It just seems like a lot of overkill

every message gets sent to everyone, and gameObjects just take whatever they need

the code is super simple, and it doesn't seem to be slow even for hundreds of units

Lets make the following changes:

Make the health bar slide down instead of just going down instantly

Make it so that when a unit dies, it fades out.

that I'm ok with, but let's say the bullet... B can't seem to get damaged before the bullet reaches it

What do you mean?

hm, I definitely seem to have two types of messages. Some take "no time" to execute because they're logical, and some take time to execute as animations and maybe later on sounds

Seems like a good idea

A charges up for 50ms, creates a bullet, the bullet takes 100ms to travel, explodes on contact for 50 ms, a death animation that lasts 200ms, a heal animation for some more

etc...

Works for me

and so a message queue could be used for animation separately to take it out of the mushy game code that spams everyone everything, and if some animations are compatible (e.g. two people firing at different targets) they can be played together

and so I could have units generate proper animations instead of animating "themselves", like "animation(UNIT_INTERCEPTOR_DEATH, 200ms, unit.x, unit.y)

BTW your css seems to fail to load

hmm? my css?

oh yeah, 404 on https://alexmitan.github.io/coding-tbg/stylesheet.css

Oh, I forgot that in there

Oh well

works anyway

BTW could you make an option to disable console.log...

I can't do anything with my console if it just fills with messages like that

fair

and switch debug

ye

But mabye make it triggerable by a keycombination

So... you type somthing on the game and it inverts debug

@AlexMitan Hold on.. that's too many replaces to make

console.log = (function() {
  const log = console.log;
  return function log(text) {
    if( debug) { log(text) }
  }
})()

Just put that in front of your scripts

just do console.log = function(){} to stop the logging

Ok

BTW I am getting noticable lag with a lot of units

Same, I think it's the Canvas redrawing everthing every frame

hence the switch to SVG

just gonna remake this in svg now

That's a lot to do :|

You can find out whether that's true from the performance profiler in your developer tools

@TheMaskedRebel I'm busy

@FreezePhoenix Fine thanks

@AlexMitan Your game used 90% of my CPU for 10 minutes

@doppelgreener Trying that out now

@FreezePhoenix Again, not really a lot to do, I just have to replace the graphics component with a SVG one

@AlexMitan Very little of the CPU is spent drawing

Crap. It's the messages isn't it

@AlexMitan Nope. This was with console.log disabled

no, the in-game messages

in a total of 22 seconds, 303 MS were idle

@AlexMitan Correct

What is your method named update

HOLY

the frame-by-frame runner

the... draw.. thing

wait, maybe it's the fixed interval

Nope

It's the function on line 45

Ok... I'm checking out your lib folder

Ok... this is bad.

Modify cmutils.pickFrom to read

?

What's going on

cmutils.pickFrom = (function() {
  const {floor, random} = Math
  return function(arr) {
      return arr[floor(random() * arr.length)];
  }
}

Every single time you run update , you are traversing the Math object four times;

Also, the game is running at "100 fps", of course it's lagging

Yeh that's way to high

It's just for scrappy "see if the architecture works" purposes

I never planned to use canvas anyway

BTW canvas is the fastest thing in your game

Every single one of your functions takes more time than the canvas operations

@FreezePhoenix once you end the profile you can see what functions have what % of time spent on them

I know

I'm gonna have to see how it runs at a proper "fps"

And with a reasonable number of units

@doppelgreener The canvas is the least of his worries - that small orange band at the bottom is the canvas operations

@AlexMitan btw do you broadcast messages to every unit or do you have some kind of event subscription mechanism set up?

The game won't have 600 units on screen at 100fps each, and won't be on canvas, and won't have this architecture. I appreciate everything but I won't optimise something that's on its way out

@FreezePhoenix yep agreed, highly unlikely canvas will be a breaking point here anytime soon

@doppelgreener Broadcast until I get an intuition what "message pipes" to run things through

@AlexMitan Why would you use SVG? That's more dom interaction and will be slower

I already ran a "simulation" on paper and it seems I need a message channel for game mechanics and another for animation

At least let me clone your repo

@FreezePhoenix Easier to code for, and more in-line with the game logic

@AlexMitan cool. Something you can instead have is objects subscribing to "I want damage events to red" or whatever, and being put in a list, then when a message is broadcast about damage events to red, only those objects are given the message.

Is optimisation of this project actually desirable or was it a throwaway? How far are you planning on going with it?

@doppelgreener Yeah, that's fair. I logged the "team kills" thing just to prove the flexibility of total broadcast: death messages get processed by the world, which takes the sender's parent, the fleet, and adds to its score

@doppelgreener I'm moving that simulation to SVG, and as soon as it runs decently I'm going to figure out the message architecture a bit deeper, if I need to. The canvas thing is a one-day throwaway

Under more reasonable parameters, 100 units max per side

@FreezePhoenix SVG also features "leaving things alone unless they need to be changed" and "timed animations instead of hand-interpolating attributes I barely know how to work with". Plus, d3 is magic

How is your pipe waiting for messages? Lowering the FPS didn't help

@AlexMitan cool, ok.

Noticeable pickup If I modify CMUtils

I noticed you never used a single async function

Async/await is pretty nice.

And is performant

SVG conversion started

@doppelgreener Got any good links for that or am I going to have to figure it out?

@AlexMitan Bit of both. Some people including me find it a little bit brain-bendy at first. First you'll want to know about promises because async/await uses them behind the scenes. Then maybe this guide will be helpful.

@AlexMitan whats up!

If you're already familiar with promises I can try to explain what async/await does.

greetings, @Duke!

Hey!

This is up!

Remember that board game I kept trying to design that slowed down a lot because I couldn't find a way to make really elegant and test fast? I'm iterating through making a web game based on it.. roughly

But now I'm more.. learning the fundamentals of all of this

i was just looking at that. really glad to see your still at it. you have some great ideas worth full exploration

you can get some very good advice on this chat--lot of experienced devs

I pushed another commit, now it's fully like the previous sim

With removal of both data and graphics on death

I'm super, super happy about this. Time to stress-test.

@AlexMitan I can't remember if I ever got your email, but hit me up at dukezhou108@gmail sometime if you think of it. (definitely would like to keep in touch)

Will do!

How's your game going? Saw that you post around here

Last I saw, you were working on an AI for it right?

It was probably a mistake to release when we did, before we were ready, but we got some good feedback and will start addressing some of the immediate issues in the upcoming release ~September. (Right now it's the problem of "too hard for some player, too easy for some players" ;)

But we likely won't have a fully "ready for primetime" version until the end of the year, best case

but I'm happy because we'll be implementing an initial "goodness" evaluation function for the AI, which should improve strength significantly

and, once we get our server working again, we should be able to do cross-platform PvP for any system we ultimately run on.

Warning, "never-made-a-game-AI-in-his-life" question incoming

Since your game has easily quantifiable state, have you guys used neural networks or something?

You could simulate your game and then classify how good an AI is by how often it wins against a random AI and other AIs

@AlexMitan that's really cool!! :)

@doppelgreener It'd seem like canvas might have been the problem after all

@FreezePhoenix I switched to 1000 units on SVG and it's alright

radical

We're avoiding NNs for now because we want the AIs to be able to function on lowest-common-denominator, non-networked mobile devices. So it's strictly a "strategy ladder" model for now, entirely based on heuristics ("GOFAI";).

Look-ahead will be next, but NNs still a way off.

Neural nets take a lot of resources to run?

I didn't think they did.

I mean, Susan on Mac in mid 90s used one.

to train they take a lot of resources

They take a while to train, not so much to execute

I don't get it

you're not training them on the client

you train them dev-side and ship them

or, you want the AI to learn from the player (like Susan did) so the training data is each game the user plays

also not a problem

Just upload the data from the games or something?

No, training a NN from a single match is very quick and simple

that's one of the cool things; training them on a dataset of a million doesn't take super long, though you wouldn't want to do that client side.

But adding the training from a single data point is just a "few" calculations.

easily done on a client

But that's only if you're interested in an AI that learns from the player.

eventually that is the goal

that can be fun, but it might not be what the player is looking for. It would mean the AI is useless to start with

Susan was easy as pie to beat, until it started learning. part of the fun of that game was watching it get better from game to game

to learn from their human player, and for humans to be able to have their personalized AI's play their friends personalized AI's

Or, you can ship with an AI that's partially trained

so it's not useless to start, just not great yet

but, initially, we want the reasoning to mirror human reasoning per the "strategy ladder"

I don't see how you know if it mirrors human reasoning

different people will approach the game different ways

maybe your own reasoning. But not "human reasoning" in general.

aah, but there are certain inherent qualities to the game that are "grokked" as a player advances

Do you know that for sure. For example, I learn Tekken combos by looking at button-press lists and practicing them

(7 year old can learn to beat those reliably)

My wife just mashes buttons and eventually learns it.

once learned, She can't even tell me what the button combos are.

so you can't say for sure what's going on in peoples' heads as they learn it

oh yeah, I've been analyzing this game from a solution perspective since ~2013

many people have different learning styles.

Some will play with intuition, without really knowing exactly what they're doing. others will calculate very carefully.

some will use a mix of both

and both play styles work, but there are certain characteristics to the mechanics that you can't get around, in particular, regional stability states (stability, epistability, metastability)

NN is the no-brainer for pure strength of play, but I'm fairly certain it won't be required to beat the strongest human players. Where I see NNs being interesting in this project is the personalization (as you've noted) and AI vs. AI in higher dimensional games, not suitable for human play

we also definitely want a string of "dumb AIs" where strategy is easily discernible

i might be wrong, but it will be interesting to find out!

@DukeZhou Also, the main point of the game is to work on a friendly coding interface kind of like Warcraft 3's to turn unit micromanagement into scripting

imagine that in an RTS you had units with triggers like "fire at most wounded enemy" or "when wounded, go behind squad" or "when on low health, go to nearest healer"

API for customizing the AI is definitely in the pipe-line, but our dev schedule is glacial atm. We're really interested in leveraging this for education, and what I hope for is that kids from middle-school on up will eventually be designing their own AIs to play, but ML and GOFAI

@DukeZhou GOFAI?

"Good Old Fashioned AI" aka "Classical AI" aka "Not true AI-since-the-recent-ML-milestones"

Ah, so rule-based, hand-written logic type stuff?

...kind of like what I hope my players will be writing ingame?

yup. imo it's still valid when approaching from an Algorithmic Combinatorial Game Theory perspective, and I think [M] is the strongest game for that, because all game states and outcomes are naturally ratios, and Sudoku itself led to difficulty classification in regard to puzzles in general

I was trying to get some of the math people to help me do a proof on a mirroring partial-solution for the even-order game grids 2x2(2x2), 4x4(4x4) etc., and then someone asked a question about the impartial Sudoku game and a PhD demonstrated the solution via "Infinite Sudoku"

(best part is, he didn't figure out the solution--his 11 year old daughter did--he just subsequently demonstrated the proof;)

what really gets me excited at present is how to distinguish between "weak stability" (regional polarity cannot be flipped with a single placement) vs. "strong stability" (regional polarity cannot be flipped with any number of placements) and how to efficiently gauge the latter

how to "attack up the resolution chain" (metastability)

how to determine when the game collapses into a tractable state so that perfect play can be achieved (does the AI agree to early resolution? Only if the outcome can no longer be affected.)

Where I see an advantage is, this generalize approach doesn't require re-training for every variation of the game mechanics and board topology. See Chaos Functions, "historical map generation"

@Tyyppi_77 but then, a twist!

(once I realized the game might be solvable to varying degrees, I needed to find efficient methods to generate unsolved variants)

@AlexMitan WHAT THE HECK IS TEH GRAPHICS