1 Point Tracking



Creative and Technical Modelling

Today we were tasked to research and create an idea to become our final model. This would show our progress with modelling and would be our best model we have created. We were given suggestions and each finalized an idea.

My idea was to create a cyborg character. However, I thought modelling the entire character would be complicated, so I decided to start with elements and build it up from there. I started with the arm. I researched different cyborg arms and images associated with the idea to get a rough image in my head of what I wanted. I then began further research into anatomy and muscles of arms. Below is the mood board I created on Pinterest.

I sketched up some anatomy drawings to help give a better image to the arm, and to get my idea on paper. I began to add small elements in such as defining the shoulder blade to use as a panel of some sort. I didn’t know how I could make the joints robotic, so after looking at some actual robot arms I thought it didn’t suit my character well, so I instead looked at ball-jointed dolls, this would make the joints easy to model and also a simpler design if the arm was to be made in real life. I wanted to create panels on the front with the muscle paths etched in, Matt suggested adding lights in between the muscles as to help separate them and make them stand out more. It would also help give the overall arm a more futuristic feel.


Here are my sketches for my idea.



Framestore Sculpt Challenge

Today we were challenged by the people at Framestore to take part in one of their work challenges, this was the Lunch Crunch. Each week they allow anyone in the company to take part in a sculpting competition using Mudbox. The only rules is a time restraint (we were allowed 2 hours compared to Framestore’s 1) and the model must start from a sphere. We were told the theme was ‘Express your unique talent’, I thought of character design and concept and my favourite game which I have played for a long time.

I chose to model the Nintendo Amiibo figure Celeste from the game Animal Crossing. She is my favourite character of the series, and the figures do not involve a lot of complex shapes and are relatively round in shape, making is easier to model.

I did run into some problems when modelling her arms as Mudbox crashes several times when trying to model them, however the rest of the model I am quite happy with. I think if the program hadn’t crashed I would have gotten further with the sculpt, maybe painting the model as well.

I would like to complete this model and will do so in my own time.

VR Experience: Google Earth on HTC Vive

Today I had a chance to test out the HTC Vive VR headset in college.

The Vive is a room-scale virtual reality using a headset to view then environment and the hand remotes to interact with the surroundings. It was created by HTC and Valve. It uses 2 sensors to create a 3D environment in which you can walk around and interact.

Image result for vive

HTC Vive

I tried out the Google Earth program in order to visit some cities around the world which I have always wanted to see.

The program includes some key landmarks such as Big Ben, Statue of Liberty and the Tokyo Tower, however there was no search feature which I feel would improve the experience allowing the viewer to visit anywhere by typing in an address instead of starting at a location and trying to figure out which direction to try and travel in.

I did however enjoy my experience with the headset as it allowed me to visit several places I had always wanted to visit. I felt more comfortable sitting on the floor compared to walking around, as I didn’t feel like I was going to fall over and the experience of looking at the streets reminded me of a Street Play rug I used to have as a child.

Turtle Model

Today we took our models into Mud box to add details to create a normal map from and bake onto a low poly model.

I found Mud box easier to use with my Wacom tablet as it game me more control on where I wanted the details to go, it also allowed me to apply different textures lighter compared to how hard I pressed the pen onto the tablet. I feel though I’m progressing with Mudbox well.


Normal Mapping a model

Today we learned how to apply a normal map to a low poly model to create detail.  This started by creating a low poly model, a cube and turning it into a chest of some kind. By using a cube, it meant the UV’s were already in place and made the process easier. I chose to base mine from a child’s toy box. I chose to add text sticking out of the box which I had some trouble with when trying to Boolean. But it was successful in the end.

We then overlapped them and followed Matt’s tutorial on how to apply the normal maps. This was easier than I thought it was going to be. This meant we could add the geometry from the high poly model, convert it into a map and apply that to the low poly model. This would give the illusion on a high poly look on a low poly model.

After finishing setting up the map. I baked the model which applied the map to the low poly model. This was successful and I am very happy with the output. This, below, shows the low poly model with the normal map applied compared to the high poly model as well as the map itself.


Different UV Maps

Normal Maps – Normal Maps are used to add more detail to a model without adding or using more polygons. It is used by manipulating the lighting to create bumps and dents in the model. Normal Maps store RGB images which correspond to the X, Y and Z coordinates of the surface normal.

Height Map – Height maps are commonly a raster image which is used to store values such as elevation on the surface of the model. This uses 3D data to create a shadow and creates depth in the model. This converts the map into a mesh which is added to the model. By using Gray scale images you can paint over it in Photoshop.


Bump Map – Bump maps create bumps and wrinkles to the surface of a model. This is caused by the map perturbing the surface of the object and using lighting calculations. This gives the model a more detailed surface rather than a smooth one without changing the actual model. They use the colour mode Gray scale.