This summarises the responses that I have received.
The original posting was to net.micro.hp: this summary goes
to that group and comp.sys.hp. (Sorry adms for the duplicate
posting rejections, but times are tough on the net right now!)
First, 3 people within HP sent me the newconfig stuff on the
tape. It appears that there is little beyond *huge* numbers of
bug fixes and extensions. FORTRAN has clearly been worked on
very hard, 2 Meg memory cards are now supported, NLS is extended,
etc. etc. The only comment I found that (fortunately) doesn't
apply to me but might be worth passing along is:
> There are two unsupported configurations in which the system
> will not boot up due to a memory controller microcode bug. They
> consist of one CPU (97043A or 97043B) and one IOP (97049A),
> combined with either twelve or thirteen megabytes of any
> combination of 97046A (the one megabyte memory board) and 97048A
> memory. All other configurations will function correctly. The
> maximum memory a Series 500 can support is 14.5 megabytes which
> would consist of seven 97048A memory boards and one 97047A
> memory board.
One non-HP person reported a bug in chatr not working with `ifprolog'
on a 550 running 5.11. I didn't understand the comment but
will forward replies to the questioner.
Finally, one HP'er whom I tend to trust sent the following to me:
>
> Howdy. I work in the Lab that did/does the S500, and administer one of the
> systems here. The 5.1 update wasn't painless but the update documentation
> ... was more complete than in the past ....
>
> Anyway, I've found a number of nice things about 5.11, nothing specific,
> just bug fixes, etc.
I particularly thank the guys at Fort Collins for taking the trouble to help.
If anyone wants, I can mail you the update docs for both 5.05 -> 5.1 and
for 5.01 -> 5.1.
I'm looking forward to 5.1, which I'm putting up in about 3 weeks.
Now -- about TCP/IP for Series 500's .... (ahem).
--
David Harrison, Dept. of Physics, Univ. of Toronto
{ihnp4,utzoo}!utcs!utfyzx!harrison