Saturday, November 23, 2019

Search For A Star Environment project

Introduction

I am participating in the Search For a Star competition. I will be solely developing a game ready environment within the next 9 weeks. I'm planning to work every day and document my daily progress in this blog. This will allow people to see my entire day to day development for this project.
The deadline is on 26th of January 2020.

As I'm writing the introduction it is the 23rd of November 2019. Ideally by the end of this week I would have wanted to establish a clear idea of what I'll be making.
The final idea will need to incorporate the following...

Supporting of the project brief

Have good art direction

Have good art reference

Be a convincing game environment

I'm currently making an art bible of all the ideas with a few pros and cons for each. I will hopefully finish it by the end of today and make it as part of my first daily post. The hardest part is finding an environment that incorporates all 4 main points. Tomorrow I will have to decide on one and this Monday (25th November) will be the first development day up till the deadline.


Schedule
Every day, any complete work must be put into the engine. Last two days of each week will be dedicated to any backup or re-doing time.

Week 1 - Mood board, Art bible, detailed rundown & info, concept artwork, photo-bash.

Week 2 - Set up Trello, set up Unreal project, make blockout, first pass textures, set up all placeholder content, make first lighting pass.

Week 3 - High & low-poly 3D modelling in Maya (priority assets).

Week 4 - High & low-poly 3D modelling in Maya (secondary assets).

Week 5 - Uv's, trim sheets.

Week 6 - Texturing in Substance Painter, materials in Substance Designer (priority assets).

Week 7 - Texturing in Substance Painter, materials in Substance Designer (secondary assets).

Week 8 - Finalizing Unreal project, vertex painting, set dressing, decals, final lighting pass, exporting & level builds.

Week 9 - Extra backup time, final screenshots, make presentation, retrospective, make sure all work is ready and submit all final files.

23/11/19

I made the idea art board and thought about what I could do. A few new ideas sprawled up like the bell tower and the piano room. I might also add the Victorian bakery idea to the list as well and shall hopefully decide on one by the end of tomorrow. All seem good in some way so it won't be easy to decide.








24/11/19

I decided to go for the Bell tower environment. In general I feel like it has a very clear and concrete motive behind it. There is a lot to play around with and I have the clearest vision for how I can make this environment work. All the direction has already been set up through the high quality reference imagery I found online. It isn't as complex and overpopulated with props like the sweet shop idea, nor is it too plain and simple like the piano room idea. I'll play to small done well. The environment can be broken down into a few main assets so I have plenty of time to refine them to a good standard over the course of 9 to 8 weeks. If I finish too early then there are many stretch goals and ways I can advance my project further which I will go over in my art bible & brief.



Things to consider from the reference:

The environment is quite small

All of it is hard surface

There isn't any decoration








North Guilford Congregational Church 360 view




Bell hung in action
A video that gave me an insight of how the bell works. I may also animate it if I have time.












Modular Wood-Shack Kit : Insurgency Sandstorm


This modular kit inspired me, I'm planning to make something similar for the structure of the room.


A similar bell someone made, gives an idea of roughly how my final focal asset should look like.

25/11/19

Today is Monday and I've made my first 2 slides for my art bible in Adobe Photoshop. It goes over the narrative and setting I chose to support this environment. 

I chose early 19th century because an old time period goes very well with an environment like this and really sets the tone. It was more religious back then and a church environment can support that. After searching bell room interiors, most of these wooden Revere bell rooms are set in north east USA, particularly the main reference 360 view image is set in the town of Guilford Connecticut.

26/11/19

On Tuesday I made 3 art boards, the first one was to establish the area and the other two go over the environment itself, I think I used the right images to give an accurate depiction of what kind of environment I'm going for.






































27/11/19

Today I made 4 more slides to give a more insight into my project. The slides are pretty explanatory.








02/12/19

Today I finished the concept and made a floor plan to show my environment in more detail. The concept goes over the colour palette and lighting a bit more, as well as the tone and feel of it. The floor plan shows a strategical layout of how this environment will function as a level in this fictional game. I decided to make the room slightly bigger than the original reference just to give more room for movement and to see the bell hung from further away as it is a large asset, much bigger than I thought as shown in this photo.

















The path from start to finish goes all around the room, leading the player around the bell to the corner where we see some props that signify that someone lives here and the diary they left that gives some insight to the story. The player then pulls the lever to activate the bell, now in reality these bells would be manually pulled by the rope at the bottom of the church, but since I'm making this one room I thought of just using a lever to activate it, a classic game mechanic. I really I wanted to try and see if I could make the bell mechanic as I feel it'd be a nice feature to complete my environment, just to give a bit of interactivity to it as if its an actual game.

04/12/19

Due to Uni work I've been behind my schedule this week. I was supposed to finish my documentation last week but I'm still writing up my plan. I have till the end of this week to complete my...
Art Bible
Brief
Schedule
Trello workplace
Blockout with first passes for all files.

I may have to just crack on with the blockout if time becomes a bigger issue.

Today I updated my material list to be more precise with what I'll be making. I made a page on the focal bell asset. The whole lever and activating bell function is secondary priority as it is more of a stretch goal. The artistic features of the environment are more important to me but it would be cool to have a functioning bell.


 I made a simple blockout over the last few days to come up with a good scale for my assets. I tried to find the right balance between it being generally the right scale of the North Guilford church but also slightly bigger to allow more room for the player to walk around. The bell looks very big but that is because it is in real life.

5/12/19

06/12/19


I decided to change my workflow to modelling and texturing each asset at once instead of having dedicated modelling and texturing periods. I feel this way it'd be easier to tick off each asset in Trello and have the stages of more important assets done and over with with less risk of spending too long in the modelling stage for all assets and running out of texturing time.

Primary (must have)
Secondary (should have)
Tertiary (could have)

The primary work will conclude the look of the room as it is in the main reference.

Trello project (work management)

08/12/19

After looking at the main reference I realized the frames of the windows are painted planks and not stone like I originally thought.  


First pass room models in progress.


Unreal project started.

13/12/19

Continued work on priority assets in Maya.


17/12/19

Bell hero asset finished low poly modelling.



22/12/19










































23/12/19

I have uv'd all priority models and added them in Unreal, added first pass Unreal materials and did some simple lighting.


























24/12/19

Today I redid and refined all the uv's for the planks, I had to re-do the mesh for the stairs and cellar entirely due to uv errors.

25/12/19

Today I worked on the wooden planks material in Designer, I worked from the reference however despite making it look like the reference image, in Unreal the material didn't look very good so my next intentions are to make artistic changes to the material that would make it look more aesthetically pleasing in engine. I'm currently trying different methods of making the material to see what works. I may have to find a different reference to work from or work solely from a tutorial. 

 <-- Reference


















26/12/19

Substance Designer Wooden Planks tutorial

I developed my other idea of taking a screenshot of the room reference and using Photoshop to turn the planks into a trim to apply to mesh uv's. This will save time and ensure that the textures are exactly the same as the reference, I did this in some of my other projects and it worked very well. However if theres no high quality image then this method drops out, I tested it out with the planks above and it didn't look good.

28/12/19


















30/12/19

Since Saturday I have developed and changed a lot. I now have two completed materials, the plank material made in Photoshop using the planks from the original reference. The second material is a wood material I made in Designer from scratch following the tutorial I've previously. I redid and fixed the lighting errors although the quality is still made so I'll try to get a few more good passes going. The ambient occlusion post process really adds a style and sharpens the scene. I removed the stair because it looked bad due to sharp edges and flatness. I since made a ladder which looks much better in the scene. I substance painted the first trim sheet but decided to leave the second one with tiling materials on. 














































01/01/20

I textured the focal bell object, its a bit flat at the moment and will make further versions of it, for it at least replaces the metal material. I had some issues with Painter not baking by mesh name so I spent a lot of time trying to resolve the issue with no avail, I made a simple bake by exploding the mesh but further work is ideally needed. Also I noticed the lighting completely changes whenever I rebuild the lighting.
























13/01/20


























Over the last 2 weeks I've been modelling props. I made the sack and the bed with Maya's Ncloth tool. The props are uv'd (except for the cloths). Tomorrow I will also make the clothes line and retopo the cloth meshes into low poly's. The cloth meshes in the screenshots are the raw high poly models at the moment. Ideally the uv's would all be combined onto one trim sheet, if not 2 in total by tomorrow. For personal reasons I've fallen really behind on this project but could still make a come back if I work more on it to make up the time.

As of 13th Monday, I still need to do the following...

Make cloth line model
Retopolagize cloth meshes and uv them
Texture all props in Painter
Import props to scene
Do lantern lighting and redo overall lighting
Make dirt decals for floor and walls. (possibly vertex paint)


(stretch goals)

Animate bell in Maya
make bell mechanic
add sound
make improved wood material passes
make improved bell texture pass

18/01/20

About a week until the deadline, I have now textured and imported all props to the project. I've set dressed the scene. I spent the entire day trying to fix lighting and there were a lot of light errors. I found that the bed was missing its lightmap uv so I had to re-import it. The clotheline had lightmap issues as well. Those were fixed but there are still black glitchy patches all across the room and I'm not sure whats causing them.

20/01/20

Today I gave it my best at making the bell mechanic. I successfully animated the bell and set the animations up in Unreal. I've managed to make the lever switch intractable by setting a keyboard binding blueprint to activate when near it. I'm currently trying to figure out how to have it activate the bell blueprint animation as well and also show a text message telling the player to press E to activate switch.



Tutorial I watched to make lever mechanic

























24/01/20

I completed the bell and lever mechanic, the lever now shows a text telling the player that it can be activated, if the player is standing near it. Here's the full blueprint for the lever. 



I was advised to start the project from scratch and re-import all assets to solve lighting problems. I was also advised to populate the other part of the room which was empty and didn't have anything there.

I rebuilt level with new lighting, and added some decals to populate the scene.

26/01/20

Today is the deadline and I have pretty much got everything to the standard I wanted it to. I didn't have time to vertex paint and add sound so its something I could expand on in my own time some other time. To replace adding sound to the project I instead made a cinematic trailer using the project sequencer and I added sound to that instead. It was alright but came out a little quite. 

Overall this is the last update for this project, the files are all packed and uploaded and I feel like I done everything I could have to make the best for this project. I originally planned to make daily blog updates but doing it weekly is still the best I've done, usually I'd start the blog at the start of the project and finish it at the end so this time I stayed somewhat consistent with the blog.









Monday, September 23, 2019

Year 3 Specialism - Environment Art Modular Kit

Traditional British Village - A House Kit



Introduction
Uni has started and we were assigned to choose a preferrable specialism and create a project based on it. I chose environment art because I liked the idea of making a modular kit and I'm also heading into environment art in general.

I was inspired to make a British countryside village. I chose this because there is plenty of influence within convinient range. Other than that I think there is a lot of potential for it to make a good looking level.

Planning (brief & art bible)
I had a perfect image in mind of what I wanted to make but the mind is often wrong so I collected lots of images of UK village houses & street ways. There were certain attributes I was looking for. Mainly to suite the style of traditional cottage-esc housing. Explored certain decorative features that convey that you are in Britain. Early 20th century to avoid modern complex details like cables and macdonalds.

Research
Along with the images I also took some photos myself, I was surprised there were good examples nearby that convey what I'll be making.



I looked at a few videos as well.

Everyone's Gone To Rapture Behind the Scenes shown me their intentions and the atmosphere they went for. I also got to see some close-ups of the houses.

Everyone's Gone to Rapture video

Material tutorials, I used the brick tutorial at some point and it really helped. I will learn from it again. Some others may help as well.

Designer brick tutorial

Cobblestone wall tutorial

Designer playlist


16th October 
This is the first time I came back to the blog since the first week of the project, reason being is that I was too busy with the work at hand. Ideally writing the blog while travelling is the best solution as I'm doing now.

The Process
The second week was all modelling. My first intention was to create very warped house parts to avoid any major sharpness to the final piece. However that was soon postponed as I was told thats not very effective as it goes against the point of a modular kit. It would have also been difficult to allign each asset and would have made up for a lot of excess time.
So I went for straight simple shapes.

The modelling took about a week, I finished off with smaller props. One of which was the plant which I made using an improvized method. I had small planes randomized using Mash function to create leaves. This was quick but needs tweaking as many planes are visibly hovering.

The next week was uv's. At this stage I had been consistently behind schedule by a few days. I organized all my assets into trimsheets. At first I had all my uv shells as tight as possible with some tileable shells out of the uv space. I then re-did all my uv's as having shells out of 0 to 1 space would need more tidying than necessary so I re-arranged my shells. There was more empty space in some trimsheets now.

For the materials I started off by making some simple stucco paint walls in 3 different variations; brick, rough (grainy) and generic flat wall. For the most part they were just normal maps and the other outputs like roughness and base colours were added later in the Unreal material editor. I did this because its faster and easier but you could argue I could have made up a better quality material with all the Designer outputs in place.

I spent the most time doing the old brick material as it was hard to recreate from the reference image because it had a lot of complexity to it.

The textures were made in Substance Painter, I prepared high poly's for baking to specific assets and made to different Painter projects for highpoly baking and non-highpoly.






In the last two weeks I created the Unreal scene which would be my street level. At first I was mainly testing out how well all assets fit together and I did have to remake some to fix some errors like gaps and wrong material sets. I tested out materials.

I then really sat down and actually attempted to create something worthy with the kit. Once I vertex painted and lit the scene I really started to feel more confident with my work.

The problem with straight edges was relatively dealt with through vertex painting and placing decorative assets around them. For example the edges of the house roofs now have planks, gutters and ceramic plates.
With the flat triange at the end of the bay window roof I vertex painted some moss to blend the hard transition between the roof and the wall materials.

Overall I am happy with the fact that it looks like a British village and complies with my original intentions.

Some more detail like decals and singular props could really improve the scene.

The level scenery could be further advanced with trees.





Today was the deadline and I've added most of what I wanted. Overall I'm glad that the kit can make something descent for a videogame.
Small decorative assets are pretty much mandatory to cover up the sharp edges, they really complete the level.

If I had an extra week then I would re-texture more materials and trimsheets. Add emissive lights to the lamp bulbs. Redo the decals and populate the scene more.

The bay window may have too much edgeware to the point that it looks stylized from far away and unappealing from close up.

I could have organized my textures more and put more effort into my materials, some materials may feel incomplete.

There are certain things I didn't add like stone fence asset, cobblestone material, pub sign decal and window parallax.
I also thought of adding window emissive decals and adding physical ivy.

Another approach I could have took is make fewer assets but with more going on, so instead of a plane wall there being extra inbuilt features.
Theres many singular assets but you can make a lot of varied houses in one area with them.















Search For A Star Environment project

Introduction I am participating in the Search For a Star competition. I will be solely developing a game ready environment within the next...