Game Dev Secrets: Extreme Design Technique!
Вставка
- Опубліковано 18 вер 2024
- Follow Isadora's Edge on Kickstarter: www.kickstarte...
Wishlist Isadora's Edge on Steam: store.steampow...
(I extreme tune everything, constantly - I love doing it so much. it really helps me with ideas, it's actually bananas how useful this is, man. even if it doesn't generate any good ideas, sometimes it's just a nice mood booster to get to watch Isadora go flying off into the sun)
The game that I'm developing as an indie game dev is called Isadora's Edge! A 2D Pixel Art platformer game, that I'm developing in the Godot Game Engine! If you're new to Godot or have any questions about Godot or game dev in general, feel free to ask me! #indiegame #gamedesign #godot #godotengine #gaming #gamedevelopment #pixelart #gameengine
Wishlist Isadora's Edge on Steam!
store.steampowered.com/app/3125320/Isadoras_Edge/
And you can follow the Kickstarter here: www.kickstarter.com/projects/inboundshovel/isadoras-edge
Thank you!
Dude idk about the final game,
But your tips and tricks during 30 sec video were amazing.
Most of them I already learned the hard way during last year.
Great content for starting game devs 🎉
Is Isadora’s name a reference to John Isidore from “Do Androids Dream of Electric sheep”? By chance?
"isadora's edge" 💀
Isadora do be edging
Dude please bring this over to consoles..I am begging you!!!!!!
I think the flurry would be comeplete if the flame effect persisted a bit longer during the animation, like a particle
Great suggestion!
A stronger end attack like a power stab would reward players for being able to find the opening to finish a flurry too
@@FaticusDolphiniuswhat if you made it like a rhythm game and for each hit of the flurry, if you also provide input it’ll do more damage, and if you hit all three of the hits well enough you’ll add a fourth hit to the combo?
@@talongreenlee7704 that would work depending on your project, imo for an action platformer that doesn’t have that mechanic anywhere else just having a full flurry rewarded would be engaging enough for the player. But plenty of games have a similar thing you mentioned baked into all of their combat and it does very well like the Arkham games
You can also do the "afterimage" effect by having a second animation layer with some transparency on a very slight delay if your move is super quick, although this depends on your art style and direction.
this was like that one trend of “multiply a random value in your game by 1000 and see what happens”
Gary did a thing.
TF2 x10 maps be like:
me jorking my weenis
I know I’m exposing myself as a theatre kid here, but the principles feel similar. I remember almost every director I ever worked with or around almost always wanted the actors to overdo it so they could rein them in. It’s easier to get a person to do less once they demonstrate they can do more than it is the other way round usually. Often what the actor thinks is “too much” really isn’t.
What a dev might initially think is too much for a game might end up becoming a huge draw for players, or you might find somewhere you can use it, like the flurry attack here.
Reminds me of advice I heard where, instead of waiting for the right opportunity and it never coming, you take opportunities until you're used to taking them and then start saying no to the ones that aren't worthwhile
Good parallel to cinema/theater, and great advice!
Yep, we do the same in music. "Make me tell you you're doing too much" is a really powerful tool for directors (in regards to dynamics for example). It's more actionable than just a vague "give me more" - it's a dare.
(also sidenote, musicians often use a theater metaphor of "stage makeup" for this same concept. Like "if you do what feels like way too much, it will sound right from the audience")
My band teacher told us the same thing
Animation does this too. There’s a fun story about an animator being told over and over that his animation was too flat and boring. Finally, he started to get genuinely upset. He was like, “You want me to go further? I’ll go so far it’ll make the director mad!” He took it in, the director watched, and then he was silent for a second. The animator starts sweating. The director slowly turns to the animator and says, “There! That’s what I wanted!” :)
I also find that the fastest way to find the best value for something (and is extremely helpful for a higher lower sort of thing,) is to get the two most extreme values, then if you choose the middle number and see if it’s too much or too little and you choose the in between for the middle number and the extreme you want it closer to and repeat this until you get the perfect value. This is also one of the fastest ways to guess proven by maths.
Your basically doing an algorithm aswell
A binary search😂
@@mojojojo6535 lol
Yep! Interesting this works for animation as well. The same concept is used when you draw key frames at the most extreme end of animations so you know how far a character will move between frames, then you can fill in the blanks with squash or stretch. Or start with squash and stretch to know how far the character will stretch (like when flattened or reaching out) and then draw the frames leading up to that. Takes out the guess work.
woah more people than just me do this
@@aykarain great minds think alike
Another neat thing is by going for big changes immediately, you can actually walk those changes back and more quickly narrow down the biggest effect that's reasonable. Basically binary search for the desired value.
There's a coding method I was taught called "A lion in the desert" (Possibly based on "How To Catch A Lion In The Sahara Desert") which basically states that you have to try a value that is definitely too high, or a value that is too low and split the value in half over and over until reaching the desired result.
Basically, you set it to the unreasonable 300%, then you try 200% (assuming we're only considering what we're adding over 100%) and then 150%, then 125% and so on. And then if the added speed doesn't feel like enough, you go backwards. For example, if 200% felt like too much and 150% feels like too little, you'd try 175%.
That's straight up a limit process, literally how mathematicians and computers find the decimal expansion of irrational numbers. And generally it's great to find a good factor.
I think a new frame during the flurry rush animation will help make the move look more smooth, Since when isadora does her slash big slash because its so fast it looks like the sword just appears from the ground to isadora holding all the way up in the air. I think adding it will make the flurry’s rush make the sequence of attacks look and feel more satisfying to pull off.
I hope make sense.
That 300% attack speed would be perfect for a whip
Exactly what I was thinking
Nice video man! I followed your journey since the start and it's incredible to see what your creation is turning into. Will definitely play the game once it comes out!
Thank you! 🎉
One other way to do tuning is to crunch the numbers to make sure they are meaningful.
If something attacks every 2 seconds and dies in 4 hits, make sure your upgrade pushes your attack speed enough that you can fit 4 strikes in that window.
It's not a good feeling when you get a 5% upgrade but it doesn't change the gameplay except for some edge cases, like making boss fight go 5% faster.
I feel humbled that I discovered this all on my own without knowing this exists. It just feels like a quicker way to tune values and understand what feels best.
I can already tell this game is gonna be sick. The way you interact with good suggestions and in general with your community gives me such faith that this game is going to be soo good.
That’s so true, I was testing my brick breaker game with no damage and then with every bounce the main character would get faster, after a while they went through stuff because of how fast they were and then I got the idea of a skill that makes you go lightning fast and go in a straight line without bouncing on enemies until you hit a wall!
The flurry attack looks a little goofy ngl sounds fun tho
if the flurry animation was refined more it couls be really cool
Yeah, that animation is just a placeholder, haha!
This is literally what happened with civ 1, there was barely any time to release, and the map was really small, so they just doubled the map size and that worked. They never would've gotten there via starting off doing fine tuning
And the quote from Sid Meier about doubling or halving things to test extremes fit that too
You make great content, Shovel.
This game is going to be AMAZING
0:23
They got Isadora on that Celeste speedrunning tech
also 0:55 is how I like to beat ganon in tears of the kingdom
"If it's worth [testing], it's worth overdoing."
Oh interesting! The version of this advice I learned has an additional step where you then take that value and do half or a third of it to get closer to the real value that you’re looking for and continue iterating from there. Great video!
we love making features absurdly powerful for the heck of it
I think it’d be cool if you could dash forward a fixed distance while you do a the flurry attack, making it a flurry rush. It would work good if there was a hoard of enemies in front of you since you could get through them all quickly
Wow Bro!
You really are doing a good job on this thing !
Proud of you and your team !
There's a famous quote from game designer Sid Meier about doubling or halving a value just to see what the extremes feel like
First heard this one when listening to people talk about board game design, so glad to see that's something transferable between the two.
Just a suggestion. I think flurry need some additional effects. Maybe air puffs in the ground escaping from the character (same as landing?). And probably some effect to show "cooldown/exhausion" after the flurry was used (some pixel sparkles rising?)
Add a easter egg where isadora just makes a flurry using that attack
The flurry would be fun to see as like a general grievous lightsaber spin
Finally now I have a real excuse to bump all my values up to unreasonable amounts for fun
I didn't know that there's a name for it, but I've been using it all the time lol. It's like you're just sitting there, thinking how you could make the controller (or anything else for that matter) feel better, and you randomly go "you know what? I'll just tripple my dash speed and see what happens". It's really helpful, cause we have this weird bug in our brains where if we stare at something for too long, it becomes hard to see the real picture, what feels right/wrong etc. It's kinda like how artists loose their sense of proportions and have to flip the canvas to snap out of it.
remember: at the end all that matters is making the game fun
The extreme tuning looks very fun and like a good chaotic laugh >:)
Its also a great way to find bugs. You might be testing within 10% of the value but not notice its all placebo and something else is resetting the value after (credit to Suspicious Developments for this one)
This is a fantastic point, it's a great bug catcher!
stats point is always the best ways to upgrade and flexibility builds for any type of player to enjoy and experimenting
"What if Attack Go Brrrr?"
"that's stupid. I love it"
I just discovered your channel with this short and watched a few more, i really like your content, keep it up!
It’s also just really fun to make the numbers big
As others have said, my first thought was binary search! I also like this approach because I will adjust things to ridiculous values just to see how MUCH it changed.
I do this in 3D a lot too: "Wait what is this changing? I bet I'll see it if I multiply it by like 300%" lol
Imagine during that flurry her hair would Ignite like the Hair of the Disney version of Hades. :D
MAN I LOVE UR VIDS
Go to town on some fooools
Since smears are seperate, it might be a good idea to have them last longer for the flurry like a sort of afterimage, as they seem to pop too much when sped up so fast.
me, a roguelike enjoyer : "unreasonable" ?
Well there's also accounting for invulnerability frames for the speed of the attack. Depending on how rapidly the damage is calculated, you eventually get to the point where you whiff half of your attacks because the enemy doesn't take damage every frame due to invincibility frames. Kirby games are very notorious for that (especially Whispy Woods, who seems to have the longest i frames due to being stationary). The balance goes both ways, but the "feel good" department is also important, and you can always just mess with the hidden damage values to make it seem like it's just a speed upgrade but you actually do damage at a faster rate ONLY when using the upgraded weapon so it feels like you're killing enemies faster.
for art from the store it looks really good lol
i mean it's placeholder art so it obviously looks out of place but it's surprisingly good
*Sets Gravity to 9000*
And that’s how the Goomba was born!
the flurry attack should 100% have its own animation beside just attacking faster imo
and just in general, if i ever make a game ill binge these shorts one by one and use all of them
Sometimes I use Extreme Tuning just to see something funny happen, but it usually never gives me any ideas to add to a game
*me increasing the variables to max to mess around in testing*
“I’m somewhat of an extreme tuner myself”
woow I never know this before, thank you
Add screen shake on the flurry, trust.
Balance by chaos!
It's fun
I like the two fast one super heavy combos.
For example if the first two connect you can launch a third slower but aoe effect attack which does extra damage.
She needs a dragon ball aura or an eye glow that would be METAL.
The transition off of the flurry looks a bit rough, so I think it might be a good idea to end it off with a downslash or something similar if you are willing to animate it.
I don't know if this was already suggested, but I think a sprint attack would be really nice! Yknow, the one where you just sprint snd lesve a trail with your sword.
Anyway can't wait to play the game!
that sounds like an upgraded attack which is cool!
I feel that for that type of attacks you need knockback on the end, it just feels and looks better
That tree saw God
One thing I really liked about Mega Man Zero was, you started out only able to make a single attack with the beam saber, but through regular use you unlocked 2-hit and then a 3-hit combo. Slightly different animations and less delay between attacks.
It's not a suggestion, and I have no idea how difficult that is to implement. But that's what this made me think of :3
Ohhh I thought this was going to be something around the multi-hit feel
Like you make the first two slashes fast then the last one starts a big slower so you _feel_ it stronger and heavier
But yeah the flurry is also nice! I agree with the particle trailing for a better feel definitely
Imagine you could start attacking an then slowly get faster and faster while continually pressing the attack button.
you should make flurry like add bonus damage based on how much hits you do se it feels even more satisfying when you hit full
Depending on game, you should try setting your speed to significantly large number. This will highlight a lot of bugs like wall clips or pass through unintended walls. After that you may change your logic or limit your character's speed
I did that once, making an archer character shoot a couple hundred arrows at a time.
300% speed feels like anime
I dont know much about game design outside of your videos, but I think the flurry should have a bit of a charge up animation. Maybe wind building around Isadora or something.
I actually do this but as a binary search. I do a high but reasonable guess and half it until I get a good feeling value. Saves a lot of time.
I think it would look really good if the last hit of the flurry attack slowed down at the end instead of keeping pace with the rest of the flurry. It would also give recovery frames if you wanted to balance it out
But idk
I did soemthing similar, but by accident when trying to decide how big the trees should be in our wingsuit game.
We started off with realistic scale of 1 but it was far too small even if it was the "correct" size, we accidently set it to 8 times scale or something while testing, and suddenly the same number of trees looked a lot more filling of the landscape and more interesting to see and fly between.
We stuck with huge trees since then.
It's also just fun in general
Overshoot. It gives ideas
At 300% it also kinda looks like a whip ^^ maybe an abbility where her sword turns into a whip? Qould be super cool
You could also approach it like this
The animation is rly fast, but the number of attacks are much less
You still get that look of fast, feel of fast
But with a trick of it not actually syncing up the dmg to match the animation
This Game looks great ngl!! I have a question since I'm new to game dev. Which resesolution does the tiles of your tileset have? I'm stuck at figuring out which resolution my player should have in contrast to my tileset.
Reminds me of spark tej. Might buy
Wonder how extreme tuning would work for enemies instead of the player. Like, “Watch out, this black and red crackling energy effect means their next attack will deal hextuple damage!” Maybe as some sort of challenge mode. Certainly more fun than just being like “this boss has 30% more health and attack” as a challenge mode
I tend to have attacks and actions with rates close to an integer multiple of the music bpm. Its a nice short hand to make gameplay feel smoother.
The music has some sonic vibes to it.
This doesn't just apply to game dev!
Extreme tuning is super useful for any variables, whether in art or data processing.
You could make a dash attack that looked like sonic rolling.
Lovin it!!
Idk why but i think if her hair and sword became orange during the flurry would be cool
OrangeDora!!!!
Something similar happened when I was testing a vehicle controller that I build, one of the test it's to do a loop on a road, but I wasn't fast enough to loop and the car kinda grip to the ceiling, furthermore I consider to let this as future upgrade instead of removing this bug
That's where binary search comes in
I do this when making music to see if the fx actually make a difference.
best channel ever
URF unlocked.
Her: "I have to go further! To go beyond flurry attack x 3....x 5...x20...."
Villain: "Foolish girl do you really think such a weak attack cou--"
_anime track drops_
Her: "FLURRY ATTACK x 1,000!!"
There's just one thing I noticed about the flurry attack is that it looks fast... but not in a cool way you know?
I might be completely wrong here, but hear me out: WHAT IF the last frame of each attack stayed in place with 80% opacity during the next two frames with 40% decay each frame?
What I mean is: In the first attack the final frame of it would leave this shadow of 80% opacity in the very next frame, then in the next frame the shadow now has 40% opacity, and in the third frame it is gone.
You know? Once again, I might be tripping here xD
BUT I think this would... look cooler, you know?
As a Risk of Rain enjoyer , just don't have limits, limitless scaling that can break the game and literally make your computer crash is the way
so, instead of boring 20% faster attack upgrade we get EX Slashes new move?
this is awesome!
Gib new moves new jump new dash new outfit and hair new everything \o/
I think that it should skip some of the frames with the attacks for the animation speed to feel more consistent
Maybe you could give her upgrades to increase her attack speed
I thought you were to suggest first starting with a big number like 256% then, finding it fast, halfing it. If 128%, too, was fast, you half it.
Then with 64%, you could either add or subtract 32%. And then you add or subtract 16%, 8%, ect.
i see you made it right, if you want an attack to really feel powerful the enemy and maybe even the playable character should recoil from said attack, as if it had a lot of inertial energy, i wonder how you made the sfx for that.
You missed the opportunity of introducing people to the notion of Binary Search in Game Design.
A next step after choosing a very high value or a very low value, would be to decide whether you think the value should be lower or higher than the current value, and double that value if it should be higher, or, halve that value if it should be lower, and keep doing that process consecutively until you reach the value that's most likely or exactly the value that you want.
That would be a killer special attack though
Nvm just watched the rest 💀
What if you make a very similar attack but the first 2 hits are very rapid, whilst the last one is very slow and is the one that deals the real damage and can be charged
Also i love how this game is turning out, keep up with the good work❤
Game looks fun, is there a parry mechanic? Perhaps you could teach us how to implement those.