a couple questions

[trueimage]

first is: which projects (if any) have "portable" or standalone clients that don't require installing, so I could use them at work etc.

second: are there any projects that could be easily modified to run my own dc project (ie brute forcing a random number generator) I am trying to write one that will be very hard to bruteforce and I wanted to be able to try to break it myself, using a distributed attack (ie 5-10 pcs)

thanks

 

[Assimilator1]

DPAD will run straight off as it doesn't have an install routine.So you could run it off a flash drive.
Though I gotta ask ,you do have permission to run it at work right?

Dunno about your 2nd question ,though your idea sounds similiar to RC5 (64?) ,& that took thousands of computers years to crack!

 

[trueimage]

I haven't asked about running it yet because I couldn't find a "portable" one, though I will ask of course.

The # of seed possibilities for the RNG wont be near as great as RC-5, it is just a personal project.

 

[BlackMountainCow]

DPAD indeed is a good option without installing stuff.

As far as #2 is concerned, I doubt that there are any that can "easily" be modified. Most of the time, the scientific application routines are not open to the public. In theory, though, you could set up your own BOINC project, if your coding skills are up to that task. Although for only 5-10 PCs that would be a lot of work to do.

 

[Ken g6]

As far as #2, if you have a command-line program that does what you want, and that can read a list of inputs from standard input and write results to standard output, you could use a program I wrote called TCPPipe. If you have a medium-sized amount of data to run (at least 32k*number of processors), that should work for you nicely.

Edit: Usually, though, for these encryption cracking things, if you know the algorithm, you can get a fairly good guess of how long it will take to crack by brute force. If you run a program on one machine for a few seconds, and see how many keys it tried, you can get a nearly perfect estimate.