The talented students and supportive staff helped Animux complete the 8 minute short storyboard in a marathon 13 hour session. PH2PC is broken down into 14 sequences excluding the beginning and end titles. We started at sharp 8AM  on Dec 19,2010 and completed around 9PM.

CCSAD helps Animux create the storyboards

The chair of  Entertainment Arts, Scott Bogoniewski ,  had given us an opportunity to talk about our project at a weekly Senior project review seminar at their auditorium.  Out of this, we were able to get 8 students who volunteered to do the storyboard on Dec 19th which would be guided by Steve Stanchfield. 

Steve and the students were in the studio at sharp 8AM and we were read the script in character where each student artist took up a character to act. We figured out the shot layouts during this reading with the insight from Steve and the students. Once we were done, we devided the shorts among the students to draw up the boards.  We took a lunch break and then the talented  students came back to draw the rest of the storyboards.  Amazing work ethic matched by their passion for their skill was very evident during the whole time.

(Front L2R) Emily Cawthon, Serene Garlow, Jayna Shropshire (Back L2R) Edward Micheals, Brandon Babcock

Renderman renderers are not as forgiving as many others. For instance if the normals are funky, this will cause all sorts of nastiness when the shaders are applied. Don’t assume that because the geo looks fine in the Blender renderer, it will work in Aqsis or another Renderman renderer.

Terminology

(I thought I should make sure we all know what I’m talking about)

• Asset - the full collection of geo to be passed between artists. The Peacock.
• Mesh – an element of geo in the asset. Left Eyeball.
• Geo - geometry in general.
• Shell – a seamless collection of Uvs.
• Shader – a program written in Renderman Shading Language to define the surface properties of the geo.

Polygons:

All polygons are to be quads. Triangles if you are really stuck.

Break your asset up into meshes that will have different surface propeties. Renderman does not support multiple shaders per single piece a geo. So you cannot have the head of the character with one shader for the fur, and then select the beak and apply a different shader. This is a really important issue. It will mean that you need to think carefully about seams, and how to connect these pieces together.

Meshes should be ‘closed’ surfaces. For example don’t just model one side of a piece of paper, model both. It takes a fraction more work, but it means that these meshes will be more versatile when it comes to shading. On a similar note, it’s worth modelling things like eye sockets for characters too. This way if the eyeball doesn’t sit perfectly - or is stretched - then the render may still look correct in the gaps. If there was no geo there, it may be too late or too hard to send the model back to get fixes made.

All normals are to face the intended direction. If on the same surface some are facing out, and some are facing in, then edges or seams will appear on these borders. You may also get things like shadows and occlusion calculating incorrectly as when they are calculating, they test for things like normal direction.

UVs:

Be patient, and take your time. There is no fast answer to setting up good Uvs.

Uniform spacing. Test your Uvs regularly with a checker map. All the checks should be evenly distributed across a mesh. It’s important for the texture painter to know that if they paint a detail in one area of the model, the same detail painted elsewhere in that model will come through the same size.

No atlas mapping. Break your UVs into logical shells. This way anyone can easily paint and amend textures. It’s not uncommon for textures to need little paint fixes late in the game, and to be able to open up and spot the problem areas quickly, is something you’ll soon appreciate. Atlas mapping is - possibly in my own opinion - only useful if you have a full 3d paint application supporting your work, like Mudbox or zBrush. Take the time to lay UVs out correctly.

Keep all geo in a single UV set. I’m not sure about how Blender deals with different sets, but I think they also need to supported in the shaders for them to work. So for now, keep them in the default set.

Get regular feedback from your favourite texture artist, on how they would like the UVs laid out. Think about the artist painting the UVs, if you have 10 fingers it might be easier to lay them all out in the same direction, near each other. This will make it much easier for the artists to paint up the details. Also, keep the different UV shells orientated vertically and horizontally. If they are skewed it becomes very hard to paint and change details.

All UVs to be kept within a 0-1 range (unless a method for dealing with UV Tiles has been developed).

Be patient, and take your time.

So in conclusion …

Just make sure you tidy up after yourself. It’s actually fairly easy to write some simple QA tools that can test many of these things, otherwise just get into the habit of checking everything before you hand off the asset to the next person in the line.  If someone has planned to do some work on an asset - and these things have not been checked - they may well end up wasting a lot of time trying to fix errors, and be well within their rights to simply send it back to the modeler for continued work. Also it’s possible that an error may not show up until a lot of work has taken place. This is why it’s very important to QA your own work.

In May of 2009, I was at the Kalamazoo animation festival in Kalamazoo, Michigan.  This is where I first heard about Prince Harming to Prince Charming.  Prince Harming to Prince Charming is an animation project led by Mark Puttnam, with the goal of creating an animated short using a completely open-source pipeline called Animux.  I met Mark at the festival and afterwards began to exchange e-mails with him. He asked me if I would like to participate in the project.  He had seen my animation demo-reel which features birds, and since Prince Harming to Prince Charming was about a turkey in a town of peacocks, it seemed like a good match.  I read the script and thought it to be quite endearing, kind of a reverse Cinderella story.  I agreed to lend my animation skills, with one hitch, it was to be animated all in Blender, a program I was totally unfamiliar with.
I had gone to college for fine arts and afterwards gone to animation school where I learned to animate in Maya and SoftImage XSI.  Blender was a program I had heard about only a few times in passing.  Nobody I knew had ever used it.  Kalamazoo was the first place where I met people who worked with it and where I started to learn more about it.  I was intrigued to find it was a 3D animation program available on the web that cost nothing to download.  Anyone who had a computer and an internet connection could have it.
Several questions arose in my mind, the first was ‘How good can this program be?’.  Maya is the program I use predominately, and even the student version is very expensive.  Maya is also an industry standard in film and television for animation and visual effects.  In the right hands, Maya is a very powerful tool.  How would Blender compare?  I started looking into work that had been done in Blender on the internet, and was very impressed at the quality of work that could be achieved.  A great example is the Big Buck Bunny short, which is comparable to anything that can be completed in Maya.
Then came my next question, ‘How is easy is it to use?’.  That was something I was about to find out.  In July, Mark told me about a 20 second short he wanted to put together for the SIGGRAPH convention in New Orleans taking place in August.  This was going to be tricky to accomplish.  The first difficulty, and the biggest, was that everyone working on the project was spread out all over the world.  Mark was in Detroit, I myself was in Austin, Texas, and we had team members as far as Costa Rica, Brasil and Russia pitching in.  What this meant is that we would have to rely on long distance communications, mostly through e-mails and chat room meetings, to make sure everyone was up to speed and on task.
The second difficulty was that I had about a month to try and learn Blender so that when the model and rig were completed I could start animating with it.  The great thing I found out about Blender, is that because it’s something of an under-dog program in an industry dominated by Maya and Studio Max, there is a large underground community of Blender aficionados all over the internet who collaborate and share with each other.  Anything you want to learn about Blender is free to read or watch in a video tutorial at one of the many Blender websites like Wiki.blender or Blender Underground.  Luckily for me I also had the Animux chat room to ask the people I was working with on the project any specific questions I had.
So I buckled down to start studying Blender and how to animate with it.  Right away I noticed that Blender was very unique from other programs I had used.  The user interface was different and relies on a lot of hot-key controls to navigate around.  Learning  various hot-key controls can be quite a learning curve.  But as I came to be familiar with the viewports and menu screens I found that while it was different, I did like the way it was all set up.  I tried my hand at some modeling, which I felt was in some ways easier to do than in Maya.  I also found the file linking and file saving systems to be very convenient.  My biggest problem however was with the IPO editor, which is similar to the Graph Editor in Maya.  This is where the animator can actually see the animation curves for translation/rotation/scale, and various other attributes.  In Maya, this editor is where I do most of my animating.  In Blender I found that it didn’t quite give the fine-tuned controls over the key frames that Maya does.  I would like to note though that in the new version of Blender that I tried out at SIGGRAPH the following month, these problems had been addressed and the IPO editor was greatly improved.
The time came when my rig was ready for me to animate with.  I was to be in charge of the peacock character, who I thought would be really fun to put some snobby attitude into.  For this short, he was to walk across a stage, behind our hero, Turkey, and give him a “look”, then suddenly jump to the front and center and begin his ‘I’m so sexy’ model poses.  Great stuff for an animator to play with.  But by this time, we were cutting it pretty close with our schedule.
I received the rig on Sunday and had to be finished animating on Friday morning so we would have enough time to render.  Twenty seconds may not sound like a lot of work to complete in a week, but for animation it can be.  And if you’re working in a program that’s new to you, it can be downright harrowing.  Working with an armature in Blender took some getting used to, but by the end of the week, I felt I had a firm grasp on the animation aspect of Blender.  Furthermore, I came to really enjoy working with it, as problematic as it was sometimes.
The peacock was completed on time, as was the whole short.  The following week I was able to see it played in front of a group of people at SIGGRAPH, which was very exciting for me.  That whole week I got to meet face to face some of the people in the Blender community, including the creator and head of the Blender Foundation, Ton Roosendaal, who was a very cool guy I might add.  I hung out with these people all that week in New Orleans, having a great time, and learning about other projects artists are working on with Blender and how they’re using it to accomplish some really great things.
I was happy to participate in this project and hope to continue to collaborate more in the future.  For me the best part about the experience was meeting fellow artists from all around the world and coming together to share a creative goal.  I had never done an internet project like this before and it was an interesting challenge.  I also got to learn a new program that I plan on further exploring.  Who knows what the future of technology will be like, or the future of art and entertainment for that matter?  I expect there will be some remarkable things coming from open-source software soon.

Here is his final character design.

Final Character Design - Pavo

Points to Ponder

1. Bean shaped body (especially the chest)
2. Wings need to arms to move around rather than sticking to the body like our preview
3. Show vanity -> we could use the plumage on his head into a hair do.
4. Stature is very important. How he carries himself. (He never looks down, if he does, he does it side ways)
5. Body has to be well kept (Shinning, clean, interesting designs on the plain body)
6. Tail has to be huge.
7. Silhouette needs to angle towards his head. (Higher than everybody)
8. Display the range of emotions/ character. Do not fall into a mold and well rounded.

List of questions to have the character answer.

• What would he have done that he has never told anyone?
• What is worse than death for this character (him)? If he is perceived as not being desirable to anyone. (Everyone should wanting him.)

Yubiy is a wise, not so old, mentor who teaches Turk how to change himself at being better at getting noticed.  Luis got his face and body correct, but did not create the tail feathers as that of an albino peacock.  He will be fixing it shortly.

Not so Final Draft - Yubiy

Get a strong purpose in the film.

She is gorgeous. Luis did an amazing job on her.

Final Draft Character Design - Bella

Points to Ponder:

• Do we want her to see past the peacock’s insecurity?
• If she can look past, the turkey’s and looks are not important? What is the reasoning for looking nice?
• She is not passive. She is a sense of “I know”.
• She always talk with a sense of confidence (seldom small talk.) and knows she can get anything ,but choose to be with who she likes.
• She is not put off by the activeness of the peacock but by the way he courts her.
• Trinkets are unique. (have explaination.)

Here is the final color concept of “Turk” our lovable that Luis Gadea put together for us based on our input. The colors are beautiful, the expression is exquiste and fits the character perfectly.

Final Draft & Color Concept - Turk

Points to Ponder:

• He is an underdog. (There should never be a doubt)
• Goal is the girl. Obstacle is the peacock and his looks.
• Underdog that never did anything wrong (should not be DONE! HERE.) Use the peacock, to show some of his flaws
• good head to body ratio.
• He is optimistic and persistent. (strike a balance between realistic expectations)
• He has inner doubt and hence needs yubiy
• is he educated?

Questions about the character

• What would he do to attain his goals?
• What would he never do to attain his goal?
• What happened in his life that started this insecurity? (helps with his reactions to things.)

Though the initial plan was to create the storyboards with David’s GIMP brushes, we had gone on to create the storyboards with just drawing on paper. One of the main reasons for this was first, the artist have not been exposed to GIMP yet. Also we had only a few brushes would mean that the artists had to create the missing brushes themselves or wait on David to create them.

We have compromised on the consistent look of the storyboard to a completed storyboard and gave into their request of drawing the storyboard in the medium they are comfortable with. However, the brushes and emotion sheet is a valuable tool in getting the expressions similar to what David has done.

Robin has reworked the script with some more change where the Casanova character was brought back into the script. The main reason for that being that most of the humor for the script was during their interaction. Also, Robin has been trying to convince us that because we have taken out the first 2 minutes of the short which introduces the turkey character, we have lost the turkey’s character depth or narrative. The audience knows nothing about the turkey character.

Bassam Kurdali has been advising us on this project. His advise was to spend time on the script and not to cut corners in the writing the script. The writing has be good for the project to work. If we have to make for the extra time we spend in writing, we could do it in everywhere else (modelling, animating, lighting etc) but never during writing. 

As Bassam reviewed the script he had the same concern about the turkey.  Bassam did like the new script that it seems tighter than the previous versions though he did not agree philosophically. The way it is written right now, it seems, the turkey character is manipulating female peacock to like him, rather than being geniune and getting her to like him for himself. This was a flaw in script, even Robin had pointed out before we made the changes.

Right now, based on the script according to Bassam,  Bella is smart, feminist, activist, into education. Pavo (Peacock) is sexy, vain, he’s all about himself first and please second. But the audience knows nothing about Turk (turkey) other than he wants Bella. So his advise for turk was to make him a country bumpkin, no education, stranger in a strange land, longing for love and extremely gullable. For Yubiy, we wanted us to make sure we make him conniving and crafty.  The key thing according to Bassam was to make sure that we have in the script that Bella knows everything about Turk before she begins to like him and hence sympathetic towards him. Try to establish right in the first scene between Bella, Pavo and Turk and keep working it through out the script. Do everything to make sure, she ends up liking Turk for who he is, rather than being conned into liking him.

Robin, is working on the script and we will place the script as it progresses on the “about” page of the site.

These brushes will be available for download along with the entire set of character brushes for PH2PC used in creating the storyboards…..including various angles and close-ups of Turk, Pavo, Bella, and of course Yubiy!

David