminaries that I've seen, they make the shots in a very generic manner. Some guy holding a broom handle for a sword wearing a bedsheet, or some such. Then the artist re-interprets it. If they would stage it better or use their imagination more creatively to make it more interesting, then they can insert what their poor staging left out. The Hildebrandt's were great at this. Photos are just a reference tool. It's up to the artist to use them correctly.

[Jasonwclark]

Well, if there are bad habits to be had, then I'm sure I have all of them.
No one ever taught me how to do it the correct way, so everything I know I've learned through trial and error, studying ref, and studying other people's drawings.

I think what your instructor was trying to say probably has the potential to be good advice, but he clearly left you hanging in the lurch there. It would be a little different if he'd said "don't draw from photo ref" and then showed you how to get around it, by putting you through a rigorous series of life drawing exercises. Or even better, if he'd said "try to avoid photo ref, but if you can't then remember to do it this way, and to keep this in mind" again with the instruction and the exercises and such. Most people don't seem to do it that way though, instead they'll just tell you "don't do it" for all the reasons listed in the posts above, and then walk away. Here's what I'm saying though - if you want your drawings and paintings to have photographic qualities, then it’s probably a good idea to start working with photo refs now. Because, just like anything else, you have to practice and fuck up a lot, before it starts looking decent. And even then someone is still bound to trash you for doing it that way, instead of some other way, just like they'll trash you for painting with Photoshop, or for having a different aesthetic sensibility than theirs, or for any number of reasons that are just as stupid. Its a lot easier to instruct in a reactionary way, and to form opinions about what not to do, what you're doing wrong etc. than it is to train your strengths. Probably for the same reason that it’s easier to break something apart than it is to put something together.

Someone the other day posted a thread about “analyzing your subject instead of copying it.” I don't know about the value of that dichotomy, but to analyze means to break down, dissect, and make sense of something through a process of division. From Gk. analysis "a breaking up," from analyein "unloose," from ana- "up, throughout" + lysis "a loosening." The conceptual antonym of analysis isn't copying (by rote), but synthesis. When you’re making a picture its important that both processes are involved, because you not only to have to break the image apart in your mind in order to understand what you're working with, but you also need to build it back up again on the page. How do you accomplish the latter? Well either you’re looking at something that actually exists in front of you (your current reference, whether that be a photograph, or a live model, or a picture in a book), or else you’re using rules and general principles based on things you’ve seen before (previous references, memories and such) which are always more hazy and veiled than the thing in front of you.

We all know that some of the most hardcore of badass artists, are the ones who can illustrate the human form from memory. If you can do that, then a lot of options open up for you, because everyone is interested in people. This is like the barrier to entry if you want to be a good illustrator, because all our stories involve people, right? They tend to involve trees and dogs and buildings too, but human beings are still the focal point for most of it. And really, if you want anyone to take your gallery work seriously when the pendulum finishes swinging back, you’ll want to be able to draw/paint compelling human figures.

So how do we do that? Basically three ways

1. Observation from life (hired models, strangers on the bus, your friends)

2. Observation from photo reference (hired models, strangers in books and magazines, your friends).

3. Construction from memory and general principles (Cartoons, Comics, Bridgman, your favorite children’s book etc.) This is easily the most difficult of the three to master, and the arena where a skilled animator or illustrator, will just run circles around someone only trained in the first two methods. I think that’s Dunn’s point anyway. If you know what you’re doing as a cartoonist then you can let your imagination run wild again, like when you were little, except that now your human figures will look halfway decent.

If you want to be truly tight, you probably should try to work with all three, because each method has its own uses, and presents unique challenges that you have to overcome in order to stop sucking at it.

Some people hated on the camera and rejected its influence on traditional art, others embraced it. You can usually pick out who’s who in the 20th century by examining the work. If you want to stage a scene with a Camera, that’s definitely going to be more involved than just busting something out in the sketchbook straight from memory, but still less involved than staging a scene for a long life drawing. In situations like that the main advantage of the Camera would be time, since you can set things up and then freeze the moment, to look at it again later. If you don’t have direct access to the thing you want to depict, if you can’t go there and look at it yourself, then the main advantage of photo ref is just that. Without it, you wouldn’t know what you were doing anyway, and would probably have to resort to guessing, so the photo is like the next best thing after direct experience. That probably applies less to the human form than some other stuff you might want to draw, but even with people it has its advantages. For example, maybe you don’t know any old guys with beards, or any beautiful women, or any Chinese kids, who are willing to model for you. In the past you might be screwed if you didn’t already know how to draw one of those figure ‘types’, whereas now you can just examine a photo. I find it hard to imagine that the old masters would have passed up a technology like this if it existed in their time. It’s just like a grid, or oil paints, or any other innovation in technology, its still only as good as the person using it.

There's a Dean Cornwell thread in the lounge right now that should be encouraging for fans of the camera. He made good use of the epidiascope.



In fact, many iconic 20th century artists made good use of photo ref; cat's we all know like Rockwell, Elvgren, Dali, and countless others with less familiar names. They might not have been as awesome as some of their contemporaries and predecessors in some respects, but clearly they were still doing excellent work.

Here's a thought - lets see some works by artists who used Photo reference effectively.

I think good examples are more instructive than bad ones, so I'll kick us off with a few proponents of the idea who are pretty impressive to me.

























http://www.conceptart.org/forums/sho...d.php?t=130468

[Dadaist]

Blame the artist for that. When they use photos, of those preliminaries that I've seen, they make the shots in a very generic manner. Some guy holding a broom handle for a sword wearing a bedsheet, or some such. Then the artist re-interprets it. If they would stage it better or use their imagination more creatively to make it more interesting, then they can insert what their poor staging left out. The Hildebrandt's were great at this. Photos are just a reference tool. It's up to the artist to use them correctly.

Perhaps you're right. I was going to say I think photo ref encourages laziness. But maybe it's already present before pencillers use it.

[Farvus]

Perhaps you're right. I was going to say I think photo ref encourages laziness. But maybe it's already present before pencillers use it.

Actually in my case it's the opposite. I'm too lazy to look for the photo reference so most of my imagination stuff is purely from memory .

[Romans6]

jasonwclark: Honestly I truly appreciate all the time you put in to that post, it really furthered my understanding on using photo reference. But it also helped me understand what it takes to truly become a better Illustrator,

So how do we do that? Basically three ways

1. Observation from life (hired models, strangers on the bus, your friends)

2. Observation from photo reference (hired models, strangers in books and magazines, your friends).

3. Construction from memory and general principles (Cartoons, Comics, Bridgman, your favorite children’s book etc.) This is easily the most difficult of the three to master, and the arena where a skilled animator or illustrator, will just run circles around someone only trained in the first two methods. I think that’s Dunn’s point anyway. If you know what you’re doing as a cartoonist then you can let your imagination run wild again, like when you were little, except that now your human figures will look halfway decent.

If you want to be truly tight, you probably should try to work with all three, because each method has its own uses, and presents unique challenges that you have to overcome in order to stop sucking at it.

I truly am taking a lot of the comments in this thread to heart, I'm at a point in my life where I really want to pursue Illustration for a career. Thanks for the further help its appreciated!

[Uli]

Very interesting questions and answers, thank you everybody!

My personal experience: I've switched to photos after trying to work with friends as life models, but that was a very time consuming process, because you need to make dates and they get tired and all that. I find it close to impossible to use life models for purely organisational (word?) reasons. And obviously I don't have the money to use a full time professional model. And with the weather that we have around here- I plainly refuse to work outside again. I got frostbite, got blinded by reflections and fell into a thorny bush - no, thanks

I feel that it's important to "digest" what you see while you paint it. I never use technical tricks like projection or tracing, I invite all the changes that happen when I just draw what I see, freehand. The idea comes first, then I try to find reference material or do it myself. I usually use more than one photo at a time, change lots of details, proportions or colour or the like. This allows me to get a more coherent look.
That said, I find it important to return to life drawing regularly, to internalize (word??) a feel for the three dimensional.

By the way- it was said that photos flatten out the image but make it more easy to concentrate on what is actually there. I went to an exhibition recently, where the illustrator said that wearing glasses does a similar thing for him, helping to turn threedimensional into twodimensional. I was wondering how my wearing different types of glasses, or taking them off, influences my technique. What's your experience?

[Uli]

PS: I think the worst problem with using photos is that you have so much time and control to copy it 100% that in the end it just looks like the photo. I mean, what's the point? And I dont mean photorealism in itself is a problem, if you want to do photorealism. It's more like the reference overwhelms the painting in some way.
N example: At an exhibition around here I saw this painting, almost life-size. It was wonderfully done. But the best parts were those that looked more painterly, like the monkeys.
I dunno what it is, but there is something irritating to a painting that looks *too* much like a photo. It feels a bit like plastic food.

I forgot to write down the painters name, he was British I believe. It shows an allegory on recovery from drugs. Here is more info about the exhibition: http://blog.fop-art.de/2008/08/20/go...ng-exhibition/

[J Wilson]

PS: I think the worst problem with using photos is that you have so much time to copy it 100% that in the end it just looks like the photo. I mean, what's the point? And I dont mean photorealism in itself is a problem, if you want to do photorealism. It's more like the reference overwhelms the painting in some way.

That can certainly be a problem, but it's just one of many that you learn to avoid hopefully as you progress. Drawing and painting from life does tend to negate that some, because when you draw from life there are hundreds of tiny changes going on every second, so your art has to be a distillation of those events into an over all impression of the scene or model. A photo is obviously static, so you can get caught up in meaningless details that you wouldn't be able to capture well from life, but it's not a fault of the photo but of the artist. Photos themselves don't cause that problem, they just enable it.

As a way to study and learn, photos are still an excellent alternative to real life. Flawed in some ways that you have to learn to watch for, but still far better than not using reference at all. As a way to produce finished art though, photos are probably best used more like you would use a figure model- as a guide more than as a bible. Pick and choose details, use it to record things you might otherwise have forgotten to include, but don't slave endlessly over it trying to capture everything.

[Uli]

...Drawing and painting from life does tend to negate that some, because when you draw from life there are hundreds of tiny changes going on every second, so your art has to be a distillation of those events into an over all impression of the scene or model. .... Photos themselves don't cause that problem, they just enable it.

I totally agree with you. I think this effect says something about the painting process itself. I like the word destillation in that context.