HP-3000 Systems Discussion <mailto:[log in to unmask]> wrote:
> Hi all,
>
> In Jan 2004 I gave an update on our SIB '04 progress. See
> http://raven.utc.edu/cgi-bin/WA.EXE?A2=ind0501B&L=hp3000-l&P=R 12868
> for the full report.

I know this is probably beating a dead horse... But here comes my stick
anyway :-)

All these nifty enhancements excite and frustrate me.

If my failing memory serves at all, something like 60%+ of all the HP3000s
ever sold were 9x7 models? (I could be way off here, but go with me anyway
;-) )

In the "heyday", I've heard numbers as high as ~70,000 HP3000s in operation
worldwide... If we're lucky today, it's probably in the  ~20,000-30,000
range?

The last MPE/iX release you can run on any 9x7 box is MPE/iX 6.5. And that's
more due to built-in features to disable the upgrade path for earlier
systems than it is for limitations in the earlier systems (though there is
obviously an increased testing cost to support older platforms). I'm even
willing to write off earlier (HPIB based) boxes, since their peripherals are
so failure-prone and hard to replace nowadays that IMHO they're not suitable
for more than hobby use...

As a "W.A.G." (tm) I'm guessing that probably 10-15% of the installed base
still running HP3000s (of all models) can run MPE/iX 7.5 ? And of those,
even fewer are likely to be interested in new-functionality patches for the
boxes.

I know HP won't release numbers on systems sold or supported, and in reality
they have very little idea of how many systems are still running out there
in any case.

At this point very few people have the option of buying new A or N-class
boxes; and those that do won't be generating any new income for HP.

It sure would be nice to see some concrete numbers from an impartial body on
just what the spread of in-use hardware is... (OpenMPE???) Cause it's
looking alot like there's some really cool development happening, but that
it'll never help probably 90% of the sites still running HP3000s.

I'm sure even the programmers working on these nice mods would feel more
satisfaction if they knew their work was going to benefit perhaps 10 times
the number of sites it currently is...

None of the systems we've got here, and only 1 of 3 I run at 3k.com could
benefit from any of these (7.5)  patches, and I'm not likely to try the
enhancements out on the one box I CAN run them on, since it would then be
incompatible with my other boxes... There wouldn't be any point in writing
any software to take advantage of these features that will only be available
on my 1 box. :-(

So I guess this long rant is basically another plea to either reconsider
making 7.5 "runnable" on 9x7 era boxes (which would seem to be less new
testing in the long run than re-testing every mod on multiple OSes); or
making all these nifty mods available to earlier releases. (I know some of
that is planned, though likely not all patches will make it to all releases;
but I would also guess that you'd likely find alot more beta-testers on
pre-7.5 releases than you do now on 7.5... so perhaps targeting patches to
THOSE releases FIRST would speed up QA?)

Some concrete numbers from the (still-running) base sure would help either
way :-)

Or am I waaayyyyy off base?

<stepping off the soapbox now ;-) >

FWIW,
  Chris Bartram

* To join/leave the list, search archives, change list settings, *
* etc., please visit http://raven.utc.edu/archives/hp3000-l.html *