Re: I know this won't work, I just want to know -why-.
- From: Willem <willem@xxxxxxxx>
- Date: Mon, 26 Dec 2005 10:52:25 +0000 (UTC)
dalek_no2@xxxxxxxxxxx wrote:
) I know that you can use redundancy data to recover an equivalent amount
) of data lost due to errors.
Yes, but you need to know which of the data was lost. If you generate a
random file, it might match the original for roughly 50%, but you don't
know which bits are right and which bits aren't.
) Indeed, this is a common practice in any
) alt.binaries newsgroup, as almost every binary post is accompanied by
) .par or .par2 files which serve this function. However, neither of
) those recovery formats fixes single-bit-errors. Or, more accurately,
) they do, but they do so by fixing a larger chunk of the data, including
) data which may have been correct and error-free.
That's because the data is checked for correctness in chunks, using an MD5
hash if I recall correctly.
However, to address compression-using-recovery-data: You can easily
calculate the theoretical maximum amount that can be recovered from a
certain amount of data, by contriving a compression algo using this
recovery algorithm, and then assuming it will break even.
SaSW, Willem
--
Disclaimer: I am in no way responsible for any of the statements
made in the above text. For all I know I might be
drugged or something..
No I'm not paranoid. You all think I'm paranoid, don't you !
#EOT
.
- References:
- I know this won't work, I just want to know -why-.
- From: dalek_no2
- Re: I know this won't work, I just want to know -why-.
- From: Nicholas Sherlock
- Re: I know this won't work, I just want to know -why-.
- From: dalek_no2
- I know this won't work, I just want to know -why-.
- Prev by Date: Re: Arithmetic coder input
- Next by Date: Re: I know this won't work, I just want to know -why-.
- Previous by thread: Re: I know this won't work, I just want to know -why-.
- Next by thread: Re: I know this won't work, I just want to know -why-.
- Index(es):
Relevant Pages
|