"miyazaki can you texture this room" "ugh but it's nearly lunchtime" "just get it done"

remembering that time I came up with a method for faking pre-rendered backgrounds

i haven't decided what this thing actually is yet

There are four enemies nearby. You press lock-on. Which enemy gets targeted and why? A) Closest distance, off-screen B) Second closest, behind player C) Third closest, in player's line of sight D) Farthest distance, but nearest center of camera

Just thinking of that time when Nintendo locked Slippy Toad On-Foot Bazooka Mode behind one of the hardest challenges on the N64, meaning most of us never even saw it 🥲

a tip I learned from Ocarina of Time - no need to model the ceiling of a cave, just throw a black quad up there

Don't sleep on trigonometry kids, one day you might need it to calculate the angle of a locked-on side hop

it's a thing of beauty 🥹

This is a skull with wheels I guess

there's something so satisfying about modelling super low poly, hacking it out fast, vertex by vertex, then immediately getting to see it in-game. it really makes you feel "in the zone"

hah yes, chains in this game are all just 2 quads. your framerate'll love it

This is how I find bugs. I call it "chaos testing"

When designing a dungeon, I'll first pick a theme (e.g. lava temple), a general layout (e.g. a tower) a key item for solving puzzles (e.g. bow and arrow), and a core dungeon mechanic (e.g. changing water level) At this early phase I'll also consider some puzzles, enemies, and obstacles that would suit this combination. I'll then start creating a 2D map/graph (something like the picture below) while also writing a story of the player navigating the dungeon beat-by-beat. E.g. "I used X item to solve a puzzle, which rewarded a key. In this room I also noticed a door I can't reach yet. I suspect it might lead to that room with the chest I saw before. I head East." If the dungeon is nonlinear, this story can split off in different directions which need to be designed as well. The goal is to tell the full story of the player navigating the dungeon from beginning to end, in a way that's logically sound, prevents softlocks and frustration, etc. At this point I don't worry about what the obstacles / puzzles / bosses actually are. I'm just focussing on the logical order of their completion. Once I'm happy with this sequence of rooms, keys, locked doors, barred doors, key items, and puzzles - then I'll move onto 3D modelling. The layout of the dungeon will change drastically when converting to 3D, but I try to keep the big story beats the same. At this point I can playtest to check my logic actually works. Then I go through designing each room, adding puzzles, enemies, secrets, etc, always careful to keep the macro-level sequence logic intact. The way I design an individual room would be worthy of its own thread. But then it's just lots of polish and playtesting!

There's a billion ways to achieve sword trails and slash effects. Here's a solution I put together this morning for my retro-style Zeldalike #gamedev

my audience is much smaller than Kaze's, but I've only ever seen positive reactions to Azaran's janky low poly graphics where all my haters at??

whenever I'm playing a new video game or watching a new movie, I always bear in mind John Updike's first rule for reviewing books (from 1977)

make small games. if you don't have the skills to hack out a polished Pong clone in a weekend, your 6-year dream game project is in serious trouble

okay, here's a compiled list of indie 3D Zelda-likes 🧵

video games: appealing to gooners for over 30 years

the older Zelda games had distinctive "zoom regions" where the camera pulls in or out depending on how cramped or open the space was - independent of camera collision. Not only practical, the effect gives a cool cinematic and extra-polished kind of feel

found a cool green man I made for some long forgotten project

gotta have a lil dungeon intro cutscene, right???

yeah the map is literally just floating above the level

my design philosophy for enemies is heavily driven by whatever's simplest to implement. I wanted some kind of crustacean enemy, but didn't want to animate legs or sideways locomotion or snapping claws, etc. So I ended up making these wheel-flea dudes

Learning gamedev is hard. It combines multiple creative and technical disciplines, many of which have punishingly steep learning curves. So motivation is often a problem early on. It's easy to feel defeated every step of the way, and giving up is the norm. Having a "Dream Game" can be useful for learning in this regard. A Dream Game can provide that boost of motivation you need to get through the hopeless times. You can channel your passion and vision into the determination to learn. Conversely, working on small games can be massively demotivating. I don't want to make Breakout or Flappy Bird, I want to make Chrono Trigger. I want to flex my creativity and my game design intuitions, I want to build something amazing. And here I am, struggling so hard to make a game I don't even care about. What's the point. This is why it's so common for new devs to work on huge dream projects as a vehicle for learning the craft. And as long as you're learning, that's great. But you must understand that this game is your sacrificial lamb. You will never finish it. Granted, some devs do manage to succeed at the "learn while I build my Dream Game" strategy - Stardew Valley comes to mind. But almost everyone else will fail, and for a few reasons. The first is that you cannot size a project unless you know what it takes to finish. How long do you want to be working on this Dream Game? 4 years? Okay, so how are you supposed to know what a 4-year game looks like? Unless you've finished a game before, you cannot know. You've likely misjudged the effort it'll take by an order of magnitude. Secondly, finishing a game requires its own set of disciplines. There's so much to learn about putting out a commercial product. Have you considered graphics options? Resolution, framerates, vsync? Audio settings? Accessibility? Localisation? Do you know what it takes to QA a game? Do you know how to configure builds and depots on Steamworks? Do you really want to be encountering all this stuff for the first time at the very end of a giant 6-year project? Thirdly, it's easy to get bogged down and stymie your learning when you have such a large game to make. You can lose six months working on animated cutscenes before you've even learned how to make a pause menu. Everything gets dragged out, and inevitably you end up remaking huge chunks of the game as you come to realise that your early work was amateur trash - as is to be expected. Lastly, dream projects provide far too many distractions from actually learning game developement. Writing lore. Drawing concept art. Dialog. GDDs. These are low value tasks, but they're so much fun that you can easily spend all your time on them. And before you know it, it's been weeks since you opened your project file. Finishing and releasing small, simple games is the remedy to all these problems. Make a warioware-style minigame, make tetris, make a shmup, make a 10-minute avant-garde experimental FPS hallucination simulator. It doesn't matter, just make a game that people can download and play, then make another one. You can still build your Dream Game. Just later. If anything, it deserves to be tackled properly, by a dev who's got some real holistic experience under their belt.

this kind of "sand burst" effect is surprisingly easy to achieve with a particle system. here I just have a cylinder with separate "size over time" axes, plus some random rotation. works for water splashes as well! (yes, an underground enemy is about to come and get you)

Heh thank you! I was originally making a Chameleon Twist clone, but there wasn't much love for it. Maybe one day I'll give this project another look! 😆

This is the most metal death in all gaming. Sure, you can find a Doom or God of War death that's more gory. But for 99% of Wind Waker you're a little kid fighting slimes and bats who die in cute puffs of smoke... and then your final act is THIS. Contrast is a powerful tool

"floaty movement" is a catchall term for when any of the following are perceived as out of balance: input latency jump accel fall accel hang time animations anim blending foot sliding ground accel ground decel walk->run->sprint transitions quick turn friction juice etc

"Where is the light coming from?" "The same place as the music."

A thread on how I animated a lil crabby sandworm entirely in code! Firstly in Blender I made some simple low-poly body pieces. The blinking eye is achieved by offsetting UVs, it's a technique I use a lot

Upgrade tokens

it's surprising how many Zelda dungeon rooms are just squares, rectangles, and cylinders. Even as hardware improved, they kept this design philosophy throughout the series Complexity comes from how the rooms link together, not the designs of the rooms themselves

if you begin learning gamedev by releasing lots of small games for a year, then transition to work on your big dream game, you'll finish it sooner than if you'd started on the big project day one of learning

Low-poly modeling is 99% figuring out the best way to split a quad

This is actually working what the heck

Mario sharing the same sprite for bushes and clouds is cool and all, but did you know the water temple has staircases on the ceiling

looks simple, but this 1-min starting area introduces the player to running, swimming, climbing vines, auto-jumping, ledge grabbing, pushing, pulling, and climbing ladders

I've been experimenting with Dark Souls and I swear it just prioritized "off-screen but in line of sight" over "on-screen and closer"!

pretty much spent a whole week just on the layout and sequencing of this dungeon. now onto designing puzzles for the individual rooms

Here it is, the most boring gamedev thread I've ever made. This one covers the glamorous topic of Autosaving, Void-outs, Death, and Respawning 🧵

silver rupees are the key to understanding old 3D Zelda design philosophy

I love these blending sections between textures in older games, before the days of splat maps. Such a cool detail that makes all the difference. Looks like they even planned the geometry to account for it

"Is that a puzzle? Is looking around a room and finding an eyeball on a wall really super fun for people?" Eye Switches aren't puzzles, they're progression locks. Their key is the Bow. Most stuff in adventure games is just locks and keys

"Benji, your sea urchin enemy is too unrealistic, in real life they don't have terrifying eldritch cyclopean eyes-- oh wait"

I'm not really an indie dev. Silksong and Hades are indie games. My games are more like those tasty snacks that Indian street vendors sell off the back of a motorbike

tutorial caves are coming along nicely! I've been sticking to the "old zelda" design philosophy of simple rectangular rooms, and it's one of those situations where constraints really force creativity. it's some of the most fun I've had on the project so far

The indie game I'm making and three of the inspirations behind it:

under the surface, all video games are just TRON

"the movement is too floaty" "the combat lacks oomph" Ah yes my apologies, let me tweak the floaty dial and the oomph parameter

in gamedev stuff always looks dumb before it looks cool. here's a big dumb capsule man. he doesn't get to be in the final game, but his role is still very important

vertex color shadows in old Rare games were wild (they did this by modelling the shapes of the shadows into the polygons of the ground, then recoloring them)

🤯the Azaran demo is now live on Steam!!!🥳 Let me know what you think if you've played it, and feel free to leave a review! thanks to everyone who's been following me on this journey so far, it's been an amazing 18 months. Can't wait to bring you the full game!!!

I consider Windfall Island the gold standard of tight overworld level design. It's small and simple yet densely packed, great verticality, interconnected in delightful ways, filled with secret little nooks and prime for exploration

is there some hidden corner of the internet filled with indie devs making awesome 3D zelda-likes, or is there literally like seven of us

Yeah I think D is the sleeper, we naturally shift our camera to aim at the thing we want to attack....

i guess you could do that with the skullcopter thing, sure

my game is okay i guess. but the algorithm bloody loves me

just some lazy piranha boys

does anyone ever actually use the "climb down" action when near a ledge, or did I just waste an hour putting this in

Okay, I've come up with an ingenious solution for handling this rare edge-case of falling between two intersecting slopes: I'm just gonna not put them in the game 🧐

bug testing is ridiculous, what am i even doing

started making actual levels with actual puzzles, almost like it's a video game or something

Spectacle Rock from LoZ doesn't just make an Easter Egg appearance in Death Mountain Crater - it actually IS the Fire Temple

looking back at this old tweet is a really nice gauge of progress. Original pic was from February 19th, compared to the final dungeon layout released with the demo on May 30th

Some new facial animations for Zayd!

with a few physics tricks and lies, you can sure make it *look* like this is water

Doors! Imagine a world without doors. Thank goodness for doors.

I love it when state machines automatically handle dumb edge cases you didn't even think about

I once worked for an mmo and saw a graph I wasn't supposed to see. A huge chunk of the studio's revenue was from subscribers who hadn't logged in for over a year and had just forgotten to cancel

✨it's me again, the Gamedev Fairy, reminding you to reduce your scope✨

mainly been working on sound design this week. probably the most overlooked aspect of game design, yet one of the most important. thank god for FMOD

"Don't quit your job" is the #1 most valuable piece of life advice for any fledgling gamedev

been tweaking combat this week, definitely tricky to get feeling good. I've added some iframes to backflips and sidehops to give them a bit more utility

the next generation of successful indie devs are currently learning their craft releasing terrible free games on itchio that nobody will ever play

"Make art that you're passionate about." The Cistine Chapel was a commission that Michelangelo absolutely did not want to do. He hated every second of it Necessity and obligation can be powerful drivers of creativity too

Argh I was just about to reveal this new item in Azaran: Islands of the Jinn - but Zelda beat me to it!!!😖

all sorts of bomb interactions

I used to be an indie author, and the state of that industry is DIRE compared to indie games. The average person barely reads, let alone indie novels. And just look up the best-selling authors of the past 5 years. It's miserable The indie game space is pretty great really

I feel like a lot of devs are worried to show anything but polished and well-edited vertical slice gameplay footage, because it might misrepresent the final product nah, show off your process, your struggles, your failed experiments, your bugs. People love a peek behind the curtain

Target switching debug (red dot = right stick)

while I'm showing off old abandoned projects, here are some dino rancher prototypes I was working on a while ago. Both a 2D version and a 3D version... 😆

vertex colors are so rad

more puzzle prototyping

working on the map system again. previously I was drawing individual sprites by tracing over the level geometry - now I'm creating the entire map out of actual polygons. it's a much faster workflow, plus it allows for SVG-style lossless zooming

faked lighting with vertex colors!

Poll results are still coming in, but there's a clear winner. Thanks everyone, here's my final lock-on target selection code

don't you hate it when you're chillin on the beach and some punk kid starts lobbing bombs in your eye

i'd love to see more people showcase their rigging. sometimes it feels like forbidden magic

Bomberlith 64

People got very upset at me for calling out a sketchy Kickstarter a while ago - it's seen as bad form to criticise other devs. I think there's generally an acceptance that sometimes your money just goes bye bye

i needed a break from the main game, so i've spent the past two weeks making a short lil side-project - Azaran: The Demon Bottle it's basically ready for release, just need to wait for all the Steam processing stuff. i'll share more when i can!

out of respect for Nintendo's patents, my game will not use creature summoning unless you count picking up a monkey and storing him with your magic gauntlet and then taking him out again. ah crap

🕹️Game: Azaran - Islands of the Jinn 🛠️Tools: Unity, Blender, Affinity, Ableton Live 🚀Started: January 2024 🎯Release: 2026

The cool thing about low poly is that you can model a boss in 30 mins