Cniles and Gotards can go seething.

>>106034180 (OP)
Name one important piece of software written in Rust.

>>106034196
troonix kernel (rusty parts)

>>106034180 (OP)
what does this supposed to represent? you really can't really beat c with rust for two reasons that I'm aware of currently atm
->language overhead for same code is never zero because rust's runtime unicode support or something like that as far as I remember
->some code is just not proveable so you either end up writing unsafe code (defeats all purpose of writing rust) or slower code than C

>>106034180 (OP)
Impressive. Very nice.
Now let's see C's card

>>106034316
the absolute fucking state of crabs

>>106034316
>now let's see [unsafe exploited riddled mess of a language]'s card
Why would anyone even bother with it?

>>106034298
>memory management
>less syntax errors
Rust is picking up steam because programmers aren't as good as they used to be.
There's too many people rushing into cs degrees like they used to rush into business degrees.
The tech market is saturated with mediocre employees who only want to collect a check and not improve as human beings.
C and C++ are objectively faster than Rust or any of the new bullshit coming out.
The fun part is when all the old programmers that got jobs doing what they liked die and younger less motivated people have to maintain all these systems written in COBOL, C, and C++.
lol desu. lmao even. quite possibly rofl.

>>106034180 (OP)
>systems programming language is faster than GC one
Wow it's nothing burger. They teach this shit on first year if CS.

>>106034298
Rust and C are equally fast in practice.
>language overhead for same code is never zero because rust's runtime unicode support or something like that as far as I remember
Rust uses C runtime(crt0 etc). Both C and Rust can run on bare metal without any runtime. You don't have to use utf-8 aware types in either too.
>some code is just not proveable so you either end up writing unsafe code (defeats all purpose of writing rust) or slower code than C
Unsafe doesn't defeat the purpose of writing Rust. The point of Rust is to write safe abstractions over unsafe code. That's why unsafe exists in the first place.

>>106034369
>I NEED MY PROGRAM TO BE EXTREMELY SAFE!
the world runs on C - most successful projects are C. imagine being this retarded. yes, rust safety guarantees are great in some areas MAYBE - or at least interesting. but the downsides are just extremely horrible.

>>106034298
Given rust has real ass inline asm and free standing support in the language and isn't hidden behind compiler specific features or flags, I'm calling this post bullshit cope. At best, c code has the advantage of gcc or clang (llvm) where rust is rustc (llvm).

You don't know what you're talking about because you're a coping baby duck bitch though so I don't expect an intelligent (You) from you.

>>106034196
sudo
https://github.com/trifectatechfoundation/sudo-rs

>>106034374
>Rust is picking up steam because programmers aren't as good as they used to be.
Please name any non-trivial programming project that never had any bugs. Feel free to pick any project from the beginning of computers.

>>106034402
>goalpost shift
oof big dog. might want to go back to the drawing board on that one.

>>106034420
What else were you implying by saying
>Rust is picking up steam because programmers aren't as good as they used to be.

>>106034180 (OP)
>language without an N:M scheduler and cycle resistant garbage collector has lower latency in both the average and maximum case
Is anyone really that surprised? Honestly?

>>106034477
crabs are fucking retarded

>>106034394
>rust has real ass inline asm and free standing support in the language and isn't hidden behind compiler specific features or flags,
NTA, but this is a really damn neat feature in my opinion. Being able to just detach from runtime, standard library and just go bare metal without any issues. C requires some compiler specific flags and it's not well defined. I don't think C++ even can do anything like that in a reliable way.
And something like #[no_core] is completely inconceivable in C/C++ world. I like how nearly entirety of Rust is declared in Rust itself, with only handful of hard coded language items. It makes Rust so easy to port and to understand really. It's one of things I hated about Ada. There, everything is a language feature, everything is magical and only explained in some random document somewhere.

>>106034234
So nothing of any value whatsoever...

>>106034389
>Unsafe doesn't defeat the purpose of writing Rust. The point of Rust is to write safe abstractions over unsafe code. That's why unsafe exists in the first place.
you can do this even in assembly. I can write a bounds checking array in C as a function returning pointer in a few lines of code. C++ has builtin safe pointers and even GC implementations. I don't know what Rust adds on top of these.
>safe abstractions
this is the most absurd thing I heard about Rust. lack of memory bugs don't make it safe, so it's misleading to call non-unsafe code safe in Rust. you prove safety the hard way by using actual theorems or automated theorem proving tools.
>>106034394
I stated my reasons in my post. I can't see you providing any backup against what you think you're claiming. At best, rust code is as fast as C.

>>106034429
I'm more concerned that you don't understand a pretty basic argument.
>C/C++ are better than Rust.
>>but anon, why are people using it
>memory management
>safer operation
>>Projects have bugs though!
>Rust is picking up steam because programmers aren't as good as they used to be.
>>I don't get your argument
If you had actually understood my first comment, you wouldn't be asking questions to my argument. You'd accept it or you wouldn't.
Rust exists because we have programmers that don't understand system processes or are too lazy to properly program. I'm guessing that the off loading to India is another indication that programming in the US is falling in quality. Every Indian professional - born and learned in India - that I have worked with has been objectively worse than the people they're replacing.

>>106034374
Cturds were never good. They've always produced walls upon walls of CVEs just trying to put strings together.
They were just lucky they grew up in an era when memory safety didn't matter.

>>106034180 (OP)
Cool. Still sticking with Odin™ thougheverbeit,

>>106034676
This. Boomers were always incompetent but they were just lucky enough to grow up in an era where their staggering level of incompetence didn't matter.

>>106034298
Yeah, also the code generated by LLVM is a little slower than GCC.

>>106034180 (OP)
>c data is hallucinated
this is your brain on rust.

>>106034676
>>106034736
>people who worked at IBM were retards
lmao honestly. you two are coping.

C is shit, rust is piss or maybe diarrhea, c is autistic dogshit though

>>106034648
cont.
I'm pro-'multiparadigm zero-cost systems language'. I don't want forced safety or anything. I don't want all the bells and whistles of Rust slowing down my runtime or compile time by default. You want safety, you use encapsulation over unsafe code which is something you can do in any language. You want 'actual' safety, you roll up your sleeves and start proving shit in your code. Saying rust code is safe is totally misleading.

>>106034821
>I'm not intelligent enough to learn advanced programming practices
>DoD contractors writting code for machines that cannot go down ever are wrong for doing it in C
This is such a weird cope from people that need to find better suited careers.

>>106034398
Do rustrannies ever write anything original or do they just transliterate the hard work of others into their freak language?

>>106034857
they rewrite, prove that rust is faster when your rewrite has 10% of the features and then everyone keeps using the Go version that's actively developed.

God it feels good being 23 and working somewhere that primarily uses C and Fortran

>>106034847
DARPA literally recommends rust over C and plans to transition legacy C code programs into rust, dumb cnile. So does the US freaking government recommend it over C
People used C for old code because there was nothing better, now there is

>>106034937
>DARPA literally recommends rust over C
I will ask about this later but I'm inclination is that we're not changing anything we're doing and you're *checks notes* full of shit again.

>>106034369
>unsafe
Skill issue

>>106034937
it's a cheap way to gain some safety. like low hanging fruit. doesn't mean that it's objectively a good idea. some people/organizations don't have to care about added compile times but for majority of developers that's not the case. any programming language forcing things down the throat is not good

>>106034648
>you can do this even in assembly. I can write a bounds checking array in C as a function returning pointer in a few lines of code
Well, no, not really. Rust safety depends on two things, runtime checks, and more importantly, static analysis. That's what borrow checker And type solver essentially are, static analysis tools. In ASM you can only really do runtime checks, since you do not have any way of declaring static information. And in order to meet Rust safety guarantees, you would need a ton of checks. I mean a lot, it would no longer look like ASM or C or Rust. It would essentially require you to write something like Java Virtual Machine.

>C++ has builtin safe pointers and even GC implementations. I don't know what Rust adds on top of these.
Guarantees. Rust has much stronger guarantees than C++.

>this is the most absurd thing I heard about Rust. lack of memory bugs don't make it safe, so it's misleading to call non-unsafe code safe in Rust
There is nothing absurd about it. Rust has some quite strong guarantees. Just like Java for example. Except unlike Java, Rust is not a managed language.

>you prove safety the hard way by using actual theorems or automated theorem proving tools.
That's not entirely true. For example Java is memory safe but it doesn't require very sophisticated external tools and theorem provers.

>I can't see you providing any backup against what you think you're claiming.
NTA but everything related to no_std, no_core and no_main is documented in rustonomicon. no_std and no_main should remove c runtime IIRC

>>106034937
DARPA can suck my cock.

>>106034298
>I am a retard with no clue about anything, here is my opinion
Who the fuck asked?

>>106034398
Wow, look how many files they have.

>>106034844
>I don't want forced safety or anything.
Then I hope you never touch C or C++, because they force type safety and stack safety among other things.

>>106034667
>you don't understand a pretty basic argument.
>>C/C++ are better than Rust.
That's not an argument. That's just your opinion.

>Rust exists because we have programmers that don't understand system processes or are too lazy to properly program.
False. In order to write in Rust you have to understand the same things as you typically learn when using systems programming languages. Rust doesn't make anything simpler, quite opposite. It makes things more complex because it formalizes everything explicitly.

Claiming that
>Rust is picking up steam because programmers aren't as good as they used to be.
Is on par with claiming that
>Static analysis and formal proofs are picking up steam because programmers aren't as good as they used to be.
Which is just naive and comes with misconception that good programmers naturally don't make mistakes. That's not true.

Good programmers do not make mistakes because they use tools that can detect these mistakes. That's how safety critical software is being made.

>>106034847
>>DoD contractors writting code for machines that cannot go down ever are wrong for doing it in C
>This is such a weird cope from people that need to find better suited careers.
They use tools and static analyzers that make C much more restrictive than Rust. They do not use idiomatic C.

>>106035241
>this shift
lol okay

>>106035294
Pardon?

>>106035130
you don't need a borrow checker to guarantee safe memory operations. actually go and try to learn what is formal safety and what are its methods are. You can prove things the hard way by actual therems and tools but that's not strictly necessary because "a function returning a pointer to bounds checking array" or a specific object with some specified invariants and post-conditions are enough for non safety critical programming. Rust looks like kids bike with training wheels. I say looks because I didn't yet deep dived into all features of that language.

>Guarantees. Rust has much stronger guarantees than C++.
>Rust has some quite strong guarantees.
no it doesn't. it just guarantees that some memory operations are safe as I said. Actual guarantees are done by formal verifications. There's a bad tendency of reductionist behavior among Rust coders that all unsafety in a code has to be memory related. They have their own respective tool Kani but they "lie" that it's for verifying unsafe code, again implying that non-unsafe code is safe.

>Just like Java for example
>doesn't require very sophisticated external tools and theorem provers
No, this is pure ignorance. Java doesn't have strong guarantees. GC/memory safety is not strong guarantees. There are JBMC from CPROVER framework that you can use to prove your Java code.

>>106035157
you have to be 18 to post here

>>106035189
>Then I hope you never touch C or C++, because they force type safety and stack safety among other things.
type safety is forced? you create your own abstract data types, or use the ones that's already available to you.

>>106034937
>DARPA literally recommends rust over C and plans to transition legacy C code programs into rust
not anymore they changed back to c after the russophobes were out of office

>>106034374
Programmers were never good, retard. There was not a single period in history where your lot of codemonkeys could be described as "competent".
Dijkstra was already writing articles about how all of you are retarded all the way back in the 60's and he did not change his opinion for the next 40 years he lived.
So yes, your kind needs to be babied around by a language since experience has shown your lot to be too stupid to do formal verification

>>106034857
The trannies don't really write anything useful, but at least big companies are incorporating rust into their software somewhat like Microsoft and Google

>>106035358
This. Dijkstra was right about everything and programmers in general are retarded at levels that few people can even comprehend. Especially the old C and UNIX programmers.

>>106035280
>He's not an engineer that has worked with DoD and is telling someone that has contracts with DoD what languages he saw
I'm not arguing with you. I am telling you. Systems used in DoD equipment primarily use C/C++. Python is also apparently a widely used for research. TRACTOR is being pushed, but it is not the transition you're claiming it is.
Go work with a contractor producing equipment for DoD contracts. You will be told to use C/C++. It was used during the development of NGAD program, it was used on the F35 program, it was used alongside ADA for the F22, it was used with ADA with the M1 Abrams, it was used with ADA for the Bradley. It's used heavily in the Navy. It's used in virtually every helicopter.
You just don't know what you're talking about.

>>106035358
>>106035381
>appeal to authority
oof. I think most engineers are retarded, it doesn't mean I'm going to argue that the systems we use are retarded. It's literally because engineering has the same problem programming has. Incompetent people looking for new technologies to compensate for their inadequacies.

>>106034180 (OP)
I taught myself rust because of your ridiculous blog posts and no one is actually hiring for it so I went back to doing side projects in C++.

>>106034196
typst
ripgrep

Just out of the top of my head but there are tons.

>>106035311
>you don't need a borrow checker to guarantee safe memory operations.
Of course. Java is memory safe and it doesn't have a borrow checker.

>Rust looks like kids bike with training wheels.
One could have said the same about any other static analysis tool. Yet, they are very important in developing safety critical software.

>no it doesn't.
It does. For example you can't mark which types are safe to be shared in C++ but you can with Rust. You can't mark the lifetime of a reference, you can't make sure there won't be a mutable access to some variable while another reference exits without runtime overhead, etc.

>Actual guarantees are done by formal verifications.
That's not true. Java is memory safe without any of that.

>Java doesn't have strong guarantees
Java is a memory safe language.

>type safety is forced?
Yeah. Except for varargs, C is a strongly typed language. It is automatically checked by the compiler and it will reject any code that for example tries to pass argument of wrong type to a function. Even if the idea is valid and you think the code should just work, the compiler will reject such program and you have to use some escape hatches to make it compile. Just like it is with borrow checker and Rust.

>>106034180 (OP)
I"m going to be 100% for real, i only hate rust because i'm too stupid to actually learn it. I know python, java, javascript, C, and lua but literally Rust is just on another level for me.

I'm thinking of dropping out from Computer Science because Rust genuinely filtered me and made me realize how much of a midwit i am.

>>106035316
>they changed back to c
[citation needed]

>>106035403
What. Are you trying to claim that DoD does not use static analysis or are you this bad at reading comprehension?

>>106035502
They teach Rust at your collage?

>>106035403
>projects from the 00s used c
WHAT? this is insane. why didn't they use rust???

the whole counterpoint to rust is
>erm im good at coding and i never make any mistakes
which is a very childish thing to say

>>106035548
>He doesn't know what block upgrades are
Do you have to wear a helmet outside due to being injury prone?

>>106035499
>safe to be shared
*Safe to be shared between threads

>>106035358
>natural language sucks for programming because people can't be specific and formalize knowledge into computer rules
>vibe coding on the rise

>>106035466
How are these important pieces of software. Where is all the Rust firmware, kernels and operating systems? C dominated in all these areas when it was Rust's age.

>>106034196
uv
Servo
SurrealDB
hyperfine
Zed
fish
embassy
tree-sitter

Don't forget to .unwrap() after killing yourself, brainlet.

>>106035758
None of these are important or widely used pieces of software.

>>106035788
>tree-sitter, servo
>not widely used
>not important
Ogey

>>106035807
No they are not. If they were deleted today nothing would change. Where are the kernels, drivers, firmware, operating systems and embedded devices in Rust? It's been over a decade, when is it all coming?

>>106034234
>>106034398
>>106035466
>>106035758
Absolutely none of these are important.

>>106034180 (OP)
Rustniggers suck my dick

>>106035894
How does Sundar Pichai's cock taste like, chromefag?

>>106035499
sorry for late replying.

I don't know why you basically ignored what I said, and continue misinformation. Memory safety is not safety. Java is not safe, and neither rust. no language is safe by default. memory bugs are a subset of bugs that could happen. and those bugs are already prevented by using best practices, modern idioms and features, so actually by being an actual CS person rather than someone who took udemy course on programming. so again, they are aiming for low hanging fruit with this 'training wheel' aproach. it's a band-aid.

>One could have said the same about any other static analysis tool.
you're comparing apples and oranges. no static analysis tool change your programming habit by being deeply embedded in workflow. I don't want to be having to type the keyword 'unsafe'. it's unacceptable.

>C is a strongly typed language. It is automatically checked by the compiler and it will reject any code that for example tries to pass argument of wrong type to a function.
I write my 'own' types. Also I don't even have to use any type. Just write under main. use var with type inference if absolutely necessary. comparing type safety to mandatory safe/unsafe keywording and borrow checking is also comparing apples to oranges.

>>106035546
No. I just tried to learn it on my own, and i realized how much of a midwit i am.

I did pretty fine in every programming class I took. But i'm just going to end up being a code monkey no matter what job i get, most likely going to get replaced by AI anyways.

>>106035499
>C is a strongly typed language
Yeah, except it's not, C is weakly typed language.

>>106034180 (OP)
Even then gojeets have to heavily skew the tests to beat Java.

>>106034720
Buy an ad, no one likes your memelang.

>>106036329
>Java, a JIT language
>faster on average but made to look like it's much slower
it got me

>>106036329
Java is only quick when the jit can kick in which is why it's so good in meme-marks. If your loop is under 4k iterations of the same shit then java isn't quick at all.

>compile rust project
>it just works
>compile c/c++ project
>need to edit the makefile and install old libraries so it doesnt error out

>>106036452
This is the major benefit of Rust over C/C++. The borrow checker etc is bullshit no one cares about. Not having to take part in header file autism and a easy to use build system is what makes it better.

>>106034374
>Rust is picking up steam because programmers aren't as good as they used to be.
That's why people use C and C++. Good programmers were the ones using (and inventing) better languages like Lisp, Ada, Pascal, Algol, COBOL, etc.

>The fun part is when all the old programmers that got jobs doing what they liked die and younger less motivated people have to maintain all these systems written in COBOL, C, and C++.
COBOL is a memory-safe, high-level language that was very innovative for its time. C and C++ are full of bugs and didn't do anything new. PL/I and Algol 68 were better languages and older. Comparing C and C++ to COBOL is an insult to COBOL.

>>106034847
>>DoD contractors writting code for machines that cannot go down ever are wrong for doing it in C
They used to use Ada, but they wanted to hire diversity hires (Russians) so they started using C.

>>106034937
>People used C for old code because there was nothing better, now there is
No, there were Ada, PL/I, Pascal, Modula, Algol, Lisp, JOVIAL, and a lot of other languages. "There was nothing better" is C shill propaganda and lies.

>>106036714
ADA 2012 is a thing.

>>106036744
C was used because it was simple, not better. Same as Fortran or Pascal.

>>106034495
How? They won. Maybe you are because you can't read a graph?

>>106035358
Nobody gives a shit about academic wankery like formal verification. Programmers are engineers, not scientists or philosophers. They should just solve problems and not prove shit.

>>106035757
>Where is all the Rust firmware, kernels and operating systems?
Pretty much all mainstream operating systems have some Rust in them.

>>106037044
Ok and windows has a bunch of C# / Javascript. Linux has a bunch of Python. All the important stuff is made in C or C++. By now in C's lifetime almost everything was being made in it. Why isn't this the case for Rust? If I were to guess it would be because Rust doesn't offer enough over C to justify moving away from the devil you know.

>>106036496
>Not having to take part in header file autism
if youre filtered by that you just shouldnt deal with os langs
its dead fucking easy
>but when its done wrong
then its done by retards and they should get publicly humiliated

how the absolute fuck can you make headers- complicated????

>>106036918
>talk about c
>c not present on the graph
fukken crabs seeing things, as per usual

>>106037275
Why can't your shitty compilers do 2 passes to save me a bunch of a work?

>>106034857
Sure, Stalwart is all-in-one email server and Rustdesk is remote control software, something that cniles never dared to create because it would get pwned constantly like sudo.

>>106037275
>shitty build systems and the clusterfuck of libraries are good, actually
delusional nocoder

>>106037363
>t.
>>106037310
why would you need 2 passes?
also:
more complexity is more code
you dont want retard ass features
unless youre a webshitter who cant be trusted with scissors, then yeah, you just NEED that interfaces

>>106037363
>>106037395
abt nocodeshitters
youre tethering extremely close to being just a vulgar troll btw
better post some actual knowledge

>>106037395
Header files are more complexity and more code. Compilers since the beginning have to know the types of everything declared in a program. You just have the compiler add that information to the object code of the program when you compile it. A lot of languages like Ada and Pascal did it decades ago. What's C's excuse for not being able to do that?

btw
to the rust community:
why cant you fukken police yourselves?
we do
C is a fucking trademark
obviously we have matters we deal with, internally

why cant you do the same?
why WONT you do the same?
is it because you know youre a passing fad, so you accept every bit of help you can get?

>>106037540
>Header files are more complexity and more code
no, header files are retarded
thats what they are
theyre retardedly fucking simple
its an impossibility to fumble header files

one has to actually try to fumble header files
>youre including pieces of stuff
>yoiu cannot invoke stuff that hasnt been included yet
thats it
thats fucking it
its the core rule of includes
wtf can ever be complicated about that?

>>106037540
or maybe youre a complete fucking retard autismo rabble garbo
and you need your hand to be held with a rigid framework

listen then
heres the solution
picrel

you do this
you absolutely dysfunctional gorilla
you need someone to order yout fucking work
lets go then
you do picrel
AND
you put all the external includes into base
which you then include in your defines
which you then include in your structs
which you then include in your globals
which ytou then include in your main header

this is retarded shit
this isnt even compsci, this is organization of work
this is shit humans did in the fucking paleolithic age my nigga

>>106034196
I reject your idea that the amount of important software written in a language is somehow a good metric of how good a language is.

>>106037678
Or you can just not do that because the compiler already has to know it so it can just save the information instead of throwing it away.

>>106037863
yeah
but not at the cost of having to deal with rust
--
yeah c is antiquated in many respects
i dont do idiomatic c, i use tons of compiler extensions
BUT
c is the best we got
and the competition didnt understand that to beat c you have to do c but better
not- something else

>>106034180 (OP)
>Graph with no context
>C not even included
>"Languages have latency"
/thread

>>106037891
>troll thread
c vs rust thread
and any thread is a good excuse for confrontation

>>106037540
>What's C's excuse for not being able to do that?
It's 60 years old.
What's Rust's excuse for still using unsafe{} in 2025 and also having a large number of soundness bugs?

>>106037891
also
>nuh dont get baited
thats how pol died

no, you make the baiter regret posting bait
like by turning bait into a propaganda thread for your product of choice lmao

you oldfags should retire already
youve got your chance and you fucked it all up

>>106037558
Based. Rustroons on suicide watch.

>>106037921
18 years of development btw
still kinda shoddy btw

wtf do crabs even expect?

>>106037945
¯\_(ツ)_/¯
they tried to do what picrel is
but in the most retarded way possible

>>106037921
>It's 60 years old.
Pascal and Algol 68 could already do that. "60 years old" isn't an excuse for any of the flaws in C because everything wrong with C was when they changed things from the traditional way. They replaced normal strings with null-terminated strings. They replaced arrays with pointers.

>>106037983
>alex, what is "legacy" for 500
also post baked bussy, crab
i wanna see some foodporn

>>106035859
https://github.com/microsoft/windows-drivers-rs
This is a library meant for efficient development of windows drivers in rust, besides all the previous windows rust development
Microsoft is a pragmatic moloch, they are often a few years late but they know when to ditch shit and so they're ditching sepples garbage

>>106037983
Microsoft also gimped the standards committee for several decades, but it does't matter anyway. The real question is why are Rustoids only concerned with one-upping an ancient and stagnant language, instead of competing the the actual competition? Zig is probably going to outpace Rust, because it's not built on a mountain of technical debt, masturbatory Safeness™ and false marketing.

>>106037540
>put includes in main.c only
>write .c files only
>one pass blazing fast compilation
ez

>>106035977
>Memory safety is not safety.
It is in the name.

>Java is not safe
Java is said to be memory safe.

>you're comparing apples and oranges.
Borrow checker is a kind of static analyzer.

>no static analysis tool change your programming habit
Lol, no. Static analyzers and automatic proofing requires you to follow very specific patterns to work. Look how limiting SPARK is for example.

>comparing type safety to mandatory safe/unsafe keywording and borrow checking is also comparing apples to oranges.
Both type safety and memory safety are kinds of safety that arise from specific guarantees of a language.

>>106037221
>By now in C's lifetime almost everything was being made in it. Why isn't this the case for Rust?
There have always been other languages used. I am not sure if at any point C was the most popular, but that's not the case for a long time.
The reason why C was so popular on start is because it could run on any shitware and there wasn't much competition. Nowadays, we have tons of languages for all kinds of use cases.

>>106038005
>>alex, what is "legacy" for 500
C changed things on purpose just so it could be "legacy" but the normal languages were still doing things the same way that you would do them if you were designing a good language now.

>>106037983
also
>they replaced arrays with pointers
no, thats autism
theres a nuance bw both bc a pointer is a reference when an array is a tangible meme-ory object
you fukken autismo dweeb
#include <stdio.h>
#include <stdlib.h>
#include <string.h>

struct s_str_a
{
size_t size;
char *text;
};

struct s_str_b
{
size_t size;
char text[];
};


int main(void)
{
char text[] = "this is a string";
struct s_str_a *str_a = malloc(sizeof(struct s_str_a));
struct s_str_b *str_b = malloc(sizeof(struct s_str_b) + sizeof(text));

str_a->text = text;
strcpy(str_b->text, text);

puts(str_a->text);
puts(str_b->text);


}


disassemble this, you filthy casual

>>106038035
c gave yo a finer control
control which is entirely lost on you

stick to toddlerlangs
fkn childproofed shit and such

>>106035758
No one uses these things. They are just hobby projects on github.

>>106038021
Why do zig drones always try to compare their toy language to actually successful projects? Weren't you guys shilling for Nim just a couple of years ago?

>>106038031
>It is in the name
C is also safe (it's in this sentence), so what advantages does Rust offer?

>>106038060
>so what advantages does Rust offer?
offset intellect to prosneedure so you can replace whites with jeets

>>106038057
Source?

>>106034196
the only important and prod ready rust software in 2025
https://github.com/buttplugio/buttplug

>>106034316
>>106034180 (OP)
>>106034389
reminder that Rust programmers FAKE BENCHMARKS to make it seem like their versions run faster than what they are replacing. you cannot trust benchmarks from rust developers EVER

>>106037291
It's not present because only retarded faggots write non trivial network code in C. Reminder the Linux kernel pretty much developes a new CVE with every feature for a reason: because C is shit.

>>106038107
reminder that all rust produced code is compiled with the equivalent of -march=native which activates vectroization
AND FUKKEN C BEATS RUST WITHOUT IT
FUKKEN LAMAO

>>106038127
>linoos is a benchmark of quality
lol
lmao even

>>106038036
>This code is going to blow his mind
>beginner-tier-knowledge.c
lmfao, I wish I was still this easily impressed with myself.

>>106038107
Why don't Cniles have their own version of hyperfine yet? :^)

>>106038151
>demonstration code to make a point
>thinks this is prod

errr
anon, i...

>>106038156
Cope.

>>106038165
cope abt what
that im smarter than you?

i mean
if you list your definitions i might understand you better...

>>106038144
Linux is *the* premier C codebase.
Microsoft pushed people to .NET because people can't write C or C++.
Linus wants Rust in the kernel because people can't write C. A lot of Linux/Unix rewrites are edging towards Rust because C is just bad.

It's over. Cope and seethe like a dumb shitter.

>>106038177
>U-uhhh Ummm I was shitposting
You're cooked lil nigga. Leave the thread.

>>106038197
>A lot of Linux/Unix rewrites are edging towards Rust because C is just bad.
Sure, but that doesn't make Rust good (it isn't btw).

>>106038021
>Microsoft also gimped the standards committee for several decades, but it does't matter anyway.
Not really. C people hate anything that makes C better. Any potential improvements to the C standard went into C++ because C people complained too much. And Microsoft didn't really care about C because they were making a C++ compiler, so making C compilers easier by gimping C wouldn't benefit Microsoft.

>The real question is why are Rustoids only concerned with one-upping an ancient and stagnant language
Because most Rust programmers came from C and C++. Rust was designed by C++ programmers from Mozilla who have decades of C++ experience. Most of the people impressed by Rust are experts in C and C++ who are amazed by how easy it is to do things in Rust. This is what you C shills don't get.

>>106038197
>Linux is *the* premier C codebase.
>others suck therefore you do too
loser mentality
and so you promote rust to create a narrative where you dont pass for a c omplete retard

>>106037540
>>106037983
Critiquing c for not copy pasting every other hll is idiocy. c is supposed to be close to assembly. Of course if has pointers not arrays, how are you supposed to shove an array into a register? If you want pascal then go ahead and use it. People used and use c because it's not pascal.

>>106038207
Rust is literally the least shit language currently out there. Keep coping though. You literally are low IQ and below midwit even.

>>106038237
>Rust is literally the least shit language currently out there.
except its perfectly worthless and the industry said as much

>>106034369
it just works

>>106034392
>but the downsides are just extremely horrible.
What are the downsides? I know nothing about Rust btw

>>106034369
>>106038247
also its self documenting (nta)
unless its branchless
if its branchless then you better comment or you will end up in the fires of programmer hell

>>106038209
At least M$ finally realized redistributing the same Visual C++ runtime a 1000 times is bad and gave us a crt/libc that at least meets C99 standards. That's probably the best thing M$ has done recently. MinGW wrapper around msvcrt was so bad.

>>106038251
its retarded on so many levels
the language creates problems which it then solves by creating other problems

its not without reason that this travesty of a language gets developed since 18 years already
THATS FUKKEN 1/2 OF THE AGE OF C++ FOR REFERENCE

>>106038282
>the language creates problems which it then solves by creating other problems
You're describing the whole history of C and C++, except with C they just create new problems without solving the existing problems.

>>106034857
>build a house out of poop
>house is good enough, but smells like poop and riddled with bugs
>eventually find a better building material
>start replacing the poop parts of the house with clay parts
>"Why do claytrannies keep rebuilding things that were already made with poop!??!?!?"

if it's not organic then it's not gonna last
once the funding dries up then it has nothing, no sponsors and no interest

>>106038209
>C people
Not sure if you can actually comprehend this, but there's no such thing as a "C person". I use C, but I'm not a "C person", nor do I think it's a good language in 2025. It's just a pragmatic tool for certain type of job. Rustoids actually are a thing though, which is bad news for Rust in the long term, because they'll inevitably move on, in order to satisfy their language gooning needs.

>>106038344
Another great example of this is rg and fd.
grep and find are smelly as fuck in comparison.
Hell, uutils works on Windows and Linux. Where are my gnu userland on Windows? Buggy ass mingw shit? lol, lmao.

>>106038237
Your bait would appear to indicate lackwit status.

>>106038155
time for basic shit and valgrind for actually determining how to speed up your program. both are written in C.

hyperfine is a weird middle ground where you just want to show off how fast something is without much productive value.

>>106038312
yeah i got no love for c++ and i dont care abt anything after c99
its all pursuit of muh safety anyways

in fact
even posix fukken c is a pursuit of corpo interests
the fucking retarded null terminated strings
the fucking unoptimized libc against the hardware it operates on
where do you think it comes from?

fukken corpo
they tried to turn c into the language that does it all and failed
and kept trying
and meanwhile they did sepples
do do the same
and failed
and kept trying
and thats how we got rust
which oh, god, failed so fukken hard
because globohomo got greedy this time and wanted a fukken programming language to be a vehicle for politics

i dont care about that
i do c 99 + extensions bc its a tool thats fits to my hand, and is adequate for the job
i dont need your bells and whistles bc i got excellent visualization capabilities
and im not gonna ruin my brain by min maxing into data retention like a fucking retard lobotomite

>>106038197
My new job is windows kernel work (it pays the bills) and it's almost all C/C++.

>>106038378
>valgrind for actually determining how to speed up your program
that's not what valgrind does at all

>>106037736
Cope, dilate.

>>106038391
>the fucking retarded null terminated strings
>the fucking unoptimized libc against the hardware it operates on
>where do you think it comes from?
Dennis Ritchie and the original C creators.

>>106038368
Name a better language. You can't. Rust is high level enough to do basically everything, while low level enough to drive shitty microcontrollers. I literally have never seen any language like it and you're coping if you disagree.

>>106034196
NVIDIA GPU driver NOVA

>>106038391
>i dont care abt anything after c99
#embed, _BitInt(), typeof() and <stdbit.h> don't interest you? The last one is an improvement, even if you were using GCC builtins before.

>>106038430
>If I double down on the point he contested, he'll probably get really mad this time and then I might actually have sex.
lmao.

>>106038443
You forgot basic shit like checked arithmetic functions too. C cucks are finally going to be safe from basic over/underflow bugs.

>>106038430
>calling anything that comes out of mozilla a language
>unironically calling it best
lol

>>106038400
valgrind/callgrind/perf (C) and flamegraph (perl) are commonly used for profiling

>>106034196
CloudFlare reverse proxy. Rust basically runs the internet.

>>106038465
>You forgot basic shit like checked arithmetic functions
No I don't. He explicitly said he didn't care about those, if you bothered to read more than the first 5 words.

>>106038464
>still hasn't named a better language.
kek.
I can make a video game in Rust
I can make a webserver in Rust
I can make a MIDI keyboard on a rpi pico in Rust

What other language is as easy to use and versatile? You can't name one and you know it.

>>106038429
no yeah
i mean
>why the fuck in tyotl 2025 do we still use null termination as unavoidable default
the whole porevious picrel was only to create a guarantee that
a) when i look through the files in a folder everything happens on the stack
b) i have 8 nullbytes at the end of directory entities so that i wont overflow bc i parse text 8 bytes at a time

picrel is how *I* deal with strings
its a solution for short-ish strings, if i were to go above ~32 chars expected len i would have picked a different solution

and thats the thing with C
these kinds of operation are the natural level of operations for c so to speak
literally a macro asm

>>106038481
Wasn't that written in Go? Or am I confusing it with something else?

>its a solution for short-ish strings, if i were to go above ~32 chars expected len i would have picked a different solution
the only difference is that i would have cast that into the appropriate SIMD type, ackshually
im justifyig my laziness i guess

>>106038465
>I need a standard function, otherwise it's impossible to check for arithmetic overflow
Interesting.
>"underflow" is definitely a real thing, when dealing with integers and modular arithmetic
Interesting.

>>106038495
>What other language is as easy to use and versatile?
C

>>106034180 (OP)
Rust is really good for low latency API servers.
https://maxday.github.io/lambda-perf/

>>106038542
>easy to use
>C
Mental illness comparable to a tranny. It's hilarious.

>>106038555
>big corpo things

>>106038514
Nah it's Rust

>>106038574
I was thinking about cloudflared

>>106038564
c is dead easy to use
thats why its used as a didactic tool in schools

its just that its difficult to master

>>106038209
>Not really. C people hate anything that makes C better. Any potential improvements to the C standard went into C++ because C people complained too much. And Microsoft didn't really care about C because they were making a C++ compiler, so making C compilers easier by gimping C wouldn't benefit Microsoft.
You must be a zoomer who doesn't get checked on bullshit very often (probably because you don't go outside).

>>106038569
API server is not exclusive to corporations. You can do it today.

>>106038587
I can only get so hard.

>>106038585
https://github.com/cloudflare/pingora

>>106038589
>You must be a zoomer who doesn't get checked on bullshit very often (probably because you don't go outside).
kek
zoom zoom doesnt know c-ultists are anarchists with the most varied backgrounds

this transcends g
c community transcends this board
and its hilarious how much the language is as much a filter as a magnet

>>106038555
Just woke up, Bruce?

>>106038587
>C is dead easy to use
Ok, so build a binary that runs on a rpi pico that parses midi frames. In rust, I just include the appropriate target with rustup, add the dependencies in Cargo.toml and move on.

You're literally delusional or have never coded in your life. Hell I bet you have never even made a basic bitch website using C.

>>106038555
>No C
Man, c must be real shit if it isn't represented anywhere in these meme benchmarks...

>>106038602
y doe?
if not for portfolio purposes...
i mean
i just dont have a reason to serve

>>106038624
I wake up at 5 usually. Today was no difference.

>>106038632
yeah its more hassle to get the framework
100% worth it bc your got ROI whenever you write a line of code

im njot learning the rust bullshit, bro
fix your fukken language.

>>106038481
Rust ruins the internet, you mean.

>>106038617
Thanks, seems like they're slowly moving away from Go to Rust. Good choice.

>>106038632
Based and embedded pilled.
t. rp2350 enthusiast

>>106038659
>its more hassle
So it's not easy to use. So Rust won, by a lot. So what language is better than Rust then? It isn't C.

>>106038674
>Based and embedded pilled.
>t. rp2350 enthusiast
post projects

>>106038564
Are you serious? What is difficult about C?

>>106038681
Tied to my real name, fuck off glowie.

>>106038676
one time hassle vs fuck in the arse whenever you wanna do the simplest things
benefit vs cost analysis:
i go with one time cost

coldest take ever
didnt even include the tranies question

>>106038644
There's no use case for C outside kernel and drivers and thus nobody writes their servers in it. The other poster has a good point, Rust is very versatile.
>>106038649
You can serve your own clients. For example back in uni days I had my own Todo list server that would sync between my phone, desktop and the server,

Man... Still no good Todo apps to this day :(

>>106038701
Not writing insecure code

>>106038707
>Man... Still no good Todo apps to this day :(
write your own anon
god gave you the potential
its up to you to fill up the skills

>>106038060
>so what advantages does Rust offer?
Stronger guarantees

Feels good being a LISPchad

>>106038107
Is the benchmark game also faked?

>>106038701
>toolchain autism
>no easy dependency management
>no standard build system or even compiler arguments
>no language features for basic shit like generic containers, vectors, strings or even fixed-size arrays that don't decay
>have to do a ritual to make compilers not insert libc calls like memcpy
>no standardized inline asm
>no good IDE tooling outside of clangd (which has other problems with gcc compiler flags + some preprocessor macros)
>no modules or even namespacing
Let's not forget that GCC is so shit you basically need 100 copies of it for every target triple you want to target. At least clang isn't THAT bad, so I can give you that.

>>106038714
Rust has the exact same problems if you want to write non trivial applications.

>>106038797
>R-rust has the same problems even though the unsafe line count is in the hundreds vs every line in C!
Giga cope.

>>106038797
NTA.
Rust safe/unsafe semantics make it much easier to avoid a lot of common pitfalls in C/C++.

>>106038791
I didn't ask about your inability to use the tooling around the language, I asked about the language itself, which you said was difficult.

>>106038707
>>106038725
(cont)
also:
>theres no usecdase for c
whenevert you need control c is a good choice.
if you know c already bc theres a ton of implicit-ish things in the dialect

but also
gpgpu ic C
if you wanna get into gpgpu
you better know C and especially stack based programs
becasue your will be working with pre-defined buffers which means offsets and meme-ory control
so cs 101 bullshit of indices and arrays, only pushed to the extreme, and asynchronous-ish

i mess around with opencl
cuda would be zero problem for me
but you better know c mechanics very well before you get into gpgpu bc its the same, only harder, kek

>>106038797
Even unsafe{} code is borrow-checked. What are you talking about? Please give me an example. Just one.

>>106038812
>unsafe line count is in the hundreds
If a GPU driver has only like 100 unsafe lines, most of them trivial one liners, then your typical large project is not going to end up in the hundreds.

>>106038816
>goalpost moving this hard
A language without tooling is delegated to the meme NIH DSLs for your random shitcode project. You lost if you're going to claim tooling isn't a factor.

>>106038824
>gpgpu
It's over for you c tranny.
https://rust-gpu.github.io/blog/2025/07/25/rust-on-every-gpu/

>>106038783
They are using a lot SIMD intrinsics, so yes.

>>106038874
>They are using a lot SIMD intrinsics, so yes.
And C doesn't?

>>106038863
>translation layer
you will always be the bottom end of the chain

>>106038831
Notice how the Cnile recoils, defeated and hurt. "I've been found out"!

>>106034507
>C requires some compiler specific flags and it's not well defined.
It's not a big deal. You don't switch between compilers frequently when working with freestanding code.

>>106038881
no
if you dont enable -march=native
or your target architecture
then it doesnt

i find it funny that all the benchmarks i saw
with rust vs c
not one enabled simd for c

rust commune is jubilating that the beat c
EXTREMELY OCCASIONALLY

BUT C WAS FIGHTING ONE HAND BEHIND ITS BACK ALL THE TIME

>>106038076
There are almost no jeets doing rust, as opposed to go.

>>106038881
>>106038905
AND ITS NOT EVEN OPTIMIZED CODE BECAUSE THE LIBC IS UTTER FUCKING TRASH (bc portabilityshit. i can explain in detail if you care)

>>106038882
Are you literally retarded? Serious question. How is Vulkan Compute, OpenCL or CUDA in C or C++ not a "translation layer" as well? Do you even know what SPIR-V is?

>>106038060
I don't even use rust for the safety features, it just is a great language overall, the safety features are the cherry on top but not the reason i use it.

>>106038881
No, the C code in those bechmarks is less optimized because we don't have a tranny army working for free.

>>106038905
You're so fucking retarded, I hope you die of aids, nigger

https://benchmarksgame-team.pages.debian.net/benchmarksgame/program/fannkuchredux-clang-6.html

This is just a random C one. Notice something dipshit?

>>106038914
yeah but autists are more retarded than jeets
who can the least can the most
or something to that effect

>>106038917
translation layer to a translation layer then
lolle
autism is a hard handicap

>>106038905
>no
https://benchmarksgame-team.pages.debian.net/benchmarksgame/program/fannkuchredux-gcc-6.html
https://benchmarksgame-team.pages.debian.net/benchmarksgame/program/nbody-gcc-9.html
https://benchmarksgame-team.pages.debian.net/benchmarksgame/program/spectralnorm-gcc-6.html
Why do cniles always lie. Why? It takes 10s to verify your claim and prove you wrong.
Why do you think this kind of jewish trickery will ever work on an anonymous imageboard?

>>106038901
Yeah, it's not a big deal. Still, having first class support for this is nice.

>>106038920
>No
Why are cniles always lying?
>>106038932 >>106038947

>>106038947
>-march=ivybridge
ok
but it satisfies only half the conditions for an equal fight
and it also invalidates the results bc theyre truncated
but whatever

>>106038947
also
>>says "jewish tricks"
>>while seated firmly on jewish cock
> mfw

Mental illness

>>106038898

>>106039008

>>106038555
C# bros....

>>106039061
I think dotnet10 is pretty good. They made some improvements to the AOT. Sadly faggots will be stuck with 8 for some time.
Never was a fan of C#, but it seems like M$ is better than Oracle's gay bullshit.

>>106039008
Notice how the Cnile is incapable of actual conversation, how his glowing hands are only capable of posting sub-90 IQ memes to try and convince other Cniles to keep using their ancestral footgun language. He makes bold claims and then refuses to engage in the resulting conversation. Even Gojeets have more integrity and are comfortable with admitting the flaws in their own programming language of choice.

>>106038831
Is that supposed to be a positive? Unsafe{} code is wickedly unsafe compared to regular C/C++ since you have a much tighter contract that you must uphold. Safe code that follows Unsafe{} has lots of assumptions, violate any one of those and your program is in mega undefined behavior territory. And if you dont need to use Unsafe{} code, why are you even writing in a language like Rust? Just use Java/C#/Go and be 10x more productive.

>>106039137
Seethe more; your anger fuels my joy.
We've had this thread so many times that it's become quite clear that nobody's making any new points or persuading anyone. Therefore there's no point in engaging with any anything that's pretending to be an intellectual point, as it is really just yet another rerun of the same damn shit.

>>106038824
Eh I'm too lazy to set up and maintain a public server and hosting/deploying the app. Way too much effort.
>>106038824
>gpgpu ic C
Well Mesa is now including opencl. Their opencl implementation is written in Rust.
https://gitlab.freedesktop.org/mesa/mesa/-/tree/main/src/gallium/frontends/rusticl
I wonder what you can do with C that you can't do in Rust. Rust has a good understanding of x86 platforms since it has its native intrinsics built-into the core/stdlib
use std::arch::x86_64::*;

// SSE
let a = _mm_set1_epi32(5);
let b = _mm_set1_epi32(10);
let result = _mm_add_epi32(a, b);
// AVX
let avx_vec = _mm256_set1_ps(3.14);
// Runtime CPU feature detection
if is_x86_feature_detected!("sse3") {
// Use SSE3 instructions
}
if is_x86_feature_detected!("avx2") {
// Use AVX2 instructions
}
unsafe fn bmi_examples() {
let x = 0b11010000u32;

// ANDN: AND NOT - clear bits set in y from x
let y = 0b11110000u32;
let result = _andn_u32(y, x); // 0b00000000

// BLSR: Reset lowest set bit
let reset_lowest = _blsr_u32(x); // 0b11000000
}
unsafe fn popcount_example() {
let x = 0b11010101u32;

// Count number of set bits
let count = _popcnt32(x); // Returns 5

// For 64-bit values
let y = 0xFFFFFFFFu64;
let count64 = _popcnt64(y); // Returns 32
}
unsafe fn fma_example() {
let a = _mm256_set1_ps(2.0); // [2.0, 2.0, 2.0, 2.0, 2.0, 2.0, 2.0, 2.0]
let b = _mm256_set1_ps(3.0); // [3.0, 3.0, 3.0, 3.0, 3.0, 3.0, 3.0, 3.0]
let c = _mm256_set1_ps(1.0); // [1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0]

// Compute (a * b) + c in single operation
let result = _mm256_fmadd_ps(a, b, c);
}
fn optimized_popcount(x: u32) -> u32 {
if is_x86_feature_detected!("popcnt") {
unsafe { _popcnt32(x) }
} else {
x.count_ones() // Fallback to software implementation
}
}

All of these are built-in. You don't lose any control at all as you can always use inline asm

>>106039259
>Well Mesa is now including opencl.
>something something written in rust
who cares
its opencl ****C****
not
opencl ****rust****
its would be honestly misguided to try to port rust onto gpgpu
its bc of all the other things that make a barrier to entry
and rust being a complex language, the industry just aint gonna bother
who c an do heterogenous architectures can deal with C(ish)
and a C(ish) compiler for a custom architecture is gonna be a fraction the effort of porting whole of rust

i mean
you could use reduced rust
but in effect youre gonna be working with c

>>106039211
NTA, but it still is better to have few critical sections where you pay a lot of attention than to have your entire codebase be the surface for potential bugs. Good Rust code has safety comments in every unsafe block/function which state all the assumption and valid use which further makes things easier to keep sound.

>And if you dont need to use Unsafe{} code, why are you even writing in a language like Rust? Just use Java/C#/Go and be 10x more productive.
Well, Java doesn't run well on my esp32.

>>106038095
lmao

>>106039259
>>106039289
>reduced rust
idk if you know
but opencl C and cuda
*is reduced c*
or it should be
i do opencl 1.2 so theres stuff i dont know
but youre not gonna be using even a pointer in opencl
its all offsets relative to your buffer and your work item

>>106034298
>language overhead for same code is never zero because rust's runtime unicode support or something like that as far as I remember
Overhead from unicode is only for string parsing, and if you know definitively that you absolutely will not encounter unicode characters, you can always work with u8 slices instead.The things that are really going to cause overhead are going to be accessing arrays and vectors in random order, because Rust has bounds checking (many other languages do this too though, and there's tricks to avoid it).
>some code is just not proveable so you either end up writing unsafe code (defeats all purpose of writing rust) or slower code than C
Unsafe isn't a "one drop pollutes the whole batch" type of thing. If it was, all Rust code would be polluted on the basis that printing to stdout requires making a system call, which requires calling code outside of Rust, which is unsafe. Rather, you should think about all questions of safety in terms of probability. If 90% of my codebase is safe Rust, that means that memory-related errors can only arise from 10% of my codebase. I am 10 times less likely to shoot myself in the foot, and if I do happen to do so, I will find the bug much quicker.

>>106034898
GEG

>>106039289
You mean the FFI? Yeah sure, but actual mesa implementation is written in Rust and exposed as a C API, sure. But many different language has bindings to it. You can consume the C API,
I'm only saying that the most used opencl implementation on Linux is written in Rust.
>>106039333
>1.2
That's really old, rusticl targets opencl 3.0.

>>106039259
>Rust has a good understanding of x86 platforms since it has its native intrinsics built-into the core/stdlib
Niggerlicious.

>>106039472
>But many different language has bindings to it.
c is still the most efficient because its the raw-est we got

>rusticl targets opencl 3.0.
i-its what i work with, okay?
more seriously- i use my gpu to do mostly binary ops so the version doesnt really matter

>more seriously- i use my gpu to do mostly binary ops so the version doesnt really matter
but i still notify my level of knowledge
opencl c 1.2: pointers have a runtime cost
dont use em. not even supported everywhere iirc

>pointers
a no
part of the standard it seems
anyways, you dont wanna be using pointers
thats additional reads

in fact
id rather emulate a pointer than rely on the framework to provide one
granted, im a grug
but you also have 3 pools of meme-ory:
global
local
and private
in cuda and opencl
each with different properties
and the pointer is gonna be an offset anyways
so id rather have control over where its stored, depending on its design-dictated availability

ev&doe id expect that to be dealt with by the framework
but i have trust issues so whenever theres something i really want done, i do manually, myself

>>106034369
???? you are making a latency COMPARISON
where is C?

>>106039944
>reading graphs failure
>common crab L
>something something they dont see colors or something

>>106038824
>gpgpu ic C
Big tech's GPGPU languages look like C, but work completely differently, the same way JavaScript looks like C but is more like Lisp. C people have such extreme baby duck syndrome that they complain and refuse to use a language unless the language looks like C.

>but you better know c mechanics very well before you get into gpgpu bc its the same, only harder, kek
It's harder because they're trying to shove C syntax where it doesn't belong. GPGPU is very simple mathematically. It's mostly array processing, so of course a language that makes you pretend arrays are pointers is going to suck for that.

>>106040009
>Big tech's GPGPU languages look like C, but work completely differently,
yeah yeah i know
kinda said it with the asynchornous part and the 3 different mem pools
>baby duck
ok so it isnt worth reading the rest
tx for the heads up

btw
have you seen those buyceps?

>>106038920
>No, the C code in those bechmarks is less optimized because we don't have a tranny army working for free.
C programmers have a tranny army working for free and a paid tranny army.

>>106040097
>no, u
that train sailed long ago

>>106040113
tbf, if you need an object-oriented language, C++ is pretty banger.

>>106039008
>heh look at me I soooo not care
>>106039041
>>106040030
>>106040113

>>106039008
Ironic. All you think about is Rust. Your picture is reversed. If you Ctrannies weren't constantly shitting up EVERY thread about EVERY language here, I wouldn't think about C at all. Maybe I'd think about C when another C CVE comes out that every language since 1960 BUT C would have prevented.