Games Production - Taking the Gun into Quixel

For the gun to look good in Unity, I had to create an Ambient Occlusion map. It would find the location of the shadows and enhance them. It gives us an extremely detailed result. There are several ways to get the AO, I choose the mental ray Option.

I started off by selecting the object that I wanted to create an AO out of. I went into the hyper shader menu and selected the Surface Shader. The next step of the process was to create an Ambient Occlusion node which we will attach to the output of the Surface Shader. I had adjusted with the settings to give me the best possible result in generating an AO map. If you look below, you will see the final settings that we ended up with to achieve the result.
  

Adjusting the Shader's to create an Ambient Occlusion. 

Setting the Ray distance for full coverage.

All I had to now was to got onto the renderer tab in Maya. By clicking on "Lighting/Shading" you end up with a button which states "Batch Baking (mental ray)". I selected this tab and adjusted the setting accordingly (look above). By clicking on the bake option, the program would proceed to do so. This took some time to get the perfect AO map for our gun, but the end result was worth the trouble. This meant that it was now ready to take into Quixel.

The Maps

The Colour I.D Map.

The Ambient Occlusion Map.

The Specular Gloss Map.

The Normal Map.

The Albedo Map.

I am happy with the way that the maps had turned out. The use of the colour I.D map was extremely important when it came to texturing the gun. As you can see, there are a variety of textures to enhance the feel of the gun and make it seem more life like. This is the first thing that I have gotten to texture alone and the result was pleasing. 

Of course there are problems with some of the textures, such as the U.V's on some of the edges weren't done to the best of standards. But at least now I know, that when it comes to producing something else of this level, what to look out for.

The Gun with a Wood Grip.

Gun form another angle.

The Gun with a Metal Grip.

The Gun with a Metal Grip from another angle.

There were several different varieties of the gun that I came up with. In the end, I was stuck on these two different result. One with a wooden grip and one with a metal grip. As this asset was going to be in constant view of the player, it required a team meeting to decide what texture we were going to keep the grip. 

Overall I am happy with the result, as this is the first time, on my own, texturing an item. There are many aspects that I would like to change, if I had the time. One such problem is the weathering that the gun has on it. I would also next time like to use the DDO painter that is built into Quixel. It means that we could paint detail onto the model. Though again, I am happy with the result that I got in the end, especially with how much I learnt from the creation period all the way to texturing it and will definitely use this method of working the next time that I create an asset for a game.
 

Games Production - The Construction and Mapping of the Props

Again, I had to work with our Props Concept artist, who concepted many different assets to be scattered around and placed into our game. I also worked quite closely with our game designer, where we decided that there were going to be two types of assets. One being the Static Objects, the other being the Dynamic Objects. The static objects where assets that weren't going to move around during the game play. This was first priority, as the game designer needed to adjust them for the game. When the game was being baked, these still assets need to be present during the process. Cover points for the enemy have to also be added behind certain assets. Dynamic objects are different to there. The Dynamic assets have to be added in after the bake as the objects will move around and have physics behind them. These assets where smaller and more durable.

After having several concepts given to me and getting real life references myself, I started off with the construction of the static assets. These assets include a table, two types of crate, floodlights and a safe. These assets needed to be low poly as they were going to be scattered around the level. I managed to achieve relatively low poly assets. Me and the other Environment Artist, Ben Matthews, felt that the other room in our game needed to be filled. So we decided that I would create a tarpooling that will drape over the edge. This, itself, is still quite low poly. I would use a new process that I haven't tried before, the use of nCloth. This was combination of the physics and animation in Maya. The result was actually pleasing and really does add to the level.

Creating the Static Assets .

The next part of the process was to U.V map the assets. Many of these assets, such as the crates and the tables, were very square, so mapping them wasnt a hard task, it just required several cuts along the bottom and sides to fully map it. The difference was that as most of these were using the materials that had already been constructed by Ben (the other environment artist). This meant that the U.V's had to be scaled so that the materials would suit the assets and not look stretched.

Mapping the Static Assets.

The U.V have been adjusted to Suit Ben's Material.

Extracting the objects required the mesh to be separate. so that in Unity, the texture that had already been made by Ben, could be applied by to the assets. As you can see from the picture above, the U.V's have had to be scaled. This was a process that required me to take the asset into unity, apply the material to the assets and have a look at whether the U.V's needed to be extended. This crate that is pictured above didn't look good with the material on it, so I had to go back into Maya and then increase the size of the U.V's. As I was using materials that had already been constructed, The U.V's didn't need to be within the U.V range 0 to 1. So the U.V's here are much bigger than that.

The First Crate Type.

I decided that It would be best if I obtained so renders of the assets that were to go in game. This will help me and the Game Designer get a realisation of what the assets will look like in game. I tried a new process of rendering. This was something that I saw in a 3D Total magazine called "RenderMan". This was a process of rendering models in Maya that the famed animation company "Pixar" use. I applied a reflective metal and adjusted the settings of the crates. I also used an environmental light in the scene. I am extremely happy with the render and will continue to use this. 

The Second Crate Type.

I decided that to give us more variety in terms of cover. So I copied concepts that I had been given from our props artist. By having the two different varieties of crate, the player will have an unfamiliar environment. This crate has an unusual geometry, so narrowing down the edges of the crate was the biggest challenge. In this render I have used the same process as the other crate. This time though, I have used a plane as a floor and applied a reflective material onto it, this again, helps me and the game designer get a visual of what the assets will look like in game.

Health Pack/Safe.

At first, there where plans in which this object was going to be a health pack that the player picks up. We all had a team meeting and decided that it would make more sense if the health pick up would be the tubes on the enemies back. So we were left with this asset and felt that we would rather it not go to waste and still have a part to play in the game. That's when we felt that it would be best as a safe that would be placed on the tables to help fill the environment.

Health Pack/Safe.

I tried a new process of rendering, putting in clear glass. I'm not 100 percent certain whether or not we are going to continue with the glass. It would involve creating an inside for the safe. Also if it is a safe, we would want to present to the viewer what the people of this facility are hiding in the safe. We will apply a material the casing that will be different to the panel. This will separate the two sections.

 

Some of the Dynamic Props to feature in the Game.

We decided as a team that we were going to create objects that will be moving in the game. They will have physics behind them and certain objects, the player will be able to interact with. Again, I was working closely with the props artist Rastislav. He gave me a list of objects to construct, with some concepts. Others I had to make up or gain reference for. 

As the objects were going to use their own materials and maps, the idea was that I would have all the assets set up on the same texture sheet and apply the maps as a material to several different models. By using colour I.D's that I used in the gun, we can decided individually what model will have which material. This will save us on texture sheets, in which we are limited on.

I have split the assets into two different sections, "Assets_01" and "Assets_02". This means that each object has perfect sized materials and its easier to manage.

The Dynamic Props getting Mapped.

As you can see from the picture above I have managed to U.V unwrap the first set of dynamic assets. All the objects that I want to texture on the same map, then, like the gun, I press the "Layout" button so that all of the U.V's are on the same map and will not overlap each other. I have tried to keep the U.V's as neat as possible, so that when I come texturing the assets, there will be no stretch or distortion.

The Colour ID Map generated for the Dynamic Assets.

I tried a different process this time when it came to creating a Colour I.D map. Using the rendering process "Turtle", I applied coloured lambets to the objects that I wanted to have different materials on. Then I opened up the renderer to and converted it to bake an albedo map with the colours still attached to the U.V's. I then had all of the objects that I wanted on the texture sheet selected and then by simply rendered the current scene. This gave me a perfect colour ID map and saved me a lot of time, instead of having to go in manually like the process that I took with my gun.

Extra Dynamic Assets to go in Game (Mainly Transparent).

Not forgetting that I also had extra assets, these where also dynamic, but I felt that it would help the work flow to split the dynamic assets in half. As I had 20 assets in total, a 10/10 split felt necessary. Again, I copied the same process as before. These objects where more transparent than anything. Ben had already made a fantastic glass texture, but I felt it needed to thicker, I took the texture and modified to create my own thick plastic material. Ben had done the majority of the work of this material, I just copied it and turned up the specular gloss opacity, so that it would appear thicker on the texture. This suited the majority of the objects perfectly.

Extra Dynamic Assets being Mapped.

As you can see from the image above, the U.V's are laid out neatly, ready for the texturing process. As I had already created the material "Thick Plastic" the majority of it didn't need on this texture sheet. If I were to do it here, it would affect all the other materials, which is something that I dont want to happen. This was discovered after going through the mapping and Colour I.D process, so they appeared on these maps, but were not affected by the materials applied on it.

Colour I.D Maps for the Extra Assets.

I also used the same process that I used for the colour I.D map before. This was because it was a very easy and quick process and managed to save me a fair amount of time. The green parts of the I.D map are the meshes that were going to be transparent, so they will not be relevant to this map.

Propane Tank.

This is a propane tank that I decided to model. I was told by the game developer, that if we had time, it would be interesting to have a object explode if you shoot it. So I went away and looked up real life references for a propane tank. Using images I found on the internet, I managed to construct a decent model of a propane tank. It will be good to put this asset in scene, even if we dont get the chance to make them explode, as it will help fill up the scene.

Lab Goggles.

The goggle have two different materials applied to them, one being the transparent plastic that I had modified from Ben's glass texture earlier and the other one being a black fabric which would represent the strap of the goggles. This has an interesting shape to it, I used the Curve tool to create the strap, which game me decent representation. The goggles, I feel, was a decent model with not too many tries in it. As it was transparent, the inside of the goggles had to be modelled as well.

Scalpel.

This is an assets that I had made of a simple surgical scalpel. As this was a research and development facility, with the addition of the chunk of flesh that is located in the centre of the map, I felt that this asset suited the project perfectly. There will be a shiny metal texture that will be attached to it.

Scissors.

I created a set of scissors to feature in our game. The problem is that for something as small as this, it has far too many polys in it. I was trying to make a real life model and at the time didn't take any consideration on how it will affect the game. So If I were to create this again, I would at least aim to half the poly count. It will also use a shiny material.

Test Tube Lab.

This was a simple model to create, but again, perhaps there are too many polys for such a small object. As a research facility, there will be many of these spread around the map, so it may affect the game play, which is something we may wish to avoid. This will use that see through plastic material that I modified. It will also come in a pair with another assets that I made, which is a "Test Tube Rack".

Test Tube Rack.

The reference that I used for this object was a school test tube rack. Obvious to its name, it will hold the test tube and display them. This was as low poly as I could get the rack, especially with the holes in the mesh. I used the same method that I previously used in the gun barrel, which gave me a great result. As you can see, the holes aren't a fully round circle, this is to save on the poly count, with the main effort behind these assets being dedicated to making them low poly, so it doesn't take many tris out of our maximum.

Wet Floor Sign.

In the concept that Razvan had created, he featured a wet floor sign, I felt that not only was this an easy asset to construct, but it was also more interesting to the scene. It has a fairly simple shape to it, with a traditional yellow plastic that will be applied to it. I will also go into the make and add text which reads "Wet Floor".

Fire Extinguisher.

The fire extinguisher was created off of a piece of concept art that was created by Rastislav. It was an interesting, more futuristic take on the fire extinguisher. So I followed the design closely and came up with a great result. This will be the basic textures that will be attached to it, Making it stand out on the grey shiny walls.

I was happy with the amount of assets that had been created, with them being scattered around the map. The next step of the process was to move on with the texturing process as all the assets had been mapped.

The tri count for all the assets in total, including the gun, numbers up to just under 70,00. This was great as it gave our other Environment artist (Ben Matthews) and Character artist (Jay Pickett) a decent amount to progress with their part of the work.

Games Production - The Construction and the Mapping of the First Person Gun

As I was one of the two environment artists, I was tasked with creating the gun and the props for the game. Rastislav Smolen is our prop concept artist. This meant that I would be working with him extremely closely, getting concepts off of him and altering them when need be.

The first task that was set to the both of us, was to create the gun that would present in the game, this would always be in the players view, so it had to have an interesting shape to it. Rastislav originally had come up with a fair few concepts, this included a shotgun and another variety for the machine gun. We decided that we would also have to look at the environment art that Razvan Prajanu had constructed to make sure that the gun would fit in to the environment.

Rastislav's Concept to follow. 

This is the concept that Rastislav had for the final gun. I was really happy with the plan he had given me, it was clear enough to follow, it also meant that the base would remain pretty much the same. There were some adjustments that needed to be made on the gun, not only to add detail, but to make sure the gun worked.

The gun has been spilt into sections, this makes it easy for me to U.V map it when it comes to it. Firstly you have the main base of the gun, this includes the grip and the handles. The changes that have been made here are that the grip has further detailing on it, with the addition of holes so that the steam can escape. On the other side of the gun, we all felt that a pin needed to be added, for animation purposes and for the gun to seem more realistic, so on the body of the gun, a gap for the pin needed to be included. Above the main body is the sight, where I have added the ridges past the scope. This again, was to give the gun a more unique feel, with it still fitting in with the environment. To the back of the main body is the Stock of the gun. There is not much change to this part of the concept, except with the addition of the soft back for the shoulder to add comfort to the gun. The Barrel at the front of the gun originally had grating in it. It would follow the concept that Rastislav had put up, but as we where limited on the tri count, so as a team, we decided that it would be best to remove this part and add a a straight barrel.

12889 Tri's for the base of Gun.

The final Tri count for the main part of the gun was 12889. This was a decent amount, as we had divided the tri's between the 3D artists and this was no way near the total that we were expecting it to be. This was due to me retopologizing the model and adding triangles into the sections that weren't going to be animated. At first, I was unsure on this process, but then my tutor reassured me that doing it will contributed towards our project.

The Addition of the Pin.

As I had divided the parts of the gun into different sections, U.V mapping the gun wasn't too hard to do. The reason why the gun needed to be mapped is so that when we apply a texture to it, it will not look blurry and will fit in well. This wasnt an extremely long process, but out of all the parts of the gun, the main body was defiantly the longest. I wanted to make sure that on the gun, the seams didn't appear on the left side. This was the side that was in constant view of the player and the seams would drastically affect how the players see the game.

The Gun Base Mapped.

One of the things that needed to be created separately was the magazine of the gun. This was because the magazine is going to be animated in the game. The magazine will be taken out by the character when they reload and a new one will be placed in the gun. As this was a fairly small aspect of the game, the magazine tri count had to low. It was also retopologized, with a few triangles added to the magazine for it to keep its shape. If I were to do this again, I would make this just a plain cube and use the normals, this will drastically reduce the polys, but as we have plenty to spare due to the low poly gun, then for the moment, this is OK.

246 Tri's for the Magazine.

As this was an extremely small asset, it was very easy to map. All I needed to do was cut it along the bottom and split the 4 section of the square. Then using the unfold tool, I could brush the U.V's and have them looking perfect. Again, this was an easy process and meant that it was now ready to send to Quixel to attach a texture to it.

The Magazine Mapped.

The last section of the gun was the pin, this was two separate meshes that were combined. The first mesh was the leaver of the pin, this was created using a torus and then stretching it along. I then got a flat cube and combined the two. The stretched cube will retract every time the gun has been fired. This will then be pulled back by the characters arms when reloading the gun. This was a later addition to the gun, as this wasnt present in the concept. Our team had decided that a pin would be required, so I constructed it and try to make it as low poly as possible. 

The pin of the gun is 676 tri's in total. This is mainly from the leaver. There is a reason for this, as when the character reloads the gun, This is the first thing that you see. Though it still is a vast amount wasted on such a small object. If I were to do this next time, I would defiantly reduce the amount, using the maps to help me here.

676 Tri's for the Pin.

The again, was extremely simple to map, as sit was two separate meshes and required very little effort in figuring out where to cut the U.V's. Once everything had been mapped, I selected all the meshes and then went into the U.V editor. Here I selected the button "Layout". This meant that all the meshes U.V's I had selected fit in coordination with each other. This has saved us texture maps, as we can texture multiple assets, with this method of texturing. Once I had all the Meshes selected and laid out, I took a U.V snapshot of the U.V map. This was created so that I could make a colour ID map. By creating a colour ID map, Quixel will recognise what part of the map needs texturing.      

The Pin Mapped. 

Below is the U.V snapshot that I created. By putting it into Photoshop, I can manually add what texture I want to go where. As you can see, the U.V snapshot gives us an outline of where the lines of the mesh. This is fantastic if you are going through the manual process that I am going through. It means that I can choose exactly where the texture is going to start and end.

U.V snapshot ready to colour for Quixel.

The picture below is the manual Colour map that I had constructed using the U.V snapshot from Maya and Photoshop. I have had to use a variety of colours so that when it comes to putting the gun into Quixel, there are many textures that will feature on the gun. The core part of the gun is a dark blue. Once this is Quixel, The Blue will be identified and a dark, perhaps black gun metal will be attached to it.

The Colour ID Map ready for Quixel.

I decided that I wanted to get some interesting renders for the gun. This was for two of our concept artists. One was for the character artist, Steve Welch, who wanted to create images of the players character and a comic book to come along with the game. I gave him one of my renders, in which he used as template and painted over the top. I was also asked for a render of the gun by Razvan, who was constructing the HUD for our game. The idea was that the gun was going to appear in the bottom corner next to the ammo count. He wanted to make it fairly transparent, so it doesn't distract the player, but makes them away of what the gun looks like from the front.

The Final Gun (Barrel later removed to save poly's).

The Final Gun.

Render from Behind.

Render from in front.

Where the Gun will be Positioned in Game.

Gun with Detachable Magazine for animation.

The object was now ready to be extracted and placed into Quixel. Everything had also been mapped, so now it was time to take the mesh with us. To do this and making sure it was easy to handle in Unity, the game engine that we were using, I had to centre the mesh. I had to select the option "Absolute Transform" and centre the pivot. In the boxes next to the absolute transform, that are labelled X, Y and Z, I entered 0,0 and 0. This took the mesh back to the centre of the scene. I then extracted the models, with the magazine and pin (not combined) as an OBJ. By centring the mesh, It makes it easier for George Burchmore, our Game Designer, to control the model in Unity.

Games Production - Researching Abandoned Facilities

We were sent off by the game designer to gather research for the game. As the idea was that your character was suppose to navigate their way around a research and development facility, I decided to go away and look at what abandonment does to buildings. This was a great piece of research, especially with me being one of the environment artist. It gives me a sense of what the rooms should look like and what the player should experience as they navigate their way around the map.

A real life Abandoned Laboratory.

An Abandoned Barracks in Berlin. 

As we had many different rooms to feature in the level, there was a huge variety in which I had to look at. Not all the rooms look the same, the time has taken its toll on these rooms. One thing we had all decided on is that the rooms have only just been abandoned, with your character being a first response to the attack. This means, unlike some of these rooms, there will be no plant life growing or rust starting to attack the machinery. 

One thing that is common in all of these pictures, is that a story is being told in every aspect of the rooms. You can evidently see something has placed items there and left them to rot, on no fault of their own. If we are to create a realistic experience with this game, then we need to think outside of the box when it comes to placing the assets, not just randomly placing objects hoping that it works. 

A mood-board of Abandoned facilities.

I finally decided to finish this part of the research by creating a mood board which looked at many different rooms, whether in an actual video game, or in real life, that had been abandoned. This is so that I could get an idea of what the effect abandonment had on our facility. One thing that is a common occurrence is the use of light so that the player has a way point to follow. Also many dark tones have been used. Staying away from the regular bright shades that many clean facilities have.

This has been a great process as I now have an idea of what the game will look like. It has also given me an idea of what makes all these images so captivating and how we can incorporate into the environment.  Also if we want to save on tri's, a modular environment needs to be constructed and in the abandoned space station images, we would have to construct one piece of a wall and then repeat over to construct our environment. 

Games Production - Game Design Document (1st version)

We were told that we would have to construct a game design document for the game. This would outline what we plan to do with the game and what inspiration we can draw on. The game design document is also essential as it gives us a plan to follow, something that we can stick to or refer back to, if we ever get lost on the project and fall off the path. In this Game Design Document, we are aiming to set out the main premise for our game and this document will act as a template for us to follow.

Example of the game design document (The X, The 4 Pillars, Story).

It was decided that I would write the first Version of the game design document, then our Game Designer, will pick it up and adjust it. This is so that our Game Designer can focus in these first few weeks to get the game going, so that the rest of the team can proceed when it comes round to him constructing the paperwork. There are many aspects to the game design document that have to be included. Below is a section of what is included in the game design document that I constructed.

-Content

-Scope

-Conventions

-The X

-The 4 Pillars

-Story 

-Art Direction 

-System Requirements

-Software

-SDKS 

-The Player/Core Mechanics 

-Controls, World 

-Landscape

-Objects 

-NPCS 

-Sound

-Game Flow 

-The Player

-Setting

-Game Structure  

-Action 

-Objective

-Graphics

-HUD

-Limitations

-Intro

-Menu System

-The Team

-References

System requirements, Software, SDKS.

Update - As the game has now changed from a story based level to a wave based, many aspects of the game design document need to be changed. This includes the "The World" and "Landscape". These are issues that need to be dealt with, as if we fall off track, we cannot follow this document as its a different type of game. When possible, it needs to be updated. Some elements can be kept, where as others cannot.

Games Production - Researching into other First Person Shooters

We, as a team had decided before the project started, back in the Christmas period, that we wanted to create a first person shooter. This was going be an extremely difficult task and would require a lot of work, but the perks out-way the risks in this case and decided that a challenge for the team would be best. This time around, we have a fairly large team, consisting of 7 members. There are 3 Concept artists (one environment, one character and one prop), 2 Environment artists, 1 Character artist and 1 Game designer.

I felt that before the project, one of the best things we could do is market research. This was looking into other games out there, primary First Person Shooters, seeing what feel we want the game to have, how we could make it fun and what our game unique.

After talking to the team, we all decided that we wanted the game to have a retro feel behind it. It had to be fast paced, where the player is constantly is on the run, enemies are less an obstacle, where the player takes their time to defeat them, more, rushing through them. One game in particular a certain member of the team referred us to, was the original doom. This was one of the first games to have the player fight enemies from the characters eyes, at the time, fully immersing them in the game. There's many things that the game does right, which we want to include, such as the pace of the game and the level design. To make our game original, there are aspects that we wish to avoid. One such is the placement of the gun. Times have moved on to incorporate a realistic feel, like you are actually looking through the characters eyes, so instead of having the gun in the middle of the screen, it will be placed on the right. We have also decided that we are going fro a 3D game, where as compared to Doom, which has 2D sprites as its enemies.

DOOM (a godfather of the First Person Shooter).

A game in particular, which caught our Game Designers eye, was the recently released game, Wolfenstein: The New Order. This is a modern game that manages to retain the feel of a retro styled game, but convert it to a modern audience. It is a remake of the classic Wolfenstein game series and manages to feel like a fast paced game, while including many themes from the original. As we are creating a 3D game, this is a good inspiration. We have many stretch goals, one being the inclusion of other guns, but if we do not reach that target, the gun that the character starts with needs to feel fun enough so that the player doesn't become bored.

Wolfenstein-The New Order

Another timed game that interests us is Quake. This is very similar to Wolfenstein and Doom, as it includes elements, such as pick-ups and and fast paced action. Again, this is a game we need to look towards as inspiration. The pick ups will be something that needs to be included in the game. It keeps the player running, instead of letting them camp in a corner until all of the enemies are defeated. They have to push themselves and take risks in order to succeed in the game.

Quake's Health Pick-ups.

Update - Due to a team decision, we have decided that in the time frame, we are no longer going to construct a level, but will now create a wave based game. It means that we can save on the tri count, the map limit and make the game run better. This is a great idea as our main issue is performance and the limitations that we are restricted to. The way that we will get around the tri limit will be to make the individual assets, such as the walls, pipes, etc and then put them into unity and duplicate them in unity. We have been told, because it is the same asset, it does not count as any tries towards our tri limit. This gives us a bit of breathing room towards the limitations.