Interview by Robert Ryan

Please give a short introduction to yourself

My name is Tajino , I m 20 and I live in Malaysia. I ve just finished Advance Diploma in Electronic Design & Multimedia in a local international art college a month ago.


How many years have you been doing 3D, and do you have a traditional art background?

I ve been doing 3D for about 5 years now, on and off as a hobby, only seriously in this past few year. I started in late 1995, playing around with Autodesk 3D Studio Release 4. Then I was introduced to Metacrations Infini-D when I entered the art and multimedia college here. The next semester, I got to know Lightwave 3D (around mid-1999) and really enjoyed it until now. Before I was exposed to computer generated art in my later years, I had a background of traditional art. As a child, I enjoyed painting tremendously, especially painting realistic sceneries and imaginary landscapes.



How do you think traditional art has affected your 3D skills? Do you feel that art training is essential in becoming a talented 3d artist?

Traditional art definitely helps a lot when it comes to making 3D computer generated imagery. I believe, a talented 3D artist, or any talented artist of other media, has what is an artistic and creative insight, such as the eye for colors, the mood for a scene, and most of all, the sense of proportions and balance. A good 3D artist needs to have all these aspects in order to help him/her create a more convincing 3D art piece.


What do you consider to be your strong points (animation, texturing, lightning. modelling), and what area do you find most interesting?

I would take modeling, texturing and lighting as my strong points. I enjoy doing them more than anything else. Animation especially in character wise , is still something new to me.


How much animation experience did you have before starting on "A malaysian friday"?

Not much to be honest. AMF was the first of its kind that I daringly tried. The main purpose of making "A Malaysian Friday" was to put myself to a test and also as a chance for me to explore deeper into the limits of animation, doing stuff like soft body dynamics, fluids simulations, and human character animation wholely in the new Lightwave [6]. Weeks before the real production of AMF, I ve to dedicated myself into the above mentioned animation aspects, making several small and quick tests to make sure that they are workable and achiveable in LW6, and that was the time that I ve learnt a lot of new things.


How did you come up with the idea of producing AMF and what are your thoughts behind it?

First and foremost, in the final semester of my course in Electronic Design & Multimedia, we were given a responsibility to research and major in a specific field that we opt. I was inclined to come up with another sci-fi approach because it is not only too common, but doesn t put up as high a challenge as simulating real-life events. This is because photo realism has the real life to be compared and to achieve photo realism is a difficult task. My first thought was to do something that I ve never done before, stuff like fluids simulation, cloths, softbody dynamics, radiosity and other features that were newly introduced in Lightwave 6.0. At first, putting all those elements together seemed like too much to handle in an animation. Then again, when it comes to animating a real-life scene, animating most of the things around us requires all those elements to be involved. Something that simulating real-life can t run away from.

Then I decided to make a typical Malaysian Friday lifestyle of an old man in a rural village area. The settings in AMF were mostly flashbacks from childhood memories of my late grandfather s village house and partly imaginary (but still entirely Malaysian) . As the village was developed into a business area and no longer exist now, I had to rely entirely on my memories to recreate and capture the mood and atmosphere needed in AMF.

How did you approach this project, did you start with storyboards etc. before turning to 3D? How much time have you spent on the project so far?

I did sketch a lot before putting it into 3D. Pencil and paper are still very vital tools to a computer generated artist. Rough images of the traditional Malaysian village house architecture, furniture and layout, the old man, the type of clothing as well as little details like the hinges on the netted wooden door, the metal bucket (I learn how little details are very sentimental because you may not be able to find them now), and whatever remains of my memories on the atmosphere and sentimental details were quickly sketched down onto paper as soon as they flashed in my mind. Though sketched out, they were only the roughs, and the main blueprint and almost exact architecture plan was in my head all the while.

Collectively after a few weeks, I managed to get a better picture of how AMF settings were going to look like when reconstructed in 3D not so later. The storyboards that I did were quite rough as in I knew that my ideas of camera directions will keep on changing as I progressed. So far I ve spent about 4 months in AMF not inclusive of the time I m taking now as holidays before continuing the next few finishing shots in AMF next month.

In what order did you finish the movie?

I started AMF by modeling, texturing and rigging of the old man, followed by the creation of important props such as the banana bunch, the bucket, etc inclusive of textures. I prefer working on objects/item basis in Lightwave rather than doing them by stages as like all modelling in one go, all texturing and so on. I think I did that mainly because I just can t wait to see how each object will look like and behave when it s finished. Most of the objects, texturing and lighting were scene/camera specific, whereby the character setup were tweaked according to the actions needed to perform in the particular scene, and this is where Lightwave 6 non-linear work flow comes in handy.

The lighting in AMF is fantastic, can you tell us a bit about how you approach the lighting of a scene?

As far as I worked on AMF, lighting means a lot, not only to bring out the details, but also to recreate the mood of a typical hot and humid Friday, aiming to make them all look as real as possible. My first thought of lighting up AMF was to use fully radiosity the entire animation. However after a few test renders, I found out that it just wasn t time for that yet with the amount of processing power that I had at that period of time. Radiosity will only be possible if only I had a render farm of 1000 x 700Mhz or higher machines in order to beat my deadline for this assignment. As a solution to the lighting issue, I ve made few radiosity rendered stills only as reference and try to faked them as far as i can.

Most of the lights used in AMF are spotslights and point lights, swarmed together to form lighting schemes which is good enough to fake radiosity. Area lights and Linear lights were kept to mininal usage as they take up a lot of processing power. Alternatively, spinning spotlights with shadow maps and moving pointlights were used in almost all the scenes to simulate soft shadows. Besides, a bunch of negative lights are also very useful to achive subtle dark spots in corners of walls.

Apart from just using lights to light up a scene, luminousity map and diffuse maps plays an important role in pushing the lighting in AMF further as well as cutting down rendering time. The average rendering time for one frame, with low-antialias motion blur (5 passes) at PAL D1 resolution was about 15  20 minutes.


You have used Radiosity in some of your still images, what where the avarege render time on those? Do you think radiosity will be the norm in the years to come or is it better to fake it?

The still "house_13.jpg" (image on your left), took about 18 hours on my 700Mhz P3 320Mb Ram machine. As long as the processing power keeps on increasing, together with the improvement of rendering engine, well i guess radiosity will most probably be a norm in the years to come. However for the time being, radiosity baking  which is possible thru Lightwave 6.5 is a real plus when realism and time is a concern.

The character in AMF is pretty much fully clothed and i`m sure you used MotionDesigner for this task. How hard was it to make the clothes behave and look real? Do you find MotionDesigner adequate for such tasks?

Motiondesigner is no doubt a very impressive piece of plug-in for soft-body simulation, however to achive the desirable effects, a lot of planning and patience are needed in advance. It wasn t as simple as I thought of at first when I tried out MD in Lightwave 6.0, MD performs extremely slow and sluggish when high polys are thrown in with collision detection on. As a work around to this, only parts of the low polygon character model that were needed in the collision detection were taken into MD. The speed and interactivity of MD was improved by using this approach of simulating the T-shirt and Sarong separately with their own collision objects section instead of throwing in the full body model. Speed and interactivity are the most important thing that we need when making minor adjustments to the cloth dynamics and getting feedbacks without having long waits and frustrations. The low poly cloth was then smoothen with Lightwave 6.0 subpatch.

For the tasks needed in AMF, clothing, sarong and even rope on the bucket are adequate with Motiondesigner as long as we have the patience tweaking the look to fit our own needs. Unfortunaly, I m still really new in character animation and the old man wasn t very up to standard. It would require a lot of time just to improve on the old man alone and back then, I just didn t have the time and too much other details to look into.

You have made the trees and botanics in AMF look and behave very lifelike. Can you tell us in a few steps how you accomplished this?

First of all, you ll need real leaves to make real trees. In AMF , I snapped a few leaf samples with a digital camera and used them as texture maps which can be found in the textures  section. Based of the texture maps, I made an alpha/clip map. Then I created the leaves plane and attached UV to it and put on the texture. By cloning and randomizing the orientation of the leaves planes and laying them onto the branches, you ll be able to get a nice tree with an acceptable amount of polygons.

For the subtle wind blown motion, multilayered fractal displacement map was applied on the leaves meshes. Motiondesigner was used in the early stage but found out to be too heavy when comes to putting together all the botanics for final scene construction and rendering.

Do you have any plans for your next project after finishing AMF?

After finishing AMF, most probably my next short animated film project would be The Death of a Stop Sign , a story written by Ahmed Balfakih , the Managing Director of Clickgrafix (www.clickgrafix.com). The approach for this incoming film is not yet decided.


So, where do you see yourself in 5 years?

Alive, healthy and happy, and still doing 3D as usual.

Don t you wish to see yourself very successful in 5 years time?

Being successful in the 3d industry is included one of the criterias that will make me a happy (mentioned above) and contented person. I will try not to rush into it. Many people is telling me to enjoy the joyride up success. I get constant reminders (this is a type of reminder that you just can t unsubscribe or put a stop to, unlike bills  where I just have to be punctual at paying them to stop the authorities from sending more) to stop, take a break and go out to enjoy food, to experience other things to make sure that I know they also exist in life. And that the air that we breathe will not be the same 30 years later, thus don t just breathe, but to take time out to really feel the air, savour the good food and realize the life around us, which 30 years ahead, no matter how much time I have then to enjoy life, money cannot buy back the air and the life that is around me today, not to mention the good food.

And yes, local Malaysian food is great, but hurry coz these traditional food is slowly dying though the years.