search

bee-san / Ares

Automated decoding of encrypted text without knowing the key or ciphers used
MIT License
545 stars 24 forks source link

Use better time estimates for ciphey / cyberchef #136

Open bee-san opened 1 year ago

bee-san commented 1 year ago 
    We should really determine the actual numbers for Ciphey instead of making them up. Ares is so much faster than Ciphey so it doesn't matter but making up numbers doesn't look good.

We should also look at comparing to cyberchef

_Originally posted by @SkeletalDemise in https://github.com/bee-san/Ares/pull/131#discussion_r1044826895_

Zephira58 commented 1 year ago

Not sure if this is good feedback. But a possible way to do it would be to run a few benchmarks across ciphey and cyberchef. Ie get a 1gb file, see how long it takes to decrypt them in groups of 5-10 for each method. Then average out the numbers and put it in a % so you can see the difference in speeds across all 3 options. Then just slap that average % per method to the time it took ares to solve.

So for example you have ae 1gb base64 file. You run that in ares, ciphey, cyberchef 5-10 times. Ares has an average of 15 secconds, ciphey at 29 and cyberschef at 48 which would mean that ares is %63 faster then ciphey and %104 faster then cyberchef. So when a user runs a new scan that takes lets say 5 secconds you can add %63-%104 to that inital 5 secconds to get a much more accurate time difference. While it wont be perfect, it'd be alot better then what we have now.