Topic: Distributed Computing For Finding The Key? (Read 223 times)

Surrido · « **on:** July 05, 2006, 02:46:00 AM »

i have been following setiathome as well as the new BOINC approach of transforming millions of PCs into one large cluster for some years now...

would it be possible to start a BOINC project for getting the xbox 360 key?

BOINC website

what do you think?

gonkle · « **Reply #1 on:** July 05, 2006, 02:56:00 AM »

[joke]
it would be easier and cheaper to hire some Mi5 Agent who steals the key direct at the M$ HQ
[/joke]

Surrido · « **Reply #2 on:** July 05, 2006, 02:56:00 AM »

is "finding" the key illegal??? I dont know. i doubt that i personally would not be allowed to krack the key to my xbox that i own.

what source code do you mean?

@ gonkle: why should it be so hard. this is a very hard key, but with enough computing power it can be done.

Surrido · « **Reply #3 on:** July 05, 2006, 03:02:00 AM »

BOINC is open infrastructure. you can get all u need there!

http://boinc.berkele...source_code.php

Surrido · « **Reply #4 on:** July 05, 2006, 03:05:00 AM »

keep me updated...

any other helpers welcome!

Surrido · « **Reply #5 on:** July 05, 2006, 03:56:00 AM »

i found some information on who long it takes....
using BOINC to crack keys

gonkle · « **Reply #6 on:** July 05, 2006, 05:18:00 AM »

if you develop a linux client, im in there and there will be enough people, but the attempt to crack the xbox1 key via distributed computing ended unsuccessfull.

Surrido · « **Reply #7 on:** July 05, 2006, 05:39:00 AM »

thats 4 years ago :-)

Textbook · « **Reply #8 on:** July 05, 2006, 10:15:00 AM »

This is for cracking the Xbox 360 PIRS signature/encryption, right? If so, count me in.

bonevichio · « **Reply #9 on:** July 05, 2006, 12:05:00 PM »

FORGET ABOUT IT!~

I've taken some excepts from previous threads (You should do some forum searching).

http://forums.xbox-s...hp/t469197.html

In a nutshell: if you multiplied the number of computers on the planet by a trillion, and made them all a trillion times faster, and networked every single one of them together and used the resultant cluster to try to brute-force RSA... it wouldn't take millions of years. It wouldn't take billions of years, it wouldn't take trillions of years. It would take unimaginably more than a googol googol years. Even if every single subatomic particle in the universe were in actuality a computer a trillion times faster than today's computers, and even if all of them were networked into one gigantic cluster, it would still take many many times longer than the total age of the universe to complete.

The sheer genius of RSA-2048 encryption is that it is a googolplex times easier to easier to create a keypair than it is to crack the resultant signature. The numbers are so big that it would probably take the lifetime of the universe to find them. Unless you start from the position of knowing what the two factors are, you'll never live to see the encrytion broken, no matter how much you apply Moore's Law. It didn't happen on the X-Box 1, and it probably won't happen on the X-Box 360

I'm tired of copy & pasting....just read the whole thread.

Havok · « **Reply #10 on:** July 05, 2006, 12:54:00 PM »

Thanks bonvichio,

I was about to post the same thing but you did such a good job only one thing left to say...

Ain't gonna happen.

Surrido · « **Reply #11 on:** July 06, 2006, 02:21:00 AM »

for a 2048 bit module you need 1015 MIPS years (million instructions per second)

one MIPS year is 30 * 10^12 instructions.

a 3GHz PC has 109 Instruktions/Second = 103 MIPS = 1 GIPS

it needs for a MIPS year 8 hours 20 minutes

we need 1015 MIPS years. that means that 1015 2Ghz PCs will find a 2048 bit key in 8 hours and 20 minutes....

those are the numers for factoring a key.

if you can factor the key you can easyly find the correct key. that would mean that it is borken and you can make unauthorised signatures...
did i make a mistake somewhere?? sounds to easy.... please help

IF we had a way to make a cluster / distrubuted computing project with THOUSANDS of PC, we will have the key before the universe blows up

Edit: we dont have a real time system here. we can crunch on ONE key for as long as it takes

Surrido · « **Reply #12 on:** July 06, 2006, 02:47:00 AM »

ok, i am totally wrong... it takes long, but its not impossible...

what happened to project x?

Havok · « **Reply #13 on:** July 06, 2006, 08:52:00 AM »

Your math is way off.

RSA512 needs 8000 MIPS-years, and RSA2048 requires
9 x 10^15 times as much.

So you need 7.2 x 10^19 MIPS Years for a 2048 bit key...

You quoted 1015 MIPS Years.. you are only off by a factor of 100,000,000,000,000,000.

Surrido · « **Reply #14 on:** July 07, 2006, 02:29:00 AM »

bad sources? the source i used said 1015 MIPS years and another source stated that 2048Bit would not last the next 10 years before it can be factored...

where did u get your calculations from?

what about probabilities? we dont need to find the key at the end of the search. we can find it earlier...
all u need is a strong CPU network. i dont know how big xbox community is, but i think it should be feasible to get a million users. with current CPU power available we would have a hell of computing power.

Author Topic: Distributed Computing For Finding The Key? (Read 223 times)