KWAB

>>105654867 (OP)
>R (73.16s)
>Python (74.42s)
Dios mio...

>>105654867 (OP)
what do the colored balls represent

>>105654867 (OP)
>JVM 8% slower than C
Why in the world would anyone bother with pajeet language at this point?
I'm talking about C of course
Official Cnile bait post

>>105654867 (OP)
what about assembly?

>>105654867 (OP)
And yet python is the superior choice

>>105654867 (OP)
This shouldn't surprise anyone? Python is an interpreted language, not a compiled one, so of course it's going to be slower because of the interpretation overhead. You don't use Python for speed, you use it because it's a braindead language that simplifies a lot of shit.

>>105655521
Assembly languages could theoretically be the fastest but it depends on how skilled the assembly programmer is and if they can use all the little register/stack tricks and weird bitwise/math wizardry. To give an example of what I mean by that is that some processors don't have bare-metal instructions for things like multiplication or division (although most assembly editors have macros for that kind of thing) and if the programmer doesn't use bitwise operators to perform multiplication or division, and if they don't use hacky math wizardry either, then you might be looking at unironic O(n) time for something as simple as multiplication through incremental addition. Most modern instruction sets have multiplication though and assembly programmers hopefully know how to do math with bitwise operations.
I don't think most assembly programmers could write code that is faster than what the C/C++ compiler writes though. Those have had decades of work put into them to make the most optimized asm code.

>>105655694
Java is also interpreted, tranny

>>105655718
Java uses bytecode though, which isn't the same as a scripting language like Python. It's basically compiled into it's own assembly language that runs on the Java Virtual Machine (JVM), which is why it's faster than pure scripting languages like Python. You are a retarded nigger.

>>105655744
Least retarded anime watching tranny shitskin
https://docs.python.org/3/library/dis.html

>>105655752
then why cant python be as fast as java?

>>105655784
Because the JVM is an amazing piece of software.

>>105655718
No it's not.

>>105655784
Believe it or not but some languages are just inherently bad.

>>105655784
It might be possible, it just hasn't been done yet. CPython's design is infamously outdated even though Python is widely used.

>>105655784
Python supposed to be for teaching people, hence the cpython implementation is supposed to be sufficiently simple that you could study it to learn c programming.

I'm not joking.

>>105655846
What does the JVM do with bytecode?
https://www.cs.cornell.edu/courses/cs2112/2015fa/recitations/13jvm/

>>105655890
Interprets it once, profiles then compiles it with a JIT, that's why neither Java nor C# are true interpreted languages or fully compiled. With AOT you're getting closer to compiled languages doughbeit.

>>105655890
>What does the JVM do with bytecode?
The JIT compiler converts the bytecode into corresponding machine code to be used instead.

>>105654867 (OP)
impressive, very nice
lets see fortran score

Retard vibe/no coder here
What other languages let me just run the scripts without compiling or fucking around

>>105654867 (OP)
Stop posting this shit.

>>105654867 (OP)
luajit beats them all but c

>>105655784
Python doesn't ahead of time compile it's bytecode like JVM langs do, also it's VM is utter garbage.This billion nested loop is a total b/s benchmark. In theory the JVM langs should be faster than C/Rust here because if the JIT/VM was any good it would realize it's doing no work and just return 0 instead of executing the program beyond how long it takes the runtime to realize the program does nothing but instead it seems it's taking 0.4s to compile the loops to native and then executing them.

>>105656011
>Python doesn't ahead of time compile it's bytecode like JVM langs do
It has since 3.13.

>>105655954
Why are you so adverse to compilation? The compiler can be your friend if you let it be. It can point out mistakes for you before they would be caught by an interpreter.
Anyways, Emacs Lisp and Perl are some other scripting languages not in that image. Emacs Lisp will send you down the Emacs rabbit hole though and you'll come out as a changed man.

>>105655784
>>105656011
Believe it or not but >>105655855 is right, Fundamentally Python's design makes is nearly impossible to achieve anywhere close to Java's performance regardless of technology, types in particular are a huge fucking problem.

>>105655136
It's a screen cap from an animation where the balls slide back and forth across the screen at a speed relative to the speed of the language at that task.

>>105655937
>fortran (1ms)
>bouncing ball is a blur

>>105654867 (OP)
>Go slower than Java
>Deno slower than Node
>PHP much faster than Ruby
those surprised me the most

>>105656042
He's not talking about jitting. He's talking about having a compile to bytecode step separate from an execute step. He's mostly wrong, since python generates .pycs.

>>105656110
java and python are both dynamic languages. Both need a jit in order to get good performance. Python has conventionally not had a jit, whereas java has. If you write a good jit for python (like pypy) it will be competitive with java.

>>105656073
>Why are you so adverse to compilation?
It takes way too long.
I would prefer it to fail quickly and paste over the debug error/print statements

>>105654867 (OP)
remember when rusttrannies used to be rubytrannies? I remember

>>105654867 (OP)
I fucking hate this image and its video so much. Wtf are the billion loops doing? If they're doing nothing the compiler will just optimize them away. actual dogshit "benchmark"

>>105656110
>types in particular are a huge fucking problem.
This is not necessarily true. Lisp literally invented dynamic typing, but SBCL is well-known for holding its own against the JVM.
Making a fast Python would not be impossible, it would just be a massive pain in the ass.

>>105656517
>I fucking hate this image and its video so much. Wtf are the billion loops doing? If they're doing nothing the compiler will just optimize them away. actual dogshit "benchmark"
t. python dev

>>105656073
Because I'm more important than the users of my code. The time saved by not needed to deal with compilation is saved by me. That's my time. If things take longer for the users, well, they should have learned to code themselves. Their time is worth far less than my time is worth.

>>105655855
>>105656110
samefag

>>105656601
Go complies 100k lines a second. For most projects its going to compile and start faster than it takes your shitty vm to cough into life. Compiled languages won.

>>105654867 (OP)
Im still just gonna use python. I'm just gonna.

>>105654867 (OP)
you should build 1000% faster ASICs only for useless loop fucking tsukuyomi

>>105656601
You'd save even more time by not writing software you have no intention of using

>>105654867 (OP)
what does C in C mean?

>>105654867 (OP)
dishonest assessment different programming languages are designed for different purposes
a screwdriver cannot be used to hammer

>>105657488
The C doesn't mean anything. It's just named C because it's the successor to the 'B' programming language. This sounds made-up and silly but it's true.

>>105657329
Bossman won't give me a paycheck that way.

>>105657538
And 'B' likely was due to it being developed by Bell Labs though some sources say it might be due to 'B' being inspired by BCPL (Basic Combined Programming Language).

>>105657667
>having a job

>>105656589
I avoid python like the plague. I'm more butthurt about kotlin being "faster" than golang.

>>105655784
Conspiracy to force people to buy more powerful hardware

>>105658964
Kotlin can compile down to C.