After some feedback, I returned to my inverted room and tried to soften the edges of the room. I tried doing this several different ways, but this was the best result that I had. I struggled to get the UV's right, since if they aren't all one piece the seams are very noticeable even if I make sure the shells line up right. So, I decided to play around with the UV's a bit since I actually thought some of the distortion looked interesting. Here is the result, I also turned on some reflections on the texture and I think that improves the appearance too. I actually quite like how the distorted tiles/bricks look, although the UV's on the wall behind the camera are very wrong and messed up. I may experiment more with this and see what I can do both in terms of rounding off the edges of the room and playing with the UV's.
For these displacement tests, I applied a skin preset onto the sphere and set the scale to 0.075 then plugged in the 'cloth' 2D texture into the displacement material channel. Like the previous displacement experiment, I played around with different channels including subdivision, displacement attributes, and the displacement material settings. I apologise if the quality on some of the images aren't the best quality, I didn't want the render time to be too high since I was mostly focused on the displacement. I enjoyed playing with this test more than the last one, I really like how odd 018, 019, and 020 look in these tests. It's very interesting to see how a few settings can drastically change what the displacement looks like.
Today I wanted to experiment more with the displacement material channel in Maya (in this case Arnold) because I found some of my previous tests using it produced interesting results. I think this overall can be very helpful with my project because I can create very unique shapes just by messing around with some settings and I can also animate the different channels if I want to. After some discussions today, I am planning on brainstorming some ideas of more 'abstract' ways to achieve what I want. Rather than always going for the literal representation of what I'm talking about in my poem I'm going to try to focus on the feelings I want the viewer to feel instead. I feel that displacement can be helpful in this regard to create unique, abstract digital sculptures for my film.
In this test I applied a 'brushed metal' preset onto the sphere and plugged in the 'bulge' 2D texture into the displacement material channel. I then played around with the subdivision/displacement attribute settings, the settings on the 'bulge' texture, and its UV coordinates. Here are some of the results for that particular test. I find them all interesting but I'm most drawn to 006 and 011.
As I move on to some other parts of my poem, I wanted to experiment with Bifrost which is used to simulate liquid. This is for the phrase when I see hospitals I smell bleach. I'm still unsure how I'm going to create bleach and if I want it a liquid simulation or not, but I felt it'd be good to at least play around with it. One problem I found using the Bifrost system is that it does not appear to have a method of moving the liquid like how nparticles do with fields or a locator.
First I tried creating a sphere and set it to be an accelerator to try to get the liquid to follow the sphere. While this sort of worked, it was very difficult to control the results and I suspect this may not be the correct use of the accelerator setting.
I then tried to create a sphere and connected it to the liquid as a collider so it would act as a shell/container for the liquid. In this case, I animated the sphere hoping it would affect the liquid inside. However, I found this method would probably look similar if I just created a sphere, added the correct shader to it, and deformed it. It had no real added details in the animation, although this could be just due to the settings I had.
Lastly I created a liquid then a significantly larger sphere surrounding it. I then made this sphere a collider and changed some settings so no liquid would leak through. I found this to be the most successful so far, but I found actually getting a detailed animation out of it caused my computer to slow down drastically and would make the mesh (once I turned off the particles and activated the bifrost mesh) very dense. Even with the mesh being dense, I didn't feel it had the details I wanted but again I suspect that is due to the settings that I had. In this test I also experimented with exporting the mesh out using an alembic cache to see if that would improve performance and give me greater control over the appearance of the liquid in regards to the material/shader.
I'm not sure how I like these experiments, but I did not expect Bifrost to be easy. It may be worth experimenting some more, but I'm unsure if it is even possible to achieve what I want using this method. I may experiment more with deformers, lighting and displacement map animation to see if I can fake the appearance of liquid bleach, but I also know that it wouldn't have as much detail as a proper liquid simulation. Below are some examples of the sort of liquid simulations I'd like to try to achieve, but I do not know how to achieve this in Maya.
I decided to return to consider what I want my mind-space/non-space to be. While I like the dome/sphere, I did want to try to create a sort of 'inverted room' based on a significant location related to this project. It took me a while to figure out how I wanted to make it, as I was trying to take influence from Rachel Whiteread, an artist who creates sculptures of create inverted architecture.
I decided it'd be best if I didn't overthink it for now at least and this was the result. I roughly modelled the room I remember and quickly UV'd it (I know it's not perfect). While it isn't perfect and I think it may be good to return and add some extra details, I think this is very interesting and I think I prefer it over a plain sphere. I rendered a quick test using Mental Ray and I hope to also take this model into Arnold so I can play with the atmosphere settings and casting some shadows (ex. like in this test) on the walls.
