On May 21, 5:53*pm, David Antliff <david.antl...@gmail.com> wrote:
> Let me try and describe the issue fully.

A quick followup for this.

After consulting with someone at Xilinx, it's clear that the .ngd file
is encrypted. Assuming it contains a timestamp, this explains why it
is always very different.

I have been told the correct thing to do is to compare the reported
Checksum (in the .par file for Xilinx 10.1, and the .map file for
Xilinx 11.1). I have confirmed that although the output from ngdbuild
and everything after this is different, that reported Checksum is the
same.

I was advised to omit the .ise file entirely from an automated system,
which was our original goal.

Thanks to everyone for your replies.

-- David.