Slashdot Mirror
DES III starts Today (Upgrade Those Clients!)
David Hallowell writes
" The DES III Contest
is starting today at 5pm (GMT). Members of the distributed.net
RC5 teams should make sure they download the
latest client
as they contain optimised code and a faster switchover to
the DES contest. The DES contest will last for a maximum
of 56 hours and the $10,000 prize will only be available if
it is cracked within 24 hours. After that the prizes are on
a sliding scale until it reaches 56 hours after which there's no prize."
For those of you who don't speak GMT, the contest starts in approximately 1 hour (as of the time of this post) Get yer client!
Do you guys have any idea what OS the Deep Crack machine runs? It would be a waste to run anything other than Linux (Hopefully Unix at least)... :-)
AFAICR a custom system controlled by a Linux-Box. Check their Website.
It's talking about a universal update at 1715 GMT. Is that when it really starts?
Identifier: DES-Challenge-III
Cipher: DES
Start: January 18, 1999 9:00 AM PST
Prize: -
Key:
IV: da 4b be f1 6b 6e 98 3d
Plaintext:
Ciphertext:
bd 0d de 91 99 60 b8 8a 47 9c b1 5c 23 7b 81 18 99 05
45 bc de 82 01 ab 53 4d 6f 1c b4 30 63 3c ee cd 96 2e
07 c6 e6 95 99 9c 96 46 5a 95 70 02 02 70 98 bd 41 c2
88 a9 f0 2f 8b e5 48 20 d2 a8 a0 6b bf 93 de 89 f6 e2
52 fd 8a 25 eb d0 7d 96 83 ee a4 2d c8 8d 1b 71
Arrgh! The ftp/http server is Slashdotted!
Anybody got a mirror?
(linux glibc2-mt)
I just put up a small mirror since ftp.distributed.net seems to be very busy.
;)
You can find some of the linux-x86-versions
(libc5/glibc2, mt or no-mt) at
http://home.student.utwente.nl/w.k.havinga
Idaho - Being to stupid to remember his password
Here is another mirror for multithreaded linux clients (glibc and libc5):
ftp://nucleus.med.upatras.gr/pub/des/
According to: http://www.distributed.net/des/proxyinfo.html
deep crack/ EEF has yet to submit blocks? Maybe they just do it at the end...
Anyone know if it is actually working?
A cracking summary is at http://www.rsa.com/conf99/des/
While according to the first link:
At the current Distributed.Net rate of 95.31 Gkeys/sec, we are doing
15338608 blocks a day.
Current as of Mon Jan 18 20:40:03 1999 GMT
the EFF already cracked it.
I'm impressed with the quality of responses to this question. Where are the legions of flamers who'd have posted stuff like "ITS a DEADICADED MaCHINE STUUPID! There shuld be LISENSES REQIRED to post to ./!!!!"? And, more importantly, how do we keep said legions from returning?
Excuse me while I bask in civilization for a bit.
Mind the Gap
My personal proxy was able to pull DES packets down, but my clients don't seem to want to start them. They were giving that same message about 1715, but that just hit and now they say something like:
[Jan 18 17:15:00 UTC] Buffer update scheduled in 7 minutes 52 seconds
Go figure...
It's a shame that the reward diminishes as the contest progresses.. Oh well. As long as we overtake the Mac team I'll be happy.
You'd think they'd provide clients more than a few hours before they start doing this, especially when their server is so overloaded you can't even connect.
Anyone got a mirror?
The world is neither black nor white nor good nor evil, only many shades of CowboyNeal.
The "total number of blocks cracked by Mac clients" has jumped from 321536935 to 1174254928 in only nine days (Jan 5th to Jan 14)!
To be clearer: 1174254928/321536935 = 3.65..!
The best thing is that I have e-mailed some of the distributed.net people (nugget first, later also dbaker,moose and silby) to ask what's going on. Imagine what have they answered? Nothing.
Am I the only one who doesn't like this?This contest ought to be finished before we have a chance to look at the stats. (Although maybe Nugget'll post the stats after the fact.) The dimishing reward provides a nice incentive to get the contest done quickly. (And shows everyone just how weak DES is.) Distributed.net's stated purpose is "to serve as a gathering point for research and projects related to distributed processing." I'd say that reworking the clients to handle a rapid changeover to a new contest took a lot of thinking and programming.
--Phil (Can't wait until v3. With source available, I might be able to run it at work.)
355/113 -- Not the famous irrational number PI, but an incredible simulation!
As distributed.net's server is hosed, you'd better try a proxy...
Like: jp-proxy.gildea.net
That goes like: rc5des -a jp-proxy.gildea.net and you're off for good! It's much faster...
since a month maybe, my client have LOT of difficulties to connect to a proxy to fetch/flush whatever rc5/des-test it is... i have a lot of bad request or bad thinggy...
if you look the stats page you'll see at the end that there's glitches!
des-iii will be "cracked" by hardware in some hours anyway
--
"Science will win because it works." - Stephen Hawking
Almost certainly nothing approaching a
big-ol' multiuser system. A system like
that, you'd probably either write everything
from top to bottom yourself, or use some
sort of standard RTOS. My money is on the
former, but I don't know for sure.
I'm a nature photographer.
Pretty impressive that they did this before :-)
the problem was actually released.
You're probably confusing this with the
contest six months ago. Check the RSA web
site for more details.
I'm a nature photographer.
at the current rate, if it is possible for the DES to be cracked within the next 30 hours. depending that the numbers don't start to go down, and that they get a slow but steady increase in users to compensate for the few users who will start to drop out...
lets see if my prediction is right...
Its spelt "L-I-N-U-X", but pronunced as "Free Beer"
Only got the Linux glibc2-mt, have fun.
Jeez, I hope I'm not doing anything illigal by distributing this myself.....
Oh well Download Here
Only got mt-glibc2, have fun
Jeez, I hope I'm not doing anything illigal by distributing this....
Oh well Download Here
My client grabbed a DES block at 17:17, finished it, sent it back to the server, and promptly went back to working on RC5.. go figure.
I always mod up spelling trolls.
Ugh, the new DES clients with MMX bitslicing locks up my AMD K6-2 300 / Win98 system hard. I have to use the older clients (build 421 Win32) that use the ByrdDES core. Unfortunately, this is 3x slower than the MMX-bitslicing code! ;)
(Oh well, I'm done with my pity party.)
Trying to join the slashdot team and keep getting server errors when I try to do anything other than view static html pages.
*grumble*
A host is a host from coast to coast...
Unless it's down, or slow, or fails to POST!
From http://www.cryptography.com/des/index.html:
When a search unit finds a potential match, it stops until the the controlling computer (a PC running Windows 95 or Linux) retrieves the key and restarts that search unit.
They have been there for a few days
Does anybody have the solaris/ultrasparc client?
I can't connect to distributed.net (/. effect)
-fb Everything not expressly forbidden is now mandatory.
I'm willing to bet it's a custom operating system. Why port Linux to custom hardware that serves one purpose, and one purpose alone: Take a whole lot of keys, and crack crack crack.
:)
Sure, linux has some great multitasking features, and it's very diverse. But it really isn't very powerful when it comes to multiprocessor support, nor do they really need task swapping...
They said it was all "controlled" by a PC, though. I would imagine that that's a unix box... I can't imagine EFF running Windows
I'm doing my part, 4 PIIs cracking in the office. I'll try to load up some more in the next hour.
Gamertag: ChrisCasey
So how does one join the slashdot team with the new clients?
Natas
http://natas.startx.org
Natas of
-=Pedophagia=-
http://www.mp3.com/pedophagia
Also Admin of
http://loki.linuxgames.com