by isp on 9/11/23, 10:08 AM with 143 comments
by oefrha on 9/11/23, 4:06 PM
package main
import (
"bytes"
"crypto/sha256"
"encoding/hex"
"fmt"
)
var (
_chars = []byte("0123456789abcdef")
_names = []string{"zero", "one", "two", "three", "four", "five", "six", "seven", "eight", "nine", "a", "b", "c", "d", "e", "f"}
_size = len(_chars)
)
func main() {
hexsum := make([]byte, 64)
for i1 := 0; i1 < _size; i1++ {
for i2 := 0; i2 < _size; i2++ {
for i3 := 0; i3 < _size; i3++ {
for i4 := 0; i4 < _size; i4++ {
for i5 := 0; i5 < _size; i5++ {
for i6 := 0; i6 < _size; i6++ {
for i7 := 0; i7 < _size; i7++ {
s := fmt.Sprintf("The SHA256 for this sentence begins with: %s, %s, %s, %s, %s, %s and %s.", _names[i1], _names[i2], _names[i3], _names[i4], _names[i5], _names[i6], _names[i7])
sum := sha256.Sum256([]byte(s))
hex.Encode(hexsum, sum[:])
prefix := []byte{_chars[i1], _chars[i2], _chars[i3], _chars[i4], _chars[i5], _chars[i6], _chars[i7]}
if bytes.HasPrefix(hexsum, prefix) {
fmt.Printf("%s\n", s)
fmt.Printf("%s\n", hexsum)
}
}
}
}
}
}
}
}
}
(Easiest optimization is wrapping the loop body in a goroutine. GOEXPERIMENT=loopvar really makes this nicer btw.)
The reply is obviously more interesting, need to come up with a lot of variations.
by rawling on 9/11/23, 10:17 AM
As always, the real WTF is in the comments.
by gtrubetskoy on 9/11/23, 3:49 PM
$ echo -n 'The SHA256 for this sentence begins with seven, seven, f, zero, a, b, b and five.' | sha256sum 77f0abb54cd09ad7b654bd5e762d7be58e7daffd1a0da6a56f5135bd667856a3 -
by Someone on 9/11/23, 10:36 AM
Bitcoin difficulty is at around 50 bits (https://ycharts.com/indicators/bitcoin_average_difficulty#:~....)
It also uses SHA256.
So, if my logic is right (is it? That seems awfully cheap to me), this is about 2^22 times as easy as mining a bitcoin. Bitcoin is at about $25k, so there are people who can find hits like this one for way less than a cent.
by nneonneo on 9/11/23, 10:24 PM
The SHA256 hash of this message begins with 534d765
The SHA256 hash of this message begins with c18b2de
The SHA256 hash of this message begins with 7fe17da2
The SHA256 hash of this message begins with a7fdc855d
The SHA256 hash of this message begins with 46eae34f1
by curtisf on 9/11/23, 5:13 PM
https://news.ycombinator.com/item?id=33704297
Generating sequential Git short commits! It's so pleasant looking I'm tempted to try using it one of my projects, but deep down I know it's not worth the hassle.
by jph on 9/11/23, 8:54 PM
The code searches permutations of increasing length. Benchmark is 8 seconds on a MacBook Pro M1 to discover the match of 182a7c9. The code is not yet optimized.
use std::time::SystemTime;
use itertools::Itertools;
use sha256::digest;
fn main() {
let digits: [usize; 16] = [0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15];
let chars = ["0", "1", "2", "3", "4", "5", "6", "7", "8", "9", "a", "b", "c", "d", "e", "f"];
let words = ["zero", "one", "two", "three", "four", "five", "six", "seven", "eight", "nine", "a", "b", "c", "d", "e", "f"];
let mut length = 2;
let start = SystemTime::now();
loop {
for permutation in digits.iter().permutations(length) {
let parts = permutation.iter().map(|x| words[**x]).collect_vec();
let sentence = format!(
"The SHA256 for this sentence begins with: {} and {}.",
&parts[0..(parts.len() - 1)].join(", "),
&parts[parts.len() - 1]
);
let checksum: String = digest(&sentence);
let starts: String = permutation.iter().map(|x| chars[**x]).collect();
if checksum.starts_with(&starts) {
println!("milliseconds: {:?}, {} ", start.elapsed().unwrap().as_millis(), &sentence);
}
};
length += 1;
}
}
milliseconds: 3, The SHA256 for this sentence begins with: zero, b, six and two.
milliseconds: 54, The SHA256 for this sentence begins with: zero, e, d, eight and f.
milliseconds: 8279, The SHA256 for this sentence begins with: one, eight, two, a, seven, c and nine.
by skilled on 9/11/23, 10:19 AM
by mgdm on 9/11/23, 7:28 PM
by chriskw on 9/11/23, 5:51 PM
"Hi! I’m a senior studying CS. My hobbies include making semantic paradoxes and my bio includes eight a’s, seventeen e’s, fourteen i’s, eight o’s, six u’s, and one wrong number"
by isp on 9/11/23, 10:09 AM
$ echo -n "The SHA256 for this sentence begins with: one, eight, two, a, seven, c and nine." | sha256sum
182a7c930b0e5227ff8d24b5f4500ff2fa3ee1a57bd35e52d98c6e24c2749ae0 -by siraben on 9/12/23, 5:05 AM
On 2015/7/6 at 00:00, I wrote this message's hash down. It started with 1337.
Thought #335552803: I love beef so much I made sure the sha256 of this message started with 0xbeef!
$ cat predestination.txt
(1991/4/7 at 00:16) <siraben> I know the hash of this message before you do!
(1991/4/7 at 00:17) <larsivi> Huh. What is it?
(1991/4/7 at 00:18) <siraben> It starts with 1337.
$ sha256sum predestination.txt
1337a4654163ab48761bf8acc75a407179e700f67f97d754b08c7afeac7da70a predestination.txtby omoikane on 9/11/23, 5:03 PM
by danbruc on 9/11/23, 8:13 PM
With 5 patterns, 24 bits and SHA-1.
The SHA-1 hash of this text starts with 051E35.
The hash of this text starts with A943BD.
The SHA-1 hash of this text starts with B6640C.
The SHA-1 hash of this message starts with C3B03D.
The SHA-1 hash of this message starts with D93717.
0AF3DE..
2AF7DF.......
3E0E50.
8EE84C.....
919025...
A57198....
B20775.....
DA525A........
ED20F4..by olafalo on 9/11/23, 9:01 PM
by nneonneo on 9/11/23, 9:19 PM
by SethTro on 9/11/23, 10:41 PM
The SHA-256 of this sentence begins with 0, three, 9, four, 8, four, and 1 amazing!
The SHA256 for this sentence begins with 0, 4, five, 5, eight, 6, and 7 amazing!
Suffix extensions (hence the "amazing!") are several times faster to try than throwing away the entire sha256 state for each attempt.
by pontifier on 9/11/23, 7:23 PM
by dragontamer on 9/12/23, 5:23 AM
-------
GPU-level brute force is pushing ~40-bits. Assuming ~1 week of solid compute, a 6 TFlop computer has ~3.7 sextillion compute cycles. Or alternatively, that's ~3 billion clock cycles (single thread) per check to reach 2^40 bits of entropy. More than easily doable for most simple brute-force computational problems IMO.
If you go multi-GPU on top of that, you'll have even more computational work available.
by ms0 on 9/15/23, 5:31 PM
by istjohn on 9/12/23, 3:42 AM
by Obscurity4340 on 9/11/23, 10:07 PM
by lionkor on 9/12/23, 6:07 AM
by sagebird on 9/13/23, 6:15 PM
I am not sure who submitted "byzBLFAM4Penguin"...
by kif on 9/11/23, 6:40 PM
by Alifatisk on 9/12/23, 10:52 AM
by Ekaros on 9/11/23, 8:18 PM
by nabla9 on 9/11/23, 5:20 PM
by Gunnerhead on 9/11/23, 3:00 PM
by narcindin on 9/11/23, 6:08 PM
by bizzleDawg on 9/11/23, 4:32 PM
- SSL
- Bitcoin (bonus: unlimited money hack if you can keep the discovery under wraps)
- Signed updates for devices
I assume these tweets are effectively brute forced given the fairly short prefix though and we're all safe
by grantmnz on 9/12/23, 12:41 AM
by nuancebydefault on 9/11/23, 7:07 PM
by wholesomepotato on 9/11/23, 4:37 PM
by matt3210 on 9/11/23, 7:09 PM
by PaulHoule on 9/11/23, 4:52 PM
by ms0 on 9/15/23, 5:27 PM
by Uptrenda on 9/11/23, 11:58 PM
by fleekonpoint on 9/11/23, 8:22 PM
by kjrose on 9/11/23, 3:46 PM
by kazinator on 9/11/23, 2:58 PM