> This same question (I think) came up last January. Here is the answer that I
> gave at that time:
Thanks for your quick reply. Sorry, I googled through the archives, but did not
find this one. The "zigzag" idea is the correct answer, as this order is
required for the JPEG DQT segment. So Live555 is absolutely correct here, but
it turns out, that the code in Appendix A of RFC 2435 omits to use the
reordered tables when creating jpeg headers, ie. it contains a bug. I just
could not believe that a RFC code example can contain a bug.
Thanks.
Best regards
Alex
_______________________________________________
live-devel mailing list
live-devel@lists.live555.com
http://lists.live555.com/mailman/listinfo/live-devel
