Checksum: Difference between revisions
From Computer History Wiki
Jump to navigationJump to search
m Proper cat |
Error detection versus error correction. |
||
| (One intermediate revision by one other user not shown) | |||
| Line 2: | Line 2: | ||
A checksum usually is simply the sum of all the data items in a block of data (hence the name). Some checksums use slightly more elaborate math; e.g. the checksums typically used in the [[TCP/IP]] [[protocol suite]] involve adding carries back in, and those used in [[PARC Universal Packet|PUP]] rotate the sum one position as each word is added in. | A checksum usually is simply the sum of all the data items in a block of data (hence the name). Some checksums use slightly more elaborate math; e.g. the checksums typically used in the [[TCP/IP]] [[protocol suite]] involve adding carries back in, and those used in [[PARC Universal Packet|PUP]] rotate the sum one position as each word is added in. | ||
Checksums typically only provides enough information to ''detect'' an error in the data. Thus it's not a [[Redundant|redundancy]] technique for error ''correction''. | |||
{{semi-stub}} | {{semi-stub}} | ||
[[Category: Computer Basics]] | |||
Latest revision as of 11:05, 2 November 2026
A checksum is a data protection mechanism, usually intermediate in power between parity and a cyclic redundancy check. (In some cases the term 'checksum' is applied to what is techically a CRC.)
A checksum usually is simply the sum of all the data items in a block of data (hence the name). Some checksums use slightly more elaborate math; e.g. the checksums typically used in the TCP/IP protocol suite involve adding carries back in, and those used in PUP rotate the sum one position as each word is added in.
Checksums typically only provides enough information to detect an error in the data. Thus it's not a redundancy technique for error correction.
