|
comd (Apr 2, 2006)
Trying a bit more than a head this time since I have all this canvas space in the advanced section. I'm just going to try to copy the photo directly this time as closely as possible.http://www.kristiaknowles.net/images/gallery/fitness/fit2.jpg I'm not sure I'll get anywhere. Thanks all for the encouragement. [Edit] This is the opposite of what I'm striving to do artistically as my goal is to draw more loosely from references (using information from the reference to make a completely different picture, not drawing the reference) and ultimately without needing references at all, but I wanted to try a shot at being unpainterly and more photorealistic: the antethesis of what I've been trying to do. To get the line drawing and major landmarks as accurately as possible, I initially started with a grid for the line drawing. This allowed me to shade and color more loosely without worrying about correcting inaccuracies from the previous stages. I don't think I quite got there as I still got lazy on the shading and didn't quite interpret the values correctly. It looks correct sized down, but not up close.
|
|||||
2draw.net © 2002-2025 2draw.net team/Cellosoft - copyright details - 0.05sec (sql: 25q/0.03sec) |
drawn in 1 hour 11 min
drawn in 28 min
drawn in 38 min
drawn in 5 hours 26 min
Nailing the reference wasn't really a major act of skill so much as a cheat. I ended up imposing a grid for the initial line drawing on both the photo and in Lascaux when doing the line drawing. That made it easy to make the outlines very quickly and get all the major features almost dead on (the slight deviations are only because I couldn't copy the grid cells exactly when making the outlines, especially in the hand on the left which I had a hard time picking out). I don't do this for all straight-from-photo pictures, but often if something is really off and I can't tell what it is, another thing I'll do is take a screenshot, bring it into photoshop, overlay my work over the original image, and see where I'm off that way by comparing the two right over another. Oh well, now I'm giving away my secrets, but I don't want people to think I have some superpowers when it comes to nailing the references: it's just a mechanical process.
So you put a grid on the ref without resizing it, and then tried to match the proportions of the grid on the ref to the grid you drew on your canvas?
This woman looks like she could whup my ass. The body, swimsuit and hair are all great, but the face is bugging me for some reason. It's not the proportions... maybe it seems a little cartoony or something. I can't quite put my finger on it.
Zack: looking at the comparison Axil made, I can see that I sort of flattened out the area underneath the eye on the left. I didn't get the shading quite right there. Also I made the nose a bit more narrow than the original picture, her chin looks kind of bumpy because I made that sharp highlight, the edges between the lit portions of her face on the left and the dark portions on the right have too hard of an edge, and her lips are pretty off. :( Also all the subtle nuances of reflected light on the right side of the face I ignored and simplified to a large region of shadow. Also the shadow region looks like it has more contrast which makes the face look sort of harder than the original. I think those are the main parts that look messed up (at least to me). Also with the cartoony look, I tried turning the lines off and painting for a while, but I deviated quite a bit from the photo at that point so as a final touch, I just reenabled the line drawing, but since my shading and outlines didn't quite match up, it sort of looks like the line drawing is floating in areas.
Doh - I'm starting to really hate this picture. I will delete it in a short while - it was too mechanically done, and it goes against my goals of drawing more loosely and without becoming a slave to the reference (or not using references at all). I just ended up sinking back to my comfort zone for this one.
Anyway, I'm going to quit that starting right now, and I'm going to make a habit to post my references more whenever I draw from them (though my goal is to ultimately not draw from references) so you guys can catch me on it and scold as necessary. :) These still work as nice coloring/shading exercises, but they hardly give me anything useful from a drawing standpoint since I'm just copying 2D shapes when I do that rather than properly seeing and feeling the 3D form, which is really what I need to be doing to reach my goal of being able to improvise and create characters from imagination. I was just having a bad week and was in the mood to just paint and not challenge myself by really taking a risk and drawing everything freehand without the mechanical aid of a grid or photoshop overlay comparisons, but really I was just copping out again. Surprisingly the one picture I have in the showcase is the one from which I used the reference most loosely of all (I hardly even looked at it except when I ran into problems), and it's sort of obvious by the way her face kept deforming with each revision. Perhaps the person who put it there was able to sense that I was hardly copying the photo in that one, but that picture is actually one of my least favorites. :|
I already think I lost something painting digitally, somehow, somewhere, and I keep dragging out my canvases and my brushes and thinking I should buy new paint and get some in my hair. (although I don't really see myself giving this up completely or not trying to slowly learn things that might make me better at it)
oh, and as far as painting from references goes, there's no need to make excuses for it. Claude Monet sat in a f!@#$% hay field from daylight till dusk looking at the haystacks to paint them, and Van Gogh went out into the uncomfortable fields to paint the workers because... well, because just glimpsing them and then painting from memory is difficult. Current successful artists hold entire huge HOORAHS where people dress up and reinact events where the artists take photographs to paint from. Chill about it, people, it's not a sin.
I must rant on grids. Please call this the "Drawing without Guilt" rant. :)
We all grid. You cannot do a realistic drawing without using proportion. To gauge the correct proportion you must grid, by eye or with lines, but you need to drop a "plumb line" and you need to see the "level line" or things won't line up. The longer and more often one draws the better their eye gets BUT every artist will need a "tool" at one time or another. I have a great "eye" (better than my drawing skills cuz i don't take time for my work) and critque drawings every day and have for the past 15 yrs as i teach realistic drawing to adults. Many times I have to use a pencil to sight or to measure proportions. Drawing is gridding.
In the Renaissance, all those folks we revere..well..let me tell you, they used devices all over the place. They used stand up 3d grids and set them in front of the subject to help them see. You can still buy those grids today. They used "sighting sticks", a stick stuck in a bucket to mark the dead center of the subject so they could judge better. And don't even get me started on the "camera obscura" which is now being thought to have had VERY widespread use...it allowed the Renaisance artist to actually TRACE the outline of the subject matter, much like todays projectors.
In my humble opinion, tracing is "cheating". Anything else, including the use of an overlaid grid is still drawing. You still have plenty of visual decisions to make. So I say, don't worry about it at all.
About drawing without a ref. Pshaw. Really. I know a very famous artist for MAD magazine. He has been drawing every single day, all day, for over 40 yrs and he STILL will go and find a ref to see how stuff looks if he has never studied that object before or drawn a figure from a particular angle. The whole point of drawing being hard is because things are so visually COMPLICATED. There is just too much visual information contained in the most simple of things to be able to draw a hyper realistic pic of it with no ref. You can draw a nice draw that "feels" real, but look closely enuff and it won't be, not really. (I am only discussing realism here.)
My 2 fav artist quotes are:
"My work has offended both God and man". That was Leonardo near the end of his life.
"If people knew how long and how hard I have worked to achieve my mastery, they would not be nearly so impressed". That was Michelangelo. :)
Sorry for the rant, i only have 2, this one and the modern art one. I won't rant again, promise. I'll be good!
If I'm using a photo, I still want to construct figures in 3D (using spheres and cylinders and so forth) rather than thinking in 2D shapes and patterns which is pretty much what I did here. In other words, I want to draw from photos the same way I draw from imagination/memory, and if I'm drawing from a photo, I want to do so in order to become better at drawing from imagination, as that's my ultimate goal. That said, I don't want to force myself to not use references when they can be beneficial, but I don't want the lack of references to prevent me from being able to draw something either.
There are several ways I know to accurately copy a photo. One is to cut the photo in half, tape one half to your paper, and then draw the other half by looking at the other half of the photo as you draw. Having one half next to the other half you are drawing gives you a half-established face, for instance, from which you can judge all the marks for the other half of the face. That doesn't really work in oekaki applets. The other is the grid method which is the most laborious and tedious to use. I hate using it, but this past week, I wasn't really in the mood for challenging myself at the drawing stage. Also a looser variation on the grid is to just draw some lines on the photo (diagonals, triangles, etc) and then try to copy those shapes on to the canvas, and then you have some marks to relate all other marks. This is more pleasant than a full-blown grid, and the lines and shapes can often be more meaningful and efficient than straight grid lines. Then for digital there's photoshop overlaying in which you can overlay your picture over the original as Axil did to compare mine, and you can see which marks are off and adjust things correctly. I have to admit I've done that last one (photoshop overlays) in between revisions for almost all of my reference-based pics, though the grid method is something I use sparingly (for this case, I just wasn't in the mood to challenge myself at the drawing stage, and I wanted to play around more at the painting stage with how I blend tones and how I establish colors). Ultimately I want to be able to paint without even using line drawings or underpaintings: I want to paint the subject straight in full color with thick opaque brush strokes, and I definitely don't want to use any of these techniques when doing that. My goals are based around speed, looseness, and improvisation and these techniques don't help at all there.
Without using any of these techniques, I have a very hard time reaching a good resemblance to the photo, though if I know the subject, I can get some resemblance to the subject even though I'm very off from the photo. An example of an exceptionally bad one is my latest doodle: http://cellosoft.com/2draw/view/65922/. In this one, I wasn't even trying to get any sort of photo resemblance (I was actually trying to do a caricature, but I suck at those - especially when I only know the subject through one photo, as in this case), and I was particularly sloppy about it. I posted the reference pic in an edit so it can be compared, but it's pretty obviously off. The way I copied that one was I basically grasped the stylus, looked at the photo, and drew without taking my eyes off the photo (not even looking at what I was drawing). That allowed me to establish the main marks quickly (which weren't 100% correct, but close enough to use as a measure for the rest of the strokes), at which point I then started using some of the basic things I knew about the proportions of an average head (eyes being in the middle of the head vertically and so forth) to make sure I wasn't too off. Note that the 37 minute timer on that one is pretty off: I spent at least half that time trying some other doodles over my original one which I hated. Anyway, that's about the extent of my copying skills when I completely freehand without doing any photoshop overlays or grids, though I can maybe do slightly better using that technique when I'm not sloppy about it (in that case, I was just wanting to paint over my original drawing which I hated, and I just wanted to put something down quickly and didn't really care as much about quality as long as it wasn't something I hated). Still, I'm not very good at copying photos freehanded, though I did the same thing starting out with Stinky, but he was a subject I knew by heart, so I already had most of his features memorized to get a good likeness even without getting close to the photo, and I still did the overlay comparisons for him to verify and adjust certain features after I made the initial marks.
I pretty much knew this picture wasn't what I should have been doing when I started on it based on my goals, regardless of the grid (though the grid made it that much worse). I set out only to draw the reference as accurately as possible and nothing more. In that sense, it's one of my biggest failures since that's the last thing I want to be doing in the future as I progress. I didn't even try to experiment with painterly techniques - I was trying to be as unpainterly as possible in this one so that it would look just like a photo. Generally I get the most comments about the painterly aspects of my works when working from photos, but this time I really wanted to try not being painterly (I tend to be painterly for economical purposes, not intentionally). I didn't even really achieve the photorealism I intended to achieve despite the use of the grid for the line drawing, so I didn't even succeed in that respect. While it's completely against my personal goals, I still admire the artists on here who can make their paintings look just like a photograph. I was hoping to achieve it here, but I think I deviated too much in the shading and still relied too much on lines which gave that sort of cartoony effect in places.