Pre-vis and Some story talk

To start, an update from Karen that I she sent me tonight!


And some explanation from Karen:

"The first square is a look at the scene when the cougar is stalking the guy and the girl. I showed them hiding behind the trashcan the cat moments away from discovering them. The second image is a possible look on what the main character experiences when the little girl uses her powers. She is shown really white and pale and ghostly. She has dark circles under her eyes and bright red lipstick. I wanted her to be spooky and still cute at the same time. I hope this helps a little. The silhouettes that are overlapping her are ghost cougars that serve as a metaphor or foreshadowing danger. the picture below that one is a close up of the girl in her strange telepathic state. The 3rd picture is the statue in the middle of the town square, if you want to show the character looking up at a statue."

There is some really strong imagery going on here. Think it is a good mix of everything we talked about last meeting, and some new ideas tossed in. Great work!! This exploration will be extremely helpful when we further solidify the story and design at our next meeting (planning for one soon).

---

Now what I have done. I may have strayed slighting from some of what we talked about, but it's a little hard not to when you translate ideas to drawings. I want to clarify nothing here is intended to be a final storyboard. This is merely my very rough attempt to loosely visualize some of the ideas that have been buzzing around in my head.

It starts by the soldier entering his home town after being away for 10 years (or so) at war. I see this as Act 1 & 2, the 3rd act hasn't been done yet:


(automaticly starts playing the slideshow. Hit pause and click through each frame)

I toyed around with a few ideas here. To start, I wanted to hint at a possible backstory for the killer cat, and how he came into existence by 2 brothers in this fantasy war torn town. More on that in a future post; I want to create a family tree of sorts showing how all the characters are related as well as a timeline.

Second, I thought it might be interesting to have the cat be excessively protective of the girl. Like in a "kill-every-who-comes-in contact-with-her" kind of way. You could think of the cat as sort of a curse for the girl. She cannot have human relations in fear that the cat will murder anyone too close. That is why she is at one point happy to see the guy, but then once she sees the cat, tries to push him away. She returns only when she realizes the cat has already set his eyes on the man to try and convince the cat to stop.

---

The biggest differences between what Karen and I have done so far seems to be with the girl. I've taken her away from that magical state to work with my different take on the story, while Karen has fully embraced the mystical side.

What we have now is choices, both with her design and what role she plays in the story. Is she a protagonist, or a victim? Perhaps both? Maybe if she was wondering soul of sorts left in the town that would work. She could need some sort of human relation to be fulfilled for her release, but the cat keeps making it difficult.

Maybe I'm just talking crazy, what does everyone think?

Some Further Lighting Development Possibilities

***Warning, more technical lighting talk ahead***


Like many of you I got around to seeing Avatar over the holidays, which naturally lead me to reading some behind the scenes interviews. One particular sentence jumped out at me in this AWN Article:

We came up with a system based primarily on image-based lights but then converted the whole system to spherical harmonics," Letteri explains. "What that meant was we could pre-compute all the lighting contributions in a scene and then put the characters and everything in with the lighting and the TDs could move the lights around.


This got me very curious. Our current solution, called Ambient Occlusion Environment Sampling, uses a very similar method to get that nice animation friendly indirect light that is one of the secretes behind great lighting. But for every rendered frame our method has to recalculate the environment light contribution, whereas the method used with Avatar does not. That is ton of wasted render time for us.

So my question was, what is this magical "spherical harmonics" and how could it fit in our production. What I found through some research is the technique also can be achieved through what is known as a Diffuse convolution envmap. It's a bit technical of a read, but I know exactly how to implement it for us. Our lighting work flow would look something like this:

1. Figure out time of day and sun direction for film.
2. Bake Physical Sky enviornment hdri with this method and apply the correct filter to the image.
3. Attach "spherical harmonics" environment to ambient slot to a white lambert for testing.
4. Light the scene with other lights as we see fit, such as a rim lights and possibly fill lights.
5. Pipe the the "spherical harmonics" into the render passes and tweak further in composting.

Ta Da!! We just potentially shaved off minutes of every frame rendered at no sacrifice in quality!!!

So to conclude, we can implement a similar lighting approach used with Avatar that can save us a ton of time. Don't know about you, but for me, that is a very exciting prospect.

Talk About an Awesome Reference

Check out the Artwork of Lisa Black.

This stuff couldn't be more spot on to what I was thinking of.

Illustrating the Melting Pot of Ideas

The general consensus right now seems to be we are working on a fantasy film, but the question is how far removed from reality do we want to take it. When brainstorming about the style, our ideas were ranging from 9 to Saving Private Ryan. I went home afterwords thinking to myself "This is one massive spread, I better get on this ASAP!" Not an easy thing to try and resolve.

I want to make sure we develop a look that plays to the strengths of each member of our team creating the assets and ultimately is fun for everyone to work on. For example Alex Diaz, with his experience in the visualization industry is incredibly well suited to model, texture, and render real world environments and architecture. On the other hand Karen Ortiz has an incredible sense of design and staging that could really give our environments and characters a creative edge that serves the story. So this illustration is my attempt at merging everything together.


Even tried to work in a little back story with the set design, explaining how this unholy creature we came up with last meeting came into existence.

I want to also document right now that we did talk about the possibility of adding some slight humor, though it would not be our primary focus. The first Mummy and Indiana Jones movies come to mind where the characters are stuck in a serious situation, but how they react and the ridiculous nature of the conflict adds a hint of humor. I think the some of the ideas flowing right now could work with that sort of approach.

On a related note, here is an illustration I did of the main girl to plan out her clothing before doing another modeling pass. Again I'm trying to make her clothing reflect the world design, and look like it is from the same universe as the Soldier's uniform design as I envisioned it many months ago.

For comparison sake:

(The bow has been axed though)

Please give your comments, especially if you do not like the direction I have taken. Now is the time to change things, as more time goes by the amount of flexibility we have will start to diminish.

Edit: Oops, I spelled taxidermy wrong, oh well.

Rigging Talk

To start, Dana I would absolutely love to get in contact with Dan Philips. We could definitely use another hand with the rigging.

I've been talking to Jonathan Soto recently about what is the best course of action with the characters. One of the main areas we touched up on was cougar, what complications are left, and ultimately if he is worth pursuing. Right now the kinematics with the rig are working beautifully, but some of the deformations need more work, though I have been in way worse skinning situations in the past. We are looking into several tools that can help straighten it out, but we both agree we cannot get so involved developing the cougar rig that the human rigs suffer.

So I suggested we use some shortcuts where we can. I little while back I made the purchase of the education discount edition of the Setup Machine 2 (TSM2), and Face Machine with some money made off my Compositing Tutorial. At the time it was mainly to throw together a quick rig for deformation tests while I was modeling, but so far it is a perfect fit for the two human characters. Of course it would be a lot cooler if all of our rigs were built from the ground up, but I think doing this would give us the time to build a really nice and unique rig with the cougar that couldn't be done with a script like TSM2. Also from what I have gather from our meetings, people are really excited about animating an animal and want to see him in the story (a post on that hopefully soon).

Also I have other reasons for wanting to use TSM2. I am no rigger, I can't tell you the first thing about setting up a rolling heal, or FK/IK switch, but I am very well aquatinted with the hassles of skinning. In that regard, TSM2 is a miracle worker. It uses a smart weighting system that literally takes a week long job in Maya and turns it into an afternoon chore. And even after that week with Maya's default tools there is no guarantee it will look as good as TSM2.

The rig itself extremely stable and functional, but the skinning is easily the major selling point.

He is not quite finished yet, but I got this all done tonight. Also if we go this route, which I am heavily leaning towards, there are certain controls on the rig I think we should modify to make it easier for the animator.

The rig works independently of the software, so we don't need to purchase licenses on campus to use it in our production. Again I would love to hear everyones thoughts.

Overcomming Rendering and Compositng Hurdles

So I would like to take some time to further talk about the important work being done on the Rendering/Compositing front. I am developing us a very professional approach similar to what is used at studios like Luma Pictures in Santa Monica, and by one particular Cal State Fullerton Alumni working in the Special FX industry. It is not just the FX studios either, here is an example from this Disney Film.

I would like to start by having a look at the artwork and pass breakdown by CSUF and Gnomon grad Sanjay Chand.

Visit Sanjay Chand's Website Here
(Again this is a sample of Sanjay work. He is an industry professional and is not in any way affiliated with our project.)

Breaking down a scene into Render Passes gives us a tremendous amount of freedom with rendering, lighting, and making color correction to our film. In simple terms, it puts the control back into the artists hands, where it belongs. I have been lucky enough to get in contact with Sanjay and get some reassurance that the way I handle passes is indeed a professional practice. We are on the right track.

Unfortunately, I have the duty of fighting through the technical barriers to ensure our approach is as streamlined and efficient as possible. This is where the DeexRenderTool I mentioned in an earlier post comes into play. Maya's own Render Pass system is extremely limited in nature and has some major bugs, so I am helping out Damien Battielle, a script writer in France, develop passes for his tool that cover our needs and work around the problems.


*Warning! I am about to get technical!*

---

Here is a breakdown of the Current Progress
Orange Passes are my contributions

Necessary passes supported
-Color
-Lighting
-Specular
-Reflections
-Refractions
-Occlusion
-Fresnel
-Zdepth
+ many more

Necessary passes Not Yet Supported
-Non Final Gather Indirect Light
-Skin
-Blinn/Phong Specular
-Material Matte Passes (needed for control over individual scene elements)

Additional Issues to Overcome
- Currently there is a 10 Pass maximum.
- Interface that incorporates more than 10 passes.
- Assigning the skin pass to a material.
- Compiling special shaders need for Maya 2010 Mac OS X.


A lot of what I have developed for him to add is difficult to implement, but I have strong faith in Damien. Thus far he has been extremely active with updating the script. The rest of the work rests in his hands, though I'm trying to help as much as I can. I'll keep us posted as the tool develops over the winter.


To conclude, here is an image of what our compositing network will look like, only done in Toxik and not Fusion. All renders from previous posts use this same work flow with Toxik:


(The model is from 3DTotal's free model section I used for testing purposes)

---

Edit: What I coincidence,I mention Sanjay yesterday and today an interview with him is released on 3Dtotal:


Good read.

Some Color Food for Thought.

A few months back I discovered this magical chart while working on my entry for the 3D art competition, Dominance war.

Now granted there are no finite rules. Much of this chart is based on cultural significance. For instance in America we tend to associate red things with anger, but in China it is a color of celebration. Now I know what you might be thinking, "Emotional ties to a color? What nonsense is Chris talking about now?" Believe me, those were my thoughts exactly when I first saw this chart. Then I started looking around and paying closer attention. To my surprise, in many cases I found it strikingly accurate.

Here is just a taste of other artist works that make excellent use of these ideas:


by Anton Lavrushkin

Look at how the red, orange and yellow create excitement where as the blue gives off a more sorrowful mood.


by AOgnian Bonev

Notice how green transitions from a more sublime yellow color to a darker value and slight hue shift that invokes a fear, showing the darker side to the soldiers.


Again, I don't want to set strict rules with color and how it relates to our film. I am just suggesting this is another tool to go in our ever growing tool chest. Color, or lack there of, can be a powerful tool to help set the mood of our story. To conclude here is an example of that:


Before and After shots of the game Zelda Twight Princess. Color changed from a desaturated gray to the colorful effect on the right half way through the game's production. Huge change in mood.

Some Great Work from Karen Okstad!

These were created before the our meeting yesterday. Somehow I missed the email she sent me. I feel horrible for not knowing of them before we had our discussion. I really love what she has done with the girl character. In particular I really like the direction of the paint-overs of my model. I have decided the hair will probably work better without the hood, hair interacting with cloth could get very messy. Perhaps we should talk about more about the sort of environment she would live in. Or would you rather go with a more desolate design like the war pictures? Maybe we could have a sort of imaginative village in the style that still contains all the destruction we talked about in the meeting.

To me it is appealing because it fits in with the idea of the fantasy European nation I mentioned at our meeting. Also with some slight adjustments and the right colors, I could see the lead Male fitting in rather well.

Winter Break Priority List

First off here are some clay renders showing off the powerful lighting, shading, and composting system I am developing for us to use. We can do some amazingly accurate sky lighting fit for animations (no flickering), let's take advantage of it!




(Created using Maya and Toxik)

Here is a list of everything that we should aim to get done during our winter break. I prioritized what needs to get done first to ensure animation work can begin at the start of next semester.

High Priority
1. Solidify the character designs to fit the new story we have developed in our last meeting.
2. Make Modeling adjustments to characters accordingly. This could be very time intensive.
3. Finish Rigging work on the body of each character. This involves a lot of cleanup to the body rigs for characters and streamlining them.
4. Create Face Rigs.
5. Create new a storyboard to plan each Act out.
6. Complete work on the Deex Render Buffer Tool to ensure we have a solid technical pipeline in place before we get serious with texturing and shading.
7. Gather a GIANT library of reference images and footage to help establish the look of our film.
8. Further brainstorm about the story, making sure everything is exactly how we want it.

Medium Priority
1. Develop a solid world design.
2. Create an animatic possibly using a combination of hand drawn and a simple 3D representations for more complicated shots as needs be.
3. Create conceptual Art to to give us a clear goal to reach.
3. Develop the back story behind the characters considering their personalities and motivations.
4. Complete textures for Characters.
5. Create an animatic using a combination of hand drawn and a simple 3D representations for more complicated shots as needs be.
6. Put in simple lights that emphasize everything.
8. Block out the stage where the action will take place.

Lower Priority, but still very important.
1. Begin working on cloth simulations.
2. Create hair, work on simulations.
3. Resolve hair rendering issues.
4. Develop back story behind the characters. Consider the motivation behind each action.
5. Begin work on the environment.
6. Some sort of introduction training session with Toxik, how it relates to our pipeline, and composting in general.


I will probably be adding to this list the more I think up. As you can see, we have an awful lot to get done. I simply can't get it all done by myself. I'm going to focus on the character modeling, story, and rendering concerns. That alone is going to keep myself plenty busy. If you are available to help please comment and mention what you want to help with. Also say if you are available in January for a series of meetings at school with Dana Lamb about developing the story, environment, and as well as some training with Bill Stewart, the production manager on one of the short films last semester.

Learning from Other Shorts

Here are a wide variety of different shorts and game cinematics to look at for both inspiration and to learn from. I purposely made sure all of these examples are drastically different.


Video 1
*Warning, not in the best of taste*

Notice how throughout the entire short there is no mistaking what the story is about. Very expressive and successful acting as well.

Video 2
*Again, some suggestive themes but we're all Adults here.*

This Short is perhaps the exact opposite of the last. It uses a lot of abstract cinematography and story conventions to add interest. Also this is an excellent example of how powerful creative lighting can be.

Video 3

The first Game intro example. Keep in mind game intros never tell the full story, just an introduction. That said this example is particularly successful introducing the world and characters.

Video 4

Take note of how they are able to establish a relationship. Not an easy thing to do in such a short amount of time. Stylistically, this is another strong example to look at for inspiration.

Video 5
This is a good example of what not to do in terms of story development. If we try to turn our short into a giant epic feature contained in a few minutes, the story and characters will escape from us. That said, what this cinematic does well that is of interest to us how it deals with having more realistic characters and mocap animations. Notice all the fast actions. They help hide the shortcomings.


One of the biggest challenges we have is working with more lifelike characters. From what I've discovered watching these shorts, we need to get very creative with our staging, lighting, and pacing to make sure our audience is not focused on pinpointing the inaccuracies. Yankie Girl is a perfect example of this. These are all things to consider for the next time we brainstorm for the story. Please comment with your opinions. Also do you think this kind of post is at all helpful?

An Overview of the Challenges of Creating a 3D Short

Seeing how we are about to embark on this journey, I thought now would be a good time to put ourselves in check. With 3D the sky is the limit, unfortunately we can also spend an eternity trying to get there.

I want to take this moment to introduce everyone to some of the complications involved with creating a 3D film. Pigeon Impossible is a 3D short 5 years in the making that was released this year. The lead creator documented the lessons he learned the hard way in production through series of video podcasts. There is a lot of invaluable lessons to be learned here:

Grab a bag of popcorn. Let's have a look:

#001 - The Production Triangle

#002 - Writing is Rewriting

#003 - Multipass Lighting

#004 - Lighting Hacks

#005 - Lighting the Eyes

#006 - Camera Projections

#007 - RGB Passes

#009 - FK Arm Setup

#010 - IK Blending

#011 - Cloth Cheats

#012 - Follow Through

#013 - Smear Campaign

#014 - Why It Took 5 Years

#015 - Anticipation

#016 - Animating The Pigeon

#017 - Pigeon Rig

#018 - Single Pass Lighting

#019 - Rendering

There are a couple of common themes throughout these videos. First 3D can be excessively time intensive and unforgiving if the storyboard, characters, and workflow are not well developed in pre-production. We could easily spend 5x longer working on the short than it should take (watch #014). I can't emphasize enough, we cannot begin until not one, not two, but all three of these conditions are met!

Luckily for us we have an excellent education in this area as CSUF students that will help us nail the story and characters down. We still have lots of work to be done in these areas, but there are no big technical problems in our way. What we have to be careful of is having a proper workflow in place, which is all to often the the forgotten factor. The "we will figure it out later, lets move onto the fun stuff" train of thought will only get us into trouble.

Why is developing a solid workflow first so important? Well it is simple, having the knowledge and tools needed to technically accomplish the 3D film prior to production ensures we never have to do the same task twice, saving huge amounts of time fixing problems and troubleshooting. That is of course is in addition to having a good game plan.

But how do we stand? Do we have the tools needed as well as the knowledge to achieve this? The answer yes... and also no, not quite yet. The biggest area we are currently lacking in is the Rendering, Lighting, and Compositing development. Just ask the other 3D Short teams at CSUF, both the Meteor Team and Buki Team will tell you tell you this is perhaps the biggest technical road block they've hit. Look at how important Compositing, Rendering, and lighting it is to Pigeon Impossible and any Studio 3D Animation for that matter, that is a gigantic chunk missing.

It is not all bad news though. I'm doing something about it. Actually so did our school and teachers by partnering with Autodesk to bring us Maya 2010. We now have in Industry grade compositor packaged with Maya called Toxik. After Effects is a great application for motion graphics and adding post effects, but it was not built from the ground up to handle the duties needed for a full-scale 3D production. Toxik is.

So that's one major potential brick wall we have avoided. Phew, so what's left? How about a proper integration with Mental Ray, our rendering engine? This is no small task, in fact it's a very monumental one. I'm currently involved with the creation of a script to do just that, collaborating with a programer from France. I am slaving to make sure it is receives all the functionality we need. Doing this on the side in addition to everything else we need to finish before the spring semester starts. When it is done, we will actually have a better lighting, rendering, and compositing system to work with than Pigeon Impossible.

I apologize if this is turned a little technical for some of you following, unfortunately it is a necessary evil behind the nature of creating a 3D film.

Welcome to the Home of the Next Cal State Fullerton 3D Animation Production!

The name is tentative for now. Feel free to email me at Crispy4004@gmail.com if you have any questions or comments.