In a 500 million record dataset a discriminator which picks 1 in a million would be expected to return on average 500 results. I’m not talking about cracking t

In a 500 million record dataset a discriminator which picks 1 in a million would be expected to return on average 500 results. I’m not talking about cracking the SHA1 of the passwords in the database since all the clears are easily available. I’m talking about the value of you sending me the first 5 characters of the SHA1 hash of your password.