+- hashcat Forum (https://hashcat.net/forum)
+-- Forum: Support (https://hashcat.net/forum/forum-3.html)
+--- Forum: hashcat (https://hashcat.net/forum/forum-45.html)
+--- Thread: Keyspace questions (/thread-10675.html)
Keyspace questions - raithedavion - 03-21-2022
Good afternoon,
Please excuse me as this is my first post. I've read about keyspace, specifically from the FAQ page. However, I wanted to know why there is such a discrepancy with keyspace values. For example, -m 14000 outputs a keyspace of 34,359,738,368. In reality, there are 72 quintillion possible combinations for the 8 character DES key.
Taking something like hashtopolis into account, the entire keyspace gets taken by a single machine because the keyspace produced by hashcat itself is so low. IE: 8x 3090's does about 450-500 Gh/s in -m 14000, so the hashtopolis assigns it to a single host. Whereas the true keyspace would allow it to be distributed among several.
So my questions are these:
A) It seems the "hashcat legacy" had a more accurate keyspace, why did that change?
B) The hashcat output is 128^5, leaving the the remaining 3 bytes worth of keyspace out (DES is 8 bytes long). Is that on purpose? Is it because of the parity bits in DES?
Regards,
Raithe
RE: Keyspace questions - Snoopy - 03-22-2022
given the DES algo yes
parity bits takes the output and "hash" it again with a way more simple algo, so if the output length ist 64 bits its in real 56 bits
https://en.wikipedia.org/wiki/Data_Encryption_Standard#Brute-force_attack
RE: Keyspace questions - raithedavion - 03-23-2022
I thought so. But in the end doesn't it still end up being 72 quintillionish combinations? ie: A single 3090 can do 56ish Gh/s on DES. So if keyspace is the only factor, a 3090 theoretically does the entire DES keyspace in less than 1s.
Obviously though, that isn't the case. A system with 8x 3090's takes well over a day to exhaust the DES keyspace. Is there a better way to distribute hashes such as DES that doesn't solely rely on the keyspace from hashcat?