Specialism Briefing

Specialism Briefing
I will now conduct a 3 week self-assigned task within asset creation (my chosen specialism), followed by a couple of art tests. I can choose between Pipeline Development or Substance Designers to base my 3 week development project. I would like to create assets from scratch to finish, so this option will develop those skills further. I will try and complete an asset to a high standard within 1-2 weeks, giving me time afterward in case anything went wrong, or to then model another asset. I shall brainstorm ideas for a couple of hours and then discuss them during my workshop tomorrow, whilst devising a 3 week plan of production.
Initial ideas for an asset to complete within 1-2 weeks:

• Necklace
• Perfume Bottle
• Tap
• Book
• Apple
• Orange
• Treasure chest
• Bomb

I would like to further my texturing skills the most, alongside getting quicker at sculpting/modelling and UV's. If I choose assets such as fruit, I could practice organic textures and getting the realism. 

We have also been asked to research job roles that we would consider applying for after graduating. This will give me more incentive to get a high quality portfolio and ensure I develop the correct skills set for my preferred area.

Job roles of interest:

• Character Artist - Reference screenshot from Naughty Dog (http://www.naughtydog.com/work/character_artist) 

• Props Artist - Reference screenshot from Foundry 42 (https://cloudimperiumgames.com/jobs/243-Props-Artist)

• Environment Texture Artist/High-Res Surface Sculptor - Naughty Dog

• Character Skinning Artist - Naughty Dog

• Environment Artist - Reference screenshot from Blizzard (https://blizzard.taleo.net/careersection/2/jobdetail.ftl?job=160011Y&src=JB-1052)

We are also researching Portfolio's for inspiration and goals. I am looking for Character Artists and 3D Modellers.

• Vick Gaza - https://vickgaza.artstation.com/
  Character Artist at Guerrilla Games
• Frank Tzeng - https://frank_tzeng.artstation.com/
  Lead Character Artist at Naughty Games
• Brian Burrell - https://bee-bee.artstation.com/resume
  Senior 3D Artist specialising in hard surface weapons, vehicles and robotics

Fruit modelling in Maya tutorial
https://www.youtube.com/watch?v=AF0qR4-bJRs

Fruit modelled and available to download free. Good reference for what works and how to texture the fruits
http://www.badking.com.au/site/shop/organic-custom-brushes/3d-fruitby-gordon-goane/

Photogramtory reference
http://gl.ict.usc.edu/Research/SpecularRoughness/

Workshop
This week we covered some basics in Substance Designer. This is my first time ever opening and using the program, and I have created a tile material.
using 2048x2048 grids, I firstly added grey scale colour nodes to the metallic and roughness nodes, and altered the value to see the effects in the 3D viewer.

 We then began more complicated set ups to make our tile-able material. We used a lot of 'blend nodes' so we could combine two nodes to get the desired effect. We made our tile shape and blended it with the tileable texture to begin with. We then made the hue of the grout and tiles different using masks from the tile-able node. Plugging most of the information into a height map and connecting it to the normals gave our materiel depth and lifted the tiles from the grout. We used masks to get certain areas having different roughness values. The latter end of the lesson became quite complex and difficult to keep up with which nodes were connected where so my material was lacking some extra aspects that would make the finish more desirable.
There are multiple function to manipulate the material, so editing later on in the process is still an option which is extremely useful.

 My final graph view

The plan
We had to submit our development plan for the 3 week project. This is the brief I have set myself:
'I hope to produce a minimum of 3 pieces of fruit or vegetables during this 3 week development project.
I aim to learn and develop my skills in Substance Designer, and further improve my ability to UV unwrap efficiently and texture in substance designer. I hope to have a better understanding of organic textures.'

This is my guide for the project, a bare minimum of what to complete and when by.
Week 1 (5th-11th Dec)

• Find references
• Model 3 fruits in Maya, high and low poly
• Begin substance designer material for the Orange or Strawberry

Week 2 (12th-18th Dec)

• Add alterations in Zbrush if needed
• Unwrap low poly model for all fruits
• Begin substance for another fruit, complete the orange or strawberry minimum

Christmas Break

Week 3 (9th-15th Jan)

• Have all materials from designer complete
• Use substance painter for any extra added details
• Render in Marmoset and UE4

Week 1 summary
I have found references for my 3 main fruits, alongside extras. I have played around in Substance Designer using the session file as a template.I couldn't get the depth I need for the indents where the pips lay. I have spoken to Chris to check the possibilities of the Designer and whether I can get my desired outcome. It is possible, so alongside workshops I shall find tutorials to make this material. I believe the Orange material will be easier as the pours aren't very deep, and if I get the correct grey scale image, I can convert it into normals.

I searched Substance Share to see how other have made organic materials and as reference. I have found a banana, watermelon and apple.  I am aiming for a realistic and sharp render, which will require a lot of different layers of textures. These resources are extremely useful as I can analyses how they have created the texture which I find the hardest part. Looking at the banana mesh that came with the texture showed me its non-textured shape. I then looked at mine and realised I needed smoother edges and more fluid bend up along the banana.
During week 2 I plan to completely finish the high poly models of the four fruits I have started. I will purchase fruit so I can precisely analyse the form and surface details. I also aim to complete the low poly meshes during week 2 so I am able to bake them in Substance. If I complete these all I will model some more fruit, but prioritise watching tutorials for substance painter and designer on organic textures.

My collection of references for my 3 chosen fruit to study and model

Week 2
This week I have polished of the high poly models of 4 fruits, and have made low poly models of most of them, in preparation from UV's and texturing. I made the base of all the models in Maya, smoothed them and then imported them to ZBrush to polish further and add finer unsymmetrical details. I plan to make the orange peel texture in substance, though I did experiment with alphas in ZBrush, though I believe the finer details won't bake as I intend so will do it within Substance Designer or Painter.

I did the same with the strawberry, importing the smoothed model to ZBrush to polish, and experimented with indents where the pips would lie.

In Maya shots with a Lambert texture applied.

Although I have chosen my 3 fruits, I wanted more variety in shape and another challenging material. I should have time to complete this also. I followed the same process as the previous 3 and can now bring the high poly models back into Maya and make the low poly meshes.

I imported the banana mesh int my Maya scene so I had a reference, as my first attempt lacked the defined points. Along with this mesh, I could open the Substance Designer file for the banana and look at a breakdown of how they got to the final material.

As the watermelon has a less repetitive texture, I brought it into Substance Painter. Using my reference images I made fill colours and added masks to pain the strips. I am happy with how this model has gone so far, and just needs fine tuning now. I used dirt type brushes to get the effect, but I shall sharpen up some edges to give it a more realistic look.

This week we wet over the fundamentals of Substance Painter using a keypad mesh given to us. I can now navigate more confidently around Painter and ensure I keep good non-destructive workflow habits. I already used masks when painting layers, however now I shall mask the groups, and keep multiple folders.

At the end of this week, I have modelled and began texturing 2 fruit out of the minimum 3 I set. They may be tweaked before the end according to any feedback given, however I am happy with the stage they are at currently.

To conclude this week I had a tutorial with Chris to go over some weak areas and discuss the possibilities with Substance Designer. Together we made an orange skin, in which I can experiment more with to enhance realism. This helped me understand some more nodes in Designer, but gave me encouragement to experiment more and fiddle with functions to get results. I also go feedback on my unwrapping of the banana. I tried multiple unwraps, all suffering from a stretched texel density in the same area. Chris showed me more tools such as the unfold and relax functions within the 'smooth UV's' tool. To sort the texel density, I needed to split my mesh and make 2 main UV island rather than one. Although that is another seam, I should be able to easy paint over this in Substance. This sorted the texel density and I am now ready to make a cage and bake.

I also discovered when doing a test bake, that my low poly mesh didn't match the high enough, so projecting this.

I had used the pre-smoothed model as the low. This would give as accurate mesh as it would using the high poly and reducing the geometry from that. I also did not account for the  finer touches I did in Zbrush, especially on the ends.

I used my ZBrush High Poly orange mesh and worked into that on Substance Painter to compare that to the designer texture from earlier. I got a very desirable outcome and feel I shall use a similar method for my strawberry. I can get more depth and detail using ZBrush for the high poly, and do all finer touches in Painter.

Screenshots from within Substance Painter

Final touches to the watermelon made the colours more vibrant and the surface more shiny.

I plan to render my fruit in a small scene, using reference from still life paintings and photographs for composition and lighting. I have modelled an engraved bowl, and quickly automatics unwrapped and baked just so I could put a basic texture on to give me an idea of what it would look like. This was my render.

Once my third fruit is done, I will have time to invest in properly topologising and baking, plus having the option of modelling more fruit.

Christmas Break (Pre-week 3)

Using screenshots of my models, I got feedback from others. Both my Orange and Watermelon are almost complete, and this feedback once applied shot lead me nearer to a final polished outcome.

It was recommended the orange have more gloss break up and a high resolution if possible. Adding more random bump placement and possibly a finger print or mark will show a little wear or handling. I will also look into subsurface scattering as this could make a nice effect when rendering.

The watermelon needs some gloss break up to add depth and realism to the surface. The high poly model could be a tad smoother.

I have added both my strawberry and banana into substance painter and baked the normal map and AO. The strawberry's leaves are one sided geometry, so cannot be seen from an underneath view. I have been informed there is a 'double sided' tick box within the mesh options in UE4, so I plan to use this. Now I can paint on one side and it should project on both within UE4. There is a small issue in this bake where painting on the leaf, slightly paints the strawberry top. I will bake the leaves separate to the strawberry to avoid this and then combine the material maps once I have finished and apply it to a combine mesh.

I spent a few hours building on the base I had baked and roughly coloured the day before. I started with the ends of the banana, laying down the colours and roughness, then using the height map to get the split in the stalk. Using cement and dirt alphas I added darker patches around the ends and randomly on the skin for the bruising. I used a few variants of colour and applied masks on all fill layers and brushed onto them with different alphas. These alphas took away the mask lightly so gave the most desired effect. I downloaded a couple of alphas from Substance Share to get the speckles on the skin. I then added a blur filter so they blended into the skin better.

Alphas downloaded from Substance Share to get the speckles on the skin:
https://share.allegorithmic.com/libraries/220
https://share.allegorithmic.com/libraries/221

 Quick UE4 renders

To conclude the project I have made the 3 minimum items of fruit I set out for, with some unfinished extras on the side. I have experimented slightly with some lighting and composition, however will continue this after the project in hopes of making a nice still life render.
I believe I can now UV more confidently and use Substance Painter and Designer with more competence and ability to experiment. An area I still need improving is topology in the models, as I struggle getting a lower poly mesh from my high poly. I have used lots of functions within ZBrush and been informed of multiple methods of making a low poly, however I need time to try these. 

Games Jam- First Experience

Games Jam 28th Nov - 2nd Dec
Following our prototype project, we participated in a 3 day game jam. During these 3 days I could apply my skills learnt through my first year and more importantly that I learnt through prototyping and blueprints. I enjoyed applying this knowledge with a short deadline, pushing us to work faster and make sharp decisions. Our brief was a single screen game (fixed camera) with the theme 'boiling point'.

After the brief Monday afternoon, we brainstormed ideas, looking into mechanics and games we liked and ideas revolving around the term 'boiling point'. With this in mind, our game ideas often went for an arcade style or mobile game.
We chose the idea of a robot overheating to expand and make a game from. In hindsight, we should have brainstormed longer and developed more challenging and entertaining mechanics. After talking through the idea, we concluded the game would be 'A old robot having to work and earn money to keep alive and pay for his repairs' . To ensure we got multiple levels and polish mechanics, we kept the scope small and planned the essential mechanics and then listed extras if we had time.
We all took a small task home for the evening, and would get a fresh start in the morning. We planned to get the mechanics in the game during that first full day, and have the main assets such as cabinets that would block out the level. 


I created 2 tables, stainless steel and wooden within the first hours of the first official day. With feedback from the whole group I kept editing until the colours were nice and the brush effects were correct for the material. I was unsure of the perspective as it was strange to me and not a true perspective, so I passed these assets onto a couple of my team, to alter to the appropriate perspective.
I mainly painted assets and gathered reference images, however I look over all other aspects so I could input feedback and keep in loop with all aspects of the game.

Alongside this, our character was in the iteration stage. We all agreed on which designs we preferred and that was iterated on further and conformed as our character. With this character, our team then did separate tasks, one made the flip book for he would animate whilst walking around the level, and someone else began doing the comic strip, which would briefly show the player the story behind the game.

 Final character design

We altered some cabinet assets and made more items that would feature in a kitchen, such as fridge freezers, ovens, and dishwashers.

When in game, we could see if the fit well. Even though we weren't all in the UE4 file at once, we often flicked round to see and give feedback on how the assets sat in the game. An example was the fridge having no depth. We noticed this, so went back and added shading, then replaced it back in the level.

 We made lots of small and quick assets to give character to the kitchen. With these, we could place them in other assets such as the fridge, and on cabinets and floors.

We prioritized sketches so assets could be used within the game even if they weren't coloured. These were coloured the next day and didn't end up being put in the level uncoloured anyway.

We ran out of time to colour the comic strip, so just used the line drawings as they still illustrated the story well enough.

 In game screen shots of some levels. 

We kept the main colours neutral so the hot and cold points stood out, and were apparent to the player they are important. Our assets were purposefully made tileable and reusable so we could make more levels with the time we had.

Final logo for the start screen.

To conclude, after the 3 days we had to make or first game in a group of 5, I have learnt a lot about time management and communication. 

Areas I feel worked and I have improved:

• My 2D skills  - I created assets from reference and alongside others so style and consistency was important. I have also learnt how to create flip books within UE4.
• Knowledge of the Health HUD - in my prototype I struggled with health bars, however since then I feel more competent and understood how our health bar was made and how the hot and cold were blueprinted.
• Teamwork and communication was smooth within my team. We were all motivated and gave valid feedback to create our first game together.

The main points of improvement for future are:

• More time exploring the idea and making sure the mechanics or narrative are interesting and how you can execute them well. We settled quickly and brought in the scope to make sure we completed in time, however our game could've been more interesting.
• Decide on an art style early on and stick to it, having reference images. The art style was mentioned and mood board and references were sent, however the style changed slowly and the mood and atmosphere of the game overall changed.
• Challenge ideas more and push each other. Even though we did not argue and got a completed game at the end, some tasks could have be completed quicker and we could have had more ideas if we would have persisted in areas and challenged each other.