Animation

The animation was very fun to do. And I felt that it required a lot of directing as opposed to designing.

The camera angles and movement of the models took a lot of thought as to where and when, and also why. for instance having the camera low, made my models look a lot bigger as they came to the camera, making them look fast, big, and powerful.

At the start I decided to make a simple starting line and lights, to show the audience what the animation was about, a race....

The idea of the lights changing similar to formula one, was to allow the audience to understand it was a race, and therefore get them to relate the two. The middle ships engines move forward slightly before the 'ship' itself, to give the audience the anticipation that it is going to move forward, But it is hard to notice in the final animation, and I believe I animated it slightly to fast so people cant really see it, it would have been better to make it slightly slower so people could see it. 

throughout the animation you can see the ships travelling at pace, with a high angled shot of the camera following the ships. this was to show the movement side of things, relying on the shadows going over the uneven terrain to show the speed, as opposed to making objects for the ships to pass, this was a good shot and the path constraints on the models and camera worked really well. 

I am happy with the fire effect when the ship crashes, but felt there was not enough reason for the ship to crash, just before the 'lazer' passing between the two engines is animated to flicker, like it is loosing power, but this isn't as visible within the shot, even after slowing the shot down.

The issue with the fire effect was placement, having used perspective view when placing you can see I did not place it precisely on the floor which would have been better, from the high angle shot it looks fine, but from the slow motion replay you can just make out it is above the ground. This would not have been acceptable, and if I had planned my animation better would not have happened. 

The Last shot of the ship slowing down uses the idea that the fins pop out from the ship first creating air resistance which slows down the ship and then allows the back half to lower. this was done to show more of the ship and what it was capable of, I would have like to have used the reaction manager but felt I was not confident enough after trying the exercise to use it within the animation. The fins are very useful and all come out to 45% from what they were. It would have been good to animate these within the turns to show how the ship reacts to a corner, but again didn't feel confident with the key frames and animation at the time to do that. In the future better planning would have allowed me to do this at the time, using a reaction manager for speed and rotation angle would have been good.

the last image shows where the bones went wrong, but unfortunately due to render times I had to leave it how it was, and also in other shots it is working correctly, so didn't understand how to change it. 

The animation was edited in windows movie maker,  the clips were compressed originally and I had to resize them too 800 X 600 in windows movie maker, I then had to dowload a free converter to change the file into an AVI file, I chose not to compress it further as it did give me the option as I didn't believe it needed recompressing.

Strengths Weaknesses Opportunities Threats

Overall my Models and Animation have been a great learning experience, I had no previous experience in 3DS Max but had heard and seen a lot of work from it.
I have Both enjoyed and hated the program at times but understand it is a very in depth modelling and animation program and wish to use it in the future, There is a lot I still don't know about it and believe there is a lot more it is capable of.

Strengths - 

• modelling advanced shapes and models, using flat diagrams and pictures.
• Creating Models using Components not one large model, this is mainly through understanding the model itself, and the components before the modelling even begins.
• Texturing components and having a good understanding of what to model, and what to leave to the texture. 
• Importing and exporting textures was simple, and had some very good outcomes.  e.g. the sky texture doesn't have a line down the middle at the beginning and end of the texture.
• Animating the models, I feel the animation side was great, and I got to grips with a lot of camera angles and techniques. I am probably most happy with the path constraints, and also the idea that one of the ships engines are attached by lines which actually bend and move using bones. It also allowed me to move parts of the ship slightly before others to give that sense its going to move. 
• Details on the ships have been both been one of my strengths and one of my weaknesses, the picture below shows some great detail on one of the models.  

• Polygon count, I feel the polygon count was acceptable for the project, as I had to have a large landscape as I wanted the ships to move with speed the terrain itself used a fair few, as I wanted an uneven realistic surface. As the landscape was quite big, it also made the sky dome quite large, which added to the count, but overall I don't feel it was too many, and I also would have cut it down further using textures for more aspects of the models then I currently have, but that does all come down to how close the camera gets to each model, and if it was to be used for other purposes... 
• Animating moving parts the model was easy, using the graph editor to make the engines continuously spin was simple and very effective. 

Weaknesses -

• Texturing my models proved harder then I expected, mainly due to the lack of skills in 'GIMP' or Photoshop. I am happy with the end results, but feel I could have spent more time on this side of my models to create a more realistic, and better models, they look to 'clean' for models which should spend there 'lives' racing around a desert.
•  Bones were a problem to begin with, spending a large amount of time getting them to work, in the end it being a simple case of re modelling the object they were attached too as it was a line and not a cylinder. 
• Finishing touches, I believe the models all though good, are not excellent, I would have wanted to add a lot more detail and possibly interior to each model, to ensure they looked correct from every single angle, there are times where you may see inside one of the engines and see something you don't wish too. The picture below shows three errors, all can be easily corrected, but the issue I had was noticing these in the first place, having rendered the scene it was only then I realized the issue, and at that stage felt it was unproductive to go back and change it and wait for the rendering again. it is also not so noticeable in the final animation. 

• Lighting I found to be an issue mainly due to editing models after I had excluded them from a light, therefore changing there grouping etc... making them included again. This created a very light looking render at one point with no realistic shadow. 
• within the animation there are stages where I feel it could have been improved, at times I lost track of seconds per frame, and this has resulted in the ships travelling at different speeds at times, It would have been better to measure a distance per second(or frame) to ensure they traveled at relatively the same speed throughout. Also during animating there are times where the ships may pass through the desert floor, this is because it is an uneven terrain, and as it does the animating for you, at times it is easy to miss the models doing this. 

Opportunities - 

• 3DS Max is a very good program to use for modelling, and animation, I feel I have not fully discovered its capabilities and want too look into the program further, but it is obvious it can be used to create very realistic still images or models, and animations. 
• Creating textures for the models could really make or break a model, Too have an artists help would be extremely beneficial to the models, someone who knew how to create anything within Photoshop would allow you to create very beautiful looking models. I wanted to make each ship look very sand dusted, but this proved too hard for me, but I believe a visual artist would be able to achieve this well, it would then be up to 3DS max to create the sand blowing across the desert, possibly with a particle system. 
• I would have liked to have used the reaction manager, so that when I turned the ships, there fins etc would have moved to create the idea they are helping the ship turn, but I spent a long time trying to make the bones system work, which meant I ran out of time to try it, Looking back this would have been a clear opportunity to improve my animation, using the reaction manager on each ship to make them move more realistic and not be so rigid. It could also be possible to slow down the engines as they slowed down etc... 
• The environment itself could have been better, I ideally wanted to create an arena they race through similar to the movie, but felt It was a bit too much work, as I wanted the animation to mainly focus on the three models designed. 

Threats -

• 3DS Max often crashed, losing work was an issue whilst modelling and animating.
• Importing and Exporting models into other files created an issue with textures, as They had the same name, renaming these textures from the start would have been the best method.
• Bones are good to use, but can bend the wrong way, in the final shot in my scene I can see one of the ropes attaching the engine to the ship bend up not down, I found it very hard to try and change and in the end gave up, the issue was that it was working the other way in previous shots and couldn't work out what had changed... 
• I feel my original storyboards were far too ambitious for my final product, Not realizing how hard it was going to be to animate and create a good finished product. In future I will continue to aim high in my work, but ensure it is achievable with hard work, As the storyboards were so ambitious I couldn't use them how I wanted too, using them more for camera angles etc... rather then following it exactly from scene to scene. 
• The Models themselves were harder to make then some, having chosen to do a pod race, I found there to be less material available online to work from to create realistic looking ships as to those in the movie, where as if I had picked some of the better known space ships from star wars, there was a large number of plans and drawings available. 

Creating an explosion

There was a few ways of creating an explosion in my scene, searching online there were two main methods to creating an 'explosion' one was using a PArray, found in the creation tab, under particle systems. The problem with using this on my models, was that because the model was moving, the particles also moved with the animation, and I could not find an effect way to stop this, I could start the Particle system at the correct frame, and even though the model was still there was still issues with the particles continuing to travel.

The second option was to use a 'Bomb'. This is found within the creation > Space Warps > Geometric/Deformable > Bomb.

Using this technique was ok, placing a bomb within the model and attaching it too the model I wish to explode. But I could not get this effect to work with the animation, So have left it as something to look into further. 

I found the most effective way of creating an explosion was too use a Sphere Gizmo, and adding a 'Fire Effect' to this. This created a very good explosion which could be timed and placed where I wanted within my animation. The model simply disappears as the explosion happens giving the appearance of the model exploding. 

lighting continued and Path constraints

Lighting my animation has been slightly tricky, with the outcome changing a number of times, the Omni lighting used was added to the scene, and everything except certain groups were excluded, the issue was if I then un-grouped or created something new, this would not be included in the exclusion. 

It is clear lighting and camera's should be set up after EVERYTHING in the scene has been created and modified, this will save time and ensure nothing is lit up by lights that should not be. 
It had a large effect on my first animation with materials that changed color in the animation being very shaded due to the lighting around it. 
I overcame that particular problem by creating another light at the end in front of the materials changing color so it could easily be seen by the camera.

I changed the lighting used from an omni light to directional spotlights and excluded most of the shapes other then the parts I wanted lighted up. My decision on this was based on the omni lighting coming through the model from certain views, where as the target spotlight lighting does not do this. Giving a better look within the animation. 

I started using path constraints with a camera, as described in an exercise, and then used the same principal to move the models within my animation, this allowed me to use the same path for each model and the camera, to create a moving shot of all the ships and the camera. This was exactly what I needed to show the speed of the ships which made it look a lot more real. 

I didn't have any issues with the path constraints, except applying a path constraint to an object and then attaching a path, this made the object unable to move. 

skinning and rigging

skinning and rigging has been quite fun to learn, and seemed very easy from the exercises I attempted. The issues I had were then re creating it within my design.

I wanted to achieve a very similar effect as too the cylinder exercise I previously did, with a cylindrical element bending when required to make the line which attaches the engines to the 'ship' look more 'flexible'.

The First issue I had was the bones were far to big for the lines, as they are quite thin and long, This meant re sizing the bones to fit within the area required. That proved to be simple to change using the bone parameters. 

Having added the IK solver it appeared to be going smoothly, and thought this would be a really quick and easy task to accomplish. The error happened after assigning the skin modifier to the cylinder, it simply wasn't working as expected, and rather then bending with the bones, the bones bent correctly, but the cylinder simply moved in the direction but stayed straight. Looking through the 3DS max help and tutorials online I couldn't see what I was doing wrong, and in the process changed and ticked a lot of unknown boxes trying to figure out the problem. 

I eventually reloaded the project to an earlier point and decided to run through it again, this time using a new cylinder, This time it worked correctly. The issue was the original cylinder was actually a 'line', which had been modified with the bevel profile modifier, this obviously created an error with the IK solver making it behave in a different way. So I had to delete the original lines on the ship as they were not a mesh which could use the bones technique and recreate them as cylinders and then re-size and move them into position. 

The end result is exactly what I was after, and I simply need to replace the texture. the bones have been attached to the holders on the ship and engines, so when the engines are moved the lines bend as shown in the picture above, this makes it look tethered and will be much easier to animate this way. 

lighting!

The main issue with lighting I have encountered is due to the sky dome added before hand. It has created a slight issue with what the light has an effect on.

The best way round this was following a tutorial around 'omni' lighting, and then adding the sunlight. It took a little while to get my head round the idea of a sky dome, and I have noticed the size of the sky dome is going to have to be very large for most designs. 

The sunlight appears to be very useful, but also harder to control then most of the other light options, placing it by compass and time etc as opposed to manually positioning. 

I have found a few tips that really help with using a sky dome, and lighting it. 

the best thing for a sky dome when working is to go into its options and cull it, so you can effectively see through some of the sky dome, allowing you to see what is going on in the scene a lot easier. 

it is also a good idea to do the same with the plane. There is a lot more in depth lighting options i wish to look into, as i want to make specific parts of my models glow for example, this uses a specific render-er known as a mental ray render-er  but, when rendering the scene at the minute it makes everything black and white, so i will have to research further into this before i can use it, it will also add a lot of render time to my animation later if i do choose to use it. 

Omni lighting is the best lighting solution I have found so far, and Can change the color, it could be possible to use this and focus it independently on the parts of the models i wish to stand out and glow in a different light.  

additional lighting -

I have now added sunlight, using the sunlight and compass, an omni light to just light the sky, and 2 omni lights as engine outputs for one of my models, it has added to the rendering time, but is a lot easier, then adding a mental ray render-er, which was a little bit too much too look into for this project, and I believe would be more beneficial for still shot rendering then animation anyway. The lights give off a nice glow for the engines. the sunlight gives good shadow for the ships as well, which shows they are hovering and will help show the speed as it travels across the sand, and pick up on the hills and dips in the environment. 

texturing

Texturing has proven to be an eye opener. 

I didn't think texturing would be so hard, but to produce good looking textures takes time, and effort. 

From 3DS Max it is possible to Map components of your model with the UV Map modifier tool.
This is a simple way of telling the texture what the object is like and how to wrap round it, using the gizmo sub 'object' of the UV map. But, you need to have a good texture to place onto the model in the first place, or its not really worth it!

I also went further as to unwrap some of the more advanced components. this is done by applying a UV map, and the the UV Unwrap modifier tool, this allows you too select faces and then unwrap the model flat, creating an image I then imported into GIMP too really allow me to add more detail to each individual polygons / face. 

Problem - Creating the models originally I feel I have no created enough separate components. Meaning it is a lot harder to make sense of the UV unwrap! especially as I wanted different sections to have a different texture on it. 

Solution - The way i got around this was too detach the sections creating separate components from them, this would not be the best way to do it, and I understand now that i should have created separate components in the first place, as to me this is a very messy way of doing it. 

Using GIMP is good, but I have not used it for a while, so have found it difficult remembering the tools, you do need to be a good artist to create good textures. 






The textures themselves are very easy too apply after the uv map has been applied, and quite often if the component is mainly one color  it is better not too add the unwrap complicating the process. 

the details - all pod racers

Each of the pod racers I have created have undergone further development with details on each pod racer. 

these were created using standard primitives and sizing shaping. One of the best techniques used here was snapping to grid, and also using the snap tool, but snapping to a face as opposed to snapping to a vertex or edge. This allowed me to create simple details on the side of engines very fast, the only trouble being with curved faces, the best way i dealt with this was too use the shape line and manually add a vertex along the curved face, whilst snapping too it, allowing it to connect to the face at a number of points kept it close to the original curve. 

the finished pod racers are good, but I personally would not add so much detail to each model. In reality I would texture some of the details on as opposed to modelling them. this would save on the number of polygons, BUT I also understand that within the game if someone needed to use a button, or if the detail required warranted it to be modeled on, then I would model it on, for example if one of the pod racers was seen very close up by a character, or is used often, it would warrant the details to be modeled, but if it was only seem from a specific distance, where the level of detail could be covered in a texture, I would assume the texture would be the preferred method. 

the details added would be of use within the final product, and is not stupid amounts of detail.
it is more items which could be used and seen close up in detail. changing the look of the model.

the third pod racer

My last pod racer has become much better then expected, This is due to my increasing knowledge in 3DS Max, looking back over my models I believe I can see the pod racers improve as I have improved with 3DS Max.

For this pod Racer I wanted to create a 3D studio within 3DS max using three different images of a pod racer, showing each profile, front, side and top. This was so I could ensure the pod racer was too scale and correctly proportioned.

This was created using planes, alligning, and then adding the pictures, The pictures orignally came from one picture showing all three profiles, which I found on the internet. I then cut out each profile in GIMP. There was a slight issue as one of the profiles was not the same size. I had too physically stretch the front image in GIMP to ensure each file had the correct proportions so it would line up in 3DS Max. failing to do this leads to all sorts of problems, which I found out from experiance when I tried to model a Plane.

Using Alt + X to turn the selection see through made it possible to really create the shapes seen from the drawings, BUT it is also important to remember its 3D, theres no point creating a square from the side view, if its round in the front view, You must have a good understanding of the drawings and the object you wish to create.

I found this section to be very tricky to get right, it was a lot of vertex movement manually, and also cloning, shrinking, cutting, and then cutting into the shape to create the desired area shown in green. the engine itself was quite simple other then that, I specifically chose less segments to each curve to get a better balance between polygon count and look.



The engine used a lot of standard primitives, the 'rings' being made with tubes, the exhaust made from a cone, and a chamfer cylinder, the front a sphere, stretched and shaped. the issue was I created the outer 'shell' which is red from the drawings, but missed the inside bit, shown within the circle, I created another curve using a tube, and then had too attach this onto the larger section and manually move the vertex to where I wanted them, I did this using the Boolean tool and union, but found a lot of vertex were left, I deleted these and then manually added faces to fill in the gaps which I had created. This was a lot of work, and In future I would have continued the curve round and manually moved the vertex and deleted where I wanted. The rings seen and the 'fins' at the bottom and back are clones, saving on the work load where possible. 

The ship itself seemed simple, but a few errors here and there made it harder then I thought. during construction I believed it to be symetrical, using tools and selecting edges which ensured to me it would remain symetrical, such as the scale to along one axes rather then manually moving edges. At some point I must have seslected vertex unknown to me and moved them slightly, as towards the end I realised it was not symmetrical.

As I had done more work I couldnt find a suitable way to go back, so I decided to carry on for now, knowing I could cut and mirror the ship later.

Things that went well -

• Boolean cutting
• Standard primitives and editing
• cloning / mirroring objects
• following drawings from perspectives
• Using Slice tool on the ship and then mirroring to create symetrical object

Things to work on -

• Vertex... at some corners it appears to have two vertex as opposed to the one, target welding helped elimanate this, but should not be an issue in future
• Saving, I must save work more often, having corrupted files and creating errors without noticing is time consuming
• Cutting down on polygons, I think there is clear ways of improving this within all three pod racers, from selecting number of segments, but also unwanted and hidden areas do not need as many, also the models are detailed, where it could possibly be textured

The final Ship -



 

The Second Pod Racer

My Second pod racer has been an enjoyable project, using my mood board I decided to go for a less round design to see if I could lower the polygon count further by not using so many curves, which to me means more polygons!

The basis was a rectangular engine with large intakes which go through to the back where there is large rotary piece.
Again I started with the engine.

Step 1 was to create the basic shape, using two standard primitives, creating the curved sides using a large sphere and the boolean technique to cut into a basic box.

Had I redone this i would have used a Chamfer box, which would have created slightly softer edges, which would have been useful for the overall look, a lesson learnt for next time! but this would increase the poly count too.

 

this picture shows the beginning of my first engine, the 'intake' areas were created by insetting and then extruding the faces inwards. it was a balance between look and polygon count for the base of the engine.

The hard point was cutting the smaller intake in the blue section as it still had the main base section sticking through, to get round this I manually added vertex and cut out the piece I did not want

The next stage was adding the grills to the intakes, I created one and then cloned it for the smaller, reshaping as I went to fit snug.

The next stage was addign the 'exhaust' where the rotary blades will sit, This was a simple Tube. The Main error for the polygon count I made was here, it has a large number of faces, and also has used a smooth modifier to make the end more smooth, as it has a slight angle, The finished exhaust looks good, but at a hefty price! In the future i would not use so many segments, and try and manually make the end smoother then I have currently, rather then applying a smooth modifier which has smoothed the entire object and made even more polygons. I may even change this before the end.


The next stage was the exhaust rotary blade, Made from a cylinder, inset a number of times, stretched and then manipulated, to creat the smooth curved circle around the base, I made a sphere again and used the boolean technique.

To create the spiral, I used two shapes, a circle, and a helix, Using the Bevel Profile modifier to make the circle follow the helix shape, thus creating the spiral. this is wrapped around a standard primitive and alligned with the allign tool.

I have then added smaller details to the side of the ship, using standard primitives and the scale and general manipulative tools.

 

 The Ropes were hard to get right, I used the bevel profile technique with a pre drawn curve to create the organic rope shape, But to make two different ropes I moved some sections, but this had a slight impact on the smooth curve look.





 
I Felt the ship itself should not have too much detail, as I intend to use textures to provide the detail, and also feel the ship itself isnt the focal point of the model, it also will not be seen as much, and also will be closer to the camera from behind, I think a good texture on the back will be all the ship itself requires, leaving the detail to the engines of the racer itself.

The final Pod Racer -

 

 

The first Pod Racer

My first pod racer model is complete. It started as a simple cone and has been cut out using the Boolean tool, it has been clear to me that components are key to future designs, rather then trying to design it all from one element. The polygon count for my model is relatively high, but was a lot higher before hand, shown in some of the earlier pictures.

I cut down the polygon count by manually deleting unneeded vertex and edges to create larger polygons, I also used the pro - optimizer modify tool, which cut down the number further, it was hard to get it right by cutting down the polygons without loosing the shape of my pod racer.

also in future I will try and cut down the number of segments within round objects, as this was a slight error on my part, I will also try and cut down the number of polygons within areas you cant see, as the detail is unneeded in those areas. for example under the engine section could have less polygons. 

I chose to create this racer from standard primitives as I felt it gave me a good base to work from, and created the shape required well, the other way of doing this would be either from the lathe and drawing a circle or extruding a circle and editing further. 

the finished results before and after the pro optimizer - 

Pod Racer progress

I chose to do a Pod racer for my first model, My aim is to animate a race using a number of pod racers. The idea of the pod racer is it is quite symmetrical, and also most pod racers have two identical engines which can easily be achieved in 3DS Max. 

My first pod racer is coming along very well, it has taken a while and started as a basic cylinder shape. the first problem i encountered was extruding a 'ring' of the cylinder to get the basic shape I wanted. I ended up deleting the ring rather then extruding it, the basic extrusion extruded each 'rectangle' individually, rather then a fluid ring, I attempted to get round this by deleting the edge segments but ended up deleting the entire ring. 

I resolved this by re-creating the polygons and edges and extruding again using the edges not the polygon faces. I also felt there was a large number of polygons at this moment in time, but felt I would cut this down at the end by deleting vertex and edges I don't really need. 

 the second problem I made myself was after I created the 'engine' part, which I created from the basic cylinder, using the inset and extrude tools, the problem being to animate, or change this became a nightmare as it was part of the basic shape, as I didn't think of this as a component to it. 

 I decided to leave this, and look into detaching parts of models to try and detach the parts I wanted, and then repair the rest. In the meantime I also created the wing for the outside of my first engine, a simple rectangle was created, then a cylinder to cookie cut the curves. 




I could not find a simple way to detach the engine component from the main model, which would have been less time consuming to repair afterwards, I chose to delete the part all together, leaving a whole, and then create the engine part again using the same method. The aim is to have this animated and spinning on its center pivot like a plane engine. it started from a circle, extruded, and then inset a number of times, and then deleting unwanted polygons, also a slight rotation was made to create the slight angle on each 'blade' 

I have also extruded and started adding details on the side, as separate components, there is also an instance of the engine at the back, to be animated in the same way, to create the effect is spins all the way through the model. I am happy with the outcome so far, but feel there is more tools to 3DS Max I can utilize in the next models. 

Week 3 Chess piece and plane exercise

The chess piece was created by adding a background image to the front view, and then tracing half the chess piece, I converted it to an editable spline to 'fillet' the curves.
it was then as simple as choosing the lathe on the spline to create a round chess piece. 

The only slight difficulty has was selecting the correct 'align' to create the correct lathe shape. 

The start of my plane has been more challenging, Creating a virtual studio is a great way of being able to model objects very accurately, provided the designs are correct, and clearly visible to make out. I have chosen quite a hard plane, which I found on the net. It was simply to test Cutting them out in GIMP and importing, but I have chosen to carry on creating the plane after this.

The issue I had was the original images were not the same size, and the drawing itself does not appear to be quite symmetrical in every way, also the quality of picture is not great and I am finding it hard to move vertex as i cant see through the object too well even with 'X ray' turned on. 

The plane itself is also very curved, which you cant see well in the three views, which has made me realized it can be used accurately, but also as just a guide...

The first picture shows my first attempt, I have since started again... 

the finished product - 

Week 2 Village

The second exercise i tried was the village exercise, using the extrude tool to create some buildings which also had basic walls inside, this was done via rectangles and extrude, i also aligned some flat roofs and extruded these further to create roof tops. 

I then wanted to create wholes in the buildings as if they had been fired at or were in decay, this was done using the 'cookie cutter' tool, which had mixed results. The first result I simply used a sphere to cut into a corner of one of the buildings, this gave a very neat looking whole in the corner of a building, which was a perfect sphere. Not happy with this i wanted to create a more unique shape. I did this by simply clicking a random shape from lines and then extruding, creating a unique shape as my cookie cutter.

The big problem was once i cut a piece out of a wall, i would also loose the cookie cutter!

The cheap way round this for me was too simply clone the object before cutting, leaving me a replacement for after. But having looked into it you can create an instance of the shape with the tool and then use that further. This appeared to be another way round the problem. Below is my finished village with added tree's...