Re: Q: How to test new hash algorithm?
- From: Martin Hvidberg <Martin@xxxxxxxxxxxx>
- Date: Sat, 28 Jan 2006 15:43:29 +0100
Kristian Gjøsteen wrote:
What properties do you want from your hash function that standard cryptographic hash functions do not provide?
None other that I need it to be running in a programming language called Avenue, this means I have to write one myself since none are included. I just need to check if two larger strings are identical, without actually sending the long stings off to the other end.
It should not be anything fancy, and quick to impliment.
I was just curious if the aproach I had in mind was hopeless or not, and wondered how to test that.
The output distribution depends on the input distribution, so the question does not immediately make sense. But waving hands, most "not-specially-chosen" input distributions get taken to a uniform distribution by the standard cryptographic hash functions.
That makes sence... I guess an optimal hash would change all allong the output string (or change all the digits) if you change just a singel char in the input.
I have inplmented the 'hash function' I had in mind and it works perfect to meet my needs - Thanks for the feed back
------8<------------- A: Because they make things impossible to read. Q: Why? A: Top postings. Q: What is most annoying in NGs? .
- Prev by Date: Re: Q: How to test new hash algorithm?
- Next by Date: Re: Q: How to test new hash algorithm?
- Previous by thread: Re: Q: How to test new hash algorithm?
- Next by thread: Re: Q: How to test new hash algorithm?