williamtatenextgen-blog

[williamtatenextgen-blog]

Synoptic Reflection

During the time that I have been working on the synoptic project with my team there have been a few complications but the product is coming together in the end. 

Communication

The communication in our team was not great to start off with. We made the asset lit as we were working so people didn’t really know what to do and we had no kinds of media messaging as well as going to see each other. To tackle this issue we started having more frequent scrums and creating a group chat so we can communicate outside of college. The scrums helped us know where we were all at and what we needed to do next. The group chat helped us be able to discuss things outside of college. Some people do still need to communicate more although these actions were taken and have shown improvement. The team needs to work as a “team” to be able to ensure the best final product and the highest mark we can squeeze out of this project.

Work Flow

The work flow has improved massively since the beginning of the project where people would be at the point of not knowing what they were going to do next or spending too long on certain models which some people are still doing may I add. I don’t mean to be negative but you have to be really uniform with the work flow and if people aren’t pulling their weight, they are holding the team back and then taking credit for something they barely contributed to. As of now most people are able to say that something needs to get done and then do it. For example we are filling the level as a product now with assets and we have room for improvements and I managed to model 4 or more models and get them all textured ready for use in a short amount of time compared to some who have barely made 4 assets in the whole time that we have been working on the project. We will be fine but it is not fair that we have to do more work ourselves to keep things in order when in reality it should be everyone contributing equally. For the people that are working to the correct standard though are doing amazing. To improve this the people need to pull their weight but what can you do.

Product

The product is looking amazing and things are really starting to come together and show the vision that has evolved over the time that we have been working on the synoptic project. One thing that is essential right now is getting sound because that will make up 50% of the product. It should be too hard sourcing out sounds or even making ones we will be able to use. There needs to be interactive sounds as well as atmospheric. Working with Unreal Engine is a lot better in my opinion over working on Unity as you do not need to code your own physics and the shaders in the engine are far superior and able to make literally anything nice in my opinion. Long story short we have a level and we are adding and modelling the final steps. Improvements would be to make sure that everything is uploaded to the drive and make sure everything makes it into the level as well as the sounds. 

[williamtatenextgen-blog]

Synoptic Project Update

Project Overview

Getting to the final stretch with the synoptic project. It has been a bit rough but we are finally getting a product together and everything working. This is the level as of now with the assets and lighting built. There are also particle effects to create a better atmosphere. All of this was textured by substance painter 2 and put together in unreal engine 4. I feel like in the time that we have had a few people could have communicated a lot better and pulled their weight but everything will be done because I am making sure to chase up the things that need doing. I know that I have put all of my effort in to getting my work done and all of it shared with my group so the product can be finished as soon as possible. I think to improve on this last stretch, those few people could maybe do a little more to push this to its final form and be a playable level. 

This is where the player will end up starting.

2nd shot

Models

My most recent models have been a giant mechanical door for the corridors and also some smaller assets like filing cabinets and lockers for rooms that are going to cut off from the main tunnel. For making the door, Instead of making two separate doors i have designed it so that you will on need one half to make the full model. On one side it has the lock and on the opposite it has a loop that the lock can enter so when two are placed in and one is rotated 180 degrees, they will fit together on both sides. This was simple to make using a semi circle and adding geometry and detail by attaching separate meshes but also using the tools given on the software (3ds Max).

Door in level

Door render

For the cabinet it was a very quick and simple asset. I made a cuboid with the correct dimensions for the level and then duplicated it. I then added detail to one making it the high poly version which I could bake down to 6 polys and plan to make it 4 because I can remove the back and bottom face as they will not be visible in game. The texturing was done using substance painter and smart materials. I played around with the levels on the masks and got the texture I was happy with.

More models will be uploaded.

[williamtatenextgen-blog]

Model Texturing - Synoptic

Here is a load of texturing I have been doing on my project for assets. Really enjoy using substance painter! Team is making great progress

[williamtatenextgen-blog]

Model Presentation

I made a professional looking presentation for my main 3D model in Photoshop which will be placed on my portfolio. I am planning to make my portfolio more uniform and well presented than last years.

Synoptic Model

I made another quick presentation of a model that I made in a short amount of time in 3Ds Max and textured for the project I am working on.

[williamtatenextgen-blog]

Main 3D Model - Matt's project

My Idea

For my main 3d model I decided to make a rifle for a video game. There wasn't really any specific reason other than I wanted to give it a shot and work with it in separate pieces. This model consists of 8-10 different parts such as the barrel or the weapon body which will make it easier to UV unwrap and texture as a whole. As I am using substance painter I am able to flatten map the meshes rather than take lots of time to sort them out by hand which saves a lot of time.

Low Poly

I set away modelling and I started on the first mesh which was the body of the weapon so I could scale everything to that piece of the gun. I did this because it made the most sense to me. I started with a cube and added edge loops and translated edges, added extrusions and got the shape I wanted for the low poly version of this model. I then did this for the rest of the weapon parts and added them all together to for the rifle. The last thing to do with the low poly was to UV unwrap them for the normal map and substance painter so I flatten mapped all of the meshes and saved them.

High Poly

For the high poly version I took all the parts and duplicated them to a different part of the scene in which I added detail on to them. I do not need to UV unwrap the high poly version as it will be baked on to the low poly version as a normal map and will be used as a texture which saves a lot of memory instead of having more trouble UV unwrapping and texturing. To add the detail to the weapon parts I used tools such as chamfer and adding edge loops for a more smooth model and then little extrusions for the smaller details. For example on the body and magazine I added little brackets and buttons like the safety and magazine release. Here is what the finished high poly version looks like and an image showing the model is in separate pieces.

Texturing

I centred both of the models in 3ds max and exported them as fbx files so I could import them into substance painter. The files were names "_low" and "_high" because substance painter recognises the files by those names when baking the two meshes and can tell them apart. I imported the low poly mesh first so I can bake the normal map on to it straight away. I did this by clicking the bake button and then changing the resolution to 2048 and then selecting the high poly model. All I had to do then was click bake. To texture the model it is kind of like using Photoshop with different layers but you can use different smart material and effects like dust or dirt. For the metal parts of the gun I used a weathered black steel and then added a dust layer on top. Some of the pieces have chrome parts on which was done my using normal material which I painted on using a brush tool on the original mesh layer with the UV map on. For the rest of the parts I added on a red stained metal which I changed the colour of in the setting to a sandy colour and then added dust on top. For these parts I also painted over some of the buttons with normal materials. The same process was used on the attachments. In substance painter you can see in the top right that the mesh is in about 10 different pieces. Here is an example of one of the normal maps, colour map and ambient occlusion.

Render

I went into render mode on substance painter and selected a background which effects the lighting on the weapon as it adapts with the selected environment. I chose the streets to get an average amount of light on to the gun so it wasn't to bright or low. I moved around the settings to improve the quality of the render then saved a few render images which I will present to Matt as well as the rest of the process making this model.

Feelings

I am really happy with the outcome of my model and it really wasn't too much fuss to create. I had originally started my main model as a dragon head that I was sculpting but there was too many issues with the mesh and would be impossible to UV unwrap with the amount of polys that it had. But I do plan on fixing it and uploading it as a separate personal project as I put a lot of time into it. But for this model I was able to quickly improvise and make a good standard model with realistic textures. I'm not sure if this would be considered industry standard but I am really happy with how it turned out.

[williamtatenextgen-blog]

Synoptic Project Overview/Update

The Pitch - Natural State

For the synoptic project this year I had to create my own idea for what I wanted to work on this year. Obviously it was competing with other projects but I stuck with what I am passionate about and what I want to see in a game and decided to express my creativity for my idea rather than make a game that everyone would want to see.

The idea that we had to pitch had to be related to our general area so I chose to use the bridges in a post apocalyptic world in which the human race was scarce and the bridges is a place for scavenging and is hard to access but beautiful in a way how the nature has taken back over the urban area. I named my project Natural Sate because I think that if the population of humans was to deplete from a disaster, it would also be for the better as the planet we live on would stop being polluted and therefor return to its natural state and because environmentally healthy. I created a pitch which I used to explain my inspiration and idea.

My idea was inspired by the last of us which is a zombie game created by Naughty Dog and I feel that in this game they perfectly captured the vision of what it would be like if this kind of apocalypse was to happen and how it would be dealt with. But most of all the visual aspect stood out to me as that is what my project is focused on more than a horror game with zombies. I also made a few pre-vis images on a story board of what the level should look like. the are just lines and it didn't need to be too complex to get the idea across. Obviously there had to be some kind of interactivity so in my idea I would like to have it so you are able to look around and pick things up as in the story this place is for scavenging. So there would be small things like keys to access different areas to add a bit of interactivity to the level.

Its nothing special on the story board although I will improve it as I want to have a better product for this part of the project. The other slides were some pictures of the bridges and then the last of us to show my inspiration and vision of what I would like to create and then the last slide being the roles that would be required on this project.

One the day of the pitch I explained my idea and sat through the other idea and after every 5 or so each individual would talk to all the lecturers for feedback and when it was my turn for feedback I got told that the name of my idea was creative and that it got their attention. Then we spoke about how I would go about the project and what the possibilities were and how to improve some of the area on the idea. At the end of it I was happy to hear everything to improve which I would make sure to implement but also I am happy with my idea because I feel I would be able to express my creativity on the project. After all of the pitches we had to choose 3 projects that we would like to work on with the 1st option being the higher priority. I chose Cameron's idea as I liked the pitch and he really expressed what he would like to see in detail and by the end of his pitch I was able to see a product in my head.

Start of Production

 After we got back to college I was assigned to the group that I chose which was the steel works and I am happy that this is the team that I got placed on as it is with people I haven't worked with much throughout the entirety of the course so it is a new atmosphere with new people.

First of all we had to decide where to start which was gathering reference images, ideas and creating a pre-vis with some sketches of what we wanted to the product to look like at the end of the year. I decided to go on our Google drive and put together an asset list which we could fill in and keep full control over the production of the product. I made different column for every part of making a finished model so that over the course of the project we will know exactly what needs to be finished and who is going to carry out the task of doing so.

This is a more updated version of the asset list to show how it works and what it contains. We have a priority system so that we make sure to get the more important things finished first such as the level and have a better chance of getting an early product. The mistake the I and everyone else made last year was that we all concentrated on making small assets than actually having a playable game by the end of the production cycle. So this time around we can set things on the asset list to high or low priority and then have something we can work with sooner than later which has been proven successful so far.

My Part

My part to play in this is to produce assets for the level so I have been modelling higher priority assets first and then the lower ones second. My first model to work on was a huge fan that is going to fit into one of the wall tiles of the level so it isn't too consistent on the style and there is more to look at. I modelled this in 3ds max and created a centre piece of the fan from a cylinder and then one fan blade from a cuboid. These were the low poly versions of the model so I the duplicated them into separate object and refined them also adding detail on that would be introduced to the model on a normal map to keep the poly count low. I baked and textured the model in Substance Painter 2 which is a lot easier than doing it on photo shop and having to mess around with the UV map in 3ds max.

I took this model into Unreal Engine and put the textures together in a blueprint material then applied it to the model and it looked photo realistic as it is the style we are going for on this project. I am happy with the outcome of the model and am amazed at how this software makes texture such an easy job compared to previously having to use Photoshop. I will be doing this for the rest of my model and uploading them in a separate update post once they are finished.

Current Feeling

I feel that our team could find a way we can all communicate effectively such as a group chat as the communication is sticking between a few people which will obviously effect the rate that we work at and how efficiently we are doing so. But we are still making really good progress and what we have put together so fare is looking really good. This will also be shown in a separate post. I am feeling a bit lost with some of the lower priority assets and what to do because rather than having a set asset list we are filling it in as we go which is bad because it slows down the production of assets because they aren't there to see what we need. Some people have things that are higher priority and are easily mass produced because on the asset list people have taken over certain sections which isn't a bad thing but it may look like I'm not producing thing which I am but just not at the same rate because making tiles for a level is simple maths and you can make so many tiles. But when I have a load of assets to make that aren't actually on the list its hard to work fast with them as this level wasn't my idea. In the next team scrum I will bring this up to try and deal with it so people who aren't on the higher priority assets have more to do.

[williamtatenextgen-blog]

portfolio model

[williamtatenextgen-blog]

Research and Modelling

[williamtatenextgen-blog]

2 Hour Model and Environment

[williamtatenextgen-blog]

Normal Mapping and Painting: Character Model

[williamtatenextgen-blog]

VFX Animation

[williamtatenextgen-blog]

Normal Mapping & Baking

To normal map a mesh you need to make a low poly version and a high poly version so I made a chair that took a short amount of time and saved it as an fbx named “chair_low” because the software I used to bake and texture the model reads the file names in a certain way. The software I used is called Substance Painter 2 and baking and texturing in this software was a lot more efficient and in Maya there always seem to be bugs with the process not working despite taking the correct steps. So after saving the first fbx I got the same model and chamfered the edges to make it higher poly and then exported it as “chair_high” and this will let me use this fbx on the normal map for the low poly model.

The next step was to go in to Substance Painter and import the first fbx in to the scene and open up the baking menu. In the menu I clicked the little file icon and added the high poly fbx in so that it will add it to the low poly model as a normal map. I changed the output size (resolution) to 1024x1024 because any higher seemed to kill the college computers. Then added an antialiasing of subsampling 4x4 in the small drop down box then clicked bake at the bottom and it started adding the normal map to the chair.

Once the process finished my low poly mesh looked like the high poly version and all I needed to do was add a texture from the massive library of materials that the software has. Substance painter is kind of like photoshop for 3D models and it has layers so I added a wood texture on to the UV map that I literally just flattened in a few seconds in 3Ds Max and it arranged it to add the correct layout of the texture on to the model and there was no distortion what so ever. This makes the job of UV unwrapping a lot easier for the more simple models as it lays everything out and knows where to put things. Usually to UV unwrap it would be a nightmare of selecting, resizing and stitching a load of faces in the UV editor but this software just allowed me to click flatten and export it.

This is what the chair looked like inside the editor but I went further to actually render the chair with an environment which the software also has a library of. This software is amazing because it changes the look of the textures slightly to fit the surroundings in the different environments and you can choose to keep and export it like that. This seems to be a lot more efficient than Maya or 3Ds Max paired with Photoshop so I went on the website and got myself a 1 year student license which will last me through this college year and serve well for my synoptic project.

This is the final outcome of the model with the normal map and texture that I made. I am happy with my work and will continue to use this process to create my environment for my synoptic project.

[williamtatenextgen-blog]

Texture Maps -

Bump Maps -

A bump map is used to create the simulation of displacement such bumps, creases, wrinkles etc on a surface. Unlike displacement mapping , the surface geometry is not modified as it is the surface normal that is modified as if the surface had been displaced. This means that Bump mapping takes a lot less resources for the same level of detail compared to displacement mapping because the geometry is left unchanged. These effects are added using a greyscale and the advantage if this is that it will be a lower memory cost and you can actually paint it in Photoshop. A method of bump mapping would be -

Look up the height in the height map that corresponds to the position on the surface

Calculate the surface normal of the heightmap, typically using the finite difference method.

Combine the surface normal from step two with the true ("geometric") surface normal so that the combined normal points in a new direction.

Calculate the interaction of the new "bumpy" surface with lights in the scene using, for example, the phong reflection model.

Another method would be specifying a normal map which contains a modified normal for each point on the surface directly. Since the normal is specified directly instead of derived from a height map this method usually leads to more predictable results.

 Normal Maps -

A normal map is used for faking the lighting of bumps and dents. This is implemented into bump mapping. It is used to add more detail to something without creating more polygons which is effective for keeping file sizes low. A common use for normal mapping is to greatly enhance the appearance of detail of a low polygon model by generating a normal map from a high polygon model or height map. Normal maps are stored as RGB images as the components of RGB correspond to the x, y, z coordinates of the surface normal.

Displacement Mapping -

A displacement map, Unlike normal and bump maps, physically displaces the geometry of a surface from a mesh to add in detail. For detail to be created on a displacement map the mesh must be subdivided so real geometry is created. The pros of a displacement map is that they can actually be either baked from a high resolution model or painted by hand. Like a bump map, a displacement map consists of grayscale values.

Now, here’s the not so great thing about displacement maps. Creating all that additional geometry in real time is extremely difficult and hard on your system. Because of this, most 3D applications calculate final displacement results at render time. In comparison to bump or normal maps, a displacement map will also add significant time to your renders.

[williamtatenextgen-blog]

First VFX Lesson - Adobe Illustrator

This lesson was just an introduction for getting to know our new lecturer so we went into Adobe Illustrator to complete a task of drawing Kurt Russell. I used only the brush tool to create my image. I added a Madelyn Monroe twist to my piece of art as we had to make them look unique. Using this software was extremely similar to the likes of Photoshop as expected because they are all from the same line of software products from Adobe. It seems to have some features that aren't on Photoshop like a perspective grid (from my knowledge) and some of the mechanics for lining things up and layers were also different but useful.

[williamtatenextgen-blog]

UV Unwrapping in Maya

Today I learned how to UV unwrap in Maya after the previous year learning it in 3Ds Max. It is indescribable how much easier it is to do this process in Maya than 3Ds Max. The shortcuts and tools are a lot easier to access and it is more simple so it sticks with you faster.

I imported the fbx file into Maya and then clicked window and opened up hypershade. I did this so I could quickly create a checker texture to see distortion that will occur while unwrapping the model in the UV editor. I did this by selecting the first default lambert in the hypershade window and then on the menu on the right of the page I selected the colour option and clicked on the checker texture which applied this to the lambert and made the new texture. After this I went into UV and opened the UV editor to start unwrapping the model.

To make the UV wrap no distorted you have to select parts separately and create seems so that in the UV map you can have a clean outline of each limb in this case. For the first arm I selected the top and then used the shift right click command in the UV editor to create a plane of the select part of the model. This lets me edit these specific faces without distorting any of the other places. I then switched to the UV tool and used the same command to smooth out the plane. I dragged unfold to the right of my screen until the plane did not move any more. I did this for both sides of the arm.

I then switched to the edge tool and decided to stitch the from of the arm together so there wouldn't be a seem there. This connected the two parts of the arm together but left some of it overlapping. To fix this I pressed shift and right click to bring up the small options menu and went back to the UV tool. I then smoothed the plane again and it separated and smoothed the plane so it was finished. I did this because I want the seem to be on the back as that part of the model will not be seen as much as the front of the arm. I did this for every other part of the body and made the seems where I saw necessary to fully unwrap the model. I then added the checker texture to the model to see the distortion on the UV map and to fix the small distortions all I had to do was resize some of the planes and rotate them in the UV editor.

Once I did this to the whole body of the character model I had the whole UV map in which I can paint in Photoshop to add the texture to the model.

[williamtatenextgen-blog]

3D/2D Environment Painting

I have to design an environment using my 3D modelling techniques and paint over in Photoshop to create the 2D image. The theme of the environment is going to be along the lines of Norse mythology mixed with future technology so there is no limit to what could be in the environment. I did some research into Norse structures and its mythology to see what kind of things I would like to design. I made a mood board of some Norse villages and landscapes. I put more into this because with future technology you don't know what might be around then so I will be creative rather than looking at other peoples views on it.

I had to do 5 small thumbnail images of environment designs in either Photoshop or on paper.

continue.

[williamtatenextgen-blog]

Art so far -

Silhouette drawing

We were all told to work on some interesting silhouette drawing that are made in a creative way. I had to make a random scribble/character shape using the lasso tool and then get the paint bucked and fill it in black. This then gave me a random shape to work with to create a character from. I drew a few different scribbles until I got some that I was happy to work with. After getting these shapes I started to add on to them one by one using a graphics tablet. I am happy that I have gotten comfortable with using the graphics tablets as it makes digital drawing 100 times easier. the first silhouette I worked on was the one that looked the most human just to try things out and see what I got from it. 

I spent about 2 hours playing around with this and adding on to it like I already did on the legs above. I then decided to change its right arm (on your left) to being in a different pose so I could make something more interesting out of the drawing. I eventually got these results and you will also be able to see some more of the silhouettes I started to change with the graphics tablet.

Custom Brush Environment

For this task we had to draw a random rock shape and add some detail to it. I drew a fairly simple rock and scribbled on it a bit with the paint brush to little detail to it and then I right clicked the image on the layer and selected all of the pixels. I then went to edit along the top of Photoshop and clicked define brush pre-set which made the rock into a brush which I could play around with to create a landscape kind of drawing. I did the same thing to get a brush that put clouds in the image. To get a good cloud brush I set it to black and white and turned the blue colour content all the way to zero as the sky was blue in the image. It was then black and which but to get the cloud to be a more visible brush I inverted the colour which made the cloud black and the sky white rather than the other way round. I then played around with the brush using different opacities and then got my final image of the landscape. It was very simple but looked nice. In my opinion I think that a more simple image is a lot nicer than a more complex drawing.

Storyboarding

For my first storyboard I had to do it really simple and try and tell a story in 4 small images. I managed to get 2 done but didn't get to finish it. The story was a man arriving to a city to find one person that mattered and I started telling this using the small images.

later on in the day we were told to come up with our own story in different groups and I decided to work alone because people had many different ideas and I just decided to move away from any difficulty in deciding and work on my own which I knew I would get finished. my story was a single man taking on a whole army and it's leader in the harsh deserts of Egypt  but the foe underestimated the man and he manages to leave victorious. I had an intro and the work up to the climax. I got to the climax of the story of the drawing and needed to think how I was going to draw the ending. but for the rest of the images I made use of the different camera angles which can tell a story without words or motion and it got the message across of the feelings and importance of things in the small images.

(inset image)