<BR>  Hello !<BR><BR>> <BR>> I conducted a few testso compare the performance of different<BR>> comparison methods.ested using string comparison, the zlib<BR>> library's crc32 checksum, and the Digest::MD5 hash. The file is<BR>> iterated over in chunks and the 1K, 10K, etc refer to the size of the<BR>> chunks. There is also a whole file measure for each of them.<BR>> <BR>> The test files were identical Ogg Vorbis audio files just below 8MB in<BR>> size (identical files should give worst-case performance). Times are<BR>> for 100 repetitions.<BR><BR>  I'm sorry, but I don't quite understand the importance of the tests: it is obvious that if you have both files at hand, it will be faster to compare them byte by byte, as you need anyway to readvery single byte to compute the MD5 or CRC32 of a file.<BR><BR>  The latters are much more handy when you have only one file at hand, that is one file you downloaded and one file on the remote server,hose md5 you provide. Then, you don't need to download the file again to make sure nothing happened during the download !<BR><BR>Or do I completely miss your point ? <BR><BR>  Cheers !<BR><BR>      Vince