in reply to Re: On showing the weakness in the MD5 digest function and getting bitten by scalar context
in thread On showing the weakness in the MD5 digest function and getting bitten by scalar context
Edit: tachyon updated to fix. Leaving my comment for posterity, but it no longer applies.
Sorry Tachyon, but you're wrong. In roughly 2^16 tries, you've got 50% odds of finding a collision - not 50% odds of finding a collision with a given result. So you have 50% odds of two numbers in the 2^16 tries colliding, but you do not have 50% odds of a number in the 2^16 tries colliding with the result of one specific output.
You've mistaken "collision with a given point" with "collision between any two points." ++ for bringing up the birthday paradox, anyways.
Sorry Tachyon, but you're wrong. In roughly 2^16 tries, you've got 50% odds of finding a collision - not 50% odds of finding a collision with a given result. So you have 50% odds of two numbers in the 2^16 tries colliding, but you do not have 50% odds of a number in the 2^16 tries colliding with the result of one specific output.
You've mistaken "collision with a given point" with "collision between any two points." ++ for bringing up the birthday paradox, anyways.
In Section
Meditations