It's only a few days before portfolio night and we're doing some last minute play testing. I worked with D'Merro and tested a lot of things/scenarios. Some examples are testing certain glitches/hacks at a certain time in a match e.g using glitch when someone dies. Here's a link to the notes that we took.
This short sprint, I'm mainly working with Randy (blog) in order to fix some of the issues with the spear animations. Now, he is pretty much doing all the fixing/animating, but I'm there to help whenever I can. I gave critique on the animations and I think that the updated animations look very good.
Credit: Randy Grout
Update
Now there is one big issue that is occurring and that is with the animations and the spear alignment. Originally, the spear was is the character's left hand and that was switched. However, when exported to unity, the alignment isn't working. In the max file it works because there is a link/hierarchy including the spear itself. For whatever reason unity is not recognizing it.
Update 2
I suggested that the spear from the max file be included when exporting. My thought is to replace the spear in unity with the max spear. They are the same except one is missing textures.
I have good amount of animations in the build, but I don't have any sort of recoil for the crossbow attacks. While it isn't much, the animation can go a long way in bringing more life into the character. Mainly a recoil type of animation will make the character look like it is actually firing/shooting and that the crossbow is acting upon the character. I chose two type of animation: standing firing and a running while firing.
I after trying to learn programming/shader work, I ended up giving up a bit on that. However, I did decide to make another base mesh. I figured that the current base mesh was a bit simple in terms of proportions/body type. I thought that I would try to make a more muscular type of character. In the end, I think the the mesh is pretty good, and I actually some of the topology is better than the current base mesh.
This sprint I decided to change things up a little. There wasn't much in terms of art that was planned. I would help if necessary. I decided to take this sprint to learn a little bit about shaders. There was a little demonstration in class a few weeks back, but otherwise I never had any exposure/experience with shaders. The whole concept is confusing since, at least in unity, it requires a bit of programming knowledge.
Update
I'm not really getting the whole programming language behind the shader videos, so I was told that it might be helpful to learn so basic programming language. I'm going to try to learn a little bit of c programming. Hopefully this will make understand shaders a little easier. However, I'll be honest, this isn't the easiest/most exciting for me.
I'm still going over a majority of the animations. I realize that a on-hit type of animation would be useful, so that the user can see the character take the damage. Since I'm using mixamo, I'm mainly slightly changing some settings for the specific animation. Only issue is, while I can edit the animation, I noticed that I have some difficulty with importing the fixed animation. For now I'll just mess with the original settings.
I'm starting a new sprint so I'm still figuring out what I'm going to do. I had discussed on possibly working on a female character, but I think that will most likely be put on the back burner. I think for now I will just go through the animations of both the crossbowman and spear man to make sure things look good.
While I had finished some hand made animations, I have been looking at other possible options. One possible option is Mixamo. It allows you to upload an fbx file and their system can auto rig your model. I wanted to try it out and it worked scarily accurately.
After I got my model auto rigged, I tried importing the files into Unity. The files worked as far as I can tell, so now I can look for more animations to attach to my model. This may seem like a cheap and quick way of doing things, which technically it sort of is, I feel like it's the most efficient way. I'm focusing on becoming a character modeler/texture artist so if I can avoid doing animations the better for me. If I have to animate, I will, but if there's a quick and efficient way to speed up the process I'll try to take that path. I then can use my time helping others/working on new things.
I finished the textures for the William Tell armor. I had based the textures off of the weapon textures. I wanted them to match or at least go together. I know in the original concepts everything was in black and white, but I never ended really intended to have the textures stay black and white. The concepts were merely to show the armor type/look. The base mesh texture is a wireframe composite material from 3ds Max, so in game engine it will only be plain gray. I'm thinking that the wireframe look can be done in unity using a shader. As for animations, I've been using some animations from last semester and trying to edit it to my liking.
Along with the armor, I felt that the crossbow could use a visual update. The previous crossbow wasn't bad, but it wasn't great either. I went for a futuristic look but also something eye catching. At the time that I was texturing I was thinking of an old project I did my first semester. It was a shooting game involving a lot of spiders. I based the textures off of that texture.
I finished the armor modeling. However, I had a lot of geometry that I felt that I could remove. In the hard surface modeling tutorial I watched, support edge loops were used a majority of the time. These loops helped with telling 3ds Max that the edge between the other two was a hard edge. Although these loops helped initially, this model is for a game, and geometry is limited, so I looked up trying to use smoothing groups on the model to achieve the same look but with less geometry.
Video on using smoothing groups for exporting in Mudbox.
I managed to use smoothing groups effectively. I was able to export my file into Mudbox while keeping the hard surface look that normally would have been smoothed out. I was able to optimize my model. Below is a comparison. Utilizing smoothing groups, I was able to cut my poly count by roughly 2500 while having little to no change in appearance.
After watching the video, I got a good idea and concept of hard surface modeling. I modeled out a couple of pieces.
After receiving some critiques on my concepts, I can start working on modeling out the armor for the William tell armor. I decided to try to take elements of the concepts and model out armor pieces. I'm trying to maintain the style but give it a futuristic look to it. It's a little challenging because I never really liked modeling out hard surfaces since I've had difficulties with it in the past; I prefer organic modeling. I looked at some tutorials to see if I can get a better understanding of hard surface modeling.
After doing research. I created several concepts for the William Tell character. I used the real-life references as my influence. I focused more on the type of armor and design of the armor than the color. Also, I realized that we want a futuristic look, but that can be accomplished through the model/texturing.
The art team is doing a bit of back peddling. We're starting with coming up with concepts for the crossbowman/rocket man. We had some concept art from the previous semester, but a lot of them weren't used. Also, there wasn't much direction when we created the original concepts. By redoing this step, hopefully we'll have a more streamline art style. I'm focusing on the William Tell character, and primarily researching armor from that era. Then once I have some good references, I can start making concepts. Here are several examples I found.
Over winter break, I had practiced my modeling a little. I created a low poly male base mesh. Previously, we had only created armor pieces and skinned them to a biped system. The main issue with that was that if any part of armor wasn't covering something, the result would be an invisible body. The crossbowman was a prime example since the helm was merely a hat. Since we didn't have a mesh underneath, the entire head/neck area was invisible. The new mesh will fix any issue of that invisibility.
The semester has started and we're going right into it. Last semester was a bit hectic. I think we'll be planning on redoing a lot of the art assets/models. The ones from last semester were ok; not horrible, but great either. This first week I'm going to have the art team rework our pipeline. D'Merro came up with a concept pipeline over the break so we'll be testing it out. Primarily, we're just going to make sure we can get from a 2d concept to a 3d model/animations; we want to get these into the game engine smoothly.
