19-Apr-2024 18:09 GMT.		 UNDER CONSTRUCTION
Anonymous, there are 33 items in your selection
[News] RC5-64 HAS BEEN SOLVED!ANN.lu
Posted on 26-Sep-2002 20:30 GMT by Christophe Decanini33 comments
View flat
View list
Your Amiga (or other) can rest now :)
Too bad the key was not found by the Amiga team. Anyway we will have more power for new contest with the AmigaOne/Pegasos/Amithlon.
Here are the details.
RC5-64 HAS BEEN SOLVED! : Comment 1 of 33ANN.lu
Posted by Christophe Decanini on 26-Sep-2002 18:34 GMT
I forgot to mention that you can find information about the Amiga team here:
http://distributed.amiga.org
RC5-64 HAS BEEN SOLVED! : Comment 2 of 33ANN.lu
Posted by Gregg on 26-Sep-2002 19:22 GMT
Isn't it entertaining that they had the answer but didn't know it for a month? There's several important lesson buried not very deep there...
Gregg
RC5-64 HAS BEEN SOLVED! : Comment 3 of 33ANN.lu
Posted by reflect on 26-Sep-2002 19:43 GMT
hm.. are there any other projects like this that has clients for the amiga?
dnetc was one of the few projects that actually covered all my clients (and I have some very rare combos). seti did a valiant effort a while, but discontinued so many clients after a while that too many of my machines was out of work. Rather not start several projects just to keep all the machines busy..
RC5-64 HAS BEEN SOLVED! : Comment 4 of 33ANN.lu
Posted by Paul Hill on 26-Sep-2002 19:46 GMT
In reply to Comment 2 (Gregg):
I never understood why people would want to spend CPU cycles cracking a 64bit
encryption method. I mean what's the point? Everyone knows that given enough
CPU power it can be done. I'm supprised it took so long. But it will only
take 1,280 years to crack RC5-72 at this rate!
Far better to spend those extra cycles doing something productive like
encoding MP3s :)
RC5-64 HAS BEEN SOLVED! : Comment 5 of 33ANN.lu
Posted by cOrpse on 26-Sep-2002 20:09 GMT
In reply to Comment 4 (Paul Hill):
"I never understood why people would want to spend CPU cycles cracking a 64bit
encryption method. I mean what's the point? Everyone knows that given enough
CPU power it can be done."
IIRC there was a prize for the winning key :) ... its also a very good blizzardPPC stability burn-in test .. my records being 2seconds -> 20 minutes from when i got it to about a month ago and now it doesn't stop until its stopped :)).
I managed to clim my way thru the rankings using my amiga and pc quite easilly but my pc had a partition eatting moment after 2 or so months so i gave up.
"I'm supprised it took so long. But it will only take 1,280 years to crack RC5-72 at this rate!"
That doesn't take into count the increasing speed of CPU's ...
"Far better to spend those extra cycles doing something productive like
encoding MP3s :)"
even better ... encoding high quality divxs or lightwave rendering :))
RC5-64 HAS BEEN SOLVED! : Comment 6 of 33ANN.lu
Posted by John Niclasen on 26-Sep-2002 20:11 GMT
Could be interesting to calculate how much electric power was used in the effort. How much coal/oil/wood was burned, how much pollution!?
RC5-64 HAS BEEN SOLVED! : Comment 7 of 33ANN.lu
Posted by Anonymous on 26-Sep-2002 20:19 GMT
In reply to Comment 4 (Paul Hill):
Actual experience tells us that people have so much to worry about that they tend to ONLY prioritise things which can be identified as a concrete threat. If your philosophy or business model requires a change in priority you will benefit from something that makes the threat concrete.
Examples of the former experience (anti-conspiracy theorists might want to consider the latter sentence too):
* OpenSSL bugs were detected some time ago. Despite the expense of having auditors find these bugs many users did not upgrade. The Slapper worm made the security threat concrete and so NOW many people have upgraded.
* Government funding for defense and intelligence, and government rights to infringe on civil liberties are dependent on a sense of foreign "threat" from the citizens. Some guys crash a plane into a skyscraper and -- just like that the majority falls into line.
* Vaccination programs for common childhood diseases are universally available in wealthy industrialised nations. Yet the %age of children who actually attend and are vaccinated increases significantly for a while after disease epidemics.
So, there were two (conflicting) motives at work here. RSA sponsored the contest to re-assure customers that RC5 is strong. If it was weak someone would quickly have won the prize. OTOH the D.net participants mostly wanted to show that RC5 can and will genuinely be attacked if the prize is worthwhile. This encourages people to support research into improved crypto and (in the past) legislation to permit universal access to strong crypto.
RC5-64 HAS BEEN SOLVED! : Comment 8 of 33ANN.lu
Posted by Mike Veroukis on 26-Sep-2002 23:02 GMT
In reply to Comment 7 (Anonymous):
That was a very nicely put together bit about human nature, politics and consperacy nonesense. I can see why certain companies might be interested in this, however the original question was really unanswered.
"I never understood why people would want to spend CPU cycles cracking a 64bit
encryption method."
Why would anyone do this? The answer is actually quite simple; they had nothing better to do (to put it nicely).
- Mike
RC5-64 HAS BEEN SOLVED! : Comment 9 of 33ANN.lu
Posted by Christophe Decanini on 27-Sep-2002 00:02 GMT
In reply to Comment 8 (Mike Veroukis):
"Why would anyone do this? The answer is actually quite simple; they had nothing better to do (to put it nicely)."
Well you can still be very busy and also keep your CPU busy.
There are millions of computers hooked on internet that spend their CPU time to idle. If distributed computing can be used to do better things than proof of concepts now why not ?
RC5-64 HAS BEEN SOLVED! : Comment 10 of 33ANN.lu
Posted by Anonymous on 27-Sep-2002 01:07 GMT
In reply to Comment 8 (Mike Veroukis):
<sigh> Is Mike actually a bot that consults a list of known wrong statements to spew? I only ask because it seems as though this is the third time in as many days that I've had to step in after he says something dumb :(
The reasons cited by distributed.net (who ought to know, they've spent more of their free time on this than anyone) include:
1. To prove that 64-bit encryption is insufficient -- exactly what I wrote, if only you'd cared to read it properly
2. To explore the feasibility of cooperative networked multiprocessing
3. Because it's fun
4. Because you can win money!
RC5-64 HAS BEEN SOLVED! : Comment 11 of 33ANN.lu
Posted by Álmos Rajnai on 27-Sep-2002 04:49 GMT
In reply to Comment 6 (John Niclasen):
"...how much pollution!?"
Now, that is an interesting question... :)
I am afraid it is nearly impossible, but would be a good opportunity to show to the mass, how much energy wasted for pure luxury.
Just for the record: to lit a normal 100W bulb takes almost 3 times more energy than continuously using an anverage processor... ;)
RC5-64 HAS BEEN SOLVED! : Comment 12 of 33ANN.lu
Posted by reflect on 27-Sep-2002 05:04 GMT
In reply to Comment 4 (Paul Hill):
why people do rc5?
there are several reasons, one is 'I can' and another one is 'I'm a stats-hooker'. some people buys flashy cars to measure their dicks. this is a serious geek-measurement for some. after wasting some 4 years on rc5, I'm determined to do something better with my cpu-time. like distributed folding that find proteins and generally helps scientists find cures for diseases.
RC5-64 HAS BEEN SOLVED! : Comment 13 of 33ANN.lu
Posted by Ole-Egil on 27-Sep-2002 07:10 GMT
In reply to Comment 12 (reflect):
And all these other, more useful projects was started because the world realized that distributed computing actually works. distributed.net started that with their crypto projects. I would rather see all these projects as packets dowanloadable with the distributed.net client than having to install new executables for each and every project, though. Would limit my paranoia somewhat ;-)
RC5-64 HAS BEEN SOLVED! : Comment 14 of 33ANN.lu
Posted by alan buxey on 27-Sep-2002 08:36 GMT
In reply to Comment 4 (Paul Hill):
One of the reasons why i did it was because it was a distributed computing
effort with great statistics and a table that showed Amiga as a computer
OS with PowerPC and 68k results ahead of many other OS/CPU combos.
it also allowed out team to be in the top 10.
OGR is still running (another distributed.net challenge) and we have
an AmigaOS client - but i'd really only advise PPC people to run this.
we're desperately looking for a client for one of the other distributed
challenges - folding@Home, SETI etc etc all need Amiga's helping them 8-)
alan
RC5-64 HAS BEEN SOLVED! : Comment 15 of 33ANN.lu
Posted by alan buxey on 27-Sep-2002 08:40 GMT
In reply to Comment 6 (John Niclasen):
....and most of these machines were on for their natural
'on time' anyway - i never switch off my computer....too
much wear and tear and heat cycles for the components.
anyway, its impossible to calculate the total energy. sure,
you can take the CPU info and calculate how much energy
was used by each CPU type at each clock speed....but what
about the rest of the system? was it a basic PC..or
did it have a 40W gfx card? did it have an energy efficient
PSU...or a TFT monitor? did the user have energy saving
enabled for other parts...or was the network card always
active..and hence the switch and cable-box always on?
it was (and still is) a very good exercise and trial for the future
of computing
alan
RC5-64 HAS BEEN SOLVED! : Comment 16 of 33ANN.lu
Posted by alan buxey on 27-Sep-2002 08:42 GMT
In reply to Comment 11 (Álmos Rajnai):
..not if the average processor is a 2 GHz Athlon or P-IV !!!! ;-)
RC5-64 HAS BEEN SOLVED! : Comment 17 of 33ANN.lu
Posted by alan buxey on 27-Sep-2002 08:44 GMT
In reply to Comment 13 (Ole-Egil):
...this is what the COSM project is about. a central
framework client which is really a small CPU-native OS
which can then run any code that is based on COSM
on the client machine....its the way this NEEDS to
go to get as many different platforms involved without
the effort needing MacOS, AmigaOS, Windows, Linux, BSD etc
knowledge. only the COSM code needs to be ported to the
OS...so, we MUST have COSM on AmigaOS !
alan
RC5-64 HAS BEEN SOLVED! : Comment 18 of 33ANN.lu
Posted by reflect on 27-Sep-2002 08:52 GMT
In reply to Comment 17 (alan buxey):
could you provide us with links to this COSM?
RC5-64 HAS BEEN SOLVED! : Comment 19 of 33ANN.lu
Posted by Anonymous on 27-Sep-2002 10:00 GMT
Excellent.
This project has proved just how secure RC5-64 actually is.
It took thousands of computers working in parallel for over two YEARS to crack one single message. A message of which (and this is the important part), they had been given the start!
Yes. Amazing. 2 YEARS and thousands of computers to crack an encrypted message when they already knew part of the message.
Remind me how that proves to Joe Sixpack how RC5-64 is supposed to be insecure?
RC5-64 HAS BEEN SOLVED! : Comment 20 of 33ANN.lu
Posted by Hagge on 27-Sep-2002 10:34 GMT
The best part is:
"Our peak rate of 270,147,024 kkeys/sec is equivalent to 32,504 800MHz Apple PowerBook G4 laptops or 45,998 2GHz AMD Athlon XP machines or (to use some rc5-56 numbers) nearly a half million Pentium Pro 200s."
8.311 kkeys/s for PPC G4 800MHz
5.873 kkyes/s for AMD Athlon XP 2GHz
needs those G5s/IBM new PPC cpu.
Anyway, G410.39 kkeys/s and MHz
Athlon XP2.94 kkeys/s and MHz
Dual G4 powermac@867MHz$1.699 gives 18016 kkeys/s or comparable to 6.1GHz Athlon XP. Maybe PowerMacs are worth their price anyway? =D
RC5-64 HAS BEEN SOLVED! : Comment 21 of 33ANN.lu
Posted by reflect on 27-Sep-2002 10:53 GMT
this was BRUTE force, there are other, more efficient ways to crack a code, which I'm sure you realize once you stop and think about it.
Not to mention special chips manufactured with just one purpose in mind, the rc5 algorithm. Anyone remember the DES-Cracker? it was pretty efficient afair, although on a different contest.
RC5-64 HAS BEEN SOLVED! : Comment 22 of 33ANN.lu
Posted by Ole-Egil on 27-Sep-2002 11:04 GMT
In reply to Comment 20 (Hagge):
Hehe, "why would Amiga go ppc, x86 is so much faster" ;-)
Ok, I realize this is the Altivec doing 90% of the work, but altivec is really not out of the question :-)
RC5-64 HAS BEEN SOLVED! : Comment 23 of 33ANN.lu
Posted by Linus G on 27-Sep-2002 11:16 GMT
Now everybody ( with a windows box ) can visit www.ud.com and join their program for finding a cure for cancer. And while you´re at it, join the amiga team.
RC5-64 HAS BEEN SOLVED! : Comment 24 of 33ANN.lu
Posted by Christophe Decanini on 27-Sep-2002 13:30 GMT
In reply to Comment 19 (Anonymous):
Well,
We had a guy in France that cracked the Visa card encryption.
If I remember well he was dividing the encryption into multiple smaller one thanks to prime numbers.
With an ATM machine, a few PCs, he managed to build a visa card that would work with any PIN, transfer the money to the seller without withdrawing it from any account.
He went to visa and ask big money to fix their encryption. He was then caught and sent to jail for having used his card buying subway tickets as a proof of concept to show VISA that he was able to crack it.
RC5-64 HAS BEEN SOLVED! : Comment 25 of 33ANN.lu
Posted by The_Editor on 27-Sep-2002 14:18 GMT
In reply to Comment 12 (reflect):
My cancer research project (united devices .. Liganfit) is currently at 42390.
I hope my results help to finding cures for this disease.
RC5-64 HAS BEEN SOLVED! : Comment 26 of 33ANN.lu
Posted by Anonymous on 27-Sep-2002 14:39 GMT
In reply to Comment 24 (Christophe Decanini):
And this is relevant to 64-bit RC5 how?
different encryption!
BTW, I'm not surprised he was put in prison - stealing money after demanding large sums of cash to tell them about the problem - theft and blackmail are rather serious crimes...
RC5-64 HAS BEEN SOLVED! : Comment 27 of 33ANN.lu
Posted by Mike Veroukis on 27-Sep-2002 14:50 GMT
In reply to Comment 10 (Anonymous):
>Is Mike actually a bot that consults a list of known wrong statements to spew?
>I only ask because it seems as though this is the third time in as many days
>that I've had to step in after he says something dumb :(
And who might you be? If you're gonna take a shot at me at least have the balls to reveal yourself. Btw, careful about your assumptions, my last post here was a couple days ago (pheonix project) and the post before that was a few days before that (OS4 idea screnshots). There's more then one "Mike" here buddy.
Look, everyone knows keys can be broken, it's just a matter of time. You can even calculate how long it would take with given resources. What's the big deal??? I just wonder how many people left their systems on all day crunching numbers for RSA. I wonder how much electricity this project burned world wide.
- Mike
RC5-64 HAS BEEN SOLVED! : Comment 28 of 33ANN.lu
Posted by Christophe Decanini on 27-Sep-2002 16:29 GMT
In reply to Comment 26 (Anonymous):
He told them for month their systems could be cracked (and offering help)
They thought he was bluffing.
Then He said he cracked it.
They thought he was bluffing.
Then he bought the less expensive stuff he could buy and tell them about the transaction so they could really check if he really cracked it.
Then he was sued.
That how in France we use very smart people :(
RC5-64 HAS BEEN SOLVED! : Comment 29 of 33ANN.lu
Posted by Mike Veroukis on 27-Sep-2002 16:48 GMT
In reply to Comment 28 (Christophe Decanini):
Funny thing is, he probably could have bought himself a few high end speakers, amps, TVs, fridges, or what-have-you and got away with it. Now THAT's sad.
- Mike
RC5-64 HAS BEEN SOLVED! : Comment 30 of 33ANN.lu
Posted by alan buxey on 28-Sep-2002 17:03 GMT
In reply to Comment 18 (reflect):
COSM can be found at:
<a href="http://www.mithral.com/projects/cosm/">http://www.mithral.com/projects/cosm/</a>
its been founded by the old main players of d.net
alan
RC5-64 HAS BEEN SOLVED! : Comment 31 of 33ANN.lu
Posted by anarchic_teapot on 28-Sep-2002 19:36 GMT
In reply to Comment 25 (The_Editor):
"My cancer research project (united devices .. Liganfit) is currently at 42390. "
Haha! 48640.. but Ligandfit runs slower than Think on my PII, so I guess this is where the big fast CPUs with hairy chests and tattoos really streak ahead.
Pity it's a Windoze-only project. There might be more hope with seti@thome, since it already runs on different systems: perhaps the CLI-only *nix version could be ported if we asked nicely and in sufficient quantity to be heard.
RC5-64 HAS BEEN SOLVED! : Comment 32 of 33ANN.lu
Posted by Phill on 29-Sep-2002 13:23 GMT
In reply to Comment 24 (Christophe Decanini):
> With an ATM machine, a few PCs, he managed to build a visa card that would
> work with any PIN, transfer the money to the seller without withdrawing it
> from any account.
You must have alot of offline machines in france.
In the UK it's very uncommon for you to be able to use a credit card without it verifying your details online. The PIN has never been stored on any cards, although a hash of the PIN used to be on some cards to reduce traffic when the number was obviously wrong. This was taken away because they decided they'd rather have it dial up when you did enter the wrong PIN :-) In the US they do online referals when you buy a sandwich.
The magnetic strip cards are too easy to mess about with, cloning cards is very common. I had mine cloned last year & had alot of hassle getting them to remove the bogus transactions.
Hopefully the new smart cards are going to stop this, not all retailers in the uk have switched over yet though.
RC5-64 HAS BEEN SOLVED! : Comment 33 of 33ANN.lu
Posted by The_Editor on 29-Sep-2002 18:01 GMT
In reply to Comment 31 (anarchic_teapot):
Ole-Egil spake:-
Pity it's a Windoze-only project. There might be more hope with seti@thome, since it already runs on different systems: perhaps the CLI-only *nix version could be ported if we asked nicely and in sufficient quantity to be heard.
Yeah ..Agreed
btw .. I'm running PIII - 1Ghz ... Ahm a comin ta git Ya !! :)
Anonymous, there are 33 items in your selection
Back to Top