Any error correction gurus in the audience?

---snip--- Hi The CRC operation is deterministic. That means one can play it forwards and backward. The are two main ways of doing the corrections. One is to break it into its prime factors and use the part with the largest span of zeros in the polynomial to determine the error mask and the other partials to determine the offset, based on the Chinese remainder theorum. It is a little complicated to describe but is what most of the crc chips used. The other method is what you are seeing, playing the crc backwards with no data coming in ( zeros for the data ) until the eror mask is seen. the applying that mask to the data. What I mean by the error mask is that in the polynomial, there is a span of zeros. If you play it backwards until there are zeros in all but the center portion where the zeros are, you have found the error mask. Shifting it until is would be at the input edge of the polynomial would give you the offset. When playing backwards, you are only un-inputting zeros. To tell you how to do it, you do the shift and xor in reverse order. Run a CRC with some data in the sum, with zeros as input data forwardwards and you'll see how on one can run it backwards by shift and xor to get the same values going backwards. If you go beyond the data size backwards without finding an error mask, the error burst was too large and could not be corrected. I hope this makes some sense? Dwight