Final word

The term has ended, we are to hand-in what we have learned and achieved. Although the final film has not yet complete, it feels like saying goodbye to an old friend who has been with us for the past 8 month. I believe a lot of the team member would feel the same with me despite we all, at some point, moaned about it.

Oh well, in the end it is the process that matters the most because I have learned a lot. Although there are regrets and mistakes and varies instants I wished I could just wind back time… in the end, it’s all part of the learning process.

I have been working on two films, ‘Pirates’ and ‘Dummy’, both entitled as the producer. Later I find the job title pretty vague as there really isn’t much of a real producer’s job to do in a student film. Luckily though, I was able to find jobs that other people have ‘left’ me to do.

For Dummy, I joined the project at the very beginning and suggested to have a life-action environment composited with 3D dummies. I think there were 3 reasons,

1. I wanted to develop my compositing skill and put some good work into my portfolio.
2. Frankly, I had just bought a professional camera and wanted to put it into good use.
3. I did not like the idea of making a 3D realism film. In the original story there were nothing we can’t achieve using live-action footages. I mean, what is the point of simply copying what we can already see into 3D?

J.B. the director of dummy agreed with me and gave me and Dean two weeks to make a compositing test. So I did. He liked it, but wasn’t sure about whether the new story which was in the process a re-development would still being able to use life-action backgrounds. That re-development took 4 weeks longer than expected. So gradually my suggestion was dropped.

In the first term, I have spent more time with dummy than pirates. It was partly because story-wise dummy was in more trouble than pirates, and also because J.B. was a very demanding person who had set the standards really high. In the end of the first term, I’ve modelled the two dummies and the room (both took over half-way from somebody else because they have failed to meet the deadline). The project however, never went too far from that point. It is a shame that after Neil and I spent so much time on the story, work out the structure, shots, details and meanings to a final version that is brilliant, in the end J.B. has lost commitment due to personal issue and decided to drop the project. I don’t want to go on too much about other people’s issues but it did affect not only me, but the whole group. J.B. later on wanted to concentrate on Pirates which I welcomed his decision, but I kept my cautions on his new promises. After all, I’ve learnt my lessons.

Pirates! A much brighter subject to talk about. We have produced some fantastic outcomes which I am just lucky to be part of it. Lewis is a brilliant director, a great concept artist and has the vision how to make a film looks good. Although sometimes he was a bit soft on people and got pushed around a little, I have learned how to help him to get the best out of the team. I know it is 10 times harder to say ‘no’ than ‘yes’ to people especially to friends, but the job has to be done. Overall, I’ve been doing modelling, rigging (weight painting), texturing, animating, camera works, lighting, rendering and compositing, and pretty much have organised all the files and folders that we are using. Frankly technically pirates is a very complex project,(the size of the supermarket, the number of objects, 4 different characters with lip-synching and the dynamics.) probably the most complex project in the room, I am very grateful that it has came out neat and OK. The reasons why we have delays are not caused by technical, but personal problems. We faced a fatal blow when our rigger went missing (who also happens to be our main technical person) which left with a huge delay in the production pipeline. But the good thing is: it has also left 3 people to learn how to rig – me, Tom and Lewis. This is surely a project that we have learned the most through out the three year of study.

This term, we have also learned the technical limitation of 3D animation which happens to be crucial, more crucial than we had thought. We have to face the fact that the college computers are almost 4 years old and they do crush unexpectedly. And there is physical memory limit of 2GB which means Maya can’t run some of our scenes (it’s a supermarket after all), which left us to spend more time to work out how to composite them back in. Although those technical problems had caused us to do more work and added more stresses over the top, I wouldn’t say it’s a bad thing. It allowed us to learn how to deal with unexpected encounters. I can only wish this one most importance skill that we have gained – problem solving skill, can be taken into account of what we have achieved this year.

In the end, I would like to thank my team for the journey we have been through. It is an experience that will stack with me for a life time. And let’s hope that we will see each other’s path crossing again in the near future.

The work of the last week.

Shot30.lighting,background texturing

Render farm.

Shot 46, failed version with motion blur.

I didn't quite get a chance to put myselft together recently, because I simply and frankly didn't have the time to. Get up at 7:30am and get back home at 10:00pm, 7 days a week...the pressure is on.

Anyway, for the past 2 weeks, I have been lighting up shots, organising files, sorting out renders and fixing up problems. Problems always pop out unexpected. And we always come to decide whether we should - go back and do it again...or...leave it and move on. Each day is so packed, but so few screen shot has been taken for blogging purpose.

Lighting in Maya. shot 30.

Compositing in After Effect. shot 30.

Lighting. shot 32.

Shot_6 final render.

Lighting WIP Shot_6

This is the orginal fog light preset in maya. doesn't cast shadow.

This fog light casts shadow. If you look at the render preview on the right for details, it actually looks pretty good! (make sure you check out the large version)

I realised the orginal model of the torch doesn't have light balt. This is what it would look like when it rolls in fron of the camera.

I have added a simple plane in the torch. 20% transparent, 0.3 glow.

Turned the intensity down a little, changed the color too.

I realised i had to add some light in although it is supposed to be night. (I joked to Lewis about making our film in complete darkness and call it Night in Supermarket.) This current lighting has three lights in Maya - one spot light casts that stright light(ban-door) across the guard's face, one ambient light for the overall environment, and a additional spot light on the guard's right to brighten up his outline.


The fog light can't cast far across the all shopping aisle, so I had to cheat for this one. Its basically two light, one fog light and one normal spot light.

Shot 46, final.

Shot 46 compelete. Bottles and shelves are both looping. motion blur applied. I kind of feel bad that I have spend too long on this 2 second shot.

Shot 46 Video WIP

Shows the 'light-link' used in the scene. (characters have independent lighting)
and reference editer - most of the shots are done this way, its better when organising files.
and rendering layers - later composite in after effect.

Stuffed but not final shot_46

So tired. I tried to cheat by looting the bottles at the background, it works..(sort of) but just need to have a bit more variety or maybe get rid of the red bottles...(sorry Ash but) they looks awful.

shot_46 new way to cheat

I've just found out another way to explain why having a little presure is a good thing.
Although I haven't put it on the blog, but we have rendered 3 different version of shot 46 - the first shot that we are trying to render via renderfarm. The very first attenpt failed, because we set our standard too high -way too high. So we reduced it and I left one over-weekend last week, there were two missing frames but the rest looks impressive. Then i checked the render time, i was not impressed. An average 8 hours per frame. For a moment i paniced, but kept the feeling inside because all the rest of team member had already jumping like headless chickens. So I paused all the other shots I am managing to render and tried to find a solution to our ridiculous render time. (remend you, its a supermarket after all!!!) The second version, with a new lighting system(5 lights parented to the camera), used about 3 hours per frame. This version, I am so glade to say, used 5 min per frame. (with a sense of relief i was able to blog)

Because for this shot, I have looped the background. It is now only 9 frames. The geometry is slightly modified but Lewis said no biggy, its going to be blured anyway. The two character's are now in different layers and with no motion blur. Montion blur(mainly for patches) will be applied in after effect.