Sunday, November 28, 2021

Week 10

Finished the building from last week. This building isn't super accurate unfortunately as the entire backside of it is invisible from the street. Also i think the roof texture currently looks pretty terrible.

All the buildings so far are just screen recordings of the openGl viewport. I made a mistake and upgraded my computer which broke some programs and as a result i can't render anything until updates are released. It's already been a month without any news so I will try to downgrade my computer before next week so that I can actually render something.

Decided to abandon the idea of the puzzle pieces as i think it doesn't really add anything to the building. I think the most important thing here is to just make it visually appealing. I also really need to speed up. So far I have four buildings which is not nearly enough for a series.

Sunday, November 21, 2021

Week 09

I think I made a mistake starting this last one as it is a little bit too big, and it turns out that i don't really have enough images to recreate it properly. Also tried to make the one from last week a little bit more complete.

Need to remove all the electrical wires in front of the building on both of them.

Sunday, November 14, 2021

Week 08

Another building, this time from Seongsu. Not completely finished unfortunately. Also the puzzle piece doesn't work as well with this building as it doesn't really fit the piece that well. I wanted to keep the pieces in a way that they actually could be pieced together but as they will be put up as separate objects i think i should focus on making each building look nice instead.

Sunday, November 7, 2021

Week 07

As previous experiments where not really suited as NFTs I've decided to explore this project again. I think that I should either try to just do as many buildings as possible to create a series and mint them at a low price. Sell them as small pieces / slices. With this way i thought it could for example be similar to the gifs below or it could be done in a more visually nice way, e.g. renders with proper lighting, reflections etc. Or i could use some time to explore ways to deform them / make them more interesting as objects on their own.

As the previous building i made is very mundane / anonymous I decided to make another building which is more unique to Korea. Thought it could be fun to do a PS1 style filter on them as they do remind me a little bit of game assets.

I could try to explore 3D models with inverted faces as well. When the faces are inverted they're invisible from one side, e.g. you could look inside 3D models from one side, and see the outside from the other / works almost like a beam splitter.

Sunday, October 31, 2021

Week 06

More experiments with volumetric video, this time with people instead of cars. I think the results are not that good, just chaos mostly. Maybe there is potential there for a nice one. I would just have to take a ton more videos and experiment to see what works.

I think the problems with these is that there's just too much going on. It would be better with less people and more uniform / controlled movement. The location i took the videos from could be better as well. When people get too close their feet are cut off, which breaks the immersion. Would be better if I could be further away so that it was possible for people to go off to the side, similar to how the cars turn.

Also, if i took more time isolating the people properly i think the results could look better as well. These are all very rough.

I think this project is easier to monetize than previous weeks experiment. With this i could make a series of different videos. Would also be nice if i could port it to web so that it could be interactive.

Sunday, October 24, 2021

Week 05

Made with three.js. The gas price is retrieved via the Etherscan API and displayed on the left in GWEI. This is then used to drive the displacement on the sphere. Everytime an action is performed on the Ethereum blockchain a gas fee has to be paid.

However, using an API is a little bit unpredictable. If any changes are made to how it works it could potentially break it entirely. As i can't make any changes to it after its minted it could prove a problem if the API stops working or if the value of Ethereum increases a lot and breaks the displacement

Another problem is that the gas price doesn't fluctuate that much it hovers around the same numbers mostly with some dips outside of this. As a result the displacement strength doesn't change that much. I could artifically create some dips to make it more random.

Going forward i could try to explore different models. E.g. maybe i could have Greta Thunberg's head (part of popular culture and also a well known environmental activist. Don't think she has ever spoken on NFTs unfortunately) instead with changing emotions. Alternatively do the same displacement on the head. Also adding a back up for the interaction would be nice as well, not just the gas price.