Aug
05

Magic Mandcandy – Can You Finish It?

Posted in Random Fluff by Nathan

Hey everyone.

Magic Mancandy I’ve done a little animation test using Bassam’s Mancandy rig. This is part of a little experiment for community participation (more on that in a second).

Click the the image to view the animation.

You will probably notice a couple of things about it:

  1. It’s a cheap rip-off of this Pixar animation.
  2. It’s stiff and lacks life, sometimes even pausing completely.

Let us ignore the first point for the sake of my reputation, and simply focus on the second.

The reason it’s stiff and whatnot is because it isn’t completed animation. In fact, it’s only a block! You see, in pose to pose animation there are basically two stages: blocking and finishing.

Blocking is the the stage where you lay in your poses and timing. You might have a couple breakdown poses if they’re important, and you might toss in a few important extras (like the hand wave I did), but by and large this stage is just straight poses and timing. This is where the mancandy animation is right now.

Finishing is the stage after blocking where you add in all those things that… well, finish the animation. Breakdowns, arcs, overlapping action, slow-in slow-out, etc. All those things that make an animation complete. Not that you don’t think about this during the block (you should), but finishing is where you actually lay them in.

So what’s this community participation thing I mentioned?

Well, I’ve been toying with the idea of having people outside the core Peach team do some of the finishing for Peach’s animation. The Peach team would do the block (like my mancandy animation), but then for some shots other people would take over for finishing. This would allow us to turn out more animation of a higher quality.

Please keep in mind that this is just an idea. Currently there are no plans to actually do this. I’m just toying with the idea.

But as an experiment toward this end, I’m curious to see what people can do with my “Magic Mancandy” block.

Here is the block: magic_mancandy.zip

If you would like to try doing the finishing for this, then go for it! And if you do, please render it either as a full render or OpenGL preview from the camera angle provided, and post a link here to your movie file.

I look forward to seeing what people come up with!

45 Responses

  1. Jonatas Kerr de Oliveira Writes:

    That´s very interesting! I´ll try it!
    I will be back! (with some polished animation I hope!) hehehe

  2. Leonardo Writes:

    Very nice idea! I’ll try it too…

  3. Tobias Kummer Writes:

    I’ll also give it a try if I find some time to do so… But I think that’s a very cool idea!

    Greets,
    Tobi

  4. Ken Writes:

    This is a very good idea – however should there not be some rules. Do you work between all the poses laid down or do you have the freedom to change something that doesn’t feel right to you. Say at frame 33 you feel the pause isn’t long enough – if you make this twice as long it will obviously have a chain reaction on the remainder of the animation.
    My guess is you don’t change anything, but do you have the ability to change small things if it’s going to help to bring Mr Candy to life.

  5. Dion Moult Writes:

    I can’t try it (still need to learn a lot more) but I’m posting this to say a HUGE THANK YOU for the pose-to-pose animation link. That was coincidentally just after I decided to do some basic character animation test. Hooray for project Peach. Most people don’t know this, but I myself am working on a project similar, but it also contains a fully documented learning process. And this is one of the things that helps it! Again, thanks!

  6. Steren Writes:

    I recently suggested this kind of idea to Ton, and I would be proud to contribute.

    We could even imagine an online call for content on a wiki, where the team asks for small help, such as objects, textures, sounds or finishing animations.

    I would like to model some accessories or background objects for example.

  7. Nathan Writes:

    Ken: there aren’t rules per se. But the block should make it unambiguous what the animation is supposed to be, and large changes to that intent shouldn’t be made.

    Dion: glad I could help! I’m hoping to post more information about animation technique in the future, in bite-sized pieces.

  8. Michael Fox (Mfoxdogg) Writes:

    Here is my attempt, i kept its snappyness and i didn’t touch his face

    http://dominomf.com/smcfiles/mfoxdogg_magic0001_0300.mp4

  9. Jonatas Kerr de Oliveira Writes:

    Here is my atempt for it… that´s what the time permits now… took me about 1:30 hours to do this (animation and render)… that´s ain´t good yet, but if there´s some oportunity to do some animation in the movie I would do mutch better and spend mutch more time on it…
    http://animacao.ufscar.googlepages.com/magic.avi
    http://rapidshare.com/files/47543762/magic.avi.html

  10. Jonatas Kerr de Oliveira Writes:

    I almost forgot! I didn´t toutch his face too!

  11. Simon Clitherow (aphex) Writes:

    I remember seeing a similar process for a commercial animation movie project, but I unfortunately forget the name…

    It was open for the public to see the progress but only in-house animators could actually contribute. It was all done online and on a shot-by-shot basis.

    Basically, the process was:

    1. animator uploads a rough draft of an animation in a particular shot (a ‘digital daily’).

    2. TD or other mentor animator gives feedback on what is wrong or could be improved.

    3. animator uploads new and improved animation, based on comments and waits for feedback.

    4. rinse and repeat, until mentor or TD declares the shot ‘complete’.

    Perhaps a similar system could be used for peach?

    Of course, it would be open to outsiders to contribute, but once a shot is ‘claimed’ or picked as best, then it should be restricted to just the mentor (in this case, bf institute animator) and the artist – which would save peoples wasted effort, etc. Just throwing my own ideas around :)

  12. Mathias Pedersen Writes:

    Is there any deadline for this? I’d very much like to contribute, but I currently don’t have the time.

    //Mathias

  13. Nathan Writes:

    There is no deadline. Just a casual experiment. :-)
    I’ll keep track of this comment thread for quite a while.

  14. amoose136 Writes:

    I’d like to try but I when I have a computer with blender and when I don’t is very random. Cool idea, I think something like this could be used for the sound part.

  15. filoman Writes:

    Great I sign in!
    Have a little time to lose, and already work a few with Mancandy
    see you later
    I’ll repost soon

  16. Pablo Lizardo Writes:

    visit my site: http://pablo.lizardo.info
    This site pretend to be a reference to the graphic designer who are working with Blender. The visitors will find things like tutorials, tips, resources and much more.

  17. Digimator Writes:

    A sound file would be nice to have.
    The voice track drives the acting performance in an animation shot.I gives the animator the character’s emotions and temperment in the given shot. And also a brief description of what the character is trying to do or accomplish.These things are a must have if the experiment is going to work well.

    All things said,I would love to partcipate too.

  18. Digimator Writes:

    I stand corrected.
    I found the reference file in the link above.

  19. Nathan Writes:

    Digimator: if the acting and character isn’t conveyed in my blocking, then I’ve failed already, and perhaps I should animate another shot. Perhaps an original one. :-P
    That said, you cannot rely on voice acting for all acting decisions in animation. Partly because a lot of animation doesn’t have dialog, but also because even when it does you need to take the acting to a higher level than just duplicating what the voice actor does. You need to really understand the character. Of course, this is all stuff that I need to work on too.

  20. Digimator Writes:

    Keep up the good work!
    Readability is a main ingredient in good
    CG.

  21. ethana2 Writes:

    Well, I’m of the opinion that the only art that cannot be performed largely by code is that of coding itself. Have you considered writing some kind of script that would smooth out the movements by joint angle? I would think it should be somewhat similar to drawing a curve around a set of points, except more complicated- but if it was done, it would make future choreography easier, no? It would also need to take physics into account to truly look natural…

  22. ethana2 Writes:

    …(perhaps even an algorithm as similar as possible to what the human brain has to do to hold a pose?

    –Compensate for all forces, constant recalibration, organic movement and constantly varying motion and pose result?

  23. brecht Writes:

    ethana, I agree it is possible to create tools to make this stage easier. There exist also advanced algorithms for motion synthesis, enhancing motion, keeping balance, incorporating tension, etc. But the results of even the state of the art are very crude still, and in my opinion would make the motion look better to start, but only save a small part of the total time needed to tweak the motion.

  24. Herb10660 Writes:

    Thanks Nathan for a great exercise. I’m a stop motion animator / director who is trying to learn a bit of CG through Blender. To have something like this to clean up is excellent training for people like me who are new to the medium. I’ve learnt loads about the principles of keyframing etc by doing this. The result is uploaded here:
    http://www.youtube.com/watch?v=ppaUlJOnN1Q
    It’s a little bit subtle but I didn’t want to change the underlying key positions too much.

  25. Goodneck Writes:

    hey, great idea!
    here’s my attempt:

    http://www.ace-studios.com/images/imagesappart/mancandymagic.avi
    hope ya like it…

  26. Morris Writes:

    haha the real problem is that he’s naked and no hair, so not much to animate. come on guys, let’s see some cloth and hair movin all around. :) I guess these peach guys probably wont have that either though, so actually, maybe this is good, but then it needs SQUASH AND STRECH. haha, ok “shut up and try it yourself, big shot” ooops sorry, just a comment..never mind…

  27. Virgilio Vasconcelos Writes:

    Hey, guys! =D

    Well… It’s a good thing to do!

    That Keith Lango tutorial is really good, but he also made another one where he wrote about a better way to achieve overlaps and such:
    http://www.keithlango.com/tutorials/overlap/overlap.html

    From the intro:
    A number of years ago I wrote my Pose to Pose- Organized Keyframe tutorial. In it I advocate a method of achieving overlap and drag on objects that- at the time- I thought yielded decent results in a quick and easy fashion. In the ensuing years I have come to change my opinion about the merits of such a workflow. In this follow up article I hope to point out the problems with my earlier ideas and propose a better solution to the challenge of creating great overlapping action and drag in your animation.”

    Cheers

  28. HolgerB Writes:

    Hi all,
    I have used the nodes editor and luminance key to combine the animations submitted. This will allow you to compare the different artwork.

    The Video Sequence Editor will allow you to do this frame by frame.

    Here is a preview pic:
    http://www.hapkido-clarholz.de/Holger/mancandy_combined.jpg
    And here is the video file:
    http://www.hapkido-clarholz.de/Holger/mancandy_combined.mpg

  29. joeedh Writes:

    my pathetic attempt:

    http://joeedh.googlepages.com/mancandy10001_0300.avi

    Joe

  30. jamie matta Writes:

    I would be much happier to support projects if the wider community can have opportunities to participate in some way.
    It also helps those who are certainly talented and would like to be involved but cant travel or only have a little time say in the weekends and evenings…
    Of course people participating this way need to meet a standard too.
    As has been said it is a way to get some experience too so the pool of eligible artists can grow ahead of the studio
    Also I think properly well it would raise the standard or at least extend the movie duration on what could be achieved by a small team in a small time regardless of their dedication.
    So yeah lots of potential
    Go for it :o)

  31. TinShed Writes:

    Nice little exercise. Its interesting the differences in each attempt.

    …And my attempt
    http://adayman.googlepages.com/tinshed_magic_3.mp4

    The blend files
    http://adayman.googlepages.com/tinshed_magic_mancandy.zip

  32. joeedh Writes:

    Heh. I forgot to render my attempt in 2.44 :/ Mancandy’s facial rigs havn’t been fixed yet for current SVN.

    Joe

  33. fatfinger Writes:

    Nice one, here’s my attempt.
    http://s79.photobucket.com/albums/j124/fingapointa/Blender/?action=view&current=magicmcandyfatfinger.flv

    cheers

  34. McSmiffy Writes:

    Here’s my effort – I spent about 3 hours tweaking, but didn’t touch the face.

    http://www.3d.vertexstew.co.uk/mov/mancandypeach.avi

    I think it could be useful to see how this process would be carried out by the Peach Team, but to be fair you’ve probably all got enough on your plates already.

    Even on a short clip like this, a detailed study of the process would be a good learning resource.

  35. Mike Writes:

    Here’s my take.
    Hope you enjoy!

    http://mbproductions.biz/wordpress/wp-content/uploads/2007/05/magic.png0001_0316.avi

  36. worlddZ Writes:

    too much blocked/paused animation.
    need improve with more dynamism and expressivity, taking care with excessive
    paused poses.
    IMO: Blender made animations need be more professional to call new users to blender. A lack of really good animations faces blender no so much apreciated to peoples who unknow the potencial of this software.

  37. Vilem novak Writes:

    Hi, I’m posting my experiment. I gave about 5 or 6 hours to it.
    I must say that I spent most of the time with playing with timing and posing – adding some between poses etc. Sorry if this is what it wasn’t supposed to be, but because the original example wasn’t simply enough readable for me, I tried to import several little motions to show more the action.The result is a tiny bit longer. I spent less on the smoothing part, since that’s what’s supposed to come after the previous. I didn’t come to a really satisfiable result, and I think I’d have to play with the sequence at least 2 days with my current skills to get something nicer. But that’s quite similar to times which are described in that pose – to – pose animation tutorial.
    Why I write this:
    I think the Idea is great, it could probably really help the project. But the system should be based on feedback.
    As described in the tutorial – 15 % plus quality in the end costs you 50 percent of whole time. And I guess that’s the part which should be really checked by the “main” animators, like pointing really well out what exactly do they want to achieve, or which parts of the spots to work on….
    I also didn’t spend more time since this is just a test and i have my own projects to work on, but I would love to spend more time on this if it would be actually for the open movie.

  38. Vilem novak Writes:

    ahh, the link :
    http://www.youtube.com/watch?v=ds8oy-SIyGI

  39. Andrew Writes:

    Two files….

    The first is a side by side shot of the key poses and the one with the breakdowns….

    http://www.digitalillusion.us/mandcandy/compare.mov

    The second is simply the one with the breakdowns…

    http://www.digitalillusion.us/mandcandy/breakdowns.mov

    I was going to play with the timing a bit but I didn’t want to mess with the key poses since those would be set by the lead. So I focused on Arcs, anticipation, secondary action , $ overlapping action. There is still some polish work needed on the overlapping action and the line of action sometimes is a bit odd but I figured I’d post what I have for now, C & C please…

  40. caleb piercy Writes:

    I’m actually trying to do what you are toying with, except for non peach related animations. I’m trying to get people online interested in working together on some short films. I have the domain name set up and the webpage will be published in a couple of weeks. I’m also working on a video editor for linux called Rasterrain if any programmers are interested in that. As for the animation studio, I have a small 10 node Blender render farm set up and if enough people got involved we would vote on animations and communicate through emails and chats. All work would be given the credit it deserves and theres plenty of room for non-animators too; everyone is invited. Check out http://www.capicy.com for more information or email me if you have any questions at [email protected].

  41. Gianmichele Writes:

    Hey guys. Unfortunately I’ve just started looking int blender, but my lack of time prevents me to really progress quickly. (BTW if anyone wants to tech me quickly how to use grapheditor, dope sheet and a general key/timeline/playblast overview it’ll be sweet).
    The thing that I’ve seen in all the animation tests so far is that you are all progressing from pose to pose and body parts move too much at the same time. Keep in mind your character body. He’s really skinny and tall, so he should have much more drag, overlap, overshoot, follow through and settle during this fast movements. Someone tried to add breakdown to the key poses and that’s good. But then you rely too much on what the computer is giving to you. THE COMPUTER IS THE WORST INBETWEENER IN THE UNIVERSE! Watch out for arcs expecially in the arms and pelvis. A lot of action are stopping without a settle and so the body seems to maintain always a state of tension.
    I’d like to animate a take myself on this, but I’m really dumb when it comes to learn a new software, so any help is really welcome.

    Gianmichele

  42. Cuby Writes:

    Great idea. Here’s my attempt at making this mancandy animation finished:

    http://dominomf.com/smcfiles/Cuby__CubyMagic.mov

    Hope you like it
    Cuby

  43. dma88 Writes:

    I think it would be a great idea to have everyone able to contribute to Peach. (wether their contributions get put in is another story..) But I think it can go farther than just animation, we can work on other steps of the 3d process, and eventually, when you’ve got that streamlined, so it’s not one huge clusterf**k, we could get involved with pre-pro (especially the script).

    It seems that the best way to do that is to have each BF person have people working under them, and have a content directory that’s on the web somewhere. Only members of BF can write or delete from it, but anybody can read it. That would make a good “official” content directory and naming convention, so everybody on the web will be on the same page. People on the web can then submit their changes, and, if approved, they get uploaded to the online content directory. It would be good to have online dailies, too.

    I’m definitely willing to help out in any way i can, but I’m learning blender right now so it may take a little time :)

    The main point I’m trying to get at with this post is the huge advantage of being able to have LOTS of people working on the project. Even after it’s released, they could be polishing it further. (A year from now there may be an Elephants Dream 2.0!!) And as the projects grow as time goes on, they can get bigger and bigger and bigger.. Especially with NO budget and NO deadline.

    -dma88

  44. yemant Writes:

    Here’s my take.
    Hope you enjoy!
    http://www.youtube.com/watch?v=BfeTV2pB4uM
    http://www.youtube.com/watch?v=OKlHP0a1iPs