Today’s 3D class was rather fun in that we learned how to use music to influence objects in the 3D realm. Always a fan of music interacting with digital media, I enjoyed seeing what sort of oddities could be created.

[Updated: Now with short video]

The audio tornado was the product of playing around and trying things that may or may not work. Often, it seems that some of the most interesting and most creative designs come out of such play. I feel that this composition is no different.

Focused around the usage of the “Sound Effector” in Cinema 4D, we spend time today experimenting with the MOGRAPH module. Having placed the effector onto a radial Clone Object (of cubes), I set up the altered properties to be size (the cubes get wider and taller) and y-position (the cubes go higher up) dependent on the graph representing the music. From there, I added in a Warp to the Clone Object (resulting in some very interesting geometry) and ran the file. The creation? This audio tornado. The film version will show the parts actually moving, but it is still rendering, so check back later to watch it! As it is, the inside part of the cubes gets squashed while the outside remains wide and the positioning of the cubes, once in a perfect circle, is highly variable and ultimately generates an image that looks like a tornado.

Finally, I added in some special lighting and there we have it! I’m excited to see how the video turns out (it was left to render overnight). All in all, I like how the piece turned out, even if it was more experimental. I need to recheck the export settings from Cinema4D as the image came out very low quality (it looks better above with some tweaking in photoshop) with lots of color banding. In any case, I thought it was a fun piece and something folks might enjoy!

The video was a little bit of a let down as things seem to move a little too fast to really be visible. I think the static image up above is a good bit more nifty! In any case, I thought I would show it as it is rather interesting still!

Yesterday in 3D class, we learned about some of the tools in the MOGRAPH module of Cinema 4D. After some play, I created this rather awesome look that I feel is rather reminiscent of Indiana Jones and the Fate of Atlantis and if you are interested, I’m selling these fine leather jackets.

This is a small image that I made using some crazy geometry in 3D. We were learning how to use the Cloner Object in conjunction with the Step Effectors.

First thing we did was create a grid using the Cloner object. I made one of 19 by 19 tubes since I rather like their shaping. From there, we were taught how to use the Step Effector to influence and modify the attributes of the grid items. In this case, I changed the scale and y-position so that ones being changed by the Step Effector would rise up higher than the rest of the grid and get bigger. Next, I changed the Step Effector to including only a spherical region and moved it to the middle of the grid so it would effect a circular region of tubes. Around then, I discovered that one can affect the falloff of the Step Effector based on a graphical representation. The falloff controls which areas are being affected by the Effector. So, I created a graphical model such that the most affected and least affected regions had no change while the mid range was maxed out in terms of scale and y-position. The Step Effector also applied the red color to the image. However, I found that red alone wasn’t enough, so I added a Shader Effector as well with a yellow color and reduced strength. Thus the yellow would be more dominant in some areas while the red would be more dominant in others. Finally, I added a visible light into the middle of the pit with area shadows to create the dark, flaming effect.

I really like the overall feel of the image and really feel like there is some boiling lava at the bottom of the pit. I wish the tubes lined up a little better rather than showing some odd lines around the edge, but that isn’t such a big thing. It was fun to make this!

This previous term (term 4), the main focus of the 3d class was on creating interesting and realistic models. While I feel that my work on the pirate ship from the previous term ultimately was more rewarding and interesting (I still need to locate the final version of that file. I’m afraid it might have gone missing), the work I did this term was fun and ultimately created a more realistic pair of projects. Hit the jump for the two.

Coca-Cola can

The first project we worked on this term was a Coca-Cola can. We were given the outer texture (the can label), a can of soda and told to get started! The first task was to create a model of the shape of the can. I didn’t find it particularly difficult to do, but ultimately I had to change around some of my splines (the lines used to create the can), which was a little troublesome. The process was to first go into illustrator and create a cross section of part of the can; such as this:

From there, we could import that into Cinema 4D and spin it to create a complete can shape. Next was the top of the can.

This part took a lot more effort to accomplish due to the complexity of the elements there. The top containing part of the can wasn’t as difficult, although it took a bit of effort to get right. Originally, i hoped to use the same technique as the body of the can and made this:

Unfortunately, that didn’t mesh well with creating an indent for where one drinks the soda from, so I ended up using a simple loft NURB to generate the shape (a loft NURB takes a set of lines and creates a shape out of them). I made circles and placed them at different depths, then linked all of them together using the NURB.

The can pull, on the other hand, was significantly more time consuming. I probably spent as much time working on just the can pull as entire rest of the can…

It all started with this shape, which took a while to get right:

From there, I brought it into Cinema 4D and implemented an Extrude NURB. Extrude NURBs allow one to give width to a spline, making it 3d instead of a simple 2d shape. Still, it didn’t quite look right with the harsh edges, so I applied fillet caps on the shape, making the edges a bit softer. I tried using a HyperNURB, but found that the softening was far too great for my purposes. In actuality, I had not learned how to use the HyperNURB properly; an issue remedied in the next project this term. Finally, I pulled down the central little half-dome there to give the impression that this was the section that was attached to the can.

Now that the modeling was complete, it was a task of adding on the materials and lighting the creation. The materials took a bit of time to add as I had to fiddle a great deal with the reflections and graininess of the metal for the can. The label went on super easily. Then, in preparation for the lighting side, I arranged them all into the shape you see in the final version and inserted a camera with depth-of-field to give a macro-like shot.

The lighting turned out a little differently than intended. I wanted to have a small light between every can, but ultimately ran out of time and they weren’t working properly (each render of the image took ~40 minutes), so the primary lighting comes from one large light behind the camera. Finally, I dropped in global illumination (creates color bleeding due to light) and ambient occlusion (which darkens out the global illumination where it isn’t appropriate).

Ultimately I was rather pleased with how the whole thing turned out. I feel the material on the metal parts of the can could still be improved as it seems much too grainy for what I think a can is supposed to look like. Otherwise, I could spend several more days on that can pull-tag, but I’m pleased with the look as it is.

The Hammer

For our final projects, we were to select an object to model and create an interesting perspective on. I was in a rather difficult position with a number of other, large time-consuming projects and a desire to make something as cool as the pirate ship. Instead, I decided to develop a final project that challenged me to do something I had no idea how to do, but also contained a simpler element to it. Thus, the hammer was chosen. The main body of the hammer is simple to model and create a material for, but the head is very difficult to get looking good. In addition, I had no idea how to even make the head, so I hoped it would be a good learning experience for me.

The body itself was a rather simple exercise in using the loft NURB with concentric circles to create the effect of the molding on the handle. It is a little hard to see, but I ended up using 3 different loft NURBs to develop the body. The first is the shiny “neck” part of the hammer. For that one, I used just two circles; one more stretched vertically and one more round and let the NURB do all the work connecting the two. The second is the “body” of the hammer that extends from the end of the shiny part to near the end of the handle. That one has different sized circles placed appropriately to give the sense of molding over the length of the handle. At the end where it connects to the shiny “neck,” there is a much smaller circle so that the handle will be flush all the way. The final part was the “tail” which finished off the hammer handle. Originally, I had included the “tail” part into the “body” section, but ran into a difficulty. The entire handle is very smooth, except for the point near the bottom where it has a sharp corner. Thus, the best solution seemed to be to split the two sections and place them right on top of one another so there wouldn’t be any split between the two. All in all, I really like how the entire handle and “neck” parts turned out. I think they look rather realistic. Then, there was the difficulty of the head.

To create the head, I found a tutorial online (for 3DXMAX, but the concept was the same) and adapted it to my particular hammer. It started with the end part as a square and used the extrude and bevel tools to extend one side of the square outward. Ultimately, using these tools and a knife tool to split the side for the nail removing section of the hammer, I was able to create a nice, blocky hammer head. From there, it was through the difficult implementation of HyperNURB that the hammer head was rounded. The way HyperNURB works is to round all edges of a shape based on a weight assigned to the edge. When one just “apples” it to a shape, the weight is a consistent 100%. However, that can create a rather gross-looking shape that will cause one’s compatriots to laugh and make crude jokes… In any case, it is possible to affect how much the HyperNURB weighs a particular edge using a special tool in the modeling toolkit. Thus, I was able to give back much of the shape to the center and back end of the hammer head, while keeping roundness on the front section and attach point of the hammer head. All-in-all, I spent a lot of time tweaking little sections in an effort to make it look more realistic, although much of it was in vain. Finally, I added the material to the hammer head. I wanted to imitate the real hammer I was modeling off of, but wasn’t able to replicate that material. Instead, I worked with creating a material that had a little reflection and a little grain. What ended up coming out when the hammer was arranged was the odd discoloration and grain on the final image. The grain effect on the front of the hammer was entirely unexpected, but rather advantageous to the final feel of the image.

In creating the final image, I included a low angle camera and a nail into the composition, hoping to create a more dynamic feel to the image. Ultimately, I added a “sky” to the image as well to give the sense that it is taking place outdoors. Amusingly enough, using Cinema 4D’s render engine, I got a really weird effect where the head of the hammer wouldn’t show up:

Neither myself nor the teacher ever figured out why. It might have been something to do with the tremendous amount of computer effort needed to make sure the head was rendered even in preview mode (I had some other effects, especially on the back of the hammer, that slowed down the rendering significantly).

In any case, I feel that the image came out “ok.” It wasn’t particularly special, nor was it terrible. I think the body of the hammer went really well and I like the camera angle I chose, but ultimately I feel I didn’t have enough time to really do what I wanted with the model and composition. I got frustrated with the head and my own inability to get it to model the way I wanted it to look, but I feel I got a good introduction to this new method of modeling an object and I can use my experiences here to create more realistic items in the future. There was also some issues with the camera. For whatever reason, it didn’t want to focus properly onto the front of the hammer head, so I got it as close as I possibly could before saving the image out. Finally, due to the weird rendering issue I pointed out with my last image, I had to print-screen out the image and crop it in photoshop… Overall, it was a bit of a frustrating project, but I did learn some new methods and techniques in order to complete it and feel that I could definitely do better work in the future.

“Twenty three” was the grad project for one of the VFS writing students. The writing student’s scripts were produced by the VFS film department and we were given the task of developing title sequences for one. Myself and Asuka picked “twenty three” out of the bunch and developed concepts for it.

(I apologize about the choppiness of the video. I’m not sure why it is so choppy and am looking into how to clean it up some)

The film itself is about a boyfriend / girlfriend that are about to meet the lady’s parents and talking about some of their fears in meeting them. All in all, it is a rather funny little film with a bit of an indie feel to it. To adequately capture the feeling of the film, we struggled greatly with concepts. Eventually we settled on the idea of a roadtrip as one of the lines from the film indicates that the couple just drove half-way across the country. From there, the idea of using a map was pretty straightforward. The execution, however, took a lot more effort. Building the entire setup in Cinema 4d. The pen and camera moves took a while to work out properly (and as I remember, I lost the file partway through and had to redo them). However, the real kicker was in the rendering.

In order to achieve the motion on the pen ink as well as the other additional elements (the drawn-on heart, coffee stain, crumbs and sun), I needed to access them directly in After Effects. Since the methodology of export from Cinema 4D is flattened image sequences, I needed to attach an object buffer onto each element I wanted to animate and render it in particular. That rendering generally took around 3 – 6 hours per object. I also had to repeat the rendering of a number of objects repeatedly due to problems or mistakes. All in all, I believe this project had somewhere near 3 to 4 days worth of rendering put into it (that’s hour-wise). I wasted a lot of time and learned a lot about the best way to work in Cinema 4D, so that was really good, but it was awfully frustrating. Extremely frustrating. However, finally, finally, all the rendering was done and then I had to face the spectre of rotoscoping all of these individual layers in After Effects. That itself took a significant amount of time, but I’m rather happy with how it came out finally. At last, I exported out of After Effects with a few additional effects added to heighten the quality of the work.

Finally came the audio. As a student of VFS, we have access to an online database of music from a company called 5 Alarm Music. I generally like to produce my own music for these sorts of projects, but for this one, I didn’t have much time available to me and I was looking for a very particular sound to the music. Searching around, I found the one you hear in the background; a nice piece of “traveling” music. At last, I added in some sound effects to reflect the car and traveling nature of the title sequence and produced the final piece seen above!

In terms of things I feel could be improved, I was never particularly happy with the shadows of the crumbs as they fell. At first, the shadows are all squares, which didn’t make sense to me, but I didn’t have much time to fix it and I had already gotten the timing right. Also, I had a lot of trouble with rotoscoping the pen ink to the pen itself; an issue that thinking ahead would have solved somewhat. The main stroke of the pen especially was difficult to match up and I eventually just re-rendered a part because I didn’t want to do rotoscoping for it (and it would have looked bad anyways). Also, I feel that the coffee stain could have been better, but I wasn’t sure what to do and ran out of time. Finally, I wish I could have added more texture to the map itself. I feel like it is a bit clean for the piece and the film itself, so a little more texture and “dirt” would give a better feel.

I think my proudest moment on the project was figuring out how to do the pen stroke. Originally I had a lot of trouble conceiving how it would be best to make that stroke and be able to animate it at the same time. Then, I was hit by a stroke of genius! Taking a screen shot of a top-down view out of cinema 4d, I made the stroke in Illustrator, exported it into Cinema 4D and applied a extrube NURB to it so that it would be a solid. From there, I attached the object buffer to it (to create an alpha channel) and rendered it out to bring into after effects. In after effects, I applied the alpha channel to the clean render to be left with just the stroke! Then I can apply a mask or whatever to the footage and all the camera angles will be kept perfect. I liked the technique so much I applied it to three other objects in the piece. That one solution made my day on this project. It was wonderful.

However, all in all, I’m rather pleased with how the project turned out. It was great and I had a lot of fun figuring out how to work between three different programs to bring the whole concept together. I hope you liked it too!