Two Late 2021 CGI Artworks

Reunion (3072 × 4096)

Medium: Blender Cycles + Hand painting. December 2021.

During my freshman year, I imagined a space colony utilizing the material of Enceladus (Saturn II). This new rendition moves the space station off the surface of the moon and bends metal and concrete into a van Braun ring space station design. It is also reminiscent of structures described in Interstellar (though that station extends axially) and The Three Body Problem (briefly mentioned in its final moments). Then I put in a hugging couple as that was what my subconscious drove me to draw in the first place.

The foreground was not intended to be barren in the earliest sketches, populated with an artificial forest. That said, I ultimately did not implement that idea to stay true to the sterile space feeling in contrast to the people in the foreground (and save GPU time). The train tracks and riverbed were illuminated with colour-varying lights in the very last minute, in order to add some futuristic vibes to the scene.

Due to the way gravity is produced and maintained, where the people stand is quite steep of a slope already. I wondered whether I should have modelled a staircase (or a fence just below even higher walls). I imagine residents there wear magnetic boots and are fairly used to the lack of a well-defined “up”.

Liberal use of elementary Blender geometry nodes is evident as the buildings in the elevated living quarters are literally generated randomly, and could have had more thoughts and care.

Below, you can see the transition from the first mesh grid of a torus to a shaded prototype. The colour tint and illumination angle were changed during later compositions as well.

An End of Humanity (2000 × 3000)

Medium: Apple Keynote + Blender Cycles. December 2021.

Somehow presented with a similar composition to the previous piece while telling a completely different story, An End of Humanity illustrates my known-to-be irrational fear that virtual realities like Metaverse will absorb a yet spaceflight-incapable civilization deeper into the crib of gratifying virtual experiences.

For the colors and lighting in the scene, I know I was strongly influenced by the dilapidated Flynn Arcade in Tron: Legacy (2010). An unattended computer collecting dust over decades, while also gating the portal into another perceived universe. Same idea really.

Looking Back at Daft Punk's "TRON: Legacy" Soundtrack -

Why the CRT display, though?

I enjoyed modeling it without referencing any particular real model, for one; I thought it’s slightly sturdier than consumer-grade panel displays in 2022, for two.

And, lastly, there also exists a physics-inspired reason — for as much computing has miniaturized and advanced, our ability to faithfully reproduce nature is still devastatingly stagnant. You don’t need to stir a cup of water, even, to melt a server-grade CPU. Simulated atom by atom, the behavior over 1ms of one Na-K pump in one of your neurons in your brain, would take hours to simulate1.

Anyway, I believe that Meta is largely a sham for capital to lure in more money. It’s about time. Still, I am alarmed by the escapist sentiment developed by some of my contemporaries. As interesting as Hollywood or your game vendors make VR seem to be, the only world we know where you can resolve turbulence, fold proteins, smash trillions of atoms, or merge black holes, is this one.

One of Cixin Liu’s pieces (Fields of Gold) inspired this view of mine. The following may be a scenario of how humanity ends. Humans never left the planet. Generation after generation, more people uploaded their consciousness into machines. Then, a few centuries later, nature reclaimed everything ever constructed. Eventually, the only footprint the species ever would leave behind, is a dusty basement underneath a collapsed building, and a computer terminal laid to rest therein. It would flash the text “Billions of people live here. Do not power off.”

Then a rat tripped over the power cord.

Prints may be available in early 2022.

This website estimates the speeds at which humanity can do to simulate individual atoms using the L-J potential. And multiplying that speed by the molecular mass of a protein gives you a (probably too optimistic) estimate of the rate at which we can simulate it.

Bonus: Full-size Workbench Renders of the Pieces