home *** CD-ROM | disk | FTP | other *** search
- Newsgroups: comp.sys.mac.misc
- Path: sparky!uunet!mcsun!news.funet.fi!funic!nntp.hut.fi!vipunen.hut.fi!tiger
- From: tiger@vipunen.hut.fi (Karl R. Tigerstedt)
- Subject: Re: CRC vs. Checksum
- Message-ID: <1992Jul28.055536.18052@nntp.hut.fi>
- Sender: usenet@nntp.hut.fi (Usenet pseudouser id)
- Nntp-Posting-Host: vipunen.hut.fi
- Reply-To: tiger@vipunen.hut.fi (Karl R. Tigerstedt)
- Organization: Helsinki University of Technology
- References: <9414.2A6FC102@psycho.fidonet.org>
- Date: Tue, 28 Jul 1992 05:55:36 GMT
- Lines: 25
-
- In article <9414.2A6FC102@psycho.fidonet.org> Jim.Kauffman@f20.n3603.z1.FIDONET.ORG (Jim Kauffman) writes:
- >Does anyone know the relative merits of checksum vs. CRC for detecting
- >error(s) in a block of data? I would like to calculate the probability of
-
- Checksumming usually means using a 8bit counter for addition. Bytes are
- received and their value is added to the 8bit counter which overflows over and
- over, when FFh is reached. This method is ok, as long as not two (or several)
- bytes in the data stream change value. Example:
-
- data bytes changed value (corrupt data)
-
- 1Ah 1Bh
- 82h 81h
- ---------------------------------
- 9Ch 9Ch (= checksum)
-
-
- CRC checking usually implements checksumming as well as a combination
- of left and right bitshifting (rotation) of the received data. This method
- would therefore detect the error example above.
- --
- ---------------------------------------------------------------------------
- "I have only come here seeking knowledge.
- Things they would not teach me of in college" - The Police
- *** Karl Tigerstedt , email tiger@vipunen.hut.fi ***
-