Orginalny news ze strony
1 Jul 2006
We have stopped our birthday search to continue with stage 2: completing a partial collision to a full MD5 collision. See the message boards for more information.
A tu post z forum:
Hello everyone!
I think everybody has been waiting for this project update.
Our server has caught up now with almost all results and I have updated the collisions page.
There are still some results in progress and they will be finished soon.
However, we will not continue with our birthday search since we feel we are ready for stage 2.
We have chosen one of the useful collisions with 8 remaining bitdifferences.
And in stage 2 we will cancel these 8 bitdifferences one by one, so that we will end with no differences and thus a full MD5 collision!!
For each of those bitdifferences we will have to go through the following two steps:
* Creating a differential path:
this is a precise description of how differences should pass through MD5 such that one bitdifference will be cancelled;
* Finding a matching message block:
find a real message block that satisfies this differential path.
Step 1 is something that is far easier to control on a simple cluster, and we are almost finished with this step for our first bitdifference.
For step 2 we really can use your help!
We will need some time to finish step 1 and start step 2. But until then there are plenty of other BOINC projects for which you can crunch while you wait. BOINC is very well capable of doing work for more than one project.
So there is no need to wait or to disconnect from HashClash.
Your help is really appreciated here!
As a final note I want to add that I always try to quickly respond to questions etc., however I am really busy with finishing my studies and writing my thesis. Project HashClash was part of my thesis and I will continue HashClash between everything else.
ZApraszam do lektury bo ciekawe :D
Ktoś coś wie kiedy dalej poleci ten projekt?
Chyba właśnie zaczął lecieć: "Today we started using our new application for our stage 2, so new work will be available." Choć z tego co widzę - to niezbyt dużo tej pracy narazie dali - status serwera projektu (http://boinc.banaan.org/hashclash/server_status.php).
Results ready to send 0
Results in progress 516
Faktycznie - nie poszaleli :)
Ooo teraz się rozkręcili:
Results ready to send 1,447
Results in progress 1,416
---------- 13:09 14.10.2006 ----------
12 Oct 2006
The last bitdifference has also been eliminated. Thanks to everybody!
To juz koniec ?
---------- 13:12 ----------
http://boinc.banaan.org/hashclash/forum_thread.php ?id=109
Przez jakiś czas nie mogłem sie nawet zarejestrować, ale sie udało. Niestety nie ma od kilku dni nowych próbek.. czytają tego posta dochodze do wniosku, że od kilkunastu tygodni :)
Ktoś wie kiedy będzie można znów coś liczyć ?
Ile trwa mniej wiecej liczenie takiej próbki na procku AMD 3000 ?
W tym projekcie od dobrych kilku m-cy WU nie widziałem - a z tego co pamiętam jak były - to niezbyt długie - takie do 2 godzin, jeśli mnie pamięć nie myli.
Projekt się zakończył (MD5), podobno mają się przerzucić na SHA-1. Ale czy to prawda to nie wiadomo :?
Ale chyba już jest taki projekt - SHA-1 Collision Search Graz (http://boinc.iaik.tugraz.at/sha1_coll_search/). BTW przedwczoraj dostałem w nim pierwszą porcję WU. Krótkie półgodzinne - dobrze punktujące.
no to sie dopisuje :)
AdNet ważne - [link nie aktualny]CZYTAJ TO!
To już oficjalne:
Cytat: "Marc"
Hello everyone,
First of all, I am happy to announce that I've received my Master's degree cum Laude!
Currently we are still working on MD5 and chosen-prefix collisions.
However, project HashClash is not likely to become active again soon.
If you enjoyed crunching for us on hashfunctions then I would like to suggest this new BOINC project working on SHA-1, the successor of MD5:
http://boinc.iaik.tugraz.at/sha1_coll_search/ (http://boinc.iaik.tugraz.at/sha1_coll_search/)
Thanks for your help!
Kurde szkoda - bo miałem nadzieję, że za jakiś czas puszczą jeszcze jakąś serię WU i zamienię moje 2,8k w tym projekcie do upragnionych 5k.