Sunday 28 December 2014

OUAN504 - Character & Narrative - Acting Up: Understanding 3D Modelling (Insight Into Polys)

OUAN504 - Understanding 3D Modelling:

Polygons, or polys for short, are the pieces that make up a 3D model. They are the shapes that play the important role of the model's foundations and are stitched to each other by vertices and edges in order to make up the final product.

But what exactly does it mean to have a set poly count limit when it comes to professionally creating models for clients? Why do high poly counts mean slower rendering and loading times? I went on a little hunt to find out more about the polygons that make up a 3D animator's life.

This image from Fallingpixel.com is an easy way of demonstrating the difference between low-poly and high-poly models. You can see that the model with a higher count of polygons has an increased level of detail. Lower counts often look blocky and are better used for models that don't have much screen time in an animation or aren't to include much interaction in a game.


If you can find the balance between a good-looking model and a low enough poly count, then it will make matters much easier if importing into other software such as Unity as most programs have a poly limit of some description for each model or set. The image below by Matt Coombe shows that sometimes you don't need to overcomplicate things as with a poly count of 6,707, you can still create a model that looks utterly brilliant and will work much better too. Personally, I prefer the model on the left with the lower poly count as it doesn't seem too "smooth" and "perfected". In fact, there are very few minor differences between the two aesthetically, whereas the difference in polygons is astounding!


OUAN504 - Character & Narrative - Acting Up: Understanding 3D Modelling (A Brief Look into Lighting)

OUAN504 - Understanding 3D Modelling:

It may be all well and good creating a fantastic looking model, but without proper lighting, a multitude of sins can stick out like a sore thumb! I thought it would be interesting to research popular tools and methods used to give subtle light to environments for a more realistic look, harsher lights, and so forth.

To begin, Radiosity, an illumination algorithm, is an addition to many of the render methods that exist in 3D modelling and animation. Unlike various other methods (Mental Ray and Raytrace are another couple of lighting algorithms), Radiosity is slightly unconventional in some respects as it doesn't focus on the usual method of "reflection of light ray>directly to an object>eye". It is based around the more natural idea that a ray from a light source will reflect on many objects and then back to the eye, thus allowing a mix of colours depending on the colours of the light source and the objects. This is particularly useful when lighting interior environments as walls and other multiple rooms may not be in the direct path of a light source, so it will heavily rely on the reflections from other objects. Although from Wikipedia, here is an image perfectly depicting the use of Radiosity with a comparison against another more unnatural lighting technique.  Radiosity uses GI (Global Illumination) which is a way of following natural light behaviour i.e. casting rays by following realistic rules. Many renderers use GI.


Many people's opinions on internet forums seem to sway more towards Mental Ray as the choice for rendering and lighting as apparently it is more predictable. Some have argued that it is all a matter of personal preference and that the only way to truly decide on the best renderer is to try them all and do tests. I personally think that it is important to test out as many methods as you can in order to have a valuable opinion on what works for you. As a student, we have only been given Mental Ray as the renderer of choice - possibly because it is reliable - and I do wish to play with many more to get an idea of what they all do.

Mental Ray has a feature called caustics which allows a ray of light to say for instance, shine onto a wall in a room and then reflect the light onto other surfaces. "A Look At Caustics" by Jeremy Birn - http://www.3drender.com/light/caustics.html - is a very helpful guide on how caustics work in animation. Below is an image from his article showing a side by side comparison on how a rendered image looks with and without caustics to enhance lighting properties. 


A Raytraced image (Left) and Raytrace enhanced with caustics (Right)


OUAN504 - Character & Narrative - Acting Up: Understanding 3D Modelling (Concepts, Principles & Limitations)

OUAN504 - Understanding 3D Modelling:

A Short History of Rendering

Way back when, early 3D models consisted of wireframe shapes that were simple representations of the everyday geometric shapes we know today. The beautifully rendered shapes and models we are able to create in this age started out with hidden algorithms. These began in the 1970's and allowed us to render curved surfaces as well as non-spherical objects.

In 1978, a wonderful gentleman named Jim Blinn who started out working for NASA creating computer graphics and animation for such things as the Voyager project. He went on to do a "simulation of wrinkled surfaces" - collgran.wordpress.com.  There is a shader named after him in Maya, this is probably why as he such an influence over 3D animation.

I wish to go on with the historical aspect of rendering and 3D animation, however, it would be more practical to look at some of issues I have come across when using 3D software and certain limitations I have experiences even though I touched on them in a previous post.

Experienced Limitations

Although I haven't been using 3D software for very long and have a huge way to go in order to at least get some sense from using it, I still believe that everyone's experiences are different and we all encounter different problems.

I find that when it comes to physically creating a model or character, it is incredibly complex to use textures with them that aren't the shaders (Blinn, Lambert etc.) as creating your own textures is a completely different skillset entirely and is baffling!

Poly count. Even though we haven't got round to creating insanely complex environments or characters just yet, I have noticed that when modelling smaller environments (grassy fields with mountainous rocks in the background for instance), if the poly count is high, it can take a good while to load up and even render. There has to be a compromise between quality of model and poly count as larger poly counts can mean in-game crashing if to be used for such purpose and long loading times.

The size of the files created when using 3D modelling is immense and can often lead to issues with inadequate computer space or processors. Not all animators are wealthy and having to fork out on the latest in computer hardware can be a bit of a setback.

OUAN504 - Character & Narrative - Acting Up: Understanding 3D Modelling (Tomb Raider - Evolution & Motion Capture)

OUAN504 - Understand 3D Modelling:

In terms of how far 3D animation has come as a process, I think it would only be fair and logical to talk about how the popular Tomb Raider games franchise has improved over the years. I remember playing one of the first Tomb Raider games when I was a toddler, and although being a toddler I couldn't make comment on the technologies of the time, I did think it was an amazing game as I simply couldn't put it down!

Tomb Raider has had a lot of stick though due to feminists screaming "exploitation". Though this hasn't (thankfully) affected the way Lara Croft has been developed as a character. This little gem was found on the website Fanpop (see image below) showing the fantastic evolution of Lara Croft over the past couple of decades since it was first released on Playstation One in 1996. I think it is phenomenal to show the changes she has undergone.


The new Lara looks much more realistic and conforms to a more natural-looking female body shape in comparison to the 90's Laras with breathtakingly thin waists and exaggerated proportions. The advance in 3D technologies is very obvious in this image and has allowed for better clothing texturing, lighting, smoothness and complexity of body shapes, and many other features. This isn't just a feature of the Tomb Raider franchise however. I plan to explore the growth of design and modelling within other animation later on in my blog.

Lara Croft's animation today is undergone using the performance capture method. Early techniques of motion capture (performance capture) were done using many cameras in order to calculate the object or person's positions within a space. There are various advantages that come with using motion capture. Some of these are as follows:

Rapid, real-time results can be accurately captured.
Traditional techniques such as 2D or stop-motion can vary in intensity of work and complexity.
Secondary animation, exaggeration, and subtle movements can be easily caught.

In contrast, the disadvantages are:

Expensive to acquire suitable software and hardware.
Specifications in order to use hardware and software vary such as capture space etc.
Unnatural movement is hard to replicate as you cannot really tamper with captured movement.
Added emphasis and unnatural movement (linear/robotic/overexaggerated) is easier to achieve with other methods.

I am actually very thankful that Lara's chest is no longer triangular in shape as it may have given men unrealistic expectations of what women should strive to look like. Thank technological evolution for allowance of more polys and better software! (Apologies.)


OUAN504 - Character & Narrative: Modelling Props (The Stars)

OUAN504 - Modelling Props:

For the very last scene of my animation, Moom is shown spinning around amongst falling stars whilst holding a T-pose. It is supposed to look like he is revelling in the falling stars and enjoying the fact that he is an "all star". I wanted to make the stars look very unrealistic and basic. I looked at many images of different types and shapes of stars and after playing a little "reference gaming", I stumbled upon some brilliant stars that would be very fitting for the animation.

In fact, I was playing the Playstation One game "Stuart Little" and saw the stars floating around in one of the bedrooms. They were bright, cheerful, very slightly 3D (the stars are only to be seen from the front view so making them too complex would be pointless), and had little shadow on them.

After creating a prism shape, I pulled in some of the vertices to give it the five-pointed star shape and puffed out some of the faces and vertices to give depth. I added some lighting to see how the light would fall and how the shadows would be cast and was very happy with results. I didn't want them too dramatic as it might detract the attention from Moom.







OUAN504 - Character & Narrative: Modelling Props (The Onion)

OUAN504 - Modelling Props:

Despite me thinking my modelling was alright for an amateur, this character in particular really threw me off guard. The onion, the character of which Moom is seen interacting with during the first scene or so wasn't exactly a demonstration of my "improving" Maya skillset. Not only did I find it hard to model, but texturing was unbelievably difficult!

The shape of the onion and the cute expression reminded me of a mob from an online game I played back in the day called Fiesta. The mob (group of monsters of which you played against) was a group of things called Slimes. I thought my onion vaguely resembled one of these. (Image below)


Or maybe even the Slimes from Dragon Quest?




I used a simple image of a licence-free onion found on Google to use for the texture as I wanted to show the streakiness of the onion's layers. I did try a simple Lambert texture and used a vague onion-like colour, but without the added detail of the onion skin and many layers underneath, it looked more like a brown excrement than an onion...

However, the face of the onion went better than expected and I still achieved the "gormlessly cute" and vacant look of which I was after. So despite the horrid onion shape (I watched many tutorials and studied other 3D onion models without prevail), I managed to model at least something. I may try again if I have time as I can't possibly make the same mistakes twice (hopefully).




Tuesday 16 December 2014

OUAN504 - Character & Narrative: Modelling Props (The Basketball)

OUAN504 - Modelling Props:

This was incredibly difficult at first despite it being a mere basketball. The reason for this was that I didn't want to apply any more images onto my models as textures as I thought they were starting to look a little shabby and without decent knowledge of texture application, I decided that it was best to stick to good old Lambert and Blinn by applying simple shaders.

This meant that rather than creating a simple ball and making it orange with black lines, I had to create the grooves on the ball by using a few tools that I had never touched before. A useful tutorial on YouTube (below) was a great method of learning this new tool of which was not available on the toolbar so instead I had to manually type the name of the tool into the command bar to select it. This tool was the "Interactive Split Tool" which allowed me to get the grooves on the basketball into a rough position on the ball. After smoothing the ball, they fell into shape.


After finishing this model, I agreed it was by far the easiest I had made. Here is my finished product.



OUAN504 - Character & Narrative: Modelling Props (The Axe)

OUAN504 - Modelling Props:

To add to the comical notion of the animation, I decided to model an axe as it fits perfectly with the lyrics ("ain't the sharpest tool in the shed" could have connotations of pitchforks, spades, and other various garden tools). However, I decided to take it a step further and turn my everyday, obvious axe into a video game-style war axe used for slaying enemies! This would work much better than a regular wood-cutting axe as the onion (of which I still have to model) is depicted looking at the axe in horror. So, a little exaggeration on the type of axe will work a treat!

Using reference once again from Google (and a couple of great tutorials on how to use boolean difference), I found an image of a war axe that I thought was easy enough to model and looked the part.


After a horrible experience with lighting and texturing this model, I think I need to get more practice in so that I can light and texture with more confidence. The first image shows how the texture didn't quite fit, and sadly I thought I could away with it. However, after pondering over this for about an hour, I realised that something had to be done. So I went into photoshop and shoved a grey background in just to fill the rest of the axe shape. This quite frankly looks worse but will benefit the animation more than leaving big blank gaps of textureless axe everywhere...

Oh well, I gave it my best shot. Modelling this was incredibly fun, though!

OUAN504 - Character & Narrative: Modelling Props (The Guitar)

OUAN504 - Modelling Props:

After a good few weeks of focusing on other modules, I have decided that it is maybe a wise idea to start modelling my props for the Moom animation as it is scheduled on my planning sheet to make a good start round about now. This will probably take a good week if I do this properly and get some good practice in animation alongside modelling the props.

My first task was a guitar for Moom. I used a simple image plane as a basis of reference to model from. I could then look at this image in order to guide me when creating shapes. I then used the same image of the guitar to create a UV texture and apply it onto the model. I admit, it looks terrible, but I have gained valuable skills in applying textures to my models and will hopefully this technique over the years. I am thoroughly enjoying model-making (not so much the animation side however!) and hope to continue 3D modelling in the future and as a hobby.

Here are some shots of my guitar. I admit that the texture wasn't fabulous and it was a rather lazy way of applying the texture to my model (applied image to model which meant it was duplicated or rather wrapped around the back showing the same image from the front on the back also!), however it was the quickest and most appropriate thing to do as in my animation, you won't see the back of the guitar of which Moom will be holding.




OUAN504 - Character & Narrative: Maya for Dummies (That Means Me!)

OUAN504 - Maya for Dummies:

As part of the course, Maya is a mandatory part of the learning process and getting to grips with this beast of a program isn't an easy task. For any stuck little minds out there sat at their desks sporting faces full of puzzled looks, here's a nifty list of the fascinating and possibly life-saving things I've learned if having a crisis in Maya. I plan to keep a note of all problems and how they were fixed be it through my own (unlikely) intellect, or through that of my tutors.

Problem One - Mental Ray has Disappeared!

According to Google, the source of wonderful help forums, you should go to Window > Rendering Editors > Render Settings and check if Mental Ray is actually present to select. If not, which was the reason why I was confused, you should then go to Window > Settings/Preferences > Plug-In Manager, checking for a thing called "mayatomr.mll ". This should be checked in order to have Mental Ray present.

Problem Two - Can't Get Back to Neutral Character Pose!

Yes, we've all done it. Playing around with Moom for too long can only entail issues where body parts go flying, end up at the other side of the planet, or worse; missing. In order to get your character back to the original pose he was created in, select all of your character, then press "O" on the keyboard for a T-Pose, or the neutral position.

Problem Three - Can't get Front View?

Simple. View the four view perspectives and click in the one you want to enlarge and view as your single view, then hit spacebar to enter that view. Done!

Problem Four - Struggling to grab all of Moom's body?

Select the little circular disc on the floor around Moom's feet and press up on the keyboard. This will select everything meaning he won't lose eyes and hands in the process!




Wednesday 3 December 2014

OUAN503 - Responsive - Individual Practice: Rewriting A Brief

OUAN503 - Rewriting A Brief:

BA (Hons) Animation - Level 05
OUAN503 Responsive - Studio Brief 1

BRIEF TITLE Taylor’s Coffee Syrup Packaging

Brief

To create a product (syrup) range for Taylors of Harrogate focusing on 3 package designs. Possibly extend this by developing the designs into a short animation.







Product
Tone of Voice

3 product designs
20-30 second animation







Sophisticated
Fun
Professional

Audience
Context

16-25 year old students of whom are not massive fans of regular coffee







To be suitable for television (consider title safe guidelines and rulers to avoid text and images being clipped off the screen)

To fit a suitable bottle type (research label sizes)

Additional information/Considerations

Consider style of animation (graphic, cartoon, painted, abstract, realistic, 3D, 2D, traditional)
Colour theory (Taylors already use complimentary colours, so sticking with this might be useful)
Programs to use when creating the animation (Maya, Flash, After Effects, Photoshop)
Flavours of the coffee syrup (conduct a short survey/questionnaire on demographic)




Mandatory Requirements
Deliverables

Product designs
Use of Taylors logo
300 word evaluation on product

Product designs
Animation





OUAN503 - Responsive - Individual Practice: SMART

OUAN503 - SMART:

As a guideline for creating successful project proposals, we were given a short lecture on the best methods of doing so. I found these very helpful and allowed me a better understanding on rewriting a brief to suit my project needs. You should always consider the original brief and use key words from it such as mandatory requirements and such.

BA (Hons) Animation - Level 05

OUAN503 Responsive

STUDIO BRIEF 1 - Indiviudal Practice

Project Proposals
10 steps to producing persuasive project proposals



  1. Set clear aims and objectives

  1. Be SMART

( Specific , Measurable , Achievable , Realistic , Time Conscious)

  1. Use words from your brief

  1. Justify the need for your proposal

  1. Describe your audience

  1. Describe your motivations

  1. Consider the reader

  1. Eliminate vagueness

  1. Visualise the ending  

  1. Assume nothing

Monday 1 December 2014

OUAN504 - Character & Narrative: Sound Recording

OUAN504 - Sound Recording:

After finally taking the time to book out the sound recording booth, I spent a good hour or so practicing different variants of my lines in a range of accents to compare which ones worked best with the narrative. I recorded a good dozen takes and after listening to them all one by one and picking out the positives and negatives of each recording, I decided on my final choice for the sound.

I noticed that some of my recordings were incredibly fast-paced and I discovered that I needed to slow my voice down. By acting out the movement that Moom would make, I was able to slow my voice down a little bit and be a little more "theatrical" in my performance. I managed to get a little bit of emphasis into my voice and portray as much emotion as I could manage.

It's quite difficult to perform when you have all these wonderful ideas in your head that come out at a million miles an hour in a rather monotonous voice, not making any sense when you listen to them back and quite possibly all being in the wrong order. I much prefer typing to speaking as you can correct everything! Despite the harsh learning curve, I then edited the sound slightly as I wanted to achieve a very low-pitched voice to go with the terrible excuse for a Scottish accent.

I did this in Adobe Audition and only slightly lowered the pitch I then exported this as both an MP3 and WAV file so that I could use both in any instance. (WAV files are compatible with Maya which was the intention for recording in the first place.)