Out of the Frame - Skill vs. Creativity

There is a tension created by our desire to be creative, and our desire to build high levels of skill.

Being creative means to do new things. When we do new things, we are beginners. thus, creatives keep returning to the state of being a beginner, over, and over again.

Skill, on the other hand, is built through a cycle of repetition, introspection, and making small changes, over and over again, often many thousands of times.

Creativity means to take different bodies of knowledge, breaking them apart, and recombining them in new ways.

Building skill means to refine your understanding of an existing body of knowledge.

How do you hit a balance between the two?

In my case, screw the balance! I'm going for the fun stuff!

I often learn a technique first, and practice it until it is a part of my tool kit that I am comfortable, or at least familiar with. Then I start exploring the creative possibilities.

The T-Rex picture above is a case in point. It is a fun picture, but it is not really creative. At this point, I am learning to use the technique. The creative part comes later, when I have practiced, probably just barely enough, to bring off using the technique in a new context.

That is also why I skimped on photo-realism. I already know how to make a T-Rex model look more organic. That is not what I was interested in here, so I skipped it. When I am comfortable enough with creating the 3D-break-out-of-the-flat-surface look, I'll bring other techniques back in, to create more creative work, with a bit more finish.

What is your approach? How do you handle the tension between being creative, and building enough skill to do good work?

Repetitions and Variations: Dana vs. a T-Rex

Dana vs. a T-Rex. Model: Petra Brewitz

In a previous blog post, I published Dragonslayer, a quick-and-dirty, but fairly nice looking, picture of a woman facing off with a dragon. I promised to write a bit more about the picture, and how I made it. I also promised to publish a few variations on the theme.

The finished version above, Dana vs. a T-Rex, features Dana, one of the main characters in the A Rift in Time photo comic. Dana was played by Petra Brewitz in the comic, and she also modeled for the pictures in this article.

We did not start out doing a Dana picture. Petra and I had been discussing doing something with a medieval theme. That takes a lot of preparation, and quite a few props, so we decided to make something simpler, just to try things out.

Technical details: The picture above was created by applying the Painted 2 + Border filter in Comic Life 3. This filter creates a painted comic book look that works very well for many illustrations.

The original inspiration for my picture came from an illustration by the Fantasy artist Joe Jusko. 

If you compare Jusko's image with mine, you will see that there are a lot of differences: Mine is, how shall I put it, more modest, his is in portrait format, mine is in landscape format, his has a medieval fantasy setting while mine is urban Science-Fiction, and so on...

My picture started out being more like Jusko's. I planned to use a dragon, and I planned to shoot vertical. As I worked through several iterations, my picture diverged more and more from Jusko's. A good thing, because those differences are what separates being inspired by, from plain old copying.

Let's look at how I built the basic composition. The picture above is built from three photos.

I photographed Petra in a parking lot. She, Julia Reinhardt, and I took a photo walk after a planning meeting (I'll tell you about it in a future blog post), and since I just happened to have a toy gun in my backpack, we took the opportunity to shoot the picture when we found a fairly clean background.

I used a single hotshoe flash placed under my camera to light Petra and eliminate unwanted shadows.

I shot several photos of Petra, with both portrait and landscape orientation, and with different lighting.

The reason I switched from portrait to landscape is that it allowed Petra to take a wider stance. A wider stance leaves more space to show the monster.

I shot the background after shooting Petra. This is usually a bad idea, because it is difficult to get the angles right.

In this case, I had measured the distance between the ground and the camera when I shot Petra (1 hotshoe flash height + 3 fingers), so it was easy to replicate the angle.

I used a 35 mm lens, and f/14. I wanted to keep everything in the picture, from Petra to the portal, reasonably sharp.

The woman sitting on the bench? I cloned her out. Quite easy to do.

Oh, perhaps I should mention: Originally, I intended to use a different, more dragon friendly environment, and something much more like Joe Jusko's original.

I had shots from a pond that I had shot for another different picture. (You may remember the brachiosaur picture I published awhile back.) I thought I would be able to use one of them.

As it turned out, all of them were shot with the camera too far from the ground. I could have gone back to the pond to reshoot, but a combination of impatience and laziness made me change my mind. Instead, I shot something close at hand, which turned out to be the brick building.

Oh, the dragon... The wings look fairly good, but the rest... I just had to try anyway.

Sometimes you got to try, even though you know it won't work.

As you can see, the dragon is truly awful. The only part that works reasonably well, is the tip of the wing in the window to the left of Petra.

How do you save a picture like this? I could put in a lot of effort in an attempt to make the dragon look more organic. An easier way is to reduce detail in the entire picture. that way, the viewers expectations change, and something can suddenly be perceived as better.

Dragonslayer

I added a quick and dirty paint effect with Prisma's Mononoke filter. The filter removes a lot of details, and hides what is wrong with the dragon.

I wasn't happy though. I wanted something a bit more realistic. There was a model I had wanted to try out for awhile, a snake, made by the Schleich toy manufacturer.

The snake version looked quite good. Technically, it is the best version of them all. I did sharpen the teeth a bit in post, and I ought to have replaced, or fixed, the tongue. Nevertheless, the picture works, except for one thing!

The story! What is the story? Why is there a giant snake in something that looks like a factory? (The dragon had the same problem, but the model was so ugly I never had to worry about the story.)

In this version, the protagonist is aware it is weird that a giant snake appears in a doorway. She expected a T-Rex, so there is a bit of a twist left for the viewer to wrap herself around.

Here is a useful storytelling trick:

When there is a completely illogical gap in your story, make your protagonist aware of it!

In the version above, the protagonist didn't expect to be face-to-face with a giant snake. She expected a T-Rex, so the viewer may still not quite get what is going on, which is intentional.

Viewers familiar with A Rift in Time will be in on the secret: It's a time-travelling T-Rex! They might also guess that the person facing off with the T-Rex is an RRC operative, and they might even recognize Dana.

Just in case you were wondering, this is what the Schleich snake model looks like. It is less than five centimeters high (about two inches).

Once I got this far, it was natural to simplify the story a bit. What would it look like if Dana encounters the T-Rex she expected?

Dana vs. T-Rex, photo version

In the version above, the T-Rex looks pretty good, but it is not photo realistic. I have solved that problem before, by adding organic textures, but I decided to go the easier route, and make a comic panel.

The T-Rex

The final version

And here we are again, with the final version...or maybe not. There are still plenty of variations I could explore.

Dragonslayer

Dragonslayer. Model: Petra Brewitz, Photo: Henrik Mårtensson

Here be dragons! My friends and I have discussed how to shoot dragons, and some other interesting creatures. The picture above is the first in a series of experiments.