For design he needed everything from the logo through the emails the store would send customers. He was really hoping to stay away from the usual over the top wackiness (or tackiness?) of the typical martial arts site. Understated, airy, and simple were his keywords.

Starting with the logo I incorporated the idea of the kamon, a Japanese family crest. Banners emblazoned with these symbols were carried by samurai into battle. Since he had already provided kanji for his company name it was just a matter of finding an equally attractive brush script to finish the logo.

Following that was design for the feel of the site. Taking the samurai idea along with his airy keyword leads easily to the blue sky header against which are silhouetted the tools of war. For the rest, simple ruled. Few strong gradients, no harsh colors, and clear action points.

Not wanting to reinvent a wheel I’d never seen or blow the sensei’s budget I went with Zen Cart as the base for the shopping cart and WordPress for the blogging software, intending to integrate the blog into the front page of the store.

Styling Zen Cart to match my design has proven to be an adventure in PHP that did not like to be modified in the ways I desired. As is always the case with these ‘highly customizable’ software packages, the authors have specific ideas about what their customers might want to do. Zen Cart is a powerful, flexible platform – if are practiced at reading other peoples’ code.

In the process of building budobuki.com I’ve learned a great deal about PHP, buffed up my HTML and CSS, and started learning SQL. There’s still work to be done but the site on target to go live at the end of March.

Iscream is a casual Flash game I’m currently working on. I’m part of a 2-man team with an amazing programmer who I’ve worked with many times before. I’m the sole artist and primary game designer (game design is never a solo task, you know) and he’s writing the actionscript code. We went with Flash and Actionscript/Flex because we both know it well and Adobe is still pushing Flash’s ability to publish to not only the web but Android and iOs as well.

We are starting prototype testing soon. This is where we will refine the basic gameplay mechanics that we will build the full game around. From there evolve the designs for reward structures, difficulty progression, and so on.

We’re planning to release to the web by summer followed by the major mobile platforms. I’m excited to learn just how well the multi-platform publishing tools Adobe advertises actually work.

Once some early concepts of the main character and main enemy had been established the next task was to make a game scene concept. Morgan had very specific ideas of how he wanted the game to feel (“thinking in screenshots” he called it) so the first thing to do was set up the scene we wanted. I started out with some free sketching while we hashed out a description over email.

The final idea was the main character looking out over an urban scene, recently attacked and still burning and clear evidence of both alien foes and active resistance. The scene should be representative of the final art style, anime, cel rendered characters and a stylized but painted background that both works with the cel shading of the characters and contrasts against it. At this point the rendering tech demo was well under way so we had a strong idea of the character rendering. The environment was a little more open. Once one of my initial scenes was settled on as a launching point I worked on color tests while the rest of the team collected feedback. I also engaged with our resident filmographer to help establish the basic idea in a more cinematically dramatic way.

Below is the final result.

For more images from the scene concept check out my screenshot album on Picasa.

After a heavy round of team brainstorming, the idea emerged that the enemy alien race would be squid-like, using powered armor to navigate the player’s home world and sensing their environment using their natural organs. Two other considerations needed to be kept in mind. First there should be a clear and consistent feel that carried across several variants of the alien/machine. Second was the general precepts of non photo realistic rendering and simple color schemes.

Once a basic idea was settled on, I had to design four variants and make model sheets for them.

For more images from the alien concept, including the model sheets, check out my screenshot album on Picasa.

Since Faerie Tale was going to have lots of characters but was still locked down to the same aggressive schedule other game releases were under I had to improvise. So rather than blowing the whole time budget on modeling and rigging half a dozen characters, I fired up Poser. Now now, hold on. I know how most artists feel about Poser, Bryce, and similar tools. But when the schedule is as tight as this one, it’s not only okay, it’s absolutely necessary to build on prefabricated material. This is something I had a very hard time as an art director convincing my staff of. That’s an interesting parallel between code and art developers. They all really want to build it themselves.

I didn’t actually animate the characters in Poser. Rather, I exported the models and made simple rigs in Max and animated them there. All of Max’ processing tools are faster for one thing. The viewports update faster, the renderer is faster. Tweaking the rigs is significantly easier and at the time, Poser rigs still caused some significant mesh folding around the joints.

Bonus Game

When building the bonus round we decided to try something a little more complex than a simple pick 1 of X game. Since I’d rendered a hobbit door in my early exploration we decided to try a Monty Hall game. Here’s how that works. The player is presented with three doors and is allowed to choose one. Only one of the doors hides a prize (since we were dealing with a bonus all the doors had to have rewards so we actually had one good prize and two lesser prizes). The host then opens one of the other two doors revealing that it doesn’t have a prize. The host then asks the player if he wants to keep his current choice or change to the other, still closed, door.

The Monty Haul game is an interesting intersection of probability and psychology. What most players don’t know is that they should always change doors. Because the host revealed a dead door the prize is twice as likely to be behind the third door as it is to be behind the door the player initially chose. The typical faulty understanding of probability suggests that the chances are even between the two still closed doors. And, given no clear advantage to changing, people are much more likely to stick with their original choice than to risk changing their mind.

In the end, partly because of the need to reward the player no matter what they chose, the translation team had a hard time understanding what was going on. Did they win the n-prize? Why did the host open that other door when i picked this one? So the Faerie Tale bonus round ended up just another pick one of three.

For more images from Faerie Tale check out my screenshot gallery on Picasa.

This was also the game on which we really started to develop the custom layout script that would be used for years after. It enabled me to do some interesting asynchronous animation on the Gleeble character in the gamble round (another first for GrandVirtual).

Another first was a real sound experience beyond a few public domain bleeps and bloops. Enter Michael Damon who would be our sound guy for another 6 years.

I set several rules for this project…

• Must be completed over the weekend, no fussing on it for ages.
• No texture tricks, painting only.
• Stick to familiar themes to make sure I don’t get analysis paralysis thinking about what to paint!

Click for Full

Then, a couple years later, I got to do it again, updating the look and feel to reflect new features, a new target audience, and a new branding.

In Mammoth Hunters I used an interesting painting technique that I’ve since heard other artists talk about. In order to make the central mammoth figure I first created a body from basic forms in 3D Studio Max. Leg sections were capsules, body parts spheres, and the tusks were quick lofts. I posed the shape-body and set up lighting in Max then rendered that out. In Photoshop I took that render and used it as a lighting and pose reference model to paint hair and features over. The technique is fast and effective and certainly easier than finding a mammoth to use as a model.

While rendering I got into an interesting argument with the product owner over the myth that elephants have four knees. They don’t.  Looking at a skeletal history it seems like the “wrist” joint has, over the eons, lengthened and thickened to the point where it looks kind of like a knee. And if you really look at an elephant you can still see the elbow. Such strange discussions we have over art.

The Butterfly Wings concept lent itself to working in a painterly style. I was aiming for a style that was somewhat impressionist but not so much that a slot player would be turned off. Impressionist painting is not exactly my forte but I knew that Photoshop filters by themselves wouldn’t cut it. And remember, 2 days is what I had to work with.

So I focused on two elements. First was to call out the idea of the impressionist paintings with the subject of the girl in the sun dress. The light, sunny-day and flowers feeling of some of the more popular examples of impressionist painting are in the breeze, the hat and the lighting. Then it came down to technical details. I rendered the girl in Max and then took her into Photoshop. I layered filter over filter until I achieved the effect I wanted, keeping every layer as I went. This is something I’ve tried to impress on every artist I’ve worked with. Keep your history!

Check out more of the slots at my Picasa screenshot gallery.