yukirij
@yukirij.kumorizora.net
Tinkering, worldbuilding, systems, and engineering.
Connoisseur of natural/military sciences, speculative fiction, and bio/ero/natural horror.
(occasional mature topics)
クリエイターやシステムの事
evebin.com · kirisame.com · dzura.com
Connoisseur of natural/military sciences, speculative fiction, and bio/ero/natural horror.
(occasional mature topics)
クリエイターやシステムの事
evebin.com · kirisame.com · dzura.com
Spent some time over the last few days re-implementing the planetary terrain system (still working on regional LoD).
I integrated the algorithm from my hex terrain generator and, while not perfect and missing biomes, the output has been reasonably consistent in quality.
#gamedev #procgen
I integrated the algorithm from my hex terrain generator and, while not perfect and missing biomes, the output has been reasonably consistent in quality.
#gamedev #procgen
September 24, 2025 at 12:40 AM
Considering adding a rectangular board view for those that do not like looking at hexagons, though I will need to substantially overhaul the visual design (this version only changes tile shapes). In particular, I find diagonal movements are harder to visualize than with hexagons.
#gamedev
#gamedev
June 20, 2025 at 10:50 PM
Considering adding a rectangular board view for those that do not like looking at hexagons, though I will need to substantially overhaul the visual design (this version only changes tile shapes). In particular, I find diagonal movements are harder to visualize than with hexagons.
#gamedev
#gamedev
Added point of interest distribution and cost-based, closed path generation.
The results are still messy and expensive to compute but should at least be workable.
Also added information displays for paths and pathing weights.
Next I want to compute remote nodes for faction placement.
#gamedev
The results are still messy and expensive to compute but should at least be workable.
Also added information displays for paths and pathing weights.
Next I want to compute remote nodes for faction placement.
#gamedev
June 4, 2025 at 8:23 PM
Added point of interest distribution and cost-based, closed path generation.
The results are still messy and expensive to compute but should at least be workable.
Also added information displays for paths and pathing weights.
Next I want to compute remote nodes for faction placement.
#gamedev
The results are still messy and expensive to compute but should at least be workable.
Also added information displays for paths and pathing weights.
Next I want to compute remote nodes for faction placement.
#gamedev
Canvas drawing was getting slow, so I converted my terrain generator to WebGL, along with a few other usability improvements.
Also modified the procedure slightly to improve map consistency.
Now looking into algorithms for terrain analysis and location/path generation.
#gamedev
Also modified the procedure slightly to improve map consistency.
Now looking into algorithms for terrain analysis and location/path generation.
#gamedev
June 3, 2025 at 2:13 AM
Canvas drawing was getting slow, so I converted my terrain generator to WebGL, along with a few other usability improvements.
Also modified the procedure slightly to improve map consistency.
Now looking into algorithms for terrain analysis and location/path generation.
#gamedev
Also modified the procedure slightly to improve map consistency.
Now looking into algorithms for terrain analysis and location/path generation.
#gamedev
Spent yesterday throwing together a hex map generator using BFS with contextual randomization.
These are larger than I would likely use for a territory conquest game, but the results are much better than my prior attempts.
(cherry-picked the interesting maps, of course)
#gamedev
These are larger than I would likely use for a territory conquest game, but the results are much better than my prior attempts.
(cherry-picked the interesting maps, of course)
#gamedev
June 1, 2025 at 5:30 PM
Spent yesterday throwing together a hex map generator using BFS with contextual randomization.
These are larger than I would likely use for a territory conquest game, but the results are much better than my prior attempts.
(cherry-picked the interesting maps, of course)
#gamedev
These are larger than I would likely use for a territory conquest game, but the results are much better than my prior attempts.
(cherry-picked the interesting maps, of course)
#gamedev
Perhaps I had a little too much fun mocking up the domain client application...
yagura.app
多分、最近のマンガやアニメの消費量がちょっと過剰になっていますかな、況して…
At least content with how much easier TypeScript+Lit has been to manage, though I have yet to implement any back-end communications.
yagura.app
多分、最近のマンガやアニメの消費量がちょっと過剰になっていますかな、況して…
At least content with how much easier TypeScript+Lit has been to manage, though I have yet to implement any back-end communications.
May 7, 2025 at 9:35 AM
Perhaps I had a little too much fun mocking up the domain client application...
yagura.app
多分、最近のマンガやアニメの消費量がちょっと過剰になっていますかな、況して…
At least content with how much easier TypeScript+Lit has been to manage, though I have yet to implement any back-end communications.
yagura.app
多分、最近のマンガやアニメの消費量がちょっと過剰になっていますかな、況して…
At least content with how much easier TypeScript+Lit has been to manage, though I have yet to implement any back-end communications.
Throwing together a simple vector graphics editor while getting used to TypeScript+Lit.
Was in need of an image format for icons and emoji for the messaging client, so this seems like a reasonable starting point.
Granted, it can only handle simple polygons at the moment.
Was in need of an image format for icons and emoji for the messaging client, so this seems like a reasonable starting point.
Granted, it can only handle simple polygons at the moment.
April 29, 2025 at 9:40 PM
Throwing together a simple vector graphics editor while getting used to TypeScript+Lit.
Was in need of an image format for icons and emoji for the messaging client, so this seems like a reasonable starting point.
Granted, it can only handle simple polygons at the moment.
Was in need of an image format for icons and emoji for the messaging client, so this seems like a reasonable starting point.
Granted, it can only handle simple polygons at the moment.
Reasonably content with some of my checkmate puzzles for my 'chess variant', though it is proving difficult to make them meaningfully different—not to mention validation.
One- and two-move puzzles are not bad (making them nontrivial is harder), but I have managed a couple four-move puzzles.
#gamedev
One- and two-move puzzles are not bad (making them nontrivial is harder), but I have managed a couple four-move puzzles.
#gamedev
February 4, 2025 at 6:58 AM
Reasonably content with some of my checkmate puzzles for my 'chess variant', though it is proving difficult to make them meaningfully different—not to mention validation.
One- and two-move puzzles are not bad (making them nontrivial is harder), but I have managed a couple four-move puzzles.
#gamedev
One- and two-move puzzles are not bad (making them nontrivial is harder), but I have managed a couple four-move puzzles.
#gamedev
Threw together a very rough concept of a 'versatile machine pistol' for the worldbuilding project.
Design has a lot of flaws (not my area of expertise), but the direction suits my preferences for a 'sidearm'.
Goal was somewhere between the MP7 and Flux Raider, with a 'less obtrusive stock'.
Design has a lot of flaws (not my area of expertise), but the direction suits my preferences for a 'sidearm'.
Goal was somewhere between the MP7 and Flux Raider, with a 'less obtrusive stock'.
January 20, 2025 at 9:48 PM
Threw together a very rough concept of a 'versatile machine pistol' for the worldbuilding project.
Design has a lot of flaws (not my area of expertise), but the direction suits my preferences for a 'sidearm'.
Goal was somewhere between the MP7 and Flux Raider, with a 'less obtrusive stock'.
Design has a lot of flaws (not my area of expertise), but the direction suits my preferences for a 'sidearm'.
Goal was somewhere between the MP7 and Flux Raider, with a 'less obtrusive stock'.
A friend printed me this rather nice set for one of the games I am developing.
Each piece is reversible to handle captures/drops, and each player has a stack of rings for piece promotion.
Playing on a board feels kind of strange, having only played it on a screen (not to mention the lack of guides).
Each piece is reversible to handle captures/drops, and each player has a stack of rings for piece promotion.
Playing on a board feels kind of strange, having only played it on a screen (not to mention the lack of guides).
December 25, 2024 at 3:58 AM
A friend printed me this rather nice set for one of the games I am developing.
Each piece is reversible to handle captures/drops, and each player has a stack of rings for piece promotion.
Playing on a board feels kind of strange, having only played it on a screen (not to mention the lack of guides).
Each piece is reversible to handle captures/drops, and each player has a stack of rings for piece promotion.
Playing on a board feels kind of strange, having only played it on a screen (not to mention the lack of guides).
Adding some simple sandbox presets to my board game to help with learning piece movements.
Still need to find a good way to illustrate/explain the interface and core rules of the game.
#gamedev
Still need to find a good way to illustrate/explain the interface and core rules of the game.
#gamedev
December 18, 2024 at 3:37 AM
Adding some simple sandbox presets to my board game to help with learning piece movements.
Still need to find a good way to illustrate/explain the interface and core rules of the game.
#gamedev
Still need to find a good way to illustrate/explain the interface and core rules of the game.
#gamedev
December 9, 2024 at 4:24 AM
After experimenting with matrix-based terrain generation, I am back to using Perlin noise; though, now with randomly generated matrices as frequency modifiers.
The results appear reasonably varied, though I still need to add regional terrain parameters and maybe some post-processing.
The results appear reasonably varied, though I still need to add regional terrain parameters and maybe some post-processing.
December 9, 2024 at 3:45 AM
After experimenting with matrix-based terrain generation, I am back to using Perlin noise; though, now with randomly generated matrices as frequency modifiers.
The results appear reasonably varied, though I still need to add regional terrain parameters and maybe some post-processing.
The results appear reasonably varied, though I still need to add regional terrain parameters and maybe some post-processing.
Once the algorithms are more usable, I am hoping to integrate them into a large-scale planet generation/exploration engine (for which my designs also need much refinement).
November 23, 2024 at 10:33 PM
Once the algorithms are more usable, I am hoping to integrate them into a large-scale planet generation/exploration engine (for which my designs also need much refinement).
Icosahedron-based, distance-weighted planetary terrain generation. Vertex heights obtained from distance interpolation of regional terrain descriptions. Far from optimized and needs better heuristics, but seems to be on the right track for my applications.
#gamedev
#gamedev
November 23, 2024 at 10:33 PM
Icosahedron-based, distance-weighted planetary terrain generation. Vertex heights obtained from distance interpolation of regional terrain descriptions. Far from optimized and needs better heuristics, but seems to be on the right track for my applications.
#gamedev
#gamedev
Hard to describe without context, but here is an unusual victory I got against the computer. The final move was swapping knight (jumps forward and back diagonals/sides) and promoted "rook" (moves along adjacent directions/jumps first tile). They also guard each other's tile after swapping.
November 23, 2024 at 9:02 AM
Hard to describe without context, but here is an unusual victory I got against the computer. The final move was swapping knight (jumps forward and back diagonals/sides) and promoted "rook" (moves along adjacent directions/jumps first tile). They also guard each other's tile after swapping.
Another image generated a while back—this one attempting to visually approximate a distribution of star systems in a spiral galaxy using a randomly addressable function rather than a simulation. Currently rather expensive, due to the use of Perlin noise, but suitable for locating single bodies.
November 8, 2024 at 5:36 PM
Another image generated a while back—this one attempting to visually approximate a distribution of star systems in a spiral galaxy using a randomly addressable function rather than a simulation. Currently rather expensive, due to the use of Perlin noise, but suitable for locating single bodies.
Rudimentary planet concept from a while ago—based on the (physically questionable) idea of a body having an axial precession rate equal to its orbital period.
Might be more interesting when I finish writing a simulation for it.
Might be more interesting when I finish writing a simulation for it.
November 8, 2024 at 4:50 AM
Rudimentary planet concept from a while ago—based on the (physically questionable) idea of a body having an axial precession rate equal to its orbital period.
Might be more interesting when I finish writing a simulation for it.
Might be more interesting when I finish writing a simulation for it.
Spent the last few months working on a web-based board game—a loose mix of hexagonal chess and shogi.
The game itself feels fairly solid, though the platform and visuals need more work before permitting a wider player base.
The game itself feels fairly solid, though the platform and visuals need more work before permitting a wider player base.
November 6, 2024 at 9:43 PM
Spent the last few months working on a web-based board game—a loose mix of hexagonal chess and shogi.
The game itself feels fairly solid, though the platform and visuals need more work before permitting a wider player base.
The game itself feels fairly solid, though the platform and visuals need more work before permitting a wider player base.