Entry and comments

Can be found here.

I violated the 180 degree rule, true. I was just thinking about the point of interest while planning the camera so I didn't see how the characters switched sides on screen. And while absorbed into the idea, it was even hard to see it can be confusing.

I went through 202 short animations in two days. And it was a good learning progress as well. While doing it, I started to see what's the exact difference between good and not so good animations. I even think it could be a good exercise to do the evaluating every month. The shorts are so short that one can easily evaluate and dig in to the piece thoughtfully.

Now when thinking about it, I could've not done anything differently. It wasn't about the lack of time, that I had enough. I just didn't really know how to proceed. Next time I'll really dig into the blocking, that seems to be the key to a good, planned animation. Starting from storykeys, then breakdowns and then the inbetweens.. Recently I also saw a clip where an animator acted out the scenes and the animation was done exactly to the reference. That also seems like a proper thing to do. Webcam, script and some space. That should do it for a while.

Entry done, 324+ hours done with Maya...

All objectives done, at last. Here is the end result for the 11 second club.

In the end I got some good feedback from Jouko, and still yesterday made some changes to the animation. 

I found out that it's a very useful thing to observe two sequential versions side by side. In the beginning I didn't do this. Now I feel that it was hard to evaluate what direction the changes have been taking the animation. It might sound like an obvious thing to do while planning and evaluating, but only now I really acknowledged the power of comparison between two versions. When I look at them one after the other, I see changes and different animation. Naturally. But only now I have some kind of a grip to evaluation: Which one is working better to my original idea? I have to compare between two almost similar clips and it's so much easier to do this, than look at one clip and think: "Is this to the right direction?"

There's something in it, for sure. I still have to think about this a bit and make some conclusions.

Oh but it was a tough month. I had to learn a lot of new stuff, lipsync in the top. I managed somehow, and my workflow wasn't definitely the best, but I managed. All objectives ticked.

Some times I was totally fed up with keyframes and the adjusting, endless adjusting, fine tuning. At times I swore to myself I'm not going to touch 3d stuff no more, ever. What a fucking waste of time, I thought. Why should I ever use this way of animating, I want to draw!

Yea.. That's how it felt at times. I didn't really have a mentor to motivate me, or guide me. I was my own boss and now I'm really glad I managed to pull this off to the end. Very happy.

Now when I've cooled down a bit, 3d does feel a lot better. :) I still have some things I want to learn, grabbing and doing some pushing experiments. Also parenting exercises with zvmasterparenting tool, that I have to learn as well. An when I go forward I have to get into modelling a bit more, rigging, texturing. . Taking stuff to Mudbox and back, all that sort..

Finally

It's been a nice experiment. It's motivating to know I can learn a new program on my own. Maya has been my objective for a long time and now I have a feel, that I am experienced enough to use it in my projects. Though in the introduction I said I try to get from 1 to 100, I think I got to .. maybe 40? 45 or something. While learning I always noticed I have more stuff to learn so the boundaries pushed further as I advanced.

Anyway, well done, Kari.
(Special thanks to the autoreference-bots, for keeping my blog alive.)

5 days to deadline

I've been now working 112 hours with the 11 second club challenge. Most work that I see in the 11 second club archives have been taking 20 - 50 hours to finish. So now I've been working more than double than the average. And is the result really good then? No, not really, when I compare it with the stuff I see in the forums. What I see in the forums, a lot of animations have a lot of dynamic and big movements. I don't. Probably because I didn't know how to block or plan the blocking properly.

At one point, I got an advice to try to make the turn of the character happen more fluently. But I couldn't. I wanted to, but there were too many keyframes connected to each other. Also used the gimbal controls to turn the character, which caused too much confusion. I could say, I went forward in the dark. Also when touching the hip control, the character popped into a very far position... So I tried to use the rest of the body to smoothen out the motion.



Now I know I have to plan the scene more carefully. Just that everything times nicely. And then I don't have to go back to fix big motions.

I'll set the whole thing aside for a while and check the animation in couple of days. If then, I don't find anything to fix, I'll post it to the forum and start celebrating new year.

WIP

I've been working now with the shot for a long time. A lot of changes have been made, and always I don't even know if the shot just got better or worse. Evaluating the piece during the same day is pretty hard, because you really can't see like you would see it the first time. So I've noticed the first preview of the day is the most important. What pops, what doesn't seem right? You see things more clearly while you see it in the beginning of the day.

It would be a very powerful way of working, if I could always empty my cache. From my head. Maybe then I could see it like the first time. At the moment I know only two ways of doing this. Sleeping and heavy drinking. And because you can't really do both at the same time, proves that they are both very powerful methods.

I found this script called ZVparenting master. It's made for parenting objects easily, so you can select two objects and parent them on the go. I don't yet know how to do it in the first place, without the script even, but this could be useful. I found it because I wanted to make a easy connection with the hands. But I never used it because I'm using referenced models, and it makes things a bit harder.

11's WIP

This is what i've done at the moment. I think it's getting easier all the time, I'm getting faster with selecting things, keyframes with shift+w and e, works.

Had to use gimbal controls with controlling the hand, otherwise it did a weird rotation for the hand. Otherwise no big problems.

I had to go back to the reference character because I colored his teeth to reddish lambert aswell. Adjusting them worked nicely, and it updated the changes when I went back to my primary scene. Which made me feel like I'm starting to get a good grip of the situation I put myself in.

lipsync and keyframes

I've been working now with the contest shot, slowly but surely. It's going forward but some days I'm having a hard time to find the goals for the day. One day I bumbed into the area of lipsync, and immediately lost my direction. I've overlooked it a bit in the past and now when I tried it, I noticed it's not easy. I have to check on some blend shape tutorials, and try to get a grip on the matter.

I found this tutorial linked in a lot of places.

He is using a MEL-code called "pose to shelf". With this you can make a facial pose, lets say a shape of mouth making an "AA" sound. Then with pressing the pose2shelf button he made buttons from the poses to his shelf. Very convienient. Seems simple and all, but I just couldn't make it work properly. I made poses, but while making other poses, the earlier poses gets messed up.. I still have to work on this, make more tests, tests and tests.

With this tool, I wouldn't have to make all the mouth shapes from the beginning for every syllable. And I guess, this is the way to do it.


And another weird thing with the keyframes. When on graph editor, I select keys and shift drag them left or right. It works properly if the graph is large enough to show a lot of change in the value. But if it's not large enough, the left or right pan resets the graph into a straight line. Which is .. a funny feature. I guess I just have to watch out for it.

Modifying colors of Norman

I had to figure out how to do this. In my scene there is going to be 2 normans and I don't want them to mix up if I change the camera angle.

I found this thread from the 11 second club.

http://www.11secondclub.com/forum/viewtopic.php?id=6615&p=1

It also clearly said that referencing characters instead of importing is the way to go. Apparently the scene gets lighter and it's easier to modify the characters afterwards.

So thats what I did, and it seems to be working for now.

2 referenced and modified Normans

Also made separate controls for Normans. Should make life easier.