Welcome to part five of my ongoing tutorial about the cover of The Dreamer issue #4.

At last!  Part five is here!  My apologies for the delays.  I have been working almost non-stop for the last two days to bring you this video tutorial.  I am extremely excited because this installment is also (unofficially) the first episode of my long-dreamt-of ArtCast.

I say ‘unofficially’ because I don’t think I am going to actually put this file out on my podcast feed for a while.  Seeing as how it is the fifth part of an ongoing tutorial, I doubt that it will make much sense to first-time subscribers.

My plan is to eventually re-release it along with video-ized versions of the first four parts (but I am going to tend to some more pressing ideas first).  For now, you will be able to see it here on my site, on Blip.tv and eventually on my You Tube Channel.

A few words about my ArtCast:

I am a big fan of ArtCasts, though I only know of a few.  I think that podcasting is a relative frontier for artists who work in still-based media.  I have wanted to do a podcast for quite a while but I have been waiting for a number of reasons.

A distinct vision for what my podcast should be has only begun to crystallize in the last six months or so.  While doing these Dreamer tutorials, I have finally realized that there is only so much that one artist can describe about his process before he has to demonstrate.  What takes me half-a-day to type and what takes you half-a-day to read can be shown, said, heard and/or seen in much less time.  …AND it is more effective!

I have been taking my time to do plenty of research on how to podcast effectively (if you’ve ever thought of podcasting, check out The Podcast Answer Man).

If you want to go ahead and subscribe to my podcast, be sure to use the feed presented in the right column of my main blog page and not the default feed (the one that is generated automatically from the blog page).

For my fellow super-geeks, the appropriate feed to use is ::

feed://feeds.feedburner.com/ChrisOatley

I am grateful to have you here.

Please enjoy part five of The Dreamer tutorial on Chris Oatley’s ArtCast:

Welcome to part four of my ongoing tutorial about the cover of The Dreamer issue #4.

For the last couple of weeks, I have been discussing the importance of making a good color-comp before you start a final illustration.  This week, I am going to discuss one of the most important (as well as one of the most overlooked) elements of any good illustration, REFERENCE.

I will begin, however, with a few words about how I shifted from my reduced-resolution color comp to the file that would become the final illustration.

As I mentioned before, I worked at a considerably lower resolution in the color comp stage so I could paint with very large brushes very quickly without bogging down my processor which makes the brush strokes lag behind my pen cursor.  When working with large file sizes and lots of layers in Photoshop, this “slow-down” phenomenon is inevitable, no matter how robust the computer running it all may be.

I avoid any potential slow-down by reducing the file’s resolution to 72dpi when I begin a color comp.  There is one problem in returning to a higher resolution for the final painting, however.  The Multiply layer on top of all of my other layers (the one with the line drawing on it) also gets reduced and I thus lose a tremendous amount of detail.

Of course, none of the lines from the drawing are showing in the final piece, but when it comes to painting certain areas with lots of detail (the characters’ faces or accessories etc…), it is necessary to turn the line drawing layer (the Multiply layer) on and off to help guide my brush strokes.  It is often necessary that the line drawing be sharp and crisp in the really detailed areas so I can clearly discern the forms when painting them.  It is hard enough  and not blurry and muddy like it becomes after I suck seventy-percent of the resolution out of it.

Therefore, I have to prepare a file in advance of the color comp.

Here is how the whole process goes:

1.)  I save a version of the full-resolution line drawing with nothing else in it.  The line drawing is on a single layer which is floating above the default ‘Background’ layer that is generated whenever you create any new file in Photoshop.  The line drawing layer is an adjustment layer set to Multiply.  I will come back to this file later when I am ready to start my final painting.

2.)  I save the hi-res file.  I also save a second version of this file with the word ‘Comp’ in the title.  I use a naming convention for all of my files so they are easily searched and identified.  My original hi-res Dreamer file (with only the line drawing on its own layer) was named ‘DreamerCover01.psd’ and my comp file was named ‘DreamerComp01.psd’.

As I work, I save multiple versions of the file in case I need to go back to a previously saved version.  This helps me to feel like I can take greater risks with my paintings because if it doesn’t work, I can always go back to the version I saved before I tried something crazy.

Also, if a file gets corrupted while I am working on it, I have a previous version to fall back on and I only lose the work that was exclusive to the most recent version of the file which is better than losing the whole thing and having to start over.

In the filename, I use the ‘0’ as a place holder because without it, if I create more than nine versions of the file, (and I almost always do) then the tenth file becomes the first one in the list when I am looking at all of the files in a folder on my hard drive.  This can be annoying and confusing when I am trying to navigate quickly through my operating system, so I label the first nine files as ’01, 02, 03…’ and so on…  They all show up in the correct order that way.  (Unless, of course, I end up with more than a hundred versions of the file which has only happened to me once and that was on a piece of animation so it doesn’t count…)

3.)  I close ‘DreamerCover01.psd’ and won’t be going back to it until after the comp is complete.  I reduce the resolution in ‘DreamerComp01.psd’ to 72 dpi.  I proceed with the comp as described in parts 1-3 of this tutorial.

4.)  With the comp complete, I merge all of the layers in the file except the top Multiply layer which contains the line drawing.  Now I have two layers, the ‘Background’ layer that now contains the entire color comp (post-merge) and the drawing layer.

5.)  I open ‘DreamerCover01.psd’.  I have both files open: the low-res comp and the original hi-res file.  I ‘Select All’ and ‘Copy’ the ‘Background’ layer from the comp file and ‘Paste’ it into the hi-res file underneath the line drawing layer.  I close the comp file.

6.)  The color layer will be very small, of course, so I use the ‘Transform’ tool to scale it up.  Because the proportions of the two files are the same, I can just use the corners of the frame of reference to register the color to the line drawing.

Now I am ready to begin adding layers on top of the comp to begin the final painting.  Because the comp has been scaled-up, it looks all jagged and gross but that’s okay because it is all going to be painted-over anyway.  The comp is, ultimately, just a visualized plan for my painting, and everything will move smoothly and quickly from here on out.

I will talk about the next stage of painting next week.  And now, I will move on to the subject of reference.  Below is some of the reference that I gathered for The Dreamer:

To begin with, Lora had gathered all of the historical reference herself.  And, let me tell you, friends, this girl is dedicated.  In case you didn’t know, she takes trips all over the eastern part of the United States to take photos and learn firsthand about the history behind her story.

As you probably remember, Lora Innes did the line drawing upon which this painting was based.  She saved me all kinds of time by providing the photos that she already had.

In many professional cases, the client for whom you are working will provide a number of reference photos (although some of them won’t even provide clear directions for the assignment, let alone reference photos).  You will, of course, almost always need to augment the provided reference with reference that supports your own point of view.

Here you can see only about half of the actual amount of reference that I had gathered by the time I was done with the illustration.  (I don’t have permission to post most of it which I gathered from Flickr.com)  You can never have enough reference and I highly recommend spending considerably more time gathering reference, planning your painting and working out the comp then you do on the actual final piece.  Prioritizing things in this way can save you a lot of pain in the long run.

Lora’s research trips enable her to get close to the historical locations and artifacts with her camera and gather information that one may never find on the internet or even at a library.  Illustrators need to be inordinately observant and we often require answers or photographic points of view that a tourist who just snapped some photos of a neat historical site and posted them on their blog would never think of.

So, the first three points to remember about gathering reference are:

1.)  In all things, be inordinately observant.

2.) Take your own reference photos whenever possible.

3.)  Gather LOTS of reference, even when reference has already been provided.

What kinds of reference are you looking for?  For this piece, I needed historical reference which was provided for me.  I wanted color and light reference to help me convey the time of day and overall mood.  I needed natural reference for trees, plants, rocks, the sky etc… and I wanted to make sure that the trees I ended up with looked like trees you would find in the actual location featured in the painting.  I spent hours on Flickr, just sifting through sunset and landscape photos of the American northeast.  I think I must have gathered fifty of these photos.

At some point after I had started the final painting, I realized that I needed to gather sometechnique reference.  This one is tricky, because it can easily lead to a rip-off of someone else’s work.  Taking that into consideration, I thought about the kind of environment that I wanted to paint, the kind of color palette that I wanted to apply and the kind of emotional tone that I wanted to convey.  I knew that I wanted the piece to have a more edgy Disney feature-animation style to match Lora’s very graphic pencil style as well as to bring the piece out of the ‘romance novel’ territory it so easily could have fallen into.

I thought of the Disney film Brother Bear and remembered how amazing the backgrounds in that movie are.  I began searching the internet for Brother Bear images and I found dozens of good ones.  I have a pretty established painting style, so I was not very worried about accidentally ripping off Brother Bear but I am always conscious of the need to have an original approach when painting.

4.)  Assess your reference needs as early on as possible.  Be patient and make sure you have enough reference to begin the final piece.

5.)  Gather reference throughout the painting process.  When you discover a need for reference during the process, take a break from painting and look for that reference.  It is never too late to apply some new knowledge that you gained from some sort of research.

Part of the reason I use Flickr.com for my reference images is because it is there that you can find very hi-res files with lots of detail, not to mention that there are many very good photographers that make their work available for download.

6.)  Always gather images of the highest-quality possible.

I explained in part one that you have to keep everything in question until the appropriate time.  This included your overall concept and approach to the painting.  Sometimes, I will be looking through reference images or observing life or doing some other sort of research and through the learning and observing process I come up with something way better than I had originally planned.

It is good to remain willing to abandon our best ideas for a better idea that comes from an increase of knowledge or experience.

7.)  Be open to new ideas that might be sparked by your research.

While you are drawing or painting from reference, do not rely too heavily on it.  Photos can make objects look very flat and they can rob an environment of its depth.  Photos taken with a flash bleach out colors and people tend to look very stiff in photos.  If you draw a person directly from a photo without applying any artistic license, the person will almost surely look like a department store mannequin.

This is another thing that the comp helps to manage.  By working out your colors in the comp, you are less likely to depend on your reference in a way that would mislead you.

8.)  Interpret life. Don’t just copy it.

After I have all of my reference gathered, I paste it all into a single 300 dpi, 11×17 (horizontal) file.  I organize the various images by color and subject.  I spread them out to fit as many images as I can into the 11×17 area.  I turn visibility off for all of the layers that I can’t fit into the 11×17 area.

When the whole area is full and I can’t fit any more photos, I merge all of the visible layers together and name that layer something like ‘R1’.  I do this over and over for as many photos that I have.  If I have text reference, I copy that all onto a text layer.  The point is to have one big reference file with all of my reference in it, but with as few layers as possible so it doesn’t bog down my computer.  On a few projects, I have had so much reference that I required two reference files because the files were so big, despite my attempts to optimize the layers.

For color reference, I try to match similarly-colored photos and paintings together for a pleasing visual harmony.  This helps to affect me, emotionally and get me ready to paint something that feels harmonious with my reference.

The 11×17 area is oriented as a horizontal rectangle so I can display the file on my secondary monitor while I have the file containing the current version of my illustration on my primary monitor for painting (my Cintiq).  The proportions of my secondary monitor are roughly 11×17, so the reference file is designed to match the dimensions of the screen.  This way, I can display at one time, the most photos possible.  Then I don’t have to keep going over to that monitor to waste time navigating around the screen in an attempt to display the photo that I want to see.

Geez.  That’s a lot of information.  I think I’m done for the day.  Feel free to ask any questions.  I actually don’t have time to go back and proofread this one, so let me know if I did or didn’t make sense.

I’ll be back next Monday with part five where I will talk about the overall style and my painting technique.  I think that may be the last of The Dreamer tutorials, so if you have some things that you want me to cover, be sure to write in and let me know soon.

Welcome to part three of my ongoing tutorial about the cover of The Dreamer issue #4.
Last week, I discussed two important aspects of the color-comp stage.  This week, I will discuss a third aspect of the color comp stage as well as how you can use light, color and detail to complement a composition.

The comp stage is the perfect stage at which to make any significant drawing or compositional changes.  This is a third aspect of the color comp and the one that I will be elaborating on today.

While painting the comp, I identified the problems that I knew I would face during the rest of the process.  I listed all of these issues last week in part 2 of my tutorial and I explained why they seemed problematic.

Just for clarity, I will briefly list these problems once again:

• I knew that I would face a challenge in balancing the blue colors in Bea’s Dress and in the flag with the pink hue that covers the whole piece.

• There were at least two compositional tangents in the piece: where the smoke met the top of the left tree and where the back of the cannon met Alan’s left shoulder.

• Alan’s left forearm was almost parallel to the picture plane and thus posed a challenge to keep it from looking too flat or awkward.

• The shape of the smoke was not fluid enough in the drawing and therefore would need to be re-shaped in order to make the composition flow a little better.

I explained the issue with the blues in great detail last week.  As I said, it was mostly resolved by the time I completed the color comp, so I won’t go on any more about that.  However, the tangent issues presented some interesting challenges, so I will start with those today.

The tangent area where the tree tops met the smoke was very difficult to resolve.  In fact, I am not certain that it is actually solved in the final piece, but I won’t go into that now.

One effect that compositional tangents have on an overall composition is that they create unnecessary tension or competition between shapes. Tangents distract the viewer of the painting from the intended focus of the painting.

This is especially problematic when the tangent exists in an area that is secondary or tertiary in importance to the main area of focus.  This is certainly the case with this painting.

Obviously, I want the viewers to be focusing their attention on the the two characters in the painting.

However, the tree top and the smoke cloud are up in the top-right corner of the piece picking a fight and causing a disturbance.  That area of the painting should be saying ‘don’t look over here, look at the main characters!’ but it wasn’t saying that.  The tree was saying ‘Come on, Smoke Cloud, back up off me!’ and the smoke cloud was saying ‘Get outta here, Tree, you’re all huge and green and pointy and you’re cramping my soft pink style…’  It was distracting.

The issue was not evident in the sketch.  It was not until the trees were painted that they began to seem too cumbersome.  I tried re-drawing the tree, making it shorter, but then it just seemed like it was trying to get out of the way of the smoke.  It looked awkward.  It looked too contrived.

I tried taking it out entirely, but there is not very much room in the overall composition to show many background elements.  That giant cloud of smoke covers most of the background.  The tree had to stay so the characters didn’t look like they were sitting in an empty wasteland.

I tried putting washes of the sky color over the tree in an attempt to push it back atmospherically.  This did not work because the size of the tree and the placement of the other trees indicate that it is fairly close to the main characters.  It is true that when objects are far away, they tend to take on the color of the atmosphere, however, the tree was not far enough away to justify many atmospheric washes.

Since none of those attempts worked, I decided to re-draw the trees and break up the big dense forms by placing the branches and green areas more sparsely.  This seemed like the best option.

Later, I also blurred the trees with the Gaussian Blur Filter in Photoshop in order to push the trees back into the distance.  I did not want to blur them too much, since, as established, they are not toofar off in the distance.

Note that I did not make this change in the comp stage.  It was not until I began painting the actual background that I re-drew the trees and painted them in this way.  It was a gamble, not firmly establishing my solution before moving into the final painting, but I was getting restless and I think it was successful enough.

I’m always trying to maintain a balance between the rules and my own personality when painting.  I had to move on before I fell out of love with the piece.  Its sort of risk-versus-reward in this case.

The same kinds of things can be said for the cannon/ shoulder tangent.  In the drawing, the back of the cannon appeared to touch Alan’s left shoulder, which was not physically possible, seeing as how the cannon is maybe fifty feet behind the characters..

When looking at the other parts of the cannon, it would appear to be sitting in its correct position, but when looking at the tangent area, I noticed that the cannon seemed to jump forward and share the middle-ground space with the characters.  This was obviously an unnecessary awkward compositional arrangement.

The cannon was unique in that it sat right inside of the cloud of smoke.  Because of this, I had more liberty to screen it back with multiple washes of the smoke color.  I hoped that I could solve the tangent problem in that way, without repositioning the cannon, but it simply didn’t work.  I didn’t make this decision, however, until well into the final painting.

The eventual repositioning of the cannon worked very well.  It really harmonized with the rest of the painting.

To the left, you can see how the shape of the smoke evolved from the drawing to the comp to the final painting.

I was attempting to get the shape of the smoke to optimally effect the overall composition.  I wanted the shape of the smoke to draw the viewer’s eye through the piece, to influence the viewer to rest their eyes on the main characters but also to maintain the qualities of smoke: random and puffy.

Granted, I was not attempting to paint realistic smoke.  I was attempting to paint a very stylized, soft, romantic version of smoke.  I tried to make the smoke look more like big soft clouds than like real cannon smoke, but since we are talking solely about composition at this point, I will save those more technical choices for a later post.

Things got especially crowded around the area near Alan’s right hand.  The outline of the smoke, the strap from the gun, Alan’s arm, all of these objects were intersecting and it took a number of tries to get it right.  Again, I was trying to keep the shape of the smoke looking random but I wanted it to complement what was happening in the middle-ground.  I am pretty pleased with the end result.

Here is where I will transition from talking about the compositional issues to how I usedlight, color and detail (or lack thereof) to complement the composition.

As a quick side-note let me say that achieving a focus on the characters in this piece isn’t too difficult because the characters are right in the center of the composition.  The real challenge with this piece, compositionally, is achieving a dynamic sensibility despite the centrally-located characters.

The finished piece could have easily felt static and boring with the characters at the dead-center.  There are a few compositional elements that help keep things interesting (the strong diagonal of the flag, the smoke, the angle of Bea’s body etc…) but any potential compositional interest could have been squandered if the painting had not been approached with regard to said potential.

Note how there is a gradient of light running from the top of the cloud to the bottom.  In other words, the smoke is brighter and more saturated with the pink light near the top and it is grayer near the ground.

This helps to emphasize the strong diagonal of the smoke shape.  The brighter area at the top of the smoke catches the viewer’s eye if they happen to be looking at that part of the painting.  The contrast between the dark blue of the flag and the bright area of the smoke pulls the eye directionally toward… …where?  You’re RIGHT!  The CHARACTERS!

The flag, of course serves the same function that the smoke does, compositionally.  It serves a much grander narrative purpose, but its compositional purpose could not be more obvious.  It creates a direct line from the brightest spot in the smoke to the top of Alan’s hat.

The trees in the background echo the same diagonal direction.  However, with this strong diagonal, how come the viewer’s eyes don’t overshoot the characters and fly off the picture plane altogether?  Contrast.  The dark brown of Alan’s hat, the relatively dark (and very saturated) brown of Bea’s hair set against the lighter cloud of smoke create a strong contrast and a visual anchor point that keeps the viewer’s attention.

In fact, this is one of the highest points of contrast in the whole piece.  I believe that the actual highest point of contrast is where the dark blue of the flag overlaps the brightest spot on the smoke, but that area creates such a strong diagonal that points right at the characters, it doesn’t seem to conflict with the characters themselves.  In this situation, the strength of that  central location of the characters worked to my advantage.

Furthermore, the characters have a certain inherent gravity that attracts a viewer’s attention.  This attention exists by virtue of the fact that they are people.  People (real people – viewers) are drawn to people (real or fictional) and this phenomenon, along with the aforementioned compositional location of the main characters combine to create a visual resting place for the viewer.

But we don’t want the resting place to become a visual stop.  Any good composition is fluid.  It keeps the eye moving around the piece with all of the compositional elements existing in harmony.  However, we don’t want an illustration to be completely caustic and lack any sort of visual anchor.

Applying this knowledge, I didn’t want the viewer’s eyes to stop on the characters.  That is why the smoke-diagonal continues and the cannon takes over the role of the flag, reinforcing this strong direction.  At this point, however, we arrive at the border of the picture plane.  We need something to bring the viewer’s eye back into the piece.

This is where the two bushes and the placement of Bea’s legs take over.  They send the viewer’s eyes down and over to the rocks in the foreground which are angled to point simultaneously back to Alan’s face and also back to the highest point of contrast (where the darkest blue of the flag overlaps the brightest spot in the smoke).

The smoke, the cannon, the trees and the other background elements are blurred so that they sit back in the distance, creating a separation between background and middle-ground.  The details in the foreground don’t overwhelm the middle-ground because the foreground consists of very similar colors.  Even though the colors in the foreground plant are very sharp and very dark, the cast shadow from the rocks helps to diminish the visual weight of the plant by reducing the contrast of the plant and the grass behind it.

The very bright highlights in Alan’s gun pull the viewer’s eyes back up into the place that all of this originated.  They don’t overwhelm the characters either because those highlights, although they are very bright, are also very small.

It is not so apparent in the reduction above, but the grass is more clearly rendered, the closer it is to the viewer.  The grass consists more of loose brush strokes, patches of color as it recedes into the background.

The trees are very loosely painted as well.  Smoke is inherently blurry and undefined.  This contrast of detail helps to diminish the background elements and emphasize the characters, which are very sharply rendered.  In fact, I actually cheated (call it ‘artistic license’) in painting the foreground elements.

I put more detail and brighter highlights into the characters than I put into the rocks and plants in the foreground.  This also helps to direct the viewer’s attention to the right place.

The cannon is slightly blurred and it was painted-over multiple times with the same colors used in the smoke to help push it back into the background, but I put some highlights and a bit more detail into it so that it would appear a little closer than  the other background elements.

There is quite a bit more that I could say about all of this, but I’m about blogged-out, so I will save the topics of gathering good reference and making the transition from comp to final until next week.

A few weeks ago, while I was completely buried in freelance, sleeping very little, working super-long hours, and feeling generally loopy, this guy just popped into my head. I got so excited. I was intrigued by his silhouette, his posture and his proportions. I quickly did a line drawing in Photoshop, saved the file and it wasn’t until today that I could afford enough time to sit down and actually do the painting.

I have another project that I should have been working on but I had kind of an unsuccessful, uninspired day with it yesterday (my drawing hand would not communicate clearly with my brain).  Sometimes when that happens, and I have enough flexibility in my schedule (ie: no killer deadlines) I can solve the problem by getting away from it and working on something else.

The whole piece took about 6.5 hours.

Below, I have included some screenshots that reveal a little bit about the process behind this piece.  I normally use lots of Photoshop layers when I paint, but this time, I didn’t use as many, just for the sake of speed.  I knew that I couldn’t spend much time on this, so I just plowed through.  I had no color comp, no real plan, only my enthusiasm.

I’m pretty happy with how it turned out.

Welcome back to my ongoing tutorial about the cover of The Dreamer issue #4.

In part one, I discussed the importance of the color comp in the formation of a painting.  I pointed out three important aspects of the comp stage:  The comp stage decides the overall emotional impact of the final piece.  The comp stage helps you to spot problems and challenges that you will face while you paint the final image.  The comp stage is the perfect stage at which to make any significant drawing or compositional changes.

Today, I will take the first two aspects and elaborate on them.

First, the emotional impact:

To the left is the final color comp.  Note that I did not spend a lot of time painting any small details like the blades of grass or the tiny metal bits of Alan’s gun.  I was interested only in the overall harmony of colors.

While it is important not to rush through the comp stage, to take your time making the right color choices, you also don’t want to spend too much time on the comp, by way of putting in too many tiny details that are just going to get painted over.  In the final piece, I don’t believe that there is even one pixel from the comp showing through.

Because I reduced the file size by drastically reducing the resolution (for this comp stage only), I was able to move fast and make changes very quickly with very large brushes.

(As a side note, for most of this comp, I used a custom brush that I was given by an art director of mine.  I customized that brush again with the brush editor, trying to get the brush so simulate an oil look.  I generally use the tool presets palette to manage all of my brushes and I only use the brush palette to customize brushes.  You can learn more about customizing brushes in Photoshop in lots of different places on the internet, but I will recommend that you check out Paul Lasaine’s blog for some really good stuff.)

I wanted there to be a pink wash over everything, even the shadows, to give the piece that warm, romantic glow.  I am a lover of diffused light, which is basically a soft glow as opposed to a strong directional, harsh light source.  Since the piece takes place at morning and there would also be a lot of cannon smoke in the air, the diffused light was easy to justify.

The diffused light obviously contributes to the romantic feeling of the piece.  You can see in the comp a purple rim light around the shadowed side of the figures which I later painted out.  It felt too contradictory to have a strong reflected light on the shadow side of the figures and a soft glow on the key light side of the figures, so I ditched the strong reflection and went for a softer pink fill light in the shadows.

I used almost no cool colors in the entire piece.  The dark blue in the flag and the blue of Bea’s dress are the only places where I put distinctly cool colors.  And this is only because I absolutely had to use blue in both of those areas – for historical accuracy and for story continuity.

Had I just used a ‘straight out of the tube’ blue for either of those colors, it would have been an eyesore.  The blues would have been very harsh and distracting, so I blocked in the blues with just enough variation between light and dark to imply the forms and then I began putting a series of pink washes over the blue areas.  I continued doing this, adding and subtracting washes of pink, layer upon layer, until they ended up harmonizing with the whole piece but were still perceived as being the correct color of blue.

The grass and plants, though they appear green in the piece, are not really green at all.  Go on, use the color picker in Photoshop and see how they come out.  The greens are just neutralized reds and pinks that appear green when placed next to the pinks in the smoke or the bark of the trees.  This relates to what I said about the blues in the sense that you have to consider both color and light separately when trying to achieve a strong emotional impact.

This is where artistic license comes in.  I was able to push the color of the grass farther away from its natural color than I could the blues of the dress and the flag.  This is because our historian-eyes demand that the blue of the flag be recognizable and our audience-eyes demand that the blue of Bea’s dress be recognizable.  We all know what Bea’s dress is supposed to look like, so I had to be more careful as a painter when ‘messing’ with that color.  The pink grass, in other words, is more forgivable.

Sure, I wanted a pink hue to everything, but I had to be selective and add more pink to certain areas and less pink to others.  Finding this balance is difficult, but lots of fun and extremely rewarding when it works.  Be patient and take your time.  Try lots of different things.  Experiment.  We all get better at this the more we do it.  Getting it wrong many times before you get it right is often more beneficial than getting it right on the first try.

I put a color gradient in the smoke.  Originally, I had the entire smoke cloud looking like it now does at the top of the frame of reference.  It was all very bright and pink.  But this made it look flat and kind of boring, even though it was bright.  I neutralized the color of the smoke near the bottom of the cloud and blended the two extremes in the middle.  This helped to create the illusion of a sunrise and an environment that extends beyond the foreground seen in the piece.

The more neutral areas near the bottom of the smoke cloud imply a cast shadow from a hill in front of the characters but behind the viewer.  It also works to complement the composition, but I will get into that in next week’s post. I’ll also talk about the color choices for the sky and cannon in next week’s post.

By the time I was finished with the comp, I realized that the emotional impact was only 90% there.  I was on the right track, but I knew that I could push it farther and make it even more intense.  I could have spent more time in the comp stage, but I was pretty confident that I knew exactly what had to happen in order to maximize the emotional impact.

I don’t recommend leaving the comp stage until you have a plan for all of your problems (at least the ones that you become aware of at this point) and your emotional impact is 100% established, but I took a calculated risk in moving on and it worked out.  I knew that I needed to saturate the color in the areas where the main key light was shining and that I needed to add some darker shadows in the foreground and middleground.  I was pretty certain that those two minor changes would really heighten the drama.

Before I wrap up for the afternoon, I will explain some of the problems and challengesthat I foresaw when working on the drawing and comp stages:

The first thing that jumped out at me were a couple of compositional tangents.

The top of the larger tree was awkwardly intersecting the smoke cloud.  This disrupted the nice, triangular composition of the whole piece.  It drew attention to itself unnecessarily and drew attention away from the main characters.

The second tangent was at a strange point where the back of the cannon and Alan’s shoulder met.  This tangent created a contradiction in depth and it made the cannon appear to be in both the middleground AND the background simultaneously.

I knew that both of those tangent issues would have to be resolved.

I have already explained how I dealt with the problem of the blues in the flag and Bea’s dress with the series of pink washes.  Granted, I would continue to tweak the hues of the blue areas, trying to perfect them and really get them set in the environment, but the problem was more or less resolved upon completion of the comp.

Alan’s left arm was also giving me trouble.  Because his arm is perpendicular to the direction of the key light, it could have ended up seeming very flat.  It could also have appeared to be floating above Bea’s back.  This challenge was a mystery.  I had no clue how I was going to solve that.

Lastly, the shape of the smoke was going to be tricky, since there would be a delicate balance between making it feel really organic and random and yet designed and deliberate.

That about covers the problems that I was in touch with at this point in the process.  I later ran into a few more minor issues with the environment among other things, but I will cover those when I reach that point in the tutorial.

Next week I will talk about using light, color and detail to complement a composition.  I will also discuss the changes in the composition that were necessary upon reaching the comp stage.  Perhaps I will also get to talk about another essential ingredient in the formation of a good painting: REFERENCE!

WELCOME fans of The Dreamer!  I am so glad you’re here.  We have been anxiously anticipating the reveal of the cover to issue #4 and now, finally, it is yours.

For those of you that are not yet familiar with this fantastic web comic by Lora Innes, head on over to TheDreamerComic.com and give it a whirl.  You will not be disappointed.  “Adventure, Romance, War.  1776 is back.”

Lora is a long-time friend and an incredible artist.  When she said I could do a painting for the cover of the fourth issue of the comic, I nearly soiled my armor.

What a great project!  I got to collaborate with a dear friend and someone whose talent I admire.  I got to paint an image that is gushing with romance and drama – an image that I could never have come up with all by myself.  I grew tremendously as an artist during the creation of this painting and I believe that we definitely came up with something good by working on it together.

But I don’t want to get ahead of myself.  I’ll talk more about the final image in the last installment of this tutorial.  Did I mention that this is a tutorial?  That’s right, for the next several weeks I will be posting (every Monday morning in keeping with my current schedule) a step-by-step series of tutorials that reveal the process behind the creation of this painting.

These Dreamer posts will be longer than my normal posts, since I want to be generous with the details in the name of education.  I will be posting about other things throughout the weeks, of course, but you can count on having Dreamer stuff to read on Mondays.  I am hopeful that this will be fun and informative for us all.

So let’s get started ::

It all starts with a sketch.  Lora and I discussed the image for quite some time before settling on a final design.

She did a few thumbnail sketches (you can see a hilarious one on the dreamer site) as did I and we eventually worked out various compositional details, placement of objects, size relationships etc..

She knew what she wanted to see, conceptually, so the overall compositional idea is hers.  Alan looking macho with flowing Fabio locks, cradling Bea in his arms, a wall of cannon smoke in the background, all of this taking place at dawn…  Obviously, the story made a few of these decisions for us as well.

Lora took all of these plans and created a final sketch, leaving it a lot rougher than her usual drawings for the comic.

Since she knew that the painting would determine the forms in ways that a line drawing would not, she didn’t waste any time going back in and sharpening the drawing like she does for her sequential pages.

I never asked her why she chose to have Bea facing away from us, but I always thought that was interesting.  …not to mention that it was really fun to paint all of that hair.

The next stage, the “color-comp” is pivotal.  The older I get, the more I mature as a painter, the more time I spend on this step.

Basically, the color comp is a loosely-painted, smaller version of your final painting.

• This is where the overall emotional impact of the piece is decided.

• This is where you discover what problems and challenges you are going to encounter along the way and begin preparing for them, although some problems and challenges may be evident in the drawing stage.  (Alan’s left arm, the position of the cannon, the tops of the trees, the shape of the smoke, Bea’s blue dress…)

• This is where you have freedom to re-draw certain things or move things around to change the composition since color and detail have a significant effect on whether or not a composition that worked in the drawing stage continues to work in the comp stage.  (Note the way the smoke changes from the sketch to both of the stages of the comp.)

I knew that Lora wanted the time of day to feel like dawn, but she was also urging me, over and over, to “make it girly.”  I haven’t seen pink light at dawn very often, but pinks and reds and warm grays seemed like the right answer to get the emotional tone that we were going for.

Lora scanned her drawing and sent it to me at full-resolution (3082×4044).  I reduced the file-size by about 75% so I could make fast strokes with huge brushes in Photoshop without my computer slowing down.  I would bring the file size back up to full resolution after the comp was approved.

I duplicated the layer with Lora’s drawing on it and then cleared the original layer.  I then had two layers, the empty background layer and the drawing on a layer above that.

I changed the ‘adjustment layer’ setting on Lora’s drawing layer to ‘Multiply.’

Doing this allows the artist to paint on layers underneath the drawing without disrupting the drawing.  The overall image maintains the appearance that there is paint underneath and the line drawing on top.I covered the whole background layer with the mauve color that you can see in the image above.  Then I began to lay in large blocks of color, just loosely trying to arrive at some sort of balance and overall color harmony.

There was a lot of experimenting at this stage, but not as much as I usually encounter at the comp stage, seeing as how I knew that I wanted this very pink tone to the whole thing.  Any decisions that you can have made at the beginning like that make the whole project easier… …as long as they are the right decisions.

You can’t know for certain, of course, that every decision you make early on is the right one which is why you have to keep everything in question until the appropriate time.  This probably sounds too vague right now, but I will try to make sense of it in future posts.

On Monday, I will discuss my specific color choices.  I’ll also elaborate on those ‘problems and challenges’ that I could foresee upon starting the comp.  I hope you guys enjoyed this and will continue hang around at ChrisOatley.com.

Thanks for reading and I look forward to hearing from you.  Feel free to email me.  Feel free to let me know how I’m doing with the communication and whatnot.  Too much detail?  Too little?  I can go on and on…  Thanks again.

:: continued in part 2 ::