Reading HP9800 tapes

[Jean-Louis Dornstetter]

Hello everyone

I used to work in the eraly 80's on an HP desktop calculator that was equipped with a thermal printer and two cartridge tape readers, plus a decent green/black display. It looks similar to HP9825 but I couldn't recognize the exact model I was working with, which I didn't find on the WEB. My problem is as follows : the mice did actually eat (!) the listing I kept of some of my junior engineer work deep in the garage. I still have the program tapes, HP200 'certified data cartridge' that are still in good physical shape and weren't exposed to dangerous conditions, and I wonder whether somebody could provide me a hint on how to read such tapes, given that I do no longer have access to one of those machines, I even don't have the precise model, and I have no clue on the (compressed ?) format used to store the lengthy HP Basic routines I would like to read the source code again.

Any help would be more than welcome...

Cheers

[Vassilis Prevelakis]

OK, first the tapes, do they look like these?

According to your description, the computer cannot be a 9825 since
this one has only a one line LED display. So you are either thinking
of a 9835, like this one:

Or (most likely as this one can have two tape drives) an 9845. This
is not the best picture, but its the only one I could find:

Now the bad news, your tapes are probably long gone -- destroyed
by age. Assuming that you can find a working 9845 you may
have a single chance at reading the data off the tape, but before
you do that, you have to make sure that

a) the rubber belt that moves the tape still works. You will have
to open the tape housing and look inside. If the belt is broken or
about to break you will have to replace it with one from a working
tape (preferably manufactured the past 10 years). This is
a not so good picture of the internals:

b) Make sure that the tape is not stuck on itself. Carefully wind
(manually) the tape from end to end (DO NOT USE A RE-TENSION COMMAND)
checking to see whether the oxide stays on the correct part of the
tape. If the oxide is falling off or sticking to the back of the
tape, then your tape is gone.

Assuming that the tape passes the above two tests you should assume
that you have ONE chance of copying the data.

Have another tape or (preferably) a disk (diskette or hard) ready
to accept the contents of the old tape and PRACTICE the copy operation
a couple of times with ANOTHER TAPE.

If everything looks OK, insert the original tape, say something
comforting to the machine in a friendly and stress-free voice and
good luck.

**vp

The pictures above are from the Classic Computers web site www.classiccmp.org.

[David Smith]

Might also be an HP85...

[Vassilis Prevelakis]

David Smith wrote:

> Might also be an HP85... 

The HP85 does not have two tape drives or a green/black screen, the 9845 has both.

**vp

[Jean-Louis Dornstetter]

Vassilis

Thanks for your thorough/detailed response. The tapes are exactly that in the upper right corner of the pictures you provide (HP 200 'certfied cartridge") and I already had done the physical integrity check, and this seems OK, so far so good. Of course, they may well desintegrte wfter the first pre-tension operation, life is sometimes a bitch.

My most 'fundamental' problem remains to either access one such machine, I found the model (also thanks for the pictures) by searching deep into my almost fossile memory, and it definitely was an HP9845B, a really rare monster I have no clue how to get one. The last sale opportunity I could track on the Web was from an Autralia-based machine in 1999 :-(

I will try to find a compatible QUIC tape reader and access the binary content, a couple of dirty hacks might prove necessary but I am not afraid with programming, and also seek for such a fossile machine in the garage of research labs, they don't easily junk their past stuff in France...

Best regards

[Vassilis Prevelakis]

Jean-Louis Dornstetter wrote:

> I will try to find a compatible QUIC tape reader [...]

There are lots of 9825 and (of course) 85 computers out there. I think their drives are compatible with the 9845,
but I not sure about data portability.

I know that the 85 will NOT read 9825 tapes, but I do not
know whether an 9825 will be able to read 9845 tapes (these
two families were quite close, e.g. they had the same expansion cards).

I have looked at the possibility of writing an assembler program on the 85 to read 9825 tapes, but at this stage this task looks hopeless.

Of course TESTING your software will also be a problem,
as you do not want to try it on the 20 year old tapes, and
(at this stage) you do not have anything more recent.

Of course, you can try recording the pulses as they are
coming off the drive and then reverse engineer the
format.

**vp

[Petr Horsky]

Jean-Luis:

I am afraid that even if you succeed to read the raw data from your tapes, everything will depend on whether your programs have been SAVEd or STOREd. The former means just ASCII files (which could be also read programmatically as a sequence of ASCII strings), while the latter are binary data corresponding with particular partially-compiled BASIC program lines. I don't see any reasonable method how to reverse-compile the latter outside the HP 9845B or its exact emulator.

BTW, as far as I know, the HP 9845 and HP 9835 data formats are fully interchangeable (again, with possible exception of STOREd programs). So an HP 9835 might do, too. They used to be cheaper, but, I think, also much rarer.

Incidentally, an HP 9845B keeps standing on my kitchen table; it isn't fully operable, though. (Or, from the user point of view, it's rather quite inoperable, as it hangs up with the "MEMORY TEST IN PROGRESS" message on screen.) Unfortunately, one of its two 16-bit processors seems to be defective. I have precisely the same problem as you have (and, moreover, I would probably like to operate the machine just out of nostalgia). I would appreciate any opportunity to borrow the boards (maybe chiefly from the left card cage) somewhere, even from a wreck of this machine, at least in order to be able to locate the defect on the board level.

Best Regards,
Petr Horsky