Bob Miller chronicles the new kind of fantastic Framestore CFC had to appeal to bring dragons to life in Dragons: A Fantasy Made Real versus their groundbreaking work on Walking with Dinosaurs.
To create the dragons, Neil Glasbey and his team had to marry images of the big flightless birds of today with dinosaurs, and also use their own imagination. All images courtesy of Framestore CFC/Darlow Smithson.
Bringing life to ancient creatures is the forte of Framestore CFC, the largest visual effects and computer animation company in Europe, based in London. Its 1999 miniseries, Walking with Dinosaurs, became a sensation leading to more specials, including The Ballad of Big Al (Allosaurus in the U.S.), Walking with Beasts, Land of Giants, Search for the Giant Claw and Sea Monsters. Now, Framestore CFC has applied its technical wizardry to creatures whose presence has been noted by cultures throughout the world dragons.
If they were real, what were they like? How did they breathe fire? Fly? Mate? Did they encounter dinosaurs? (That is, other dinosaurs, since the term dinosaur was invented in 1841. What were dinosaurs called before 1841? Dragons!)
These questions are explored in Dragons: A Fantasy Made Real, a 90-minute docudrama that sees its U.S. premiere on Animal Planet on March 20, 2005, at 8:00 pm EST, followed by Dragons: The Magic Behind the Scenes at 9:30 pm. A 100-minute version of the special was first broadcast in Germany on Dec. 1, 2004, called Dragons World Unglaubliche Entdeckung im Reich der Drachen. In the U.K., Channel Four telecast the special as The Last Dragon on March 5. A Region 1 DVD edition will be released on April 5, 2005 as Dragon's World: A Fantasy Made Real.
For Dragons, a team of seven animators were managed by lead animator Neil Glasbey, who discussed his contribution to the project in a phone interview.
Bob Miller: Please tell us about yourself.
Neil Glasbey:
Well, Im a lead animator at Framestore CFC. I was the lead animator on the Dragons project, which was sent here about five years, now. I originally started working on Walking with Beasts many years ago [2001], which was just after Walking with Dinosaurs, then progressed on to the two specials, which were Land of the Giants and Search for the Giant Claw. We then did Sea Monsters. I then worked on Harry Potter and the Prisoner of Azkaban. We did the Hippogriff on that one, which was great fun. And then on to Dragons. Im presently working on another project, Walking with Monsters. Of course its going to be great, but I cant tell you too much about it, Im afraid.
BM: Of all your projects, what is the most satisfactory animation that youve done?
NG: Certainly the most exciting, and satisfying would be Dragons, simply because theyre such interesting creatures. Ive never really seen anything done like that before. Theres been lots of dinosaur [films], which have all been fun, but the dragon side has been wonderful especially like having a dragon fight a T.Rex. Its everyones dream to do something like that. To be able to do it was fantastic.
BM: In terms of personal satisfaction with dragons, is it your favorite because youve advanced enough in skill to bring something to life more convincingly something different than the subject matter?
Last year, Framestore CFCs Neil Glasbey led a team of seven animators in the making of Dragons.
NG: There were two things. One was, I had been animating for a few years now. I was capable of making them more believable than perhaps what I had done in the past. Experience, and shared knowledge and so forth. And two, we had developed very quickly a relationship with the director [Justin Hardy], who gave us fairly free reign, to allow us to [manipulate] the characters in a way we felt appropriate. And to have that freedom was unusual. It was great. To think of these things in your mind and to give them the life that you thought was appropriate, rather than being, you know, overly-directed (as sometimes happens), to have that freedom was unusual and wonderful.
BM: What was your training background? What school did you attend?
NG: Many years ago I was a technical illustrator at Bournemouth College of Art and Design. And that was doing technical-based drawings for things like car manuals, maintenance manuals, that sort of thing. That doesnt sound too related to 3D. But the good thing about that was I was able to pick up the idea of how objects worked in 3D space, which was obviously useful for animation purposes. A few years later, I started gaining an interest in computer animation and then all things based on the 3D side of the computer, to work on FX coming from film, inspired by films like Jurassic Park. I taught myself to a certain level. Then I managed to get a place at Bournemouth University and the NCCA, which is the National Centre of Computer Animation. Did the Masters Degree in animation there, which I thoroughly enjoyed, and then went from there working at a company called Peppers Ghost, in which we did a childrens TV series, which was Tiny Planet. From there I went straight to Framestore CFC. Framestore CFC was my first real job in the animation industry and Ive been here ever since and thoroughly enjoying it.
The locomotion of the dragons was tricky. Digital artists had to cope with the musculature of the front shoulders working with the legs and wings, which is entirely inaccurate in evolutionary and biomechanical terms.
BM: So how did you come to work for Framestore CFC?
NG: I was lucky I already had some friends that already worked here. I had to use my student film reel because it was only a few months after I left from university. I was at the Peppers Ghost place. They told me about this project they were working which was called Walking with Beasts. I sent my reel in and I was asked in for an interview because I was a friend of the animators. On the back of my reel was a recommendation from friends, and managed to get a position.
BM: How did Framestore CFC acquire the project?
NG: I would imagine it would have been through our reputation gained from our previous works, certainly our work on Dinosaurs. I first heard about it probably about two years ago, which was when the idea was given to us. I think the same production company that used us on Walking with Dinosaurs and possibly through Impossible Pictures decided to approach us with the idea of doing it. We were the obvious choice, really, to do that, because of the knowledge base and skills to bring that sort of program to life.
BM: How did you become involved?
NG: It was good timing, really. I had just finished working on Harry Potter and the Prisoner of Azkaban, on the Hippogriff. That came to an end about a year ago, just when Dragons was gearing up. It was going to be six months between Potter finishing and our next big animation project starting, and Dragons fitted nicely into that. Plus we had also heard about the project and it was something I would have loved to work on, so I put my name forward. It was a combination of the two, the timing and me putting my hand in there, I think.
In the special, dragons appear in four basic types:Prehistoric: Bipedal with wings one adult female, one old stallion, one juvenile male (also T.Rex and Pteranodons for this segment) Marine: quadruped with fluke on tail one adult male Forest: land-based variant on Aquatic dragon one adult male Mountain: quadruped with wings one adult female, two adult males, one juvenile
Neil viewed remarkable footage of eagles mating and came up with ideas on how to show the courtship between flying dragons.
BM: Your dragons were an amalgam of several kinds of animal behavior. What animals did you use for inspiration?
NG: These were quite tricky ones because dragons are unusual creatures for a start. Theres nothing alive today thats that close. I think for the sundragon, which was the primeval dragon, the first one; it really was a combination of the big flightless birds of today, like the ostrich. Inspiration we got from the allosaur and the T. Rex. On the ground they were supposed to be like live and nimble creatures.
In the air it was a different matter. With these creatures how to visualize its flight was difficult. There really isnt anything similar today. I dont think anything ever existed like that. Really we used a lot of our imaginations and [used] anything that looked correct, that looked pleasing to the eye.
For the medieval dragon, which is a four-legged creature, and with wings as well, we decided it was more canine. So I used a more dog-like approach to it, which gave it sort of an athletic and quite appealing nature. But I also tried to keep more reptilian tones to it because, again, these were reptiles. As for flight, I used ideas we picked up on the Hippogriff, which was a similar proportion creatures, although obviously the Hippogriff was a totally different creature with feathered wings. That kind of style, that posing, worked quite well. Even though the Hippogriff was a horse-eagle type of creature. This is more a dog-lizard kind of thing. The overall styling, I felt, worked rather well married to the two.
The animators looked to the chest of a horse for design inspiration.
For the aquatic and Chinese dragons, they were tricky ones because they were supposed to be serpent-like. They also had that kind of almost dog-like look to them, to the heads. Because theyre such long sinuous creatures, its really a case of trying to do a lizard-meets-mammal style of walk, with that nice sinuous, lizardy serpent look to them.
Every creature we could imagine, wed take a bit from each one of them.
BM: I understand that the mating ritual of the eagle was used as inspiration.
NG: Yes, thats correct. I had never actually seen that before. I was shown a short video of this eagle mating. I just couldnt believe it, that they actually did this, flying together at an incredible speed in locked talons spinning around each other. It was a fantastic thing to watch. We thought that would be perfect for the medieval dragon mating ritual. So we applied those techniques. Obviously the difference is that the eagles have two legs; our dragons have four legs. So we combined the style of what the eagles did and put it on to our falling creatures. And I think the end result worked really, really well. What was remarkable about that because I had never seen the eagles do this before was not something I would have thought of, ever having our dragons doing. Once we saw it we thought that was perfect. We put it together and we thought the shots looked great because of it.
Inspiration came from the following:Combustion: beetleWarning colors: mothSize: dinosaursExtra limbs: fruit flyWing structure: batMineral quarrying: elephantGas production: hippoCourtship: eagle
BM: Ive seen the clip. It looks great. I noticed a handheld camera shake as they were descending, which I thought was a nice touch.
NG: Yeah. Thats the sort of thing that gives it the feeling of an entity chasing them down and you really feel that theyre plummeting. It was a fantastic sensation of speed, I thought. Its those little touches that really do enhance the program.
BM: How were you able to convey personality and individual traits in your dragons beyond just simple animal behavior?
The Framestore CFC team delivered 167 shots some 35 minutes of CG in just 25 weeks.
NG: Yeah, thats always the trickiest thing to get across: the emotion rather than just simple locomotion. Making the creature move from A to B is one thing. Making it move with a certain feeling is it sad? Is it happy? Is it running from something? Is it chasing something? These are tricky things and the technique I always use is trying to put myself in the creatures position. If I was this creature in this situation, how would I be feeling? What would I be doing? Would I be scared? Would I be happy? And its trying to think your way through those situations that I think, if youre trying to empathize with the creature while youre animating it, that reads across well when people watch it. You get the impression that somebodys thought their way through this instead of making it move from A to B.
BM: In directing your animation team, how much latitude did you give them in terms of bringing personality to the performance of the individual creature?
NG: I try to give them as much latitude as possible, because people always tend to surprise you if you give them a bit of free space when they work. Thats one of the best things about this industry. Everybody has their own ideas and their own personalities, and I think they put that into their work. Its so much better. It was always a case that we basically knew what the creature had to do in each shot. The director was telling us the choreography of the shot. That is, where the creature should be at a certain point, how it should be placed. The style of how it got there was something I tried as much as possible to leave to the individual animator, so that they could put their own style of their work into it. I think that came across very strongly and it worked out very well, indeed. Though really, its kind of about 50% is where we know the creature is going to be in a shot and what it should be doing, and the rest of it is made up from the individual animators imagination.
Framestore CFC used the knowledge it gained from its work on Walking with Dinosaurs to help create the models for Dragons.
BM: Presumably you had to make sure there was some consistency in the performance.
NG: Of course. Yes. Theres a certain amount of storytelling that has to be done in each shot. Continuity in each shot is very important. There were some shots that we knew where the dragon had to be heating rock from the side of the mountain (he was on the side of the mountain at this point) and he had to fly out of the shot. Now how he got from one of those positions to the other, I had to [allow] as much as possible the animators decision because, again, as I say they come up with styles and the way they were doing it that often surprise you. I know its a joy to watch because of it. But we always knew it had to be done in a certain style and at a certain point. So there were certain limitations that certain goals that had to be met within each shot, obviously, and then, there was that certain area of freedom that gave the beautiful fluidity to a lot of the shots. Theres a noble style to a lot of it. That worked out very well.
BM: Were you able to do any animation yourself on the project?
NG: I did, yes. I mainly did the aquatic and Chinese dragons. I did the underwater swimming for the aquatic dragon and I did a lot of the land-based shots for the Chinese one. And they were great fun because it was like nothing Ive worked on before, cause they were so serpent-like, and so slinky, and it was nice to be able to do this nice underwater serpent-like animation. It was great fun. I also got the chance to work on the medieval dragon. I did the basic walk cycles and the raw locomotion of how the creature gets from A to B which gives a base for the other animators to build from.
BM: What was your opinion of the animation in Dragonheart and Reign of Fire and how do you compare it with your work?
NG: I thoroughly enjoyed both of them. Dragonheart is a wonderful film. I love the fact that Draco has so much character. But the difference between Dragonheart and ours, is that Dragonheart is very much a character piece, and ours is trying to explain that these things are real creatures and they move in a more realistic way. But we certainly studied Dragonheart and got inspiration from that, because it was more of a character piece as opposed to creature piece. And Reign of Fire, again, was also a film I thoroughly enjoyed.
Anything with good nasty dragons like that is good to watch. But the dragons from that were fundamentally different from ours, in that they were very much airborne creatures. Ninety percent of that was in the air. Once they were on the ground, they were almost like great big giant bats walking along. Which I thought worked great for them but again, wasnt the sort of thing we were looking for. But certainly, some of the movements we were inspired by the two films. But we looked to them and decided they were two extremes and we were kind of going somewhere in the middle of it. (That would be the best way to put it.) Get some of the character from Dragonheart so you can empathize with the creatures, and some of the locomotion from Reign of Fire.
Glasbey and his team were able to produce the creatures for Dragons quickly because of the companys previous work on the Walking with series combined with not having to deliver tiny exquisite details for the small screen.
BM: Regarding the locomotion, CGI supervisor Alec Knox said that working with the dragons was tricky because the musculature around the front shoulders had to cope with working legs and wings, and thats just not really on, in evolutionary and bio-mechanical terms.
NG:
These were tricky creatures because the way they were built, because theyve got the wings and especially the medieval dragon with four legs and wings also. Its tricky to work out how the muscles worked. For the sun dragons and the primeval dragons, it was easier because we used if you looked at a bird, its got those huge pectoral muscles. We were able to take some of that and the thighs and the hips and so forth, and looked at some of the modern-day flightless birds like the ostriches. But the medieval dragon was a tricky one. So, there really was a bit of scratching of the heads and thinking on that. I think we came to the conclusion to use probably the front end of a horse. If you look at the chest of a horse, its got those big muscles from around there, so I guess thats probably where we got most of our inspiration from for that one.
BM: Interesting! In The Making of Dragonheart, Rob Cohen, the director, was talking about how he was inspired by horse musculature.
NG: It makes sense because if you look at the front end of a horse from the face to the neck to the top of the legs, theres a huge mass of muscle there. Its what wed have in our shoulder area. That works perfectly for the look of the dragons in our show. Certainly the four-legged ones. Yeah. I can see where he got that idea from as well.
Knowing how to frame the dragons saved large portions of time. Glasbey knew that little things like keeping the feet out of the bottom of the frame or hiding behind an object wouldnt hinder the story or overall look of the piece.
BM: Were you inspired by his approach or was this a conclusion that you developed independently?
NG: Well, you always look for as much reference as possible. If somebody else has looked at that and come to the conclusion that thats the way they were going to do and develop such a good system then obviously we looked at that and think theyre probably along the right lines. I think we had to develop ours in a slightly different way cause our creatures were so different. And they behaved in such different ways. I think theres only certain ways you can approach such a thing because, theres only certain ways in which a character can move, and strapping a huge pair of wings on a horse or donkey-shaped creature is going to work in a certain way. And that is, to have the motion of the muscle wrapping around the front end of the shoulder and down into the chest area, so the horse was the perfect starting block, as it were.
BM: Lets talk about the production timeframe of Dragons. When did it start?
NG: It was just over a year ago, late February/early March [2004]. We got the first storyboard through and started doing an animatic for it, to bring the storyboards to life. Our animation deadline was in the middle of July. We finished a little bit before. The animation always finishes a little bit before the actual project, cause then it gets handed over to the tds.
We did the animatic in just a few weeks and then they go away on the shoot to get the live action, which they bring back to us. A fair chunk of the initial time would have been the animatic, building the models, making animation rigs, getting the whole pipeline set up. Then once the rigs were done we had the live action from the shoots to work with, the actual shots sorted out. Thats when animation started going full on. Totally, we had seven animators to do the shots. I was there right from the start, guiding the animatic, making sure the rigs were set up for animation purposes.
BM: According to the press material, A 35-strong team from Framestore CFC delivered 167 shots some 35 minutes of CG in 25 weeks. By comparison, Industrial Light & Magic took over a year on Dragonheart, with a shorter amount of screen time for Draco, 23 minutes.
NG: Yeah.
BM: This is amazing. How were you able to do something faster than Industrial Light & Magic?
NG: Weve had a few years to work on it since then. Plus we had all the knowledge base we gained from the Walking with series Walking with Dinosaurs and Walking with Beasts. And now we know where we can cut those different corners and make those little compromises that dont affect the look of it onscreen, but dramatically increased our speed in production. Also, our dragons arent as strong a caricature as Draco. We didnt have to put all those tiny, tiny little nuances in as they had for Draco in Dragonheart. So again, that saved a little bit of time, because those tiny exquisite movements that bring it to life on something as big as the silver screen do take an enormous amount of time. We were able to compromise by not putting quite as much detail into that. So, we probably got about 90% of the data that they did on Draco, but that extra 10% takes up so much time. I think its a combination of the knowledge weve gained over the years and those sort of savings to get so much work done quickly.
Also, the more enjoyable the project is, the harder people work on it. Theyre getting so much more out of it. This is a fantasy. Everything that we do is an enjoyable thing because its just a great way to work. Youre making wonderful pictures on TV or film screens. Its a fantastic industry. So, people tend to throw themselves into, and put more of themselves into their work. That especially came across on the Dragons project. I cant exactly put my finger on why that is. Everyone who had worked on it just seemed to enjoy it so much they put their extra effort into it. Thats where we managed to achieve what we managed to achieve in such a short period of time.
BM: How did the techniques developed in the dinosaur projects enable you to get results at a faster speed?
NG: Really its a case of knowing how we can frame a creature in such a way that it tells as much of the story as a director would want. It looks as great on the screen because they save us time. And that was usually little things like keeping the feet out of the bottom of the frame, or hiding behind an object. Those tiny little actions, those interactions that take so much time to do but dont really add that much to the story or the overall look, we were able to work around a few of those. Its little techniques like that that save large portions of time.
BM: Was this [restaging] done at storyboard stage?
NG: When we blocked things out, when we actually got the live-action plates back and put a rough model of the creature and see how it would fit in the frame, we were kind of always aware of frame it in such a way that would save time. Usually we had a pretty good idea of how the shot was going to work out before we started the animation. That is, the choreography of the shot and the framing of the shot. So once you started animating the overall position of the creature and the framing changed very little from the block stage, the earlier stage.
BG: How has CG technology progressed since Walking with Dinosaurs and what new developments were you able to incorporate into Dragons?
NG: As far as our models and animation rigs go, our rigs are now much more complicated than they were many years ago. Thats really a combination of how much faster computers are today and they can handle that much more information, and throw it around in real time, which is essential for animation purposes. We can, to a certain degree, fake real muscles and give the impression of underlying solidity of the creatures, which we were unable to do many years ago because of the speed of computers and the complexity of the software. So really, its a combination of computers becoming faster and software becoming much more versatile. Also, theres so much more in software these days. Also, the knowledge base is that greater than it was a few years ago. People learned from project to project, and added to what they learned before and that helps enormously.
Since Walking with Dinosaurs, computers have increased in speed, and with projects like Dragons, the rig has become far more complicated.
BM: You mentioned the improvement of speed in computer performance. How much faster, would you say?
NG: When I first started, I was using computers that were 200-300MHz speed. And I think the ones we use today are near to 3GHz. So the actual processing power has improved by a factor of 10. And with graphics cards becoming much faster we can give that much more screen information. So were probably using computers that are ten times faster today than we were six or seven years ago.
But then, our rigs that we do are that much more complex. One offsets the other. If the computer becomes that much more powerful, well use all of that power by making the entity that more complicated and realistic.
BM: Does that mean theres no appreciable decrease in production time because of that?
NG: Pretty much. Yeah. I mean, we can do as many shots today as we did five or six years ago or even longer. The difference is, they all look that much more realistic today. Whether we can actually do more than we did a few years ago, then I doubt it. I think the actual amount of time it takes to do a shot remains pretty much stable. Simply because if weve got all this extra computer power, well use it to make things more realistic. And that takes up more time. One offsets the other, really.
BM: What further advances do you think need to be made in CG animation and do you foresee that happening?
NG: Theres so many little things that there are always nice to have. The one that springs to mind, really, is the way the animator puts information into the computer on to the actual rig itself, to actually create the rig to move. Because essentially what were doing is acting. The animation rig we use is an extension of the motion wed like to deliver. So inputting information through a mouse and keyboards seems strangely crude. Its effective, but strangely crude. So I think that would be the area that would be great to improve in. Not just a mouse and keyboard, but some other physical input device that links directly to the rig. Theres been many attempts to try this before. I would certainly think that perhaps that would be the one thing that would benefit us the most, more than anything else.
BM: Now, were you able to recycle the models used in Walking with Dinosaurs and Dinotopia? Like, say, the T. Rex?
NG: We couldnt actually recycle the models but what we could do is take the knowledge we had gained from those and put it towards the models we had created for Dragons. Because they were such similar creatures, obviously. Theres a T. Rex in Dragons like there were in the Walking with specials. But they were two separate companies and two separate programs. So you see we cant really use those models. But we can [use] the knowledge gained from the original builds of those creatures and take a direct step from that to what we created for Dragons. It was certainly a help to have done them before. So it sped up that process.
BM: So what improvements were made on these models?
NG: Well, for a start the models are a much higher resolution. Theyre much more detailed. Because as I said earlier on, the computer power that we have available is much higher now. So we gave more detail in the texturing and the modeling and so forth. Basically we have higher definition models and higher definition textures. Were able to go that bit closer to the creature on screen. The biggest change over the years is the level of detail on the creature. Its a lot higher than it was.
BM: Thats interesting because on Walking with Dinosaurs they used puppets for the closeups. So in this case, in Dragons, you really didnt need to use puppets, did you?
NG: I do, yeah.
BM: So in this case, in Dragons, you really didnt need to use puppets, did you?
NG: Thats very true. I mean, theres some things where its better to have an animatronic as a puppet, and things like the dead juvenile medieval dragon is physically interacted with. Its better to have that stuff. I dont think wed ever completely replace animatronics. Theres a place where theyve always been needed. Were certainly capable of going so much closer than before. But of course the downside of that is, the closer you get to a creature the longer it takes to render. So, it does become expensive, time-wise, to go that close to a creature. But it certainly does hold up very, very close, now.
BM: Speaking of high-resolution, was the CG in Dragons rendered in a high enough resolution to play in HDTV format?
NG: I believe it was rendered at PAL [720x576]. Its not HD format, no [NTSC is 640x480]. We could probably transfer to it but it wont use the full resolution of HDTV format [1920x1080], because thats up near film, and once you start rendering in those resolutions it becomes so much more time-consuming to render.
Thats a technical question. I deal with the animating side, so I dont get too involved with those issues. I certainly believe the models could hold up to that resolution.
BM: So once TVs and DVDs All become high-res capable, you could up-res the models in Dragons, then?
NG: We probably could do that. Thats an interesting point, actually. HDTV is 2-1/2 times the resolution of PAL, I think. Youre up to film resolution for rendering. And of course as you do that it takes an awful lot longer to render. So theres going to be some interesting decisions made when that becomes the basic format. Hopefully by then computers will be fast enough to swallow the overhead in time.
BM: What else about your work in Dragons that people should know about?
NG: It was thoroughly enjoyable to work on. It has a look that I dont think that I had ever seen on TV before. And I think the quality of it was that much higher than was to be expected which was great. I think that everybody got really, really into it; they put their all into it. And the end result was, I was extremely pleased, and everyone who had worked on it was extremely pleased.
Since 1985 Bob Miller has written numerous articles covering the animation industry for publications such as Starlog, Comics Scene, Comics Buyers Guide, Animation Magazine, Animato! and Animation World Magazine. He was storyboard supervisor for MGMs Lionhearts, Courage the Cowardly Dog for Stretch Films/Cartoon Network, Megas XLR for Cartoon Network, and lately, the Say it with Noddy 3D interstitials for Make Room for Noddy, coming this summer to PBS Kids. Bob won a 1999-2000 primetime Emmy Award certificate for storyboarding on The Simpsons episode, Behind the Laughter. Bob serves on the board of directors at the International Animated Film Society, ASIFA-Hollywood.
