Meet the Artist: Spencer Cook

Become a member of the CGSociety

Connect, Share, and Learn with our Large Growing CG Art Community. It's Free!

THREAD CLOSED
 
Thread Tools Search this Thread Display Modes
Old 06 June 2007   #61
Originally Posted by CB_3D: And one more from me ;-)

In the Superman space shuttlethrow sequence I saw a musclesimulation running under the Supermanmesh.

Does the Spiderman deformation use Muscle simulation, and if so, is it the muscle system that directly deforms themesh? Or does it only serve as an orientation for the animators?

Or do the Spiderman puppets use only weighting?

Specially in Spiderman, when his limbs go into extreme positions the jointdeformations were perfect, congrats.

If there´s automated compensation/musclesliding etc by what tecniques is it achieved?

Thx again


Hi again CB_3D,
Spider-man's deformations are all point weighting from Koji Morihiro. He set it up for the first movie and we have been using it ever since. I'll let him know that you enjoy his fine work.

thanks,
Spencer
 
Old 06 June 2007   #62
Thank you for your reply. One more question, slightly similar. You may not know the answer, but do they also solely use maya for the modelling (and zbrush of course)?

Once again. Superb Work. Keep it up!
 
Old 06 June 2007   #63
Hi,

I am sorry to ask rigging related questions, but this is my only chance to get some info from you!

ReInsisting on CB_3D 's ques : You said that Koji Morihiro used all point weigting on spiderman Rigs. I would think though that surely there is something else going on ,on top of weighting that helps maintain such believable volume/and gives smooth deformations. Is it some blendshapes that were triggered based on joint rotations/pose or some other technique. If its all only weigting and no shapes , then was it based on multiple joints being pushed/pulled internally,depending on the pose(which would be so difficult to setup m sure) ? Pls share the brief theory behind the setup if posssible :-)

And GREAT work, I was amazed by the technical output of the movie! It continues to inspire me to do much better work everyday!

Thanks for your time




 
Old 06 June 2007   #64
Great job with Spiderman and Matrix. Each one is ground breaking in its own way and I love all of them. What the next project?
 
Old 06 June 2007   #65
Hello, I just wanted to say what a great film and congrats on such a success. And as many have already posted here, I just wanted to say that my favorite part was the sandman birth. I loved how soft the sand felt and how he was trying to come to grips that "okay im made of sand now.....just need to pick myself and move on" lots of determination in that scene and as a growing animator I hope I can pull something like that off in the future. Great work! Also just wanted to thank you for giving us your time here. Thanks.

-Bryan
 
Old 06 June 2007   #66
Hi Spencer,

First I want to thank you for your time...

Well I entered VFX field almost 2 yrs ago, I work in Lebanon(Middle East).

The field is evolving here but still not as professional as what u guys do. I'm really eager to learn and improve in this field more. I would like to ask you, is it in any way possible for me to do training in one of the post production companies in the US? If yes how can I reach any, if not then what's your advice for me in order to improve and reach the pro level..

 
Old 06 June 2007   #67
Here goes some critics about Spiderman III, its just my opinion thou.

Story was the weakest among the three films, I'd say the first film had the strongest storyline and the most clear to follow, but the third film has too much material and very little time devoted to each as if barely scratching the surface, never letting the audience really get involved emotionaly with anything. For example, the story with the sandman and his daughter and wife, so little screen time is given to them while it could have been a great subject to go deep with.

On direction, while I may not be quilified to criticise, as a mere member of the audience, I found some of the action sequences too crowded, the cutting and continuity was sometimes too hard to follow, specialy on the fight with harry and pete in the alleys, it was like they tried to pack in too much goodies into that sequence that it spoiled it all.

On animation, there was some minor stuff, one during the sandman born scene while hes getting up and falling, i think it would have needed more polish, it felt to me that it was somehow rushed in animation. also some of the shots with spidy just screamed computer animation, as they were too clean and perfect to fit into the live action set. As it is with CG animation in live action films, sometimes you need to add those little imperfections, like a few unclean arcs or maybe mess up those slow in/out a bit to make it more life like.

animation of the sandman in the final battle could have been better, I would have personaly added more weight and force to his punch. it was there but could have been pushed more to add more impact.

Overall, I enjoyed the film but the first spiderman remains my favorite.

Last edited by GoldenCamel : 06 June 2007 at 02:58 PM.
 
Old 06 June 2007   #68
Originally Posted by LavenPillay: Firstly, Mr Cook, HUGE kudos to you an the ENTIRE ImageWorks team. I think it goes without saying that you guys are almost "benchmark" level with regards to the entire end-to-end VFX pipeline.

Secondly, Questions :
1. For the intro fight with Goblin and Spidey, I'm assuming that that was a mix of
a) Live Actors
b) Digital doubles
c) Photographic/Video back plates
d) CG'ed buildings
e) real buildings
Tricky question, but could you tell us what some of the desicion-making process is like for a shot like that ?
I'm thinking its not just the technical considerations, but also time and money budget etc.

2. What were average render times per frame for really complex shots ? I'm just trying to compare to things like Narnia where Aslan scenes took about 9 hours to render!

Thirdly, and i dont mean to contradict what you said, but just suggest a more probable method :



My understanding is that for shots like that, where it appears that time stops and the camera does a "fly around" the subject, what they do is setup an array of cameras (sometimes over 30-40, more ?) in a ring around the subject.

Then, they fire the cameras off all at the same time.
(say, when a person jumps up and reaches the height of the jump)

The result is that you have multiple frames of the subject from different angles, but at the same time. So, you just play those frames back in sequence and it appears that a single camera is moving around the subject in freeze-frame.

Hope that answers your question, akhenaten.

And thanks again for taking the time out of your obviously crazy schedule to answer our questions, Mr Cook.


Hi Laven,
The Peter/Goblin fight (we called it pg sequence for short) was a mix of everything you mentioned. Some of the decisions of what technique to use were based on things like; how well do you see the actors, is it possible to shoot on location, is it possible for the actor or stuntman to perform the action. To clearly see the actors, some of the action was performed on a stage in front of a bluescreen. In animation we could manipulate the bluescreen plate to enhance the motion. In some cases we comped the actors face onto the animated character. This was a tricky process that involved animating the character then using that motion to program a motion control camera to photograph the actor's face with the proper perspective changes. For some of the more complex action that no person could perform and no camera rig could photograph we went all digital. We tried to mix up the techniques and see the actors as much as possible. Time and money always figure into the decisions as well.
As far as render times; we generally render a shot in separate passes (and comp them together) so it's difficult to quantify a time per shot or per frame. One general rule though is that if an element takes more than 2 hours to render then every effort is made to reduce that time. There were some shot elements that took as much as 7 or 8 hours to render. I hope that answers your questions.

thanks,
Spencer
 
Old 06 June 2007   #69
Originally Posted by TheJinj: While it's true that what you describe was truly the method pioneered by a French company for a Rolling Stones clip and then used for the bullet time sequence in Matrix one, I don't think someone could fly over an exploding truck and have that captured by multiple stillframe cameras. It was digital.

Spencer, Thanks for answering all these questions. Here are mine:

1)I have heard that begining animators have a very hard time in the big companies. What's life like for a begining animator in Imageworks? Is any slack given or even possible to give with the tight deadlines? Or do you hire only battle-tested animators that are expected to dive into production from the first minute?

2)Coming from a smaller, generalist team, and hearing your pipeline description here, I wondered what modelers do after the animation has started. Are they all fired after the first couple of months of production? Do they make the animators coffee? Or do they zip around from project to project, working the first month or so on each?

Thank you very much for your time and patience. Your kindness to the community is as inspiring as your work.

Adam.


Hi JinJ,

Thanks for the questions, here you go;

1. We are always on the look out for talented and enthusiastic young animators. It's important to have a mix of veterans and beginners on each team. Where else will the beginners get the experience to become veterans one day? We have a training department here that helps get new folks up to speed on software, tools, pipeline, etc. It also depends on the time frame for the show. We probably wouldn't have too many junior animators on "911" shows since we do have to push stuff out quickly. However, there are shows with a schedule of a year or more where juniors have time to ramp up to speed. Also, a junior animator would probably do more simple shots in the beginning.

2. At Imageworks, once a modeler is done on a show we box them up and put them in storage. I'm kidding of course, ideally what happens is that they move on to another show. If there isn't another show in house they can take a vacation, work on some internal projects or go to work at another studio. It all depends on the indivdual and the state of the facility at the time.

hope that answers your questions,
thanks,
Spencer
 
Old 06 June 2007   #70
Originally Posted by scook: Hi Laven,
The Peter/Goblin fight (we called it pg sequence for short) was a mix of everything you mentioned. Some of the decisions of what technique to use were based on things like; how well do you see the actors, is it possible to shoot on location, is it possible for the actor or stuntman to perform the action. To clearly see the actors, some of the action was performed on a stage in front of a bluescreen. In animation we could manipulate the bluescreen plate to enhance the motion. In some cases we comped the actors face onto the animated character. This was a tricky process that involved animating the character then using that motion to program a motion control camera to photograph the actor's face with the proper perspective changes. For some of the more complex action that no person could perform and no camera rig could photograph we went all digital. We tried to mix up the techniques and see the actors as much as possible. Time and money always figure into the decisions as well.
As far as render times; we generally render a shot in separate passes (and comp them together) so it's difficult to quantify a time per shot or per frame. One general rule though is that if an element takes more than 2 hours to render then every effort is made to reduce that time. There were some shot elements that took as much as 7 or 8 hours to render. I hope that answers your questions.

thanks,
Spencer



I just found out some new info about render times from Michael May, one of our coordinators. Some of the very complex sand shots took as much as 12 hours per frame to render.

Spencer
 
Old 06 June 2007   #71
Mr. Cook,

To start, I just want to say thank you. You and your team have done a magnificent job with the Matrix movies, as well as the Spiderman movies (super huge ridiculous spiderman fan since I was born). Your fx are greatly inspiring and amazing! I have always had a passion for animation and I am getting ready to attend AAU online at the age of 26 (military service after highschool). I feel like I am a bit behind the curve because I have very little knowledge in the 3D animation world (messing with the downloadable version of 3DS Max as of late) and the animation/production language used. If you wouldn't mind, I would just like to get your advice on some of the possible obstacles/problems that I may encounter as someone starting out "this late in the game" so to speak. Thank you ahead of time and keep the amazing work coming! Take care!

J
__________________
Captain Hero - "Save Yourselves!"
 
Old 06 June 2007   #72
Spencer,
Congrats on great work and great career.

Joe Piazzo
SVA Film, 1984
 
Old 06 June 2007   #73
Originally Posted by amitabhverma8: hello their

recently i watched the the 3rd Series of Spiderman Movie. First let me congratulate you the special effects and the camera work is quite outstanding and top notch. so here are my question every thing is quite brilliant. what i would like to ask is that their something called the x factor you know the character like venom and sandman and the darker spidey quite didnt reach at the level of the xtreme goblin aka mr harry osborne. i mean the projection to the character and the way he performed his task is quite impressive in comparison to the other characters. secondly why is that the part of darker spidey is so less. i mean from the point of view that he was projected so much in the trailer and in the posters and in teasers.

secondly what is the difference of doing a shot in a digital world in comparison to doing it in a real world. how do you project so much information and surrounding in so much small time in a digital world.

thirdly what does it take to become a smooth animator and rigger. where can i get the best trainning who do you recommened

finally. do you think their is going to be another chapter in the series. i thought when spiderman 3 was pounded by sandman and venom their was an intresting shot in which he was nearly again caught by the symboite. and i thought wow may be he will go the darker side or turn into something that we never expected. so what is your opinion is their spiderman 4 happening. ?

thank you

amitabh


Hi amitabhverma8,

Thanks, I glad to here you enjoyed our work. Thanks for the questions.

I think your first question is more of a personal preference. We animated all of the characters the best way we could to fit their indivdual personalities.

As to your second question, the amount of screentime for dark Spidey was based on the story that Sam Raimi wanted to tell. The advertising did focus alot on dark Spidey, mostly I guess because it's a new element that audiences didn't see in Spider-man 1 and 2.

To create a photo-real looking environment digitally it's best to use as much real photography as possible. Scott Stokdyk, our visual effects supervisor, tries to design fx shots to include as much real elements as possible. Most of the time it's a mix of plate photography, hi resolution stills, skillful model making, texture painting, matte painting, lighting and a critical eye for detail.

The best advice I can give to improve as an animator is this; animate what interests you not what you think other people want to see. Whether it's cartoons, photoreal monsters or spaceships, you'll do better work and be happier if you follow your interests. Also, study real world reference. Act out an action in front of a mirror or, better yet, videotape youself and study it frame by frame. If you are animating a creature, find an animal in the real world to get ideas from. Find nature footage of that animal and study it frame by frame and apply that to your animation. Read animation books like "The illusion of life" or Richard Williams "The animator's survival kit" for basic animation principals. Most of all...animate, make mistakes, learn from your mistakes, be your own harshest critic and study human and animal body language.

Finally, I'm pretty sure there will be a Spider-Man 4. I don't have any inside information on this yet but a basic rule in Hollywood is that if a movie makes money the studio will want to do it again. There are still tons of great Spidey stories to tell.

thanks,
Spencer
 
Old 06 June 2007   #74
Originally Posted by wotanist
Hi Spencer,

I was wondering if there were any sequences that the director demanded which were too difficult to pull off and were dropped because of technical limitations, or would you say that you've met even the most outrageous demands?

My second question is about the CG Spiderman and Sandman? How many animation controls each do they have approximately, and how long did it take for rigging them?

Thirdly, how long did it take to render the most complex scene in the movie? Amongst the characters, I'm assuming that with Sandman's complex shaders, he would have taken the longest to render. Whats the size of the renderfarm?

Finally, What were the specs of the Workstations the team was working on? Despite working on high end workstations, were there any scenes that may have crashed the workstations while working?



Originally Posted by scook: Hi wotanist,
Most of the considerations for whether a sequence can be done or not involve time and money. There are very few things that can't be done technically given enough time(which equals money). I remember on the first Spidey movie Sam wanted to have a more massive aerial battle between Spidey and the Goblin in the Times Square sequence. It was ultimately simplified because of the expense. Sam got his aerial battle in #3 though.
Concerning your other questions, I'll have to get back to you with the specifics. I don't have that info handy at the moment.

thanks for your questions,
Spencer


Hi again wotanist,

I can't give you too many specific answers to your other questions but I'll pass along what I can. The most complex sand scene took close to 12 hours per frame to render. We use a linux os on our workstations. Scene crashes are an unfortunate part of working on a computer. Sorry I can't be more specific.

thanks,
Spencer
 
Old 06 June 2007   #75
Originally Posted by seema: Hi,

I am sorry to ask rigging related questions, but this is my only chance to get some info from you!

ReInsisting on CB_3D 's ques : You said that Koji Morihiro used all point weigting on spiderman Rigs. I would think though that surely there is something else going on ,on top of weighting that helps maintain such believable volume/and gives smooth deformations. Is it some blendshapes that were triggered based on joint rotations/pose or some other technique. If its all only weigting and no shapes , then was it based on multiple joints being pushed/pulled internally,depending on the pose(which would be so difficult to setup m sure) ? Pls share the brief theory behind the setup if posssible :-)

And GREAT work, I was amazed by the technical output of the movie! It continues to inspire me to do much better work everyday!

Thanks for your time






Hi seema,

Thanks for the kind words, I'm glad you enjoy our work. The best person to answer your question is Koji himself, so, ladies and gentlemen...Mr. Koji Morihiro;

It was all done with point weighting, basically.
Most of the deformation is done with Maya clusters applied directly onto the geometry
and there are also a few Lattices whose points are driven by clusters.

There are several hundreds clusters placed over Spidey's body.
And each of them works as a virtual muscle whose movement is
triggered by the animation joints via Maya expressions (equations).
For example, there are clusters placed and weighted around his biceps that get triggered by
the elbow rotation, and as a result these cluters give a nice bicep bulge effect.

The idea behind this technique is that I was convinced the best way to get the most believable deformation
was to have geometry precisely sculpted(weighted) at as many poses and have them interpolated as real muscles and skin would move.
To achieve both precise shapes and realistic interpolations, I found combinations of clusters and expressions work very neatly.

 
Thread Closed share thread



Posting Rules
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts

vB code is On
Smilies are On
[IMG] code is On
HTML code is Off
CGSociety
Society of Digital Artists
www.cgsociety.org

Powered by vBulletin
Copyright ©2000 - 2006,
Jelsoft Enterprises Ltd.
Minimize Ads
Forum Jump
Miscellaneous

All times are GMT. The time now is 12:53 PM.


Powered by vBulletin
Copyright ©2000 - 2017, Jelsoft Enterprises Ltd.