First to Dance

Before we even were set the assignment for the vertical slice, I was approached by Michael and Martin with the offer to work with them on a still unnamed game project for the second half of the year. At this point, all that was known about the game was that they wanted it to be set in a desert like environment so that we could play a lot with light and make it beautiful, and that the main character was to be a Native American woman. The idea was to step away from the trends of dark, horror games that have been made by students on the course recently and attempt to make something set during the day in an open space and also, play as a female protagonist who is from a minority. I got asked to be a character concept artist for this project and of course help with anything else I could. As I quite liked the idea, I immediately accepted. It was still early, but I thought that it would be good.

We still had time before the start, but we had several group meetings where we were discussing where we were planning to go with this project. It was all quite ambitions and we were aware of that. It became clear when we even thought about the idea of continuing it over the summer and do it is a main project next year. Still early plans though.

We wanted to make our game as historically true as possible. As part of our research we an arrangement with the museum where they had an exhibition on Native Americans we planned on visiting and talking with the people there, trying to learn as much as we can about the culture and the way the Indians lived.Unfortunately we never heard back from them so we didn't get to go. However, we had a trip to London, where we learned a lot of things, took tons of reference photos and we finally began to outline our ideas. The few documentaries we watched helped as well.

After we decided on a time period, which was around 1850, the time of The Trail of Tears, I went ahead and did some initial research on the different tribes and how women were viewed in them. I learned some really interesting things.For example that actually the land owners were the women. Which was perfect for our purposes because we wanted to have a conflict in the narrative that revolved around land and land being taken away from the Native Americans, it would make sense for our protagonist to go and seek justice if it belonged to her. Then I also learned that women warriors were also not that uncommon. Which again fit really well with our initial idea of having a strong  female protagonist that wouldn't be afraid to fight and would have the skills and courage to do so.

In the end it was decided that we are going to have our protagonist, named Ayita which means first to dance, a descendent from the Cherokee tribe, trying to hunt down a white man who had illegally taken her land and forced her and her people out to find a new home. The start of Ayita's journey is her jumping off a train in the middle of nowhere to go and visit a friend who would give her information on how to settle the manner with the white man.Unfortunately, as she approaches her friend's ranch, Ayita finds out that it has been infiltrated  by bandits. The goal is to sneak in the ranch unnoticed and save Ayita's friend. The vertical slice would end there, with Ayita witnessing her friend being killed in front of her eyes by the boss of the bandits which would make her snap and go on a killing spree~

That was the basic idea we had in mind. Of course it was too ambitions for us, so we never expected to reach that level. We were happy to even settle with just a beautiful environment and a character to walk around with, no gameplay involved at all. We got criticized for going down this route, but it was only our last resort. We just wanted to try and see how far we could go with this project in the time given.

Having all this information, the initial research and reading I did, the gameplay to keep in mind, I was finally ready to start designing our main character. As a start I went back to researching just that this time, it was for hunting visual references. I needed to settle on a style, I needed to find out exactly what Native American women were wearing around this period, what colours, materials, how they worked and etc. Finding references wasn't easy and I really needed to dig deep to find what I wanted. I even got into some official archives when I found authentic black and white pictures of Native Americans from that time. The modern pictures from the museums helped with choosing the colours and learning more on how they used to wear their clothes. Then I had to not only look at the clothes, but how they wore their hair as well. I found this really interesting tribe which used to have their hair divided in two giant buns on the sides. I loved the idea of that haircut and I even gave it a try drawing it with a little photostudy. It was useful for future reference. Also a good exercise to try and nail down the general Native American face features.

As an inspiration for her general figure as an agile and strong girl I looked at Korra and how her design was executed. Wider shoulders, strong arms, a confident posture, I learned a lot by looking at her and thinking what I could do with Ayita to achieve something similar. I took some references of fit girls and studying those I designed her general body shape.

Being fascinated by all the amazing outfits I found out they wore, I really wanted to give Ayita a traditional long dress with tassels and beads and feathers~ It would be authentic and a good way to represent her as a respected Native American woman. However that kind of outfit would be highly impractical. Not fitting for her mission and not fitting for us since we had to think about animating her as well. She was to be agile and silent so she could quickly run, hide, jump over fences, climb walls, crawl under places and all these other action-y things. I had to create my design keeping all those in mind. Unfortunately I couldn't find any references of women warriors to help me with that so I had to improvise.

I started changing the design little by little, consulting with Michael the whole time as he had to model and animate the character in the end. We decided to give her shorts so she could move around easily even though judging by the research I had done there was no indication of them wearing similar things. I would say the final design ended up not being so authentic, but just very inspired by the Native American culture. I tried to keep as much as I could from the traditional designs, trying to avoid the cliché representations on female Indians.

After I had the outfit finalized it was time to move onto the face. After having a long discussion with Ewan about paintovers and techniques that AAA games use to speed up their processes, but still produce amazing results I decided to give it a try myself. For Ayita's face I took photos of quite a few women who I thought might be fitting and by cutting and stitching I made this hybrid face which I later painted over and blended into one adding the hair and accessories. Everyone was quite happy with the result and even if the actual paintover process and the idea behind it doesn't really appeal to me, I was glad they were pleased with what I had produced. That is when I decided that I'd continue using this technique for this project and produce all the characters in the same way.

I had Ayita's face, I had her body, it was time to combine them and prepare to hand in for modelling. I drew her in the tyical T-pose from the front and side and gave it to Michael. While he was modelling I used the time to think about what textures I wanted to use on the outfit. I went back to my references and picked quite a few materials from there, sampling the colours as well. I used the front view to mock up a textured version of Ayita to see how she would look in the end. At that point it wasn't clear yet if I was going to texture her or not, but I had offered my help and that was something I'd happily work on.

While Michael was modelling, I decided to help a little bit with literally the smallest things- the accessories for Ayita's outfit. I had concepts  for those prepared and after gathering some photo materials and references I managed to recreate them in 3D form. I mostly used planes with transparency and a normal map to achieve the effect. I stopped working on those when I realized that it was all getting a bit chaotic and it was really hard to judge what colours to use without having the whole character textured yet. I decided to wait until we have those before I continue. It would be much easier for me then to make the whole design and colours work together.

Except the main character I was also asked to design the face of a bandit as Michael was already working on the model for one. At first I did some sketches and concepts trying it a bit hard to decide on a style. Realism is not my native ground so I had to experiment quite a bit to make it work for me. In the end the paintover technique did the job. Just like with Ayita I picked some photos an combined them before painting over, creating the bandit's face.

I didn't have to do concepts for the bandit's outfit as Michael already had an image in his mind and was working from it. However I was asked to do some accessory models for the bandits. A hat was one of them. I made the base mesh and then handed it over to Michael for a quick sculpt before doing the textures and normal maps. As I didn't have a model of the bandit at this point I just used Ayita to see how the hat would fit on a human head.

An exciting character I had to design was the leader of the bandits or the boss, like we would call him. He was to be in the final room with Ayita's friend , being the one to pull the trigger on her. For him I had quite a bit of freedom as no one had really expressed any opinions and how they wanted him to look like. So I created this image in my head of a man who would be a bit more intelligent than the average bandit, he would always pick the best loot so he would get the fancy clothes and he would actually look a bit neater from the rest. I wanted to give him a serious face and I was even thinking that he could be a former ranger of some sort, someone more organized and clever who became an outlaw and gathered his own little group of bandits who would do the dirty work for him. Having that in my head I went on a hunt for references. I was mainly looking at rangers and some cowboys to get what I needed for creating the leader. I started doing a paint over for his face when I was asked to switch working on something else as the boss was no longer a priority.

I did some concepts for the environment as well. Especially for that I gathered a lot more references and got some brushes for Photoshop to speed up my process. After playing around with the brushes a bit, I did a few photo studies of images of plains to get the feel of it and the right colours.

Then I did some concepts for the tunnel where the train was going to disappear into the darkness. 

And some other doodles and notes I made as reference for when drawing and modelling wood planks.

The thing I worked on next was the ranch. Martin already had a design for it and a 3D mock up model which he then gave to me and asked me to do some concepts around it and think about the textures. I decided to once again follow Ewan's advice and go down the paint over route. I gathered textures, spent quite some time editing and tiling them, and then used Martin's render to lay them on top and make them work in the 3D space. For the landscape I used a painting concept I did earlier instead of a photo because I thought it still worked quite well. I spent a couple of days just working on this ranch concept, laying textures over each other, cutting from photos, gathering references and just stitching it all together. And I was quite pleased with what I produced in the end. It actually turned out to be very useful for later on when I had to do the actual textures of the ranch as I already had the images saved from when I did the concepts.

For the texturing of the actual ranch, I was sent the unwrapped model from Martin. But it was left to me to resize and arrange them to fit the 1x1 texture space. Annoyingly, this took me a lot longer than expected. Almost longer than doing the actual textures. It was quite the challenge to take a UV shell and to have to judge how to fit it in the square while keeping in mind that you need to leave enough space for all the other 100 UVs Prioritizing larger and more obvious objects was essential to this. For example I made the front of the ranch much bigger than the back because that is what the player will be looking at more and that is where most of the details, such as the windows and doors, will be going. And all the UVs needed to be unique because at the end, i had to make a normal and AO map. It wasn't easy, but after a lot of hours of tweaking, moving and resizing, it all fit well in the end. After I had that I used Maya to bake the AO map. I had to fix some bits and then I started making the diffuse texture. I mainly used images I already had from the concept art, except that I tried to tile them better and make them more varied on places to make it less obvious they were tiled. Having unique textures was useful for that. After the diffuse was done I merged it with the AO map to make the final texture. All I had left then was to make the normal maps. I used nDo for them and after tweaking a bit, the ranch was ready. Because of the nature of the materials out of which the ranch was made out of, we didn't need to have a specular map.

An addition to the ranch which I had to create were the cellar doors which were actually the final destination and then end of our game level. Ayita was to sneak around the ranch, avoid the bandits and make her way into the building through the basement. Modelling the doors and the base for them wasn't too hard except a really annoying problem I had with the pivot points. I managed to work my way around it and simply textured it in the end. I also had accidentally flipped the normals on a couple of the objects and I found out that when the AO map didn't bake correctly. That was an easy fix, only that I had to wait again for the normals to bake before merging them with the colour map. The doors were supposed to be animated in Unity, so that wasn't something for me to worry about.

I did some other models as well, such as the top part of the well. At first I made it out of simple boxes and after I was satisfied with the structure, I went in and added some variation by adding a couple of edge loops and doing some twists and bends to make the boxes look more interesting and resemble wood. For the textures, I used some I already had from the ranch so that I could match the style. One would imagine that a thing like the well would be made out of leftovers from the ranch anyway. Just like with everything else, the baked AO map was merged with the diffuse and done.

Other small objects I modelled and textured:
- Hay barrels
I made 3 types of hay, one barrel and 2 piles of hay.

- Chopped wood

- Ax

- Hay Pitchfork

- Shovel

- Bucket

I also helped Martin with unwrapping and texturing some more of his models. I did:

- The bridge

- The cart wheel

-The rain barrel
I actually did the concept and blueprint for that one as well.

- Rail tracks

As the main 2D artist another job I had to work on was the GUI. From the start I was trying to come up with a cool title, following advices from different people. The initial course I took with that was to look at Western films and other games to see how they have handled the problem. That's how I came up with our placeholder logo, but later I was told that we would need to change it. Martin no longer wanted the Western feel because technically our game wasn't a Western, it took place in the Wild West, but our main character is a Native American Indian. We wanted something more fitting.

Due to other more urgent things I had to work on, I kind of abandoned the title and left it till the end. I tried a couple of more things, but unfortunately I was only limited to the fonts Photoshop had because I was working on the uni machines and we don't have the privileges to install fonts. Which was quite unfortunate because I managed to find some really interesting Indian inspired fonts. I was going to try and imitate one of them by hand, editing one of the simpler Photoshop fonts, but then I was asked to work on something else.

I was asked to do the mini map. Unfortunately I wasn't allowed to make it all fancy and pretty how I had it in my mind, but that was fair enough since we didn't have much time left anyway. I made a simple circle, an arrow to indicate the main character and a red dot with a radar for the AI enemies.Very simple and basic.

I also had a couple of ideas for the start menu screen, but we didn't even need those because Martin had something completely different in mind and my concepts were irrelevant.

One of the main things we were going to have in our scene was a train. We were going to have a cutscene at the beginning with our character sitting and waiting inside one of the wagons before jumping out of the train and heading off to the ranch while the train drives away in the distance. I spent a whole day just researching trains from that time period, gathering image references and diagrams to help me make the blueprint for the train which Martin was then going to model. We even had a small scale toy model of a train which was trusted to me to use as a reference for drawing. Making the blueprints of something as technical as this was very tedious and took me quite a bit of time. I even went and found out what the sizes of the elements needed to be and marked them on the blueprint.

The last day I was handed Ayita's unwrapped model for texturing. I had to really rush that because obviously I had no time to do a proper job. I quickly laid out the UVs, without thinking too much and then used my concepts to get some placeholder textures. Ideally I would have spent a couple of days just working on that, finding more fitting textures, tiling them properly and putting them all together.The face would need a lot of tweaking and painting as well. I also realized I barely have any experience with texturing cloth and organic models. If I had the time, I would have probably looked for a couple of tutorials just on that because I would imagine it must be a bit different from texturing hard surface models.

I'm sad I couldn't finish the textures, but I'm not disappointed or mad. I didn't have time, but I'm not blaming anyone for that. We had so many technical problems with the character model that it's completely understandable why it took so long. I'm only sad because I wanted to have it textured for Michael's sake so he could see his model come to life. I probably will finish it anyway, even when it's past the deadline. I'll add the details I modelled for it as well and then we could pose her nicely and make renders.

So in the end my list of jobs consisted of:

- General Research
- Reference image gathering
- Character concepts
--- Sketches
--- Paintovers
- Environment concepts
--- Photo studies
--- Paintovers
- Asset concepts
- Asset models
- Asset UVs
- Asset textures
- Character textures
- Character accessory models and textures
- GUI concepts
- GUI textures