Final Year Project - Week Three: (13th February - 19th February)
- Ryan Harlee Jones
- Feb 20, 2017
- 6 min read
Final Year Project - Week Three: (13th February - 19th February)
Monday:
Monday as a day was very much classed as a work management day, I started the week with my morning session where I updated my blog as i usually do. The blog entry was uploaded and accomplished on time with plenty of daylight left to use.
I decided that my LinkedIn profile was in dire need of some updating, so I spent some of the day going through the website and their updated page formats to update my profile accordingly. I added new relevant experience, with correct links to my new blog / website, along with some other things.
Fig 01. Sketchfab Account Homepage (Jones, 2017)
I then continued to tweak some pieces on my Wix website, I uploaded another set of models for display on Sketchfab, these models were for the AGD and Journeyman pages, so that I could display them on my website via embeding.
I also at the end of the day applied for Expotees, making sure to correctly apply all relevant information to my profile so that I would at least be prepared for the application deadline.
Tuesday:
On the Tuesday I created the 3D asset of the surrounding wall for the average sized doorway I had planned for my scene. I decided to create this asset without attaching all of the pieces to the mesh via welding. Instead I had the extruded pieces hovering and intersecting the base mesh, this way I could get the assets accomplished quicker, allowing me to see how they work in the scene, and if they work well together as a whole.
Fig 02. Door Surrounding Wall (Jones, 2017)
After the doorway was modelled, I decided to test a high to low poly bake yet again, to see if the result that i got beforehand with the pillar was going to happen again. Once I had gone about correcting the cage around the asset and then rendered the normal map, the normal map worked much more efficiently compared to the attempt before hand. I found that this was because of the fact that I had added chamfers to the edges of most of the asset, I had previously been warned that when baking normal maps that it doesn't pick up too well on complete right angles, so I decided to put that to the test and found that the result is much better.
I decided to have a look at the separate pieces that created up the normal map, by looking at the various channels that I could edit. I hid all layers except for one and then added extra paintwork to it to then see how it affected the model, the results were originally hard to understand until thinking about how they would all work together in unison. I came to understand that each of the channels R, G and B all had a specific role in the creation of depth illusion, and I now on a base level at least, understand how a person would go about editing a normal map manually. After doing these tests I decided that I had enough time in the day to create another asset, I created the 3D asset for Artefact that would hover in the centre of the room, ready to be put into my dump scene for testing.
Fig 03. Platonic Artefact (Jones, 2017)
I found that the scene was already looking quite busy, between the pillars, floor, artefact and wall piece, although at the same time I decided that the issue was to do with the scale of some of the assets, specifically the artefact and pillar which were making the scene look quite cramped.
Wednesday:
On the Wednesday I created yet another 3D asset, this piece was the doorway's side wall, which was an addition of length for the wall piece. This piece took a little while to design, although in the end kept with the overall style.
Fig 04. Side Wall Panel (Jones, 2017)
After adding the asset into the asset dump scene, I found that the overall scale of most assets were consistent. Although, the wall pieces didn't quite manage to meet each other at the corners, and the pillars and artefact were taking up to much space in the scene. These were minor things that needed to be changed, although if not sorted out would still affect the overall scene.
Thursday: The Thursday was a day filled with learning new things, I went into the labs and started getting on with more of the 2D designs in anticipation to create more assets.
Fig 05. Portal Surrounding Wall (Jones, 2017)
While midway through creating the door frame asset after creating the 2D image, I started discussing projects with some of the surrounding students. One student talked to me about how I could create some of my other assets with a much more efficient poly count, compared to my pillar asset which came out at around 22'000 tris.
His technique consisted of taking all of the extruded sections which were detached from the base mesh, and then creating the panel mesh which they would originally intersect with, between the extruded mesh, this way I could very easily take control of the wire-frame without getting confused, as I connected the bits together I could optimise the frame as I went. Although, to me personally the way I was being showed looked slightly messier than what I was used too, this method did look to be much more efficient on the tri count.
The rest of the day I ended up creating more 2D images for the upcoming assets, the Portal door wall asset had a 2D image created for it, along with the start of the Portal doorway image. After more discussion with some other lab members, I ended up going home to look at some online magazines which were free to view called Vertex. I used this time to research and scroll through what I could that seemed relevant to me, this way I would have more reference and a better idea of what I was heading towards technically.
Friday:
On the Friday I went ahead and started playing around with the normal maps again, I decided to bake a normal map test again for one of the side wall panels, along with applying a very basic diffuse texture, and attempting to learn about how to apply displacement maps other than using one of the modifiers in 3DS Max. Once I had seen the result, it didn't look awful, but I found that a lot of the charm and detail of the environment I was creating was taken away from it. This allowed me to decide that I needed to model in certain sections of detail, rather than attempting to normal map them in.
Fig 06 , 07, 08. 2D Asset Designs (Jones, 2017)
I created three more 2D images, by finishing off the Portal door wall image, candle image and earth chunk image. I then decided to create a debris piece that will float around in the void beneath the environment, which was just a simple cube with some added extrusion detail.
Fig 09. Debris Cube Asset (Jones, 2017)
I spent the rest of the day doing more research into the Vertex magazine, I found an interesting section about 2.5D texturing, which I debated looking into. This technique is where the user locks the camera in a certain orientation to view a world, similar to the camera that Diablo 3 and League of Legends use. Then via the camera angle, the artist textures with that in mind, to their advantage. This is done by painting in the depth, since the camera will not turn around to reveal the illusion that is being given by the way it's painted. For when floor hits the walls, so that the illusion isn't broken, on the flat floor that the plane is with the texture, they just add a little geometry to help the illusion stay in place. The example used was floor boards, so the artist just pulled up one board so that it was not flush with the wall, helping to not break the overall illusion.
Saturday / Sunday:
The weekend was more of a rebooting period of time for me, I spent most of the two days doing what I want with only some more research placed in between, although this research was only done through reading more magazines such as 3D World and the Vertex magazine. Over the week I was feeling very disheartened and drained with my lack of progress, and knowledge on the pipeline I was undertaking. I healed the majority of my negative mental state by spending the two days with friends and doing what I wanted to do to eliminate any bad emotions so that I could start the new week fresh.
Overview:
This week in overview was fairly poor, I feel like a lot of the work I accomplished was just to waste time, or done because it's some of the easier tasks to undertake. Although this mindset may be from feeling disheartened from progress and over stressing myself about where I currently stand. I plan to speak to one of the tutors on Monday that specialise in environment creation so that I can get some feedback and maybe get some help with what directions I need to be taking to better my progress. There is only so long I can be at the stage of uncertainty before I have to get on with creating the large chunks of work.
- Ryan Harlee Jones
Comments