Here we go, I am pretty much done with that shot. It is still in French but I added some subtitles for those who want to understand what this is about.
The audio clip came from a funny french candid camera show my friend animator extraordinaire and BYOA organiser Samy shared on Facebook.
In that clip the host was pretending to be a tattoo artist until, looking at the work he had started on their body, people realised with horror he didn’t know how to draw. Fortunately the needle was in fact a felt pen.
I really liked the audio as it couldn’t get more genuine, presented with an opportunity for a “take/gear change” and allowed for something broad without looking overanimated.
The original character was standing fairly still but I believe entertainment doesn’t come from under-animation or over-animation but instead in stylised choices situated at the edge of both. A good showreel should demonstrate understanding in broad and subtle acting. This will be my broad piece.
I will leave it in playblast for now and do a full render if I do an other pass. My goal wasn’t to have the best shot ever but just to get more experience with some of the concepts I rediscovered studying “the work of the masters” few weeks ago and hopefully get a junior or even a regular animation job in a french feature animation studio.
For the technical nitty gritty, I used my world orient FK spine again, world orient head and arms. I really like that workflow as it allows me to work in a layered method and keep adding levels of detail even when in polish. For my next close up shots I will be using an IK spine instead but I am still unsure how Bishop handles that.
For the nose I did something a bit fancy that I think is participating to make the face look more organic. In real life, if you observe people talking you will notice that the tip of their nose, nostrils and cheeks keep moving up and down following the opening and closing of the mouth.
Bishop has cheeks controls but with my custom nose, moving the nose control up and down wouldn’t have looked realistic. The skin of the face doesn’t just move up and down shifting the nose, it actually slides over its cartilage and the skull, causing a stretch of the soft parts of the head. (the ears also move actually)
Instead, I created a custom blend shape where the side and lower part of the nose came down simulating the pull of the skin according to the opening of the mouth. The effect is very subtle right now and I would have liked to push it further but after breaking the rig several time and using up few evenings with technical things I think this will do for that shot, time to move on, I think it is a nice close up shot with very stylised facial animation.
Alexandre Belbari, one of the ESMA graduates who worked on the very good looking Mythologique short film just sent me a link to his latest showreel. You will finally be able to see a bit more animation. Really nice stuff, I am looking forward to seeing the entire short film at my favourite Animation festival!
The shots start around 2:46 into the reel, thanks Alexandre!
For those of you who still haven’t added Speaking of Animation to their bookmarks, SOA is a new blog focused on animation and specifically the work produced at Dreamworks.
SOA celebrates its first birthday this month and after the great podcasts with classical to CG animator Ted Ty, How to train your dragon directors Sanders and Deblois…. They are, this month, interviewing my AM mentor and Dreamworks Megamind’s Head of Character Animation Jason Schleifer.
Jason followed a very interesting path as he didn’t start in 3dstudio dos, Lightwave or even Imagine like most of us but instead went straight for SGI workstations and Wavefront softwares just before they merged with Alias and created Maya. What was supposed to be an initial training so he could teach the other students became the stepping stone that lead him to a solid career at Weta then PDI, the San Francisco branch of Dreamworks.
I could definitely relate that that interview as it brought me back to the years where Bay Raitt and Martin Krol were kicking ass and Mirai’s polygonal and edge flow paradigm gradually got adopted… (ripped off was the term used at that time) by all the major 3d software makers, eclipsing Nurbs and Patches for organic modeling. Some feature studios still use Nurbs and patches nowadays but this is mainly due to pipeline legacy than actual superiority.
Back in the day, Bay Raitt was running Spiraloid, the first “traditional CG modeling” forum dedicated to polygon modeling. The forum is now down but you can find a quick video of Bay Rait modeling Gollum on the Lords of the rings’ Two towers second DVD. There was also “One afternoon with Bay Rait”, one of the first timelapse modeling videos that shook the modeling community in 1999.
Regarding Nichimen/Izware’s Mirai and Nendo, I remember a great video by Martin Krol showing how revolutionary the software was but I can’t find it right now. I will post it when I get home , I have a copy on one of my hard drives back in the UK.
Jason is also mentioning a close up shot by my classmate John Comey where one of the characters from his short film is pressing a button. John hasn’t got his shot online unfortunately.
Having worked with Weta, Jason has had to face the usual question about Motion Capture to which he gave a good answer “in its place, it could be okay”. A previous SOA interview on Avatar was very enlighting in that sense. Most animation purist don’t like Motion capture but for certain things like VFX or realistic games there is just no way around. Staying away from those two is actually a good way to avoid using it :-)
The interview is also about how he approaches his role as Head of Character Animation. Many leads, supervisors or anyone wanting to get into management in feature animation can definitely learn from.
At the end of the podcast, Jason brings up some very interesting points on performance, mentioning the “What if” game and how useful video reference can be, not really as a crutch, to use Milt Kahl ‘s legendary quote, but more to bring subtlety or ideas we would have never thought about.
Victor Navone had a very interesting post on video reference a while back where he first blocked a shot from his own imagination then shot video references in the Pixar hall. The video reference had so much more subtleties that he decided to integrate more of it into the final shot.
Jason Ryan interview
Ted Ty interview
Q&A with Jason
There is probably one frame of overshoot missing for the tip of the index and possibly a need of overlap for the other fingers but the next shots will be much harder to polish so let’s call that one DONE!
The lighting is temporary, I am now using Mental ray and this is foreign territory for me. A lot of testing and research will be necessary.
The Area light simulating the sky has been switched to a direct light to get a more realistic look but causing the loss of the soft shadows. I don’t know how to use depth map shadows in Mental ray yet. I will also try to frame the skeleton’s shadow a bit better and brighten it in the next pass. It would also be nice to get the guy’s hand shadow come into screen before the actual hand to prepare the audience for what is coming next.
Ah, I havent used any global illumination or occlusion yet, this requires much more testing than I have time for at the moment, my priority is to finish the animation before Xmas and have an amazing showreel for january 1st.
Regarding the materials. I am using a quickly customised misss_fast_sskin for the guy’s hand. It works fine for that shot but not for the next ones. I have in mind something similar to what they did on “Tangled” (gorgeous movie) or what Malcon Pierce used on his AM short film.
A tweaked leather procedural material has been applied as a bump map to the skeleton to give the feeling of a weathered material, this only shows in Full HD. I will probably need to add some 2d textures to create more dirt on the lower part of the skeleton unless I could use a procedural gradient, is that possible?
For the cinematography, I am wondering if I will use a cross fade at the begining then use a slowly focusing camera or maybe a slight pan, truck in…. we will see.
Ah ah!!! I was about to forget! I have in mind to have a shredded piece of cloth around the shoulder of the skeleton, some sort of bright blue hawaian shirt but I am not too sure how to go about it yet. A cloth simulation and a shredded looking opacity map could do the trick but I don’t know how Mental Ray handles opacity map with Depth map shadows.
[update]I am back to raytraced shadows and the cloth simulation was much easier than I thought :-)
Unfortunately there is still an area where professional riggers, TDs and technical skills are required, stylised animation.
Here is a very intriguing rigging showreel I just found, I have no idea how he does that but that is very exciting!
and while we are on the topic of rigs and autorigs, check out a making of from a short I still haven’t found the time to feature and very similar to “Salesman Pete”: “Meet Buck”. A wacky looking shortfilm coming from Supinfocom and currently touring animation festivals.
Well let’s feature a short video of “Meet Buck” since I will never find the time to write a full article about it.
An interview with Simon Pegg just emerged on IGN. In that interview the Shaund of the Dead star is talking about the great downloadable game I was working on few month ago at EA: “Spare Parts”
Check it out, there are plenty of new clips/concept arts from the game and even ….. some of the cutscenes I worked on!!!! :-)