File: No maid dress.jpg (125 KB, 814x814)

125 KB JPG

>>108101763
>no maid dress

>>108101763
I look like that

File: d2b1e584c1278f8ad51085677(...).png (1.99 MB, 1000x1367)

1.99 MB PNG

>>108101763
DOS-level kernel (single task and text mode)
I need to write programs for it now

>>108101939
My advice: use DOS Extenders and/or Unreal Mode with 32-bit registers.

>>108101997
It's for ARM

File: test.png (305 KB, 640x480)

305 KB PNG

@thefaggotthatcomplained
I fixed the colors you stupid fuck. Fuck you.
Now everything is working. Lets clean up, make it beautiful and then write a go web frontend to show the camera

>>108102049
What about that green vertical stripe on the left?

>>108102116
I assume that's just the wrapping of the frame, because there is at least one delay missing in the pipeline.
Just have been too lazy to read the datasheet of some IP yet

>>108102155
>one delay is missing
>but everything is working

i don't understand why man pages just can't include a copy of the struct they use.
no i don't want to grep through includes
no i don't want to google it.
just put your fucking struct in the man page. fuck.
>bbbbut it may be platform/version dependent!
i don't give a fuck.

I have figured out how to (re)do my script to format/style the CMD window an extremely easy way. I have planned it out. now I will sniff nail polish remover that I found in my house and kode (skript)

>>108103223
yep, forever stuck at C89 in that respect
one of the nice things about statically typed languages newer than C is that they really try to make their standard libraries kind of consistent around types and they have documentation about them. this is true of FP langs, C replacement memes but also even Java.

>>108103223
Some of them do, like man dirent.h

File: __happy_meek_umamusume_dr(...).jpg (132 KB, 850x1197)

132 KB JPG

did you know, all numbers * 2 = an even number
that is because, in binary all even numbers end in = 0, and shifting bits one to the left makes it end in "0" and so makes it an even number

I plan to take advantage of this by doing

cmp i, int(129);

this when it is a NOT will make a positive number negative and add/subtract 1. by doing this it can either (A: find if the number was even/odd and/or B: +1 to the final number after the cmp), and ADD the cmp to a second integer to do subtraction to get the first part of the result of (dec(129) + 1) + the number which can be either odd OR even

>>108103908 (You)
>result of (dec(129) + 1)
*

result of (dec(128) + 1)

Since my day job directly relates to PLT, I think I have the 'semi-authority' to write a 'book' which I intend to publish on Z-Library, Anna's Archive and Libgen (Z-Library lets you advertise your books, did you know that?).

So, the book is titled "Compilonomikon" and it's a 'semi-teach-to-learn' book about construction of programming languages. A half-literate-program, half-vade-mecum. In fact, the subtitle of it is "A Vade Mecum for the Passionate".

If dumb fucking retards like Rob Nystrom and Thorston ball get to write books on PLT, and have it published by legitimate publishers, then I think I am at least, in some way, entitled to write a book on PLT and publishing it on Z-Library?? Right.

Plus, "Compilonomikon" excels at where Nystrom and Ball fail at: notation. I'm going to write a book chokful of notation. Because this is a teach-to-learn book, and I still kinda suck at notation (I dropped out of college after 3 + 2 semesters, the only training I have for reading notation is from what I studied in high school --- and funnily enough, what today is taught at undergrad is why I learned at HS 14 year ago).

I'm going to write it in my own typesetting language, Dukette.

This is my personal project. For my day job, I'm developing a shim for Cppcheck's add-on interface. More 'abstraction' so to speak. Remember, a shim, not a polyfill.

File: 1768228230405810.jpg (69 KB, 634x634)

69 KB JPG

>>108103908
You might be interested on arm (and riscs generally) cmp and sub are synonyms
>CMP{cond} Rn, Operand2
>The CMP instruction subtracts the value of Operand2 from the value in Rn. This is the same as a SUBS instruction, except that the result is discarded.
maybe you can use your idea to combine them

File: chatgpt.jpg (66 KB, 960x540)

66 KB JPG

umm sisters... claude wrote a c compiler in rust

https://github.com/anthropics/claudes-c-compiler

apparently ran for days and cost $20,000 in tokens

I just had Claude Code implement a feature for my game I work on in my spare time. It burned all my daily Pro tokens and completed the job but produced a steaming pile of shit that technically works, but badly. It will probably take me hours to build on top of this and unfuck it and I just might do it all from scratch.

Okay? Where's the hype coming from? This is rather bad. Do people just ship shit like this or is there like a whole other layer I dont understand about this.

I would literally prefer typing in the browser chat interface so it can give me a small function or class, at which it can be pretty decent.

>>108104101
Does it work?

>>108104136
sort of, but not in the way anthropic claimed...
https://harshanu.space/en/tech/ccc-vs-gcc/

D does not allow delegation and inheritance for structs, only classes.

**This is the correct design choice**.

I bet a few of you have ever even touched D, and that's because it hides under the shadows, waiting for a victim, like me, to gnaw its fangs at. And it takes months of "Come on, would a D project help me portfolio?" to split free.

I went on D's Discord server, and asked them "Is Walter Bright Jewish?" and they banned me. But this is very important. For reasons that I won't enumerate (involves an extant family member getting killed by a certain outfit who also own one of my ISPs, and are major share-holders in my other ISP) I am pro-Jew now. And this is my belief that Jews are smarter than other ethnic groups. Think about it, there are 15 million Jews, and there are one billion jeets (some of whom are Jews, but I shall digress, we all know the best Jews are the AshkeNAZI Jews, followed by Persian Jews, followed by Sephardic Jews). Walter Bright looks really Jewish, and D is a good language, because it's created by a Jew.

I just love watching Walter Bright's talks. He's salt of the earth. He also gives me lots of ideas vis-a-vis CCDev.

Give D a try, would not hurt.

>>108104120
I get LLMs working by giving them corpora of books and papers. I convert PDFs to text files and concatenate them. I have two directories for corpora, `txt-literature` and `txt-docs`. The latter contains text-ified documentation for frameworks and libraries, and the former contains scientific papers and books.

if I don't give them these corpora, they give me nasty garbage. If I give them this corpora, they give me good code.

**Still**, I use this code as stencils for my own code. I never copy-paste their code, I have it write a book on the subject I want, and I convert it to PDF, and I patch-and-paste whilst reading the code from the PDF.

Go on Google Scholar and grab papers on your task.

>>108104120
>Where's the hype coming from?
From uninformed people.
AI isn't useful for implementing anything non-obvious. It's more suited for explaining code you don't know, writing throw-away scripts, generating (rough) types or schemas.

>>108104045
nice nature does NOT come in 3rd ever
also I am turned off of programming in asm (still going to learn a bit of it to help learn computers) because there are two different architectures and my code might not work on some computers
except gameboy asm
https://github.com/pret/pokered/
or maybe (like I think you said?) combining them together, writing my own .sub to use in code so that I can write my own version of the asm lang and use that

hmm asking AI for how it REALLY works in [asm instruction] to do that, then writing .routine for each instruct, which then does the thing that it would do but on the logic level (so it is mostly compatible with all)

>>108104146
Claude Opus 4.6 is still my bae. I just had it generate the AST for Smalltalk. It gave me great code. Something I would have to waste hours making myself.

But two things:

1. I gave it a corpora on Smalltalk;
2. I gave it a small go-ahead sample;

This is what it gave me:

https://pastebin.com/0HDQXCxf

Beautiful. Remember, I explicitly told it to use `std::variant`. The retarded way to make ASTs in OOP languages like C++ is delegation. But std::variant fixes this problem. It lets you create sum types quite easily. In other words, use parametric polymorphism instead of subtyping.

If I've read TAPL correctly, subtyping and parametric polymorphism are isomorphic. But one sucks, whereas the other one does not.

Good thing is, this is a zero-cost abstraction. But a functional languages does not guarantee it.

Rust enums are true sum types. That's why Rust is a good language, I just don't use it because I have projects in Rust, but not C++ (most of my projects are in vanilla C).

>>108104136
It allows for

int x = "bla";

And let's not even talk about optimizations.

>>108104197
I think the correct (technical) term is 'non-trivial'. And you're right. That's why I make books with AI. Not programs, unless it is wholly trivial.

Funny thing is, I built half of a C compiler with Claude 4.5 two months ago.

I narrowed down the scope to make it follow lcc's architecture.

It's a useful prompt session. It weered away after 8 responses.

Of course I did not have 23 million to waste. But it was educational.

I gave it a lot of stuff, including lcc's book (by Fraser and Hanson) and lcc's source code.

If you wanna make a C compiler, try to make something minimal like lcc.

cosine
ine = x;
e = y;
cos(ine)
sin(e)
>
true?

so it would seem...if:
cos(y) ' plot this
when the wave function(s phase) is at its peak (-1 & +1), it is e (number) away from the last time on (x) it intercepted in the "0" before
for each peak in cos(y) ' cosy cosy
x = x ' equals current position - e
x = currentPosition - e = currentPosition = x + e ' last position plus the distance the wave has travelled, + for positive wave direction
thgoughts?

is there an algorithm to fetch any arbitrary digit of pi? like if I wanted the 11 trillionth digit?

>>108104656
digit = 0;
for (i = 0; i < 11000000000000; ++i)
{
digit = <calculate the next digit of pi>;
}
return digit;

>>108101763
Why the AI slop

>>108104672
>wtf why is my cpu usage at 100%

>>108104656
https://en.wikipedia.org/wiki/Bailey%E2%80%93Borwein%E2%80%93Plouffe_formula

Every file can be compressed as an offset into pi where, through pure random chance, the sequence of numbers consituting the file's binary occurs.

>>108104656

maybe write pi approximation to file then grep for 11 trillionth character

digit = 0;
for (i = 0; i < infinity; ++i)
{
digit = <calculate the next digit of pi>;
}
return digit;

what would this do

File: 1745468327298320.png (351 KB, 839x768)

351 KB PNG

>>108104725
>the offset is larger than all information that can fit in the universe
>"compression"

>>108104735
waste electricity

>>108104741
He's right though, it's compressed incredibly well.
It's just not quick to compute.

>>108104197
Exactly, I would never copy-paste any AI-generated code bigger than a few lines. Another thing I find it useful for is "How is this usually done?" type of questions

>>108104764
Assuming that the digits of pi are truly random, calculate the average offset into pi until an exact sequence of a 1mb file occurs.

Protip: the bits required to represent the offset will be greater than 1 million

>>108104170
>**

>>108104817
what if you get lucky and the offset is exactly 0? now you just turned 1 million bits into 1 bit

>>108104817
>Protip: the bits required to represent the offset will be greater than 1 million
Yeah, but you could just store that number in pi too.

File: Screenshot 2026-02-09 222128.png (77 KB, 706x483)

77 KB PNG

>>108104817
not him but I asked this cringe question and got a cool result later on in the prompt (check the timestamp btw)
didnt post it to not hog all the thread but, did you know:
The "Feynman Point": At the 762nd decimal place of $\pi$, there is a famous sequence of six 9s in a row (999999)
and funny image at bottom

>>108104725
Proof?

if pi contains an infinite amount of information than how come I can represent it with one unicode character π huh?

>>108104725
the files binary in dec? in decimal numbers? converted to dec

File: 84b66ec3f7335d4b.png (85 KB, 280x280)

85 KB PNG

What if there was an audio codec that used "audio palettes", i.e palettes of waveforms similar to color palettes in image files?

seems the difference between cos(y) 1 and cos(y) 2 is pi by the way

>>108104828
>It's even bigger than the first number

>>108101763
I want to see the outline of her erection on the front of her skirt!

>>108104886
number are numbers my dude, you could take every bit making up a file and interpret that as one big ass integer if you wanted to

File: 1747799154161141.png (102 KB, 241x228)

102 KB PNG

>>108104939
>certain numbers are illegal

>>108104939
you cant do this in vbscript though you can have very long longs
there is also money type, which is accurate with no floating point error!
the script will also take a variable and put it into memory to use it, so it cant be raelly big

>>108103908
Please stop posting this image, it's genuinely disturbing to look at that deformed midget arm
I will refrain from AI art social commentary but holy shit it hurts to look at

>>108104988
Anon look at her knees

>>108104957
The types your language/hardware supports are irrelevant. Mathematically you can just keep addings bits one at a time and getting bigger ints.

>>108104988
man its just the perspective, an illusion
ill stop until one day I randomly forget to not post it and I hope you arent here

>>108104952
even 1 is illegal because it's an index into the FBI's library of illegal material
and remember when timestamps became illegal in Youtube comments?

Yo /dpt/ you guys are smart. I flunked some weird junior programmer test. Help me out with this one. I got tripped up and couldn't wrap my brain stack stuff and greedy stuff. Even when I ask AI its like 20 if statements. Is there any clean way to do this?

def snake_case(s):
   pass

s = "HTMLMicroServicAS";
assert snake_case(s) == "html_micro_service_as"

is he tackling the 1mb file into pi issue

I would program something but I have no ideas of anything pracitcal because everything has been made by other people

0 is an even number (ends in 0 in binary). thoughts?

if 0 is an even number
0 * 2 = 0. this is an even number
if 0 is odd
0 * 2 = 0. not an odd number (impossible)
thoughts?

File: 20260112_124909.jpg (980 KB, 1920x1200)

980 KB JPG

>>108101763
>What are you working on
the coding for this cute little fire hazard

>>108105066
what was your non-AI attempt?

>>108105137
who said 0 is even?

>>108105219
me in my post

File: isnt-that-cute-but-its-wrong.gif (102 KB, 220x165)

102 KB GIF

>>108105224

0 = (0(~even)* 2 = 0 (even)) !~ (0(even OR ~even)* 2 = (0 (~even))is true

>>108105263
You are trying to speak Math, but you are using code.

>>108105234
if 0 = even then * 2 = 0 = even
whether 0 = even OR ~even, it has to = even when 0 * 2 and therefore even can ONLY != ~even
&
0 AND 00 = 0
(0 * 2 = 00)
+ >>108105282 both ending in 0 in binary/mod(2) system, so 0 = even
let 0 mod(2) with a (max?) range of 0 - 99 (or 00?, if not 0 - infinity (hehe))
0 * 2 = 0 in this modulo, meaning 0 is even

>>108105305
again, >>108105282

write this up in a proof notation, citing the proper axioms, and I will read your slop.

File: snake_case-ocaml.png (67 KB, 864x648)

67 KB PNG

>>108105066
>if statements

>>108105325
thinking about that: mod(2), (0+1) * 2 = 2
mod(2), (0 + 1 * 2) * 0 = 0; 2 = 0 = i
i = (o for odd ev for even here
i = ((2 * 2 = ev) (~2 * 2 = ev), ~((2 * 2 = ev), (~2 * 2 = ev)) = even

>>108105377

# snake_case "H";;
- : string = "hh"

instead of the "" case,

  | s when String.length s < 2 -> String.lowercase_ascii s

>>108105396
oh is this a meme?
responding to my post using a chatbot?

>>108105419
let 0 = 2 inside of brackets and 0 is even
if 0 = 1 (or any odd number) inside the brackets then it still becomes an even number, i.e. when (0+1). if
0 + i * 2 = even
0 * 2 + i = odd
0 * i = even number
0 behaves like "2", not "1", due to it being an even (and positive maybe) number

if mod(2) 0 IS or WERE an odd number, then infinity would be even, and infinity-1 would be even, because 0 = odd and 1 = odd is two odd numbers in a row. think about that

>>108105464
>if false, ..., think about that

>>108105494
how would numbers start at odd, odd, then go , ... even, odd, even, odd ... and not end on even even? thats just wrong

>>108105464

# let iseven n = 0. = n -. (2. *. float_of_int (truncate (n /. 2.)));;
val iseven : float -> bool = <fun>
# List.map (fun n -> n, iseven n) [1.;2.;3.;4.;5.;infinity;nan;neg_infinity;0.];;
- : (float * bool) list =
[(1., false); (2., true); (3., false); (4., true); (5., false);
 (infinity, false); (nan, false); (neg_infinity, false); (0., true)]

all of those weird numbers are, appropriately, odd

>>108105527
by your own code 0 = even is true

>>108105527
yes, obviously. That's elementary school knowledge, encoded in iseven

infinity, NaN, and -infinity = false because in vbscript true = -1
so when you do it you get (-1) * (-1) or true * true = false, because -1 = false. if this was false then it would be 0 * 0 which is false (-1, because its true, a negative number). iseven(true) = FALSE, so true = -1

>>108105066

def snake_case(s):
    rs = ""
    underscore_next_loop = False
    for i, c in enumerate(s):
        if(len(s) > i + 1):
            rising_edge = s[i].islower() and s[i+1].isupper()
            falling_edge = s[i].isupper() and s[i+1].islower()
        
        if(falling_edge or underscore_next_loop):
            rs += "_"
            
        underscore_next_loop = False
        if(rising_edge):
            underscore_next_loop = True
        
        rs += c.lower()

    return rs

s = "HTMLMicroServiceAS";
assert snake_case(s) == "html_micro_service_as"

this works I guess, idk what you mean by "brain stack stuff and greedy stuff" and "20 if statements"

>>108105632
UnboundLocalError on 'h'

>>> snake_case('ABc')
'a_b_c'
>>> snake_case('AbBc')
'_ab_b_c'
>>> snake_case('AbCDe')
'_ab_c_d_e'

vs.

# snake_case "ABc";;
- : string = "a_bc"
# snake_case "AbBc";;
- : string = "ab_bc"
# snake_case "AbCDe";;
- : string = "ab_c_de"

>>108105661
yeah, I got lazy with edge cases

def snake_case(s):
    if(len(s) == 0):
        return ""
    if(len(s) == 1):
        return s.lower()
    
    rs = ""
    underscore_next_loop = False
    rs += s[0].lower()
    for i in range(1, len(s) - 1):
        rising_edge = s[i].islower() and s[i+1].isupper()
        falling_edge = s[i].isupper() and s[i+1].islower()
        
        if(falling_edge or underscore_next_loop):
            rs += "_"
            
        underscore_next_loop = False
        if(rising_edge):
            underscore_next_loop = True
        
        rs += s[i].lower()
    rs += s[len(s) - 1].lower()

    return rs

s = "HTMLMicroServiceAS";
print(snake_case(s))
assert snake_case(s) == "html_micro_service_as"

>>108101763
What's the whitest programming project one can undertake?

>>108105770
visualbasic script

I cannot delete posts this often
your own DDoS tool written in visualbasic

>>108105749
actually this version still has bugs since I tried to optimize the beginning and end but I can't be assed to fix it beyond this

>>108105770
for demographic reasons, the older and less popular the whiter. Forth, SML, Prolog, Ada.

>>108105813
what are some failing cases?

>>108105826
aB

>>108105818
Forth has to be the single most white thing ever made.

>>108105770
Probably some sort of facial recognition software, numeric simulation software for military engineering or something along those lines that helps the IDF kill more people

>>108105813

def snake_case(s):
    rs = ""
    underscore_next_loop = False
    for i in range(len(s)):
        rising_edge =             i+1 < len(s) and s[i].islower() and s[i+1].isupper()
        falling_edge = i != 0 and i+1 < len(s) and s[i].isupper() and s[i+1].islower()
        
        if(falling_edge or underscore_next_loop):
            rs += "_"
            
        underscore_next_loop = False
        if(rising_edge):
            underscore_next_loop = True
        
        rs += s[i].lower()

    return rs

couldn't help myself, should be good now

File: 1769537655455669.png (9 KB, 208x110)

9 KB PNG

>downvoting doesn't decrease their score
I fucking hate the modern internet

>reddit-esque downvotes have become meaningless
Good. If it makes redditors angry, it's good.

>>108101925
Prove it, you saucy minx

holy shit either 2 or -2 * 0.5 = 1

File: 1768386593759897.png (22 KB, 621x353)

22 KB PNG

>>108105928
also today's problem is terrible, I spent like 20 minutes thinking if there was a dynamic programming / prefix sums solution before checking the hints and being informed that the only solution is brute force.

>>108105973
Yeah see that's not actually what's happening, you're getting the worst of the reddit upvote system without any of the benefits. "downvotes" are there to help users police and filter out trash content. But since they do nothing and there is still an "upvote" / reputation ("karma") system, it encourages users (jeets) to spam low iq karma farm bait posts. And there is no way for the community to police this shit, because comments are either sorted by most likes or most recent, and both methods fail to provide any useful information.

>"wow i just came out as a troon to my dad and he beat the shit out of me!!!!"
>800 likes
>gets promoted as the top comment
>even if 3,200 people dislike it, it doesn't matter because it's still the top comment
>continues to get more likes
>no penalty to the user's score because dislikes don't affect him
>he proceeds to spit out another gorillion bait karma farm posts

File: snake_case-python.png (77 KB, 993x528)

77 KB PNG

>>108105928
downvotes on youtube comments do literally nothing, btw. They don't even get sent to the server. It's a purely in-client placebo that tempers some of the useless seething and internet fights that people would get into if not for this illusory retaliation.
>>108105859
nice. The OCaml also fails on aB. Since I don't care as much, have an OCaml version of your Python

Swift is such a pile of shit. Not because the language is bad per-se but because the tooling they ship with it is terrible. The Linux lsp-sourcekit leaks about 500mb of memory every 30 mins or so, it appears that this has been a bug for over 2 years as well with no fix in sight. Multi-platform my ass.

>>108106075
from the xcode bitching from the past couple of threads, maybe this is the standard for all platforms?

((n * -1) * 0.5) * -1) = even number
n = 0
((n * -1) * 0.5) * -1) = 0

>>108106082
No idea. I tried it out on my Ubuntu install and an Arch VM it has the same problem on both them. I'll try it out on a Windows VM and see if it's a linux only problem. Digging around on Github it seems that the Swift devs know what's causing it but 18 months later haven't fixed it.

I've been reverse engineering a game's image format and although some of the work was already done, the last few unknowns I discovered myself which was very fun.
I don't really know how to use ghidra, I just got lucky finding the function and using the garbage C-like code I got enough clues to figure it out.
What I found very shocking is how good AI seems at this, I copypasted the garbled mess ghidra shows and it correctly identifies LZSS decoding, wtf?
I've been thinking about reverse engineering more of the game but now I'm hardstuck because I can barely fiddle with ghidra and I have no idea how to learn more of it

actually everyone knows that 0 = even
couldve got something done today

>>108106705
>it correctly identifies LZSS decoding
That would imply you have inserted your own LZSS-encoded images into the game? After all, this is reverse-engineering.

File: screenshot.png (148 KB, 1312x1204)

148 KB PNG

>>108106755
I don't understand your comment
Picrel is what I'm talking about, it had the context that it has something to do with image data but getting almost exactly right from this mess is impressive

>>108106818
>I don't understand your comment
You don't seem to understand anything.

>>108106921
Whats that supposed to mean

>>108098350
This was a good idea, I replaced the bitmap allocators I was using for segregated sizes >= 8 bytes and <= 32 bytes with segregated slab allocators using explicit free lists. Went from ~5 MiB/s to ~10 MiB/s. If I pack the pointers I can use the same allocator type for 5-6 byte allocations, and I can do segregations up to larger sizes as well which will probably further improve performance (at the small cost of fixed size overhead per segregation).

File: test.png (317 KB, 640x480)

317 KB PNG

>>108101763
>What are you working on, /g/?
I'm hustling and cleaning up the software side for camera handling.
I did all kind of hacky scripts to get it running and capture an image. I'll try to wrap this all up in Go now and make a cute webfrontend.
Enabling camera (not initializing) and then getting stats from hardware works!
Now I2C module to get rid of the bash init script

Still gotta do these steps to get ONE picture

# ./camctrl/mycamera
2026/02/10 10:17:42 Hello, World!
2026/02/10 10:17:42 Camera not enabled yet. Enabling now. Please execute I2C setup
# ./perfect_setup.sh
OV5647 camera configured
# ./camctrl/mycamera
2026/02/10 10:17:55 Hello, World!
2026/02/10 10:17:55 Frame count: 474
2026/02/10 10:17:55 Line count: 960
2026/02/10 10:17:55 Pixel count: 1228800
2026/02/10 10:17:55 Error count: 0
# ./nocap

>>108108740
>still stripe on the left side

>>108109179
Yes I won't care about that.
I do have a raspi camera v2 here (imx219 sensor) that is much better, i just dont have the time rn to go through the datasheet and find out the settings

who was the mongrel that decided octal literals should be denoted by a 0 prefix and not a o0

what a fucking RETARD its not like someone would ever attempt to prefix a base 10 number with 0s right? FUCKing retard im becoming a rust tranny. Where can i buy the cheapest socks?

>>108109644
Every good language uses 0o for octal

>>108106705
Dont ask "what can ghidra do now?" ask "what do I want ghidra to do now?"
you probably want to learn ELF / PE formats (or whichever binary medium you're working with here)
generally when people dont know what to do next it isn't because they lack ghidra knowledge its because they lack knowledge of what theyre working with.

once you know what YOU want to look at / reverse next then its very simple to google and figure out if ghidra has any features related to that

I'm back. I cannot sleep. My supervisor has told me to fix my sleeping pattern or else I'm out.

I usually stay up for 5 days, with some intermittent slumber, then sleep for 36 hours straight.

I take Ritty. Lots and lotsa Ritty.

I found this tool called 'tangler' on Pacman to extract away code blocks from Markdown files. Now it does not seem to work. I have to do make one myself.

>>108109785
goodluck schizoid

>>108109644
>who was the mongrel that decided octal literals should be denoted by a 0 prefix and not a o0
mistake happens, the better question is why dimwits have been using this syntax when making a new languages
>>108109658
based.
0x, 0b, 0o
"\xff \x{fff} \b{1101} \o{377}"

>>108109793
Thanks f.a.m. I have a date (with a girl) tomorrow. It's odd that, now that everything's gone to shit, my miserable hunk of shit life's picked up.

I've also quit Youtube. Every time that i think of the slop it fed me on the through of its recommender system, I just gag. Doctors should develop some sort of framework for addiction to recommender systems. Perhaps, as soon as people stop calling them 'algorithms'. I cringe every time someone uses the term 'algorithm' to refer to recommender systems.

File: Screenshot_20260210_181017.png (448 KB, 993x1494)

448 KB PNG

I write all my documents with NeoVide now. Its animations help me focus. How did they make it anyways? It uses my Neovim config. But it's an X application?

>>108109986
What's a twitter application? Does it run inside of a tweet?

>>108110233
He means X11 you dense motherfucker

File: 4f798ab1514479d5ebbd19902(...).jpg (462 KB, 850x1017)

462 KB JPG

>120 clones in the last two weeks
>no visitors
Who the hell is cloning my repo?

>>108110349
That was me

>>108110304
What's that? Is this something like xbox?

>>108110349
Now you know how LLMs operate.
Your fault for being retarded enough to post your code into public repositories.

File: 86e518bdb7fd1eafaf4cdc86e(...).jpg (114 KB, 675x988)

114 KB JPG

>>108110400
That's a good explanation
Not that I care about the anti-AI evangelism part

>>108110414
I was assuming your appreciation for the process was low.

File: 294px-Iggy_Death_OVA.gif (203 KB, 294x225)

203 KB GIF

>>108110400
Not either anon but this makes me wonder
Is the reason LLMs are so bad at coding because they're trained on my old shitty code?

>>108109986
>Celeste is
>Celeste has
>Celeste uses
Does it now? Where's the implementation?

>>108110518
nta but some people document before they code/implement

File: 9bcf80ea5857785ee9f670238(...).jpg (320 KB, 1692x1772)

320 KB JPG

>>108110487
My godcode is what keeps the US economy from exploding

>>108110549
>tone deaf
tiresome

No they don't, this doesn't make any sense and the pic rel is neither a specification nor a documentation.

You can write the specification before writing the implementation.
You can document a program after writing the implementation, but you don't document it before hand.

>>108110632
>what is speculative execution
>what is out-of-order evaluation

>>108110690
STFU idiotic bot

>>108110805
STFU idiotic bot

>Ask AI on how to modularize my project, because I finished hacking phase
>Don't even read the text except of a few words and just do it how I find it cute
Why don't I just stop using AI? It's stupid even opening the page.
Sure sometimes it can help, but most times I clearly don't need any of this.

>>108109986
>>108110690
>>108111073
"Celeste" only exists in your mind. cope
You're like the nigger on reddit that post "here's my compiler that is such and such" except that the github repo is literally empty.

>nigger
*niggers

File: small-chalkboard-with-sch(...).jpg (210 KB, 1300x956)

210 KB JPG

>>108111129
Your like*

>>108110518
I've literally written "in this document, we propose...". PROPOSE.

I have a directory full of these proposals. They serve two purposes, sort out my intentions and plan the project, and also, something to feed to LLMs to ask for feedback, and some prototype code.

>>108111159
dumb fuck

>>108111185
and then you wrote "Features OF Celeste"
Celest is ... , Celest has ..., Clesete does ...

>Celeste has a JIT, a concurrent GC, a module system.
No it doesn't.

example where i = 0; i != even: infinity
such as when the status (boolean) is no longer "alive", I mean the time AFTER the system has reached maximum entropy such as when k -/+ k = 0 the "possibilities" are no longer infinite but 0 (the opposite of complete entropy with the calculation including +1)...
when k +/- k = 0 and entropy has reached its max, if k = +/- 0; = k +/- k (1 more "step") then (infinity + 1) = 0 & != even

File: Screenshot_2026-02-10_19-34-11.png (80 KB, 1291x791)

80 KB PNG

Made this cool backtester with monte carlo in python/tkinter. To help me calibrate my crypto bot