Why aren't there complete tripcode databases?

Why aren't there complete tripcode databases?

>If you use 8 characters and you choose from alphanumeric characters (A-Za-z0-9) there are (26+26+10)^8 = 62^8 = 218340105584896 possible combinations, against a GPU with 130 million tripcodes per second it will take about 19.4 days running non-stop to exhaust the keyspace.

That seems easy enough.

Attached: 1529780709617.png (870x467, 58K)

Other urls found in this thread:

github.com/meriken/merikens-tripcode-engine-v3
twitter.com/SFWRedditGifs

You know the DES tripcode is 10 characters and not 8 right?

Then it would be just 62^10 instead. Not a big deal.

But what’s the hash algorithm?

It would require more disk space than there are atoms in the universe

hmm yes, 20 days times 62 and 62 again is 210 years.

>62^8 vs 62^10
>not a big deal
user...

>GPU with 130 million hashes a second
>roughly 74724 DAYS
>>NOT A BIG DEAL

With 1000 computers dividing the work evenly that's 2.5 months

Ahh yes, rent 1000 GPUs for 2.5 months just to find that one tripcode. Then the tripfags switches to SHA-1 12 character tripcodes and all your work is ruined.

No company or richfag is willing to spend 100s of thousands of dollars to exhaust the tripcode set for it to be made public and "free"

It's definitely possible, but why would someone make the effort?

>Not a big deal

Attached: 1543298303897.png (650x581, 284K)

Jow Forums only takes the first 8 characters of the password. It's still higher than 62^8 though.

>It's still higher than 62^8 though.
Because of the weeb charset?

>wasting a single computational cycle on tripfags

They could have the pass as trip and no one would care

Has anybody tried to "crack" the secure tripcode system? As I understand it, there's a secret salt added to the user input that gets hashed. But if you have precomputed (or a fast way to compute) hashes, you can check for ${yourInput}${somesalt} and see if it matches the hash by Jow Forums. Then you'll know 4chans salt.

P.S yeah, the salt can be really long, but maybe we can exploit something in the hash algorithm.

Input space is the number of 8 byte strings so 256^8 = 18446 Quadrillion
However output space is the number of 10 alphanumeric characters plus dot and slash strings so (64^9)*16 (last character only has 16 possibilities instead of 64) = 288 Quadrillion
62^10 is actually a pretty close estimate log wise to the latter.

Jow Forums doesn't have salt; the salt is computed from the 2nd and 3rd character of your key

Then how is it more secure than the nonsecure version?

Salt alone doesn't make that trip secure because it's fairly easily attacked. But it's a 12 characters SHA-1 tripcode instead of the 10 characters DES one

>2.5 months of no vidya and huge electricity bills because i'm mining trips for Jow Forums
no thank you

you can gather secure tripcodes and bruteforce them to see what hash generate the insecure tripcode that coincides with it
as soon as you have two correspondences or more you can start deriving the system that builds the secure one, and thus be able to generate secure tripcodes yourself without having to query Jow Forums
of course, this isn't guaranteed to work at all since the hashing function could be absolutely different and in no way related, but you never know desu

rich miners have this equipment and it now useless
where is GPU trip explorer?

How the fuck are you planning to store them? You need like 2PB of space to store 62^8 tripcodes.

github.com/meriken/merikens-tripcode-engine-v3

It's actually closer to 2300PB.

(You)

218340105584896 of 10 bytes strings is 2PB roughly.

It's not 62^8 though; see

Not every byte can be used for the trip.

can use rainbow_tables to make it 65536 times less (or more)

Which doesn't matter at all because input space is vastly bigger than the output space and you need a table of at least (62^9)*16 lines to store the end result.

No, you can't. You need to store keys for each of of the 288 quadrillion trips. That's the point of a rainbow table

rainbow tables can reduce needed space in 65536x (or more)

No.

yes

Do you even know what a rainbow table is? Rainbow table doesn't save space because it's at the extreme of a space-time trade off.

You don't seem to know what a rainbow table is

How much would you estimate that to cost?
What's the projected return for this investment?

The only tripcode that matters is already known. Nothing else matters now

Attached: c08.jpg (1486x1058, 263K)

rainbow tables can reduce needed space in 65536x (or even more and more with modern SSD disks)

How would storage medium affect needed space to store something? You just might be too dumb for this board

by speed of search in database