Dan Miller
@dan0mighte.bsky.social
game dev. Software engineer. Systems obsessed.
he/him
check out my work and blog: http://linktr.ee/fraculation
he/him
check out my work and blog: http://linktr.ee/fraculation
I found an issue and would contribute a fix, is the website open source?
October 28, 2025 at 9:01 PM
I found an issue and would contribute a fix, is the website open source?
Back on my game graph bullshit again. This shows the graph taking shape gradually via a random walk process.
October 28, 2025 at 12:54 AM
Back on my game graph bullshit again. This shows the graph taking shape gradually via a random walk process.
It is cool to see visually how games in the same family can be played in mostly the same way but only up to a point! Here I select 4 nodes (shown in Teal) and play those 4 games win. They are forced to merge into one game before they get to the win state
October 16, 2025 at 3:53 AM
It is cool to see visually how games in the same family can be played in mostly the same way but only up to a point! Here I select 4 nodes (shown in Teal) and play those 4 games win. They are forced to merge into one game before they get to the win state
@limeblossom.bsky.social gave me the idea to treat each box as unique objects rather than as interchangeable with each other. This creates more states so its a bit harder to analyze. But it also creates a ton of symmetries which I think make it quite pretty!
October 9, 2025 at 2:47 AM
@limeblossom.bsky.social gave me the idea to treat each box as unique objects rather than as interchangeable with each other. This creates more states so its a bit harder to analyze. But it also creates a ton of symmetries which I think make it quite pretty!
Got a cool video capture of the graph solver running now that its optimized into the range of realtime
October 8, 2025 at 11:17 PM
Got a cool video capture of the graph solver running now that its optimized into the range of realtime
conveniently, due to the nature of the simulation bi-directional edges attract stronger than single-direction. This chain is composed of small groups which you can navigate between easily, but it appears one can only progress in one direction between the groups. (without hitting a dead end)
October 6, 2025 at 6:43 PM
conveniently, due to the nature of the simulation bi-directional edges attract stronger than single-direction. This chain is composed of small groups which you can navigate between easily, but it appears one can only progress in one direction between the groups. (without hitting a dead end)
I made it 3D with a custom solver, and its starting to look promising. I'm curious about whats going on in this little chain separate from the rest of the mass. There must be something interesting here!
October 6, 2025 at 6:41 PM
I made it 3D with a custom solver, and its starting to look promising. I'm curious about whats going on in this little chain separate from the rest of the mass. There must be something interesting here!
This level is only a little bigger but its state graph is enormously larger (9000 nodes vs 30). The player has many more unique ways to solve because there are fewer dead-zones for the boxes
October 5, 2025 at 6:04 PM
This level is only a little bigger but its state graph is enormously larger (9000 nodes vs 30). The player has many more unique ways to solve because there are fewer dead-zones for the boxes
Playing around with analyzing how many ways there are to solve a simple Sokoban. Orange = no boxes on destination, Blue = all boxes on targets.
October 5, 2025 at 5:58 PM
Playing around with analyzing how many ways there are to solve a simple Sokoban. Orange = no boxes on destination, Blue = all boxes on targets.
It can't answer this but it can write code which can. crazy world. IDK what it needed panda's data frames for though LOL
August 8, 2025 at 2:43 AM
It can't answer this but it can write code which can. crazy world. IDK what it needed panda's data frames for though LOL
Garden update! The hollyhocks really took off this year. Letting garlic go to seed again. Peppers and basil doing great, tomatoes a little slow but getting there
July 12, 2025 at 9:07 PM
Garden update! The hollyhocks really took off this year. Letting garlic go to seed again. Peppers and basil doing great, tomatoes a little slow but getting there
Garden! All the wild grapes from my batch of wine 2 years ago are really starting to take off. Probably going to have to cull them hard soon but for now I love em all, they literally tie the garden together #garden
June 26, 2025 at 5:31 PM
Garden! All the wild grapes from my batch of wine 2 years ago are really starting to take off. Probably going to have to cull them hard soon but for now I love em all, they literally tie the garden together #garden
Played around with debugging visualizations
June 24, 2025 at 3:42 AM
Played around with debugging visualizations
Basil transplants: in! Can't wait for the pesto and other basiling this summer
May 25, 2025 at 11:15 PM
Basil transplants: in! Can't wait for the pesto and other basiling this summer
Adding some stats to save slots really make Doggos in Dungeon feel like a Real Game! #screenshotSunday
May 4, 2025 at 5:55 PM
Adding some stats to save slots really make Doggos in Dungeon feel like a Real Game! #screenshotSunday
Have you ever wanted self-invoking functions in C#? No? Well we got em anyways:
May 3, 2025 at 6:25 PM
Have you ever wanted self-invoking functions in C#? No? Well we got em anyways:
I gave em dice to play with! Who do you think won this one?
April 11, 2025 at 12:29 AM
I gave em dice to play with! Who do you think won this one?
I love graphs. This is a graph showing the win rate of every generation of gladiators when they fight against gladiators from 10, 100, and 1000 generations ago. I wonder if there's a standard name for this, seems like something that generalizes to any genetic algorithm!
April 9, 2025 at 8:19 PM
I love graphs. This is a graph showing the win rate of every generation of gladiators when they fight against gladiators from 10, 100, and 1000 generations ago. I wonder if there's a standard name for this, seems like something that generalizes to any genetic algorithm!
I added a "Attack the enemy with the lowest armor" gambit. They started using it! The team on the left looks like they have evolved to take down healers first based on their armor. And with a little luck, they do quite well in this fight
April 8, 2025 at 8:07 PM
I added a "Attack the enemy with the lowest armor" gambit. They started using it! The team on the left looks like they have evolved to take down healers first based on their armor. And with a little luck, they do quite well in this fight
60s of extremely exciting autobattler content, enjoy. These guys are getting some smarts! Everything about them, including their brains, is determined by genes, evolved through 7.5 million simulated battles in this case. Some of them have learned to focus on only one enemy at a time!
April 8, 2025 at 4:37 PM
60s of extremely exciting autobattler content, enjoy. These guys are getting some smarts! Everything about them, including their brains, is determined by genes, evolved through 7.5 million simulated battles in this case. Some of them have learned to focus on only one enemy at a time!
After nearly 3000 generations, these little guys have figured out a strategy that lets them focus down one of their opponents at a time
April 2, 2025 at 10:50 PM
After nearly 3000 generations, these little guys have figured out a strategy that lets them focus down one of their opponents at a time
Had an awesome Severance finale party: Coveted egg bar, and post it note theory voting (spoiler free images). We had 4 confirmed predictions to 18 unconfirmed predictions 🥲
March 22, 2025 at 4:35 AM
Had an awesome Severance finale party: Coveted egg bar, and post it note theory voting (spoiler free images). We had 4 confirmed predictions to 18 unconfirmed predictions 🥲
Tried making a terminal shader in windows terminal today. Looking pretty good so far, just about ready to copy this over to my work PC
March 16, 2025 at 1:55 AM
Tried making a terminal shader in windows terminal today. Looking pretty good so far, just about ready to copy this over to my work PC
Got #AdventOfCode Day 1 Part 2 working. Starting to get the hang of using Effects as a form of dependency injection, this is feeling comfy. The syntax is really flexible as I get the hang of it #functionalprogramming #kokalong github.com/dsmiller95/k...
February 26, 2025 at 2:57 AM
Got #AdventOfCode Day 1 Part 2 working. Starting to get the hang of using Effects as a form of dependency injection, this is feeling comfy. The syntax is really flexible as I get the hang of it #functionalprogramming #kokalong github.com/dsmiller95/k...