Gigabyte 2048-bit Encryption Key.

[Eclecticos]

http://img167.imageshack.us/img167/305/ultratpmsg9.jpg

GIGABTE Ultra TPM provides an added layer of security by allowing users
to store their digital signature key on a USB thumb drive, so when they step away
from their system, they can take the key with them, locking up their data
and preventing unauthorized access while they are away.

A computer capable of one million instructions per second
would have to work for 1014 years to crack the key.

Additional Info (http://www.gigabyte.com.tw/FileList/NewTech/2008_mb_newtech/article_10_ultra_tpm.htm).

[Airbozo]

http://img167.imageshack.us/img167/305/ultratpmsg9.jpg

GIGABTE Ultra TPM provides an added layer of security by allowing users
to store their digital signature key on a USB thumb drive, so when they step away
from their system, they can take the key with them, locking up their data
and preventing unauthorized access while they are away.

A computer capable of one million instructions per second
would have to work for 1014 years to crack the key.

Additional Info (http://www.gigabyte.com.tw/FileList/NewTech/2008_mb_newtech/article_10_ultra_tpm.htm).

I worked with a cryptographer at SGI who said most of the estimates of key breaking were based on a linear approach to breaking the key, when in reality most key breaking algorithms do not take a linear approach, rather a random approach.

I was given the example of opening a 4 digit combination lock. If you start at 0000 and increment 1, it will take a while to reach 9999, BUT the combination lies somewhere between those 2 numbers (usually), so the time to actually find the correct combination would be some fraction of the time to reach 9999. If you randomly try different numbers, the time to find the correct combo is decreased. The guy holds several dozen patents for algorithm's and software and 2 for musical instrument circuits.

[blueonblack]

If you randomly try different numbers, the time to find the correct combo is decreased.


Ok, I'm confused. If the number is somewhere between 0000 and 9999, and it has an equal chance to be any number in that range, and the two methods operate at the same speed (calculations per second), why does it make a difference what order the numbers are tried in? Each attempt should have an equal chance of success.

(This is probably why I don't hold any patents.) :)
________
Amber Trichomes (http://trichomes.org)

[Luke122]

If it's random, you COULD hit the number on the first try.

[TheGreatSatan]

If you cut one of the chip's leeds would it be disabled and you'd have access to the system?