Tuesday, June 30, 2009

thumb-boards

While working at Big Idea I had the opportunity to work with Tod Carter- an amazingly bright and talented story artist. When he was working out a sequence for a video title Tod often used simple little thumbnail sketches. He'd cram 10 or more shots onto a single piece of paper, no formal board templates or anything. While these were intended as first pass rough sketches, more than a few times the thumb boards were so clear and functional they just scanned those, blew them up and cut them into the edit- no full clean up board needed. I'm sure it helped story development to move faster (saving money & time- two very important things in the world of lower budgets), but whether I was animating, supervising or directing I never felt like we were getting short-changed in the process. The thumbs were clear and they gave us plenty to work with.

That made a big impact on me because I'm always intimidated by the act of drawing. I can draw all day long (and even do OK at it), but the moment the drawing becomes "serious" or needs to be on par with some arbitrary skill level I just shut down. So thumb-boards work perfectly for me. I've boarded a number of films and projects the same way over the years. Quick, rough and simple. Like everything else about this project, immediacy, speed and simplicity are the goals.

As you can see, the drawing is utterly lacking professional skill. But they don't need to be fancy or amazing. I'm not out to impress anybody. The drawings just need to be clear. 
  
  
 

I've been trying to keep my Twitter account updated with things as I move along.

Saturday, June 27, 2009

Schmitty

 
  
Otto's roommate, Schmitty, and his most prized worldly possession-- his 1974 vintage arm chair.

Thursday, June 25, 2009

Tweet!

Well, seeing as I do stuff online I feel somewhat compelled to have a Twitter account. So on the sidebar you'll see my Twitter updates, and you can follow me there. I'm gonna try to keep it updated with progress notes on my various up and coming projects. Other people have production blogs for their projects (plogs), so this will be my production twitter. Pwitter? Kinda has a ring to it.

Anyhow, I'm currently working on making another short animation using Otto, along with a new character. It's not musical, but will have dialog and such. It's more of a real look at the characters' personalities. The plan (for now, at least) is to make several short-shorts of 1 minute or less rather than one long short. Longer shorts (4+ minutes) have a kind of intimidating heaviness to them. They take a long time to complete and often need to justify their length by being a bit more complex to hold one's attention. So the inner unspoken fear is that it will be a year or three of my life wasted if the short isn't absolutely perfect. The result is paralysis by analysis (aka.: The Lango Secret Joys of Myopia Syndrome) So I've written up a bunch of short shorts instead. They seem less important that way. I've read that some artists have a similar fear of their expensive fancy Moleskine sketchbooks. The books are so nice and cool that they are intimidated by them, so they avoid sketching in them and prefer to sketch on notepads or other cheap media. I can definitely relate with that. I once had a really nice sketchbook. I did one page in it, very badly drawn and terrible all around. Never touched it again. My doodles are still mostly on the back of church bulletins, Post It Notes or random pieces of paper lying around. So these shorter shorts are gonna aim for that 'back of the envelope' spirit. All part of the unpolished immediacy vibe I've been striving to grab lately. In many ways Twitter updates are like this, too. Short, fast, less drawn out ordeals. Good for quick stuff.

I'm sensing a trend here.

All the same, follow me if you dare. Or care. :)

Tuesday, June 23, 2009

Funky -- rigging details

Thanks to everybody for their great feedback on the funkiest man alive clip. It's cool to see that it resonates with people. The great fedback has emboldened me to continue with plans to do more stuff along this line.

OK, as promised here are some details about how I created the funky animation. First up I want to explain the rigging. It will help put the animation process in context. None of this stuff is really revolutionary. It's just using various old techniques in a way that suits the particular style of design and motion I am aiming for. I doubt this exact combination of techniques would work on more "normal" CG characters or styles. The demands of those styles necessitate a different approach. Ultimately any good rigger understands that the end defines the means. For me the driving principle is shape control. The second priority is ease and speed of animation. I want to have fun while doing this, not die of torture. Both the style of animation, and the manner in which I create the animation, allow me more freedom in "breaking the rig" because I never really have to worry about showing it breaking or coming back together to the "correct" form in a literal manner. It can be broken for a frame or two, then be 'normal' again and the audience doesn't ever see the transformation. Without that nasty business exposed the audience just accepts the image for what it is.

The Body
At first I tried to rig the character using a hybrid veggietales approach. With the VeggieTales characters we used lattices with the various lattice point rows clustered together. You can see the basic idea in this image...


Properly networked with dozens of math nodes under the hood (aka: "magic") for automatic counter adjustment the basic control can be boiled down to a few controllers. For Veggies you could literally do 75% of your primary acting animation with one controller for the head...

There are no bones in a VT rig. I figured since Otto (and previous incarnations of him) had a simple gumdrop shaped body, perhaps this would be a good way to rig up the main body. The attractive part was the nice squish-able volume and simplicity of using one control to get a lot done. Well, that and I hate, hate , HATE defining skin weights for joints. So if I could do it with a lattice, all the better. Sounds great in theory. Not so much in practice. I wanted to do things with these characters that they never dreamt of doing with Veggies. The primary downfall was that it was hard to get good isolation of control over a specific region of the torso without it affecting neighboring portions too much. Say I wanted to push the belly out to the left while leaving most everything else where it was and having a good flow of the mesh as it deformed along. The deforms just didn't work right. I tried about 6 or 7 different variations (each more complicated and automatedly counter-forced than the one preceding it) before I scrapped the idea completely and just went with simple joints. The key to this was a GREAT plug-in by Kickstand Labs called Stretch Mesh. It's amazing. Stunning. Life altering. Magnanimous. Scrumptious. Punctual. I cannot use enough superlatives to describe how awesome this plug-in is. Go check it out at their website and see the examples. It made bones a simple pain free solution. They have my undying love.

OK, beside me having a serious man-crush on the Kickstand dudes, I ended up with joints for the body. There aren't many joints in the body, but each one has a control for it. The control can adjust position, rotation and scale. So the body is pretty much totally "breakable" at any point along the way- something that ended up being easier to pull off using bones instead of a lattice. All the better to get finer control over silhouettes and shapes.



The Limbs
The "hybrid" part of my original hybrid-Veggie approach were of course the limbs. I rigged those pretty much straight up. I attached them to the body using a MEL script called djRivet. It uses the hair system in Maya to stick things to a mesh. So the arms & legs are stuck on the body like they're plugged into a hair follicle in the skin. Weird, but you riggers know what that means. That part works like a charm. Always did. There were some arrangement & parenting issues so that stuff would have the proper following properties, but nothing major. A simple IK switch between chest space and world space is enough.
Speaking of IK/FK-- I have evolved to the point where all I use is IK arms and legs no matter what. I'm just comfortable that way. It's much easier & faster to adjust the arc or path of a hand in motion when all you have to worry about is one control object- the hand IK control. Adjust it to clean up the arc, done. To fix a hand arc or whatever in FK you need to find the right combination of adjustments from the hips, thru the chest, down the arm- in a rig like Otto that can be 5 or 6 controllers. So I skip the FK and stick with IK. I've worked this way for years now. Life is good.

The arms and legs follow with the body, but the shoulders & hips behave like they have their own IK and can be pulled off. This allows them to be positioned as needed to maintain the proper shapes in motion.

The bendy deforms on the arms and legs are pretty simple. I originally wanted to see if I could live without them (less controllers = faster animation), but I needed the shape control, so I added them in. As per standard practice I break the rigging into two distinct functions: Control and deformation. Control is your garden variety IK system on a joint chain- it's how you articulate the character. But this joint chain is not what deforms the mesh. Deformation is bound to a completely different set of joints. These deformer joints are then constrained to follow along with the control joints. So when I move the IK controls, the skinned joints pull the mesh along. Again, pretty standard stuff. This allows the skin to be deformed in a macro sense (with the hand or foot IK control), and on a micro level with extra controls that pull the skin joints out and away from the control ones. Not cutting edge, but it works.
 
 

The hands and feet have some extra bend deformers on them that let me kinda shape them more elegantly with a cleaner line than the joints system alone. They're nice simple attributes on the hand or foot control. This really comes in handy for the fingers. For simple hand stuff it's easier to adjust a few attributes on the hand IK control than the shape the fingers joint by joint.


The Face
The face bits are modular- they're not modeled into the body mesh, or even into a single head mesh. The eyes are separate, the nose is seperate and the mustahce is seperate. As are the teeth and tongue and ears. They're stuck on the body using the same rivet/hair follicle approach as the arms and legs. They're all able to be pulled off the body or re-arranged on the mesh to suit the shape needed. Again, this isn't rocket science so much as it's an attempt to make the design fit the end goal animation and visual style. I pretty much stole a page right out of Jim Henson's playbook.


As for the under the hood stuff for the face, it's really simple. The controls drive the position of clusters on a seperate copy of the body mesh. These clusters are simple because, once again, the Stretch Mesh plug in makes them blend so smooth and easy that I don't need to set up some complicated system of sub clusters that are blended between various master clusters, blah, blah, blah. Stretch Mesh. Have I told you how awesome it is? Anyhow, the result of this "off to the side" face is then applied back onto the main body via a blendShape morph. The reason to keep the clusters and face deform stuff seperate is to keep things stable. Clusters are infamous for double transforms, for having pole alignment problems when their parent or contrainers are rotated, etc. Better to leave them parked off in a hidden group where they stay nice and still while the results are piped back onto the main body.

The facial controls can either ride on the face, or with the flip of a switch be pulled off the face into a more typical face control panel. For close up shots the on-the-face method is great cuz it's like sculpting the expression. For shots further away the panel version is easier to work with.
 
 


The Triggers
OK, the last little bit of rigging fun for today was my trigger set up. Many folks are familiar with Hamish McKenzie's zooTools suit of MEL scripts. Hamish is an insanely bright guy- and a real cool one, too. He's currently performing signs and wonders for Valve. His trigger tool is really spiffy for turning objects into buttons which can perform a variety of functions quickly. So I created a series of buttons for each character that set keys, hide visibility of controls, select or reset various controller groups (body, mouth, eyes, all). I have a quick little script that hooks these to the frame mask of my own home made camera tool. This way I can quickly click on a button to perform various functions quickly and it stay out of the way, but always right where I can get it. Even if the camera moves these buttons stay stuck to it.

OK, that's it. Like I said, nothing here is ground breaking or mind blowing. I'm just using whatever technique best fits the desired end result.
More on the animation process later.

Wednesday, June 17, 2009

Funky!

Otto's coming out party, with some back-up help from Garfo.



Here's a high-res QT version for you to download and frame through if you feel so inclined. There are some interesting things going on here. Smears, mixed 1's, 2's and even 3's, lots of kinetic transitions that aren't precisely literal, static parts mixed with moving parts, etc.

I'm really pleased with the way this has turned out. The style of animation is right in that sweet spot I've been aiming to hit for the last few years. I like how the energy comes through the rough edges. The idea is to let the viewer sense the hand of the animator showing through, to let the immediacy of the creation moment remain perceptible, even if it's not perfect.

I worked on this in my spare time over the last 3 months. Otto's rig underwent about 3 or 4 major revisions/rebuilds along the way. I'd start animating, hit a roadblock or find something that just didn't work the way I wanted, then I'd back track, gut his rig and rebuild it, move forward, etc. For example, the first singing Otto part is from a completely different rig than the second part. When all was said and done I finally ended up with a rig that's really solid and fun to work with, as well as a fun little proof of concept on the animation style.
If folks are interested I can post info detailing the animation approach. 

Tuesday, June 09, 2009

Just for laughs

Michael Sporn posted a copy of a great, great article on his blog today. Working for Walter Lantz in the 1930's.. writen by Leo Salkin. I absolutely love stuff like this. An excerpt...
Out of the nonsense Walt (Lantz) would select the stuff that could be made into a film: comedy bits, funny lines, gags. The cartoons of that period were still being concocted and assembled in much the same way Mack Sennett had made live-action comedies: “Charlie, there’s some kid auto races going on down in Venice—grab a cameraman, go down there and see if you can come up with something funny.” Or, “Hey! They just drained Echo Park Lake, it’s all mud, that oughta be funny as hell!” That’s what we did. We took a locale, an occupation, a situation, or the basic premise of a popular feature and did a lot of gags, strung them together, built in a chase, and got out in under seven minutes.
There was no market testing, or who is our target audience, or will a sponsor buy this? Come on, fellas, it’s just comedy. Get some laughs. Walter was the judge. What he thought was funny was what got up on the screen.
Go read the rest.

ps: this is why you don't write off good blogs, even if they fry your biscuits now and then

Made me laugh

Saturday, June 06, 2009

Ahh, memories...

Of my life as a departmental manager in an animation studio...

Tuesday, June 02, 2009

At least I'm not alone

It's nice to know I wasn't the only person underwhelmed by Up. In addition to some commenters here, Michael Sporn and Michael Barrier both were fairly critical of the film's faults. Both gentlemen have a long history of accomplishment, as well as critical thought and analysis of animation. Each have strong opinions, as well. Some say that puts them into the crumudgeon category. I wouldn't agree with that label. They are definitely fans and want animation to be great- but they have their standards and they don't waver from those. I certainly don't always agree with their views. I'd be afraid of the person that I always agreed with. Yet I find their blogs to be required reading, especially if you wouldn't normally share their opinions. Always seek out different points of view- it makes you a more well rounded person and keeps you from becoming intellectually inbred. It also makes you less likely to be used as a tool.

I'm thankful for the chance to have a different perspective. I'm finding that my own views on animation -- especially the big studio feature film variety -- have evolved a lot more now that I'm not in the belly of the animation studio biz like I used to be. That culture can be claustrophobically limited and tends to be very self re-inforcing (a dangerous hallmark of any brand of fanatacism, by the way). You always end up working with the same people, or people who know people, again and again. Everybody is one level removed from everybody else. So there is a strong, unspoken pressure to not rock the boat, to be as vanilla as possible. Keep potentially caustic opinions to yourself because you never know what thin skinned person (or friend of said person) you might offend by expressing an original critical thought. Plus there is a tendency to elevate the profession of character animation to levels of importance and significance that are, frankly, just silly. One artifact of this is to navel gaze about the smallest of details, which is tiresome to me and misses the point. There is a noble benefit to self improvement and refinement of one's craft, but do we really need to study video reference of the tongue in slow motion in order to convincingly animate an "L" sound? Really? So often CG film animation misses the forest for the veins in the leaves on the twigs of the branches of the trees.

At this point I don't feel a need to work in the big animated film biz. Been there, done that, got the receipts to prove it. I'm just in a different place. Our family loves where we live and I thoroughly enjoy teaching and tinkering with my own experiments in animation. I don't worry as much about somebody getting their tighty-whiteys in a twist because of something I have to say. I think the very fact that Barrier and Sporn have no stated desire to work in the big studio animation film biz is significant. It's good to have informed outsiders chiming in. Perhaps there are benefits to having informed former insiders who are now almost outsiders chiming in, too.

Monday, June 01, 2009

my review of Pixar's "Up"

Spoilers ahead, so skip this if you want to save the experience....

I'm a sucker for emotional films. I'm a softie romantic. As a kid I sobbed my way through Snoopy, Come Home. I sniffle at every single chick flick I've ever seen. Schindler's List and Saving Private Ryan devastated me for weeks. Literally. About Schmidt left an impact crater on my life that still moves me (It's a great film that shares a lot with Up- it's about a widower going on an adventure to find himself after his wife passes away, establishing a relationship with a young boy along the way). So films usually have a powerful emotional affect on me. But for some reason this one didn't hold. I think I must be the only human being to have experienced this (based on all the rapturous and breathless reviews I've read so far), but Up never grabbed me. It's not like I came in wanting to not like the film. I really liked Pete Docter's first film, Monsters Inc. Emotionally I found it very rewarding when I first saw it, and the scene at the end with Boo still gets me when I watch it (like I said- I'm a softie). And I'd heard how Up was Miyasaki like in many ways, so I was looking forward to that vibe. And by all accounts Pete's a great guy, too. So I was expecting to like this film. But something happened along the way to Act 2.
I got left behind.

The 4 minute Ellie/Carl life montage felt superficial, cliche'd and- frankly- rushed. It was like "Look here's the character you're supposed to care about. See how sad it is? Let that music roll over you. You want this man to be happy, right? Good. Now- let's go fly that house!" Their relationship had no real drama. Unfortunate things & disappointments happened in their life, but none of them came from their relationship. The deepest pain and the greatest joy in life come from those we love the most. What if their childlessness caused real tension in their relationship- a wounding that needed to be redeemed and resolved? Couples who deal with sterility don't sail through very cleanly with a "There, there" pat on the hand and a supportive smile. There's wounding that goes on. You could have put that in there -- even in a montage. That sincerity would make me connect with Carl & Ellie. Remember how powerful the argument scene in The Incredibles was? That sold me on those people. Film is about expressing something of the human story in each of us. Nobody lives Carl & Ellie's story. Nobody. And that's why I felt it was a hollow moment. All that aside, even if I accepted their blissful relationship at face value, I couldn't get my heart around Carl. Carl seemed merely a spectator in Ellie's life. From the moment we see her as a child she was a catalyst character, to me Carl was a cypher. Ellie lived, Carl watched and reacted. Even the most touching thing any character ever did in Acts 2 & 3 were the words Ellie wrote in her book to Carl. I found it difficult to engage emotionally with such a character. If you're going to put the character development efforts on fast forward, then you better have some meat on that bone. The blue balloon & cross your heart symbols were just too "on-the-nose" for me. If this were done in live action- shot for shot, action for action with the same memes and symbols it would be panned as maudlin and ham-handed. If performed by Will Farrell it would have been pure comedy. In Up apparently it's the height of filmmaking, I guess. For me it was just too obvious, too by-the-numbers. Like a lot of animated filmmaking done today it seemed formulaic to the core. Like I said, I think I'm the only person who experienced the film this way. Maybe I need to see a doctor.

All that aside, I was surprised by this one discovery. I found that if the film didn't grab you in that montage, the rest of it didn't hold up. That montage was an all-or-nothing moment. If you're not in by the end of that, you're out for the rest of the film. Most folks were hooked in that early montage, and that emotional investment papered over a lot of weaknesses in the film. Which is really how films are supposed to work, because every film has holes that need to be papered over by the audience. It's called the suspension of disbelief. I guess I was Up's kryptonite.

Since I wasn't emotionally engaged I found myself just watching the film without the tears I shed in Act 1 operating as a lens that affected how I saw everything that followed. I found it to be an interesting experience to see a film from such an objective viewpoint. As a result I noticed a number of things that just didn't ring true. There was a big disconnect in the film regarding pain, danger and mortality. The rules weren't the same for every moment. In one moment wounding pain is real and meant to elicit sympathy or have fear for the welfare of the character, in another it's a gag meant to make you laugh and in other moments things that should have caused significant physical pain and injury had no effect at all on characters. Normally you'd just roll with it in animation, but they purposefully chose to show wounds with blood, welts and cuts, along with the frailty of age. That's taking things up a notch. These mean things when you are trying to establish the rules for how the audience interprets your world. This inconsistency kept me from knowing what to feel about injurious possibilities. I had to wait to have them interpreted for me. In storytelling we call those "bumps along the ride". Things that toss the audience out of the moment.

A significant item that held me back from buying into Carl's story as it unfolded was that I had a hard time buying into Carl, the old man. After the first act he stopped moving and acting like a slow, creaky, feeble 78 year old widower who needs a walking cane. He became a physically strong, high endurance, highly flexible action hero who can take a beating and keep on ticking- brushing off any ill effects from a fall or being trampled, tossed and otherwise roughly handled. I get it- his adventure revitalized him. But there's a difference between a revitalized old man and a strong young man. For all the comparisons to Miyasaki that people have for Up, this is where Miyasaki handles things way better. His frail characters live their adventures in their frailty. The girl transformed into an old woman in Howl's Moving Castle moved and acted like an old woman for the majority of the movie (until she started transforming back into her younger self, that is). The children in My Neighbor Totoro were limited children the whole film, experiencing their world as children, not action heroes. Up dispenses with human frailty in exchange for action set pieces. How many hanging one hand grabs to save oneself from falling to one's doom can a 78 year be expected to pull off? How many times can they be thrown down from heights and not break a hip? (my 76 year old mother in law broke her hip falling off a bed. And she wasn't some wilting flower of a woman, physically, either). And that's only one example. There were scores of them. I'm not the kind of person who picks nits in movies over physical impossibilities (ie: the entire Michael Bay filmography), but this is the guts of Character Animation, folks. Sure the moves exhibited a form of technical polish and solidity, but they were hollow of meaning- they weren't believable in any way because they were not true to an old man in any way. It wasn't character animation (ie: within character, expressing a unique person in every way)- it was movement to keep up with the gymnastics of adventure. Aside from a spare gag or two jammed in to occasionally remind us of his age, Carl as an elderly man just didn't ring true for the last 3/4ths of the film. If you want me to buy into a character's story then I need to buy into the character's being. This wasn't a story about an old man on an adventure saddled with the limitations of his old-man-ish-ness. That would be an interesting and intriguing film if you ask me. Instead this was an action adventure flick with a strong athletic character in an old man costume who occasionally acted old for effect. Stanislovsky's rolling in his grave. Isn't this why we pan CG 'performance capture' films? Some may say I'm quibbling. Maybe I am. But that really kept knocking me out time and time again. And it wasn't just Carl who had this problem. The antagonist Muntz behaved the exact same way. Set him up as an ancient old man (20+ years older than even Carl himself), have him move like Errol Flynn in his prime. Russel was little better in his 9 year old boy-ish-ness.  I will say that there was one genius scene with Russel where he's complaining about being tired of walking like a real 9 year old boy would. It was the one moment in the film that I felt like a real character existed there. I'm sorry, but for all the raves over how great Pixar's character animation is, I found this whole thing astonishing.

I thought the most honest character in the film was Dug. His mind was simple and he was really just supposed to be a comedy relief character, but everything he did and said and emoted felt like it came from an honest place inside of him. He was a dog, he thought like a dog, had dog feelings and motivations and reactions and movements. When Dug was on the screen everything about him was sincere. His scene on the porch after the house lifts off again was amazing. He wasn't just a dog, but a real character. The animators did a magical job of expressing him as a character. In other words, I bought Dug's part in the story because I believed in Dug, even though the filmmakers didn't seem to try very hard with him. But I did not buy Carl's or Russel's story, despite all the efforts to make me choke up over them.

Artistically, I'll have more thoughts on the overall aesthetic approach and how it related to the story in another post. Quite often Up was a visually stunning film.

Anyhow, let the flames begin. I expect no shortage of scorn for posting this. *sigh*

Post Script: It's been two days since I watched Up. I haven't thought about it one bit since I wrote this review on Friday. The film continues to have had no impact on my life. I can say this is not usual for me. For whatever that's worth.