Project Postmortem

The wondering Lands of Alice

This project was creating a level with a team of six, the theme was Alice in wonder land and we had to reference the source materials given to us by the British library. We had to develop our own style guide and constraints to fit the engine. It was Suggested that we use the side scroller element of UE4 to help us with engine.

This project I really feel I pushed to learn about and take the reigns on the things I want to develop. These skills included: Art Direction, texture artist, 2D Level concepting and over painting, 3D level concepting, 2d character concepting, character art direction, model reviewing, texture reviewing, unwrap reviewing, packing and creating, particle reviewing and assisting, Engine texture assistance. My roles have been based on over seeing parts and keeping textures and styles consistent in such a complicated level.

Most of my contribution to the level has been texturing and concepting at the start. Some of it has been correcting models, redoing unwraps and art direction for things like size and style of particle effects. My major jobs where art direction, especially with texturing and colours. My secondary role in this was overseeing character creation, like denzils mushroom and cat. When it came to modeling I mostly took first pass models from others so they could work on engine and I could fill in for our group lack of understanding when it came to engine. This meant I did a lot of re-unwrapping and re-topologysing. However it worked out really well as the quality of texture is high because I had enough time on it, but the texturing is also consistent because while we all did some mine was around 70% of the level.

Because we where assigned groups for 6 to work with for 3 months we had a lot of team meetings to start with, this allowed us to get to know each other and what we wanted to learn in the project. With this the first stage of development began, we decoded the brief and tried to sum up the ideas that we all had floating in our heads as individuals. This was via a white board to start with. It allowed us to see what ideas linked, what ideas people had the same and what we could strike off the mass list of things we’d like to do. Once we narrowed it down we all made mood boards and storyboards to match the consensus of ideas. It soon became apparent that the roles given to us didn't suit our skills so we divided labour more efficiently to suit what people where good at.

For instance i'm not a strong environment conceptor so everyone pulled together to help and we all worked hard together to visualize the story, the layout, the blueprints needed and the overall style. Next time I think I will really practice my environment work because it is very lacking, My biggest issue was working on setting from scratch, as colouring them didn't pose the same issues.

I started off with story boards (shown below).

Refined my first idea for a section of level layout.

Rough shorter story board after talking to the group.

  

It soon became apparent that as a group we had leaned towards underground. Once we agreed on the shorter board above we started to look into style for textures and level assets. This involved white boxing and paint overs but also traditional concepting and style guides.

One of my first style mood boards

One of my first mock ups of how layers of planes could create backgrounds like in theatre.

The style guide I came up with for how paper, hand painted and 3D could all fit together, looking into modified books, theatre sets and hand painted 3D assets.

One of my first ideas for how the planes and realism could fit together. 

Looking further into the style above but with more colour and lighting.

One of the last paint overs of blockouts, this time focusing on colour and lighting further. This was something I coloured but didn’t sketch. As I mentioned people worked to their strengths so I worked on colour and art direction with amber at this stage. Looking back really I should have been able to take a stronger lead on this but I'm not convinced that would have been more successful.

As I worked on so much of the paint overs and art direction I also created this level layout design from the block out and some paintovers. Naturally a lot of this ended up needing modifying and sketching over to fit but it gave us a solid colour, style and size reference for modelling the level, not to mention asset list and working out what blueprints or particle effects we wanted. 

Once we had this all together many people started modeling from the white box, however because of the plane aspect of our level and how me and amber worked as a pair to get one small part working to check style, I worked on texture right from the word go.

The first set of textures and a paint over using these planes to show where they should go

These are examples of how I worked with people using engine to show how models, placement. In hindsight I think there should have been more of this kind of communication throughout the project 

Before                         After

A lot of my work after concepting was either continuing with making textures for planes we knew would be used in game, like the backgrounds and trees, or correcting textures (as above).

Weeks 5 -  Easter

Examples of textures

Here is one of the many sheets I put together using the books from the British Libraries references and our own libraries books. The orange bit would glow in engine giving it the look of dusk without creating a too dimly lit level.

I think that textures are one of the really good things about my work on this project. I've worked with scanned materials from our library here at DMU along with hand painting textures and I feel they all fit together well which isn't that easy given the way they are completely different styles normally

Example of my texture setup I did in engine:

The crystals I created by looking at the best way to have a translucent texture but still retain an amount of PBR. This wasn’t easy but by changing the setting in the material as well as using specularity I got the right look for crystals.

The cat that glows a bit to help visibility in its dark environment.

Although I learn more about engine texturing and got quicker I competly let myself down by not doing roughness and metalness and normal maps and not setting them up. We did this so I had more time to fix things and paint textures, which some members where very weak at doing, but I didn't learn enough about engine on this project because of it and I feel like I will have to counteract this by doing a lot of PBR work over summer to learn how to do it quick enough to spend longer on an albedo.

 So, after all that, this is how it came out!

I think a lot of the work I did shows up because it’s texturing but I do feel like my input was not as much as it could have been for some areas like modelling. I’m also very aware that all the effort I put into rewrapping and fixing silly mistakes like nested verts and bad topology went completely unnoticed by the group due to the nature of correcting someone’s work that’s in progress rather than giving it back in a spot the difference like fashion. However part of me is pleased because even though I wasn’t directly modelling the amount of correcting has meant my skills are still getting used and I’m still an efficient modeller in 3DS Max. I feel that for the most part my skills in art direction and texturing have come along so far that the pain of group work has very much been worth it.

Five Things to work on:
texturing speed, working on own maps whole way through, critical paint-overs throughout project, understanding more in engine, my time management.

Five Things that where a success:
My hand painted textures, helping unwrapping, art direction improvements, helping retopologize models, working together in a way that played to our personal strengths.

One thing learnt:

How to use matinee to animate in engine!