Life drawing update

This year, I made a very conscious effort to attend as many life drawing classes as I could. Aside from the mandatory classes on Tuesday, there were sessions running all day every Thursday. Not only did this provide a good release from course work, but also gave me lots of opportunity to practice and improve my life drawing and pencil skills.

For a long time while life drawing I have been very focused on capturing gesture and good proportion, usually doing no longer than five minute drawings. These open sessions also gave me the chance to focus on more slightly longer poses, these are a few done in minutes or so.

Container City, Post mortem

At the beginning of the project we all had some very different ideas of where the project could go, it took us a while to finally arrive at an idea that we all liked and agreed upon. Even so, I don't think that any of us had any real idea of how it was going to turn out visually, as naturally we all had a different idea of how the final product was going to look. In the end, I think we ended up with a very nice looking project however much we may have winged it along the way. A lot of the assets turned out very nice, and worked together to create a cohesive whole. The modularity of our container city was also a strong point in our work as most of our assets were modular and fit together very well. However I think things could have been a lot better if we had established a more solid visual direction earlier on other than "lets make Kowloon walled city but with shipping containers". Its a simple concept, but again, I think everybody has a different vision in their mind of how that one specific thing could look, especially when that idea only exists as a statement.

I suppose what I'm trying to say is that we had to get to the very end of our project before we knew how it was really going to look, which isn't always a good thing. We could have done more concept art earlier on in the project to get a solid visual goal, but it wasn't as simple as that as we also had to manage our time effectively as well as think about other elements of the project such as the game design, playability and other aspects of the engine work. These things took up a lot of time for most of the team. There was a point of relief for us during the stage of our project when the level finally came together, as up until then we weren't sure how it was going to look. We ended up with, what I think, is a fairly nice looking project, and one that also plays out smoothly for the player.

Container City part two

When first starting to construct assets for our container city we had a few hiccups with the modularity when trying to set a standard for the scale we would use. We decided on a final scale that would export with the correct proportions in engine, this way we wouldn't have to mess around scaling assets in Engine and disrupting the modularity.

I modeled the primary container that would be used as the main component for building the level. When building the level, the container was broken up into segments; front, top, back and sides. This meant we could save a lot of tri's on areas where we wouldn't see the full container, which helped our fps immensely. I also modeled numerous assets for populating the level such as an oil drum, pallets, gas tank and modular shop shutters to be used in place of the sides of some containers. Since the container would be the primary modular asset featured throughout the level, this was modeled first and used as a scale reference when modeling all the other assets.

When modelling the container, I thought the corrugated edges on the side panels were far too sharp. Chamfering the corners meant using far too many tri's, so I decided to have the corrugated side panels all on the same smoothing group and bake a very high poly chamfered version of the side panels into the low poly. This made the corrugated sides looked more like the sides of a real shipping container with much softer light hitting the edges. Rather than have another albedo texture for a different coloured container, I created instances of the original material in Engine to create colour variations. Since this project, I have learned a lot more about the high poly baking workflow in Max, and would probably have taken some different approaches when creating my assets.

Below are some final shots of our level:

Container City, Week One

First project back after Christmas tasked us to create a section of a city or inhabited area created entirely out of shipping containers. This was a group project and we needed to show at least some small element of interactivity and playability in the level.

On the first day we sat down as a group and discussed various ideas on themes and locations for our container city. We came to around 26 final ideas that we liked, including prisons, rebel bunkers and a drug cartel, but the one we eventually decided on was a container city that would be inspired by the Kowloon walled city in Hong Kong. A densely populated and ungoverned fort like city built entirely from layer upon layer of makeshift buildings and scrap that was demolished in 1993.

What attracted us to the idea of a container city based upon Kowloon was the dirt, grittiness and dark dystopian nature of the city and its maze of alleyways filled with makeshift shops and houses. Kowloon is a city built upon layers. We thought this would make a great game level with lots of potential for detail, material definition and verticality.

Another aspect of this project we were going to have to learn was modularity. This is a very simple concept in Game design and refers to how individual assets can be reused and recycled in various ways throughout a level or game. This can include simple assets such as separate sections for networks of pipes, or larger more complex sections of dungeons or buildings. In our case, we were going to employ modularity by having a primary container repeating throughout the level that was going to be edited in different ways via other smaller assets such as windows, pipes, shop stall etc.

A good source we used to learn about modularity was Bethesda artist Joel Burgess's blog post on Modularity in Skyrim 

Another large part of this project was going to be the level design. This is something none of us had touched upon as our course is largely focused on creating art and assets. We struggled for a good while debating our level design, and decided to blockmesh a few of our ideas and get some tutor feedback before deciding on a final layout.