Here it is again.

thx, damn that was quick.

I was just wondering whether there is a downside to perfecting the tables.
E.g would it decrease the probability ?

So if i had

sha1_loweralpha#1-8_0_5500x13450000_#000.rt
....
....
sha1_loweralpha#1-8_0_5500x13450000_#009.rt

and i perfected them into:

sha1_loweralpha#1-8_0_5500x49754263_#000.rt

would there be any difference in the hashes they can crack.

It will decrease the success rate, but improve your cracking speed. You can come up to the same success rate with perfect tables, but you will need to generate more chains than regular tables.

hi,

May be a stupid question, but i really don't understand why if i have "normal" tables on which i erase the duplicate chain , that will reduce the success probality?
are duplicate chain not exactly the same?

Thanks
Alone

Only the last link needs to be the same for a chain to be a duplicate. Most duplicate chains are very different until then get towards the end.

Example:
a->h->y->l->f->s->m->6->d->r->l->6->9->n
o->t->l->e->n->5->p->q->d->r->l->6->9->n
u->6->v->r->k->e->a->h->7->e->j->9->z->i

When there is a collision like this at d then the rest of the chain is the same. There can be collisions that are not at the same chain position like h, but these do not end necessarily end with a chain collision. The reason the success probability drops is the fact that you are losing the beginning of the chains. As you can see in this example if you removed the first chain you would no longer be able to crack y since it only appears in the first chain.

I think the "http://www.sisecurite.fr/articles_et_actualites/Retirer-les-doublons-des-Rainbow.html" under linux is better then rtperfecter ... auto rename files

hi,

May be it is a better remover, but it don't run under windows....

Does it has a sens to rtperfect file with different index?
for instance:
sha1_loweralpha#1-8_0_5500x13450000_#000.rt
....
sha1_loweralpha#1-8_0_5500x13450000_#009.rt
sha1_loweralpha#1-8_1_5500x13450000_#000.rt
....
sha1_loweralpha#1-8_1_5500x13450000_#009.rt

to
sha1_loweralpha#1-8_0_5500x13450000_#009.rt or sha1_loweralpha#1-8_1_5500x13450000_#009.rt

Thanks

Alone

It only removes duplicate chains, not merging chains.
It doesn't perfect the tables.

someone can give me an example of correct command

I tried but whenever its worked

Hi there,

I hope this will not be too off-topic, but I'm comming from the sticky thread topic841.html, and after I've read 3 pages of this thread, I've decided to post because I'm totally lost.

What the hell is "a perfect table" ?
Thanks

Note; I've tried to use the search function, but the term apparently is too common to be searched.

Edit: Just read the FAQ, section 4.9. Thanks

But is the perfect tool only remove duplicate from a single file, or from the whole set of file ?

From what I've understand, the command to run the app should looks like this:


The first %u will write the number of chain in the file
The second %03u will write the file number.


But from now, I get this:


First: Tring or Trying ?
Second: What I'm suppose to do if he try to write unsorted data ?


Edit: I just ran rsort.exe *rt on the folder and all table have been sorted.
Perfecter give the same error.

Hi. Firstly I'm new here and I'm also rather new to the concept of rainbow tables so please forgive me if I'm not 100% correct with everything. I tried rtperfecter on a test table 1-4 chars, mixalpha-numeric + symbols + german characters. I'm using winrtgen. So I started with default settings and tweaked them a little. Since I'm generating some larger tables on my desktop I tried to make this smaller set using my notebook. So I lowered the number of hashes per file to a number which left me with 64mb per file. This was 95%+ success, but for such a tiny data set I wanted 100% and seeing as my notebook has a core2duo I thought I'd make 2 files for the tables (2x64mb). Obviously this is way over 100% so I lowered the chain length (this speeds up the use of the table, while lowering success, correct?) To the lowest number that was still 100% (1102).

So, I have a few questions about perfect rainbow tables. I understand that "duplicate" entries are removed, sometimes this means removing perfectly valid entries that "look" like they are duplicates? So when using this table (2x64mb files) I notice it was REALLY slow. Slower than a 1gb table spanning 4 files that I was also using. I tried rtperfecter on the two files, it told me how many unique chains were left in the table. I tried using it and it was blazingly fast (as I would have hoped) So, finally, here are my questions:

1) If I open winrtgen and use the same chain length setting, modify the chain count to the number "left" in the perfect table is the success chance it display that of the perfect table I just generated?

2) If I generate 2 more files (002/003) can I perfect these into the "new" one? The output was 1.7mb, can I use rtperfecter on the 1.7mb perfect table and the two "extra" 64mb files to merge them into another perfect table (with a higher success rate, possibly?) or do I need to keep the other 2 original files and use rtperfecter on all 4?

3) (Almost the same as #2) If I have a friend generate 2 file, and I generate the other two (Let say I generate 000+001, my friend generates 002+003). Can my friend use rtperfecter to reduce the size of the table to make it easier to send over the internet (I understand it will have a lower success rate) then I use that rtperfecter to perfect the combination of my perfect table and his?

Also keep up the good work. The rtperfecter tool seems fantastic!