New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

feature request: use Moore's Law to calculate long time estimates #1553

Open
roycewilliams opened this Issue Apr 1, 2018 · 4 comments

Comments

Projects
None yet
5 participants
@roycewilliams
Contributor

roycewilliams commented Apr 1, 2018

When requesting a long-running attack, time estimates can be quite long into the future:

Session..........: hashcat
Status...........: Running
Hash.Type........: scrypt
Hash.Target......: SCRYPT:1024:1:1:MDIwMzMwNTQwNDQyNQ==:5FW+zWivLxgCWj...yFcfo=
Time.Started.....: Sun Apr  1 12:53:48 2018 (3 secs)
Time.Estimated...: Fri Aug 25 23:29:29 8175 (6157 years, 146 days)
Guess.Mask.......: ?b?b?b?b?b?b?b [7]

(this is the example 'hashcat' scrypt hash)

It would be nice if Moore's Law could be taken into account for these time estimates. Since hashcat sessions can be saved and restored, it stands to reason that a long-running hashcat job could be moved to different hardware every 18 months or so, such that a hashcat job could run for hundreds or thousands of years. If a highly determined attacker is using this approach, the current time estimates are inaccurate.

Update: Huang's Law may be more applicable for this purpose.

@sundhaug92

This comment has been minimized.

sundhaug92 commented Apr 1, 2018

How'd one predict performance-improvements over such a long stretch? Moore's law seems to already be slowing down but we don't have that many years to extrapolate a model from.

@matlink

This comment has been minimized.

matlink commented Apr 1, 2018

Moore's law has an end. Transistors size is limited by electrons size. You could however imagine a similar law according to parallel computing grow.

@RealEnder

This comment has been minimized.

Contributor

RealEnder commented Jun 15, 2018

For sure Time estimated have to be kept for current user's HW IMO. Maybe another line with that, but you may see coming the next request about quantums (with OpenCL support) ;)

@sundhaug92

This comment has been minimized.

sundhaug92 commented Jun 15, 2018

@RealEnder Fair point but I'm not sure QC would substantially benefit hashing-algorithms, which is usually what hashcat deals with

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment