Not that bright....just borne out of many many trials of working with 93C46's over the years and finding something that worked well and minimized processor time overhead bit-banking the serial interface.

BTW,,,,If anybody is interested in looking at the test code I wrote to validate the robustness of the check codes across small data sets when one byte is compromised I would be happy to email it to you. I have recently updated this to test packets of 5 bytes, 4 data bytes plus 1 check byte. It is written as C code and I compiled it as a WIN32 console applet using Microsoft VC++ Ver 6. I let it run for 12 hours and it did not complete !!! Too many combinations I guess.

Michael Karas