Following on from the last blog post, I decided to at least attempt to model the Bear. It was important for me to at least try and modeling the characters, probably since the game has been very personal for me. Unfortunately, I got a little too focused during the modeling process and failed to take any screenshots during the build (continual documentation is a skill I’m still trying to master), however I thought I’d write another blog post about my experience learning to rig and animate the bear asset.

I began with my new bear model. I used many references online to try to achieve the low-poly aesthetic while it still being recognizable as a bear.  I received mostly positive feedback from my colleagues - some adjustments here or there (and one who said it looked like a big pig….) . I was fairly content with my first attempt at character modeling in Maya.

With no experience in rigging or modeling, I began the way I always do -  by watching a lot of video tutorials. I picked up some very useful tips, including the correct way to line up the skeleton. The other thing I learned was that rigging a quadrupedal was going to be significantly more difficult than a bipedal. 

I began with the front right leg, then the back right leg and mirrored these for the left side limbs. I then created the spine and the head.

After connecting the limbs, to the spine, the last step involved binding the skin to the skeleton rig. This was a fairly straight forward process, however I feel as though something went wrong in the rear half of the rigging process. I didn’t realize until after I bound the skin to the rig that the rear half was merged just above the hip (a problem I’d been specifically trying to avoid).

The next blog post will be my experience experimenting with both the Maya and Unity animation tools.

During class we were able play test each others games and offer feedback to our colleagues on their games. As well as getting hands on time with some of their incredible work, I was able to get some great feedback on the current build of my game. I had 3 play testers during this session (M, D and S), and found some common complaints, as well as some differing opinions.

M was already familiar with the project from earlier play testing sessions, but it was very interesting to hear their feedback on the progress made. Some of the feedback included:

- The player has too many HP to start with. There should be very real consequences to being attacked by the bear (Note: I had intentionally upped the player HP for the play testing process since the game was very punishing).

- The lights not coming back on after the player dies is strange. It creates a sort of consequence for the players action but was this intentional or not? (Note: It was not. Please see Blog 07 - Sean Harkin: Code Breaker for further details). 

- There should be something to make it clear to the player they are in range to take damage, possibly an animation or audio cue.

D was very excited by the aesthetic of the game, and gave compliments on the environment models I had created. They also make some incredibly loud and shrieking noises when they are excited playing games. Some of the feedback included:

- They were put off by the lack of HUD. There are very little UI elements intended for the final build, however this player may have been used to more traditional game interfaces.

- When D died, they thought it was a great effect that the lights didn’t come back on. When told this was unintentional, they suggested that it might be some sort of mechanic tied to a perfect play-through. 

- D died very quickly - they suggested a higher starting HP. 

- Due to the camera position and player controller, they found it slightly confusing to control the character. (Note. This was very important feedback as the games mechanic is all about approaching the Bear in the correct way, the player should feel they have good control over the avatar)

S is very avid game enthusiast so I was very interested in what they thought of my game. Some of the feedback included:

- Again, S thought the player died too quickly and there was no clear indication of the player taking damage.

- They also had some critique of the player camera and controller mechanisms. They suggested a more top down approach, which would keep the 3D aesthetic but require less work to change. (Note. The top down suggestion was tried, but ultimately reverted to changing the camera and controls entirely)

The play testing session was incredibly helpful, as though I had conducted a few before hand, these players were familiar with the game and the mechanics (such as M) and therefore were less critical of some of the controls which were clearly an issue for new players. This told me I would either have to create an explicit control mechanic, or change my controls to be more ‘user-friendly’.  Also, the lack of animation (which was not intentional, just not yet implemented in that build) was clearly an issue - but there was a very clear need for some sort of indication of the player taking damage; be that audio, animation or a UI element. Overall, this session highlighted the areas which I should concentrate on in my development.

I have extensive 3D CAD modeling experience from my work as an industrial designer - however I learnt fairly quickly that this was in fact not in any real way helpful. If anything this actually proved to be a massive hindrance - in CAD modeling programs you work with solid shapes - everything you create has a depth; but in Maya everything you make is just a series of surfaces. That changes the the entire way you even think about how to create your models. 

To start with, I wanted to see how transferable my skill set would be. I downloaded and opened up Maya, and very quickly discovered they were in fact not. I spent a not insignificant time with some online tutorials, and began to learn the basics. 

Wanting to make the most of time (read: trying to find an easier and quicker way to do this) I considered creating my models in a more familiar software package and transferring these into Maya. After a little research, it seemed this was in fact possible (if not widely recommended according to the Maya and Unity forums). I wanted to experiment myself and see what kind of results. 

Unfortunately, this was not as smooth as process as I had hoped (pun fully intended). After bringing the model in and trying to reduce and smooth to get something that looked as intended, I was not successful. The biggest issue I came across was the extremely high poly count on the models I was exporting from Inventor to Maya, which would be difficult to render later. I decided that although this might be possible in some cases, it would probably be a better use of my time to concentrate on Maya.

Model created in Inventor

Model exported to Maya

Poly count reduced by 20%

Poly count reduced by 25%

Poly count reduced by 30%

After returning to Maya, with the help of some more online tutorials and spending time getting to know the basics, I was able to create some nice low poly assets for my environment. I would like to spend the time required to improve my skills and create the Player Character and Bear NPC assets myself, however I feel this may be outwith the time constraints I have, so I am also considering purchasing some assets to complete the rest of the game.