Jump to content

Compute Missing Pieces?


rompom7

Recommended Posts

Is it possible (using uTorrent or any other software) to compute missing pieces of a nearly completed torrent?

For example... You were downloading a text document that says something like:

hello this is an example

And you got first 50% ("hello this i") of it before all the seeds left, and you couldn't get any more of the data.

Using the hash and the size of the data, can you compute the rest of the data?

hello this iaaaaaaaaaaaa: failed (computed hash != original hash)

hello this iaaaaaaaaaaab: failed

hello this iaaaaaaaaaaac: failed

...

hello this iaaaaaaaaaaay: failed

hello this iaaaaaaaaaaaz: failed

hello this iaaaaaaaaaaa0: failed

hello this iaaaaaaaaaaa1: failed

...

hello this iaaaaaaaaaaa8: failed

hello this iaaaaaaaaaaa9: failed

hello this iaaaaaaaaaaba: failed

hello this iaaaaaaaaaabb: failed

...

hello this is an examplc: failed

hello this is an exampld: failed

hello this is an example: success!

(of course this is a very simplistic example, only taking into account every alphanumeric combination, not every binary combination)

It iterates through every combination until the hash is equal to the hash of the complete data.

I'm not exactly sure how BT works, but if each piece has a separate hash, then it would be faster to compute each missing piece than all the missing parts at once.

I can imagine it taking a reallllly long time to compute the missing data (depending on how much data is missing), but sometimes theres those things that you just really need and you can't find it anywhere else.

Link to comment
Share on other sites

As you said, it's too computationally intensive to try to look for a collision, and a collision might not necessarily mean identical data (though I guess it generally is). Additionally, you can't be sure if the incomplete piece you're trying to guess the remaining data for is already corrupt, and it'd end up being a total waste of time for the computer to try to figure out how to complete poisoned data.

Link to comment
Share on other sites

Yes it would be very computationally intensive, but if there was an estimated time remaining (until all possible combinations tested), and maybe a progress bar of total combinations tested / total combinations possible... Then a person can see if they can be bothered trying to guess the data...

This is sorta turning out to be a feature request, which wasn't my intent.. Just wondering if anyone has heard of any program to do such a thing... Otherwise I'll work on a little something myself.

EDIT: And, I'm back... I did the math.. It would take a really REALLY long time to compute even 100KBs of data.. 5-6 years @ million guesses per second... I'm gonna put this on my NOT to do list.

Link to comment
Share on other sites

Archived

This topic is now archived and is closed to further replies.

×
×
  • Create New...