ZTree.com  | ZEN  | About...  

 Index   Back

[Discuss] 23, 302, 77164   [Discuss]

By: Walter Rassbach       
Date: Nov 03,2000 at 09:00
In Response to: [Discuss] 23, 302, 77164 (John Gruener)

> [snip]

> What I meant was, in the context being discussed,
> if the file sizes were different you would already
> know that the files were different. So we would be
> applying this CRC test only to sets of files that
> were all the same size. CRCs that matched on files
> of different sizes, (and of course there could be
> many), would not enter into the CRC comparison.

Yes, but files of the same size result in less random CRCs...

> This is the point that I, (and I think Rudy), were
> trying to get to. There would not ordinarily be
> tens of thousands of files all the exact same
> size. That would be a very special circumstance,
> and then of course the CRC would not suffice.

> Very interesting.

> Yes, I think we'd have to assume that no one was
> tampering with the CRCs.

That is not an assumption I would be comfortable with -- It may not be intentional, but *anything* that puts out patterned data is essentially tampering with the CRC disributions...

> Well I didn't feel it would affect me, but I took
> the opportunity to trade in the processors anyway.
> So I guess that's a good point. ;-)

> - John

865 views      
Thread locked
 

Messages in this Thread

 
96,637 Postings in 12,231 Threads, 350 registered users, 58 users online (0 registered, 58 guests)
Index | Admin contact |   Forum Time: Mar 29, 2024 - 5:04 am UTC  |  Hits:62,393,093  (7,879 Today )
RSS Feed