21 Jun
2009
21 Jun
'09
2:05 p.m.
-----Oorspronkelijk bericht-----
Van: cctalk-bounces at classiccmp.org
[mailto:cctalk-bounces at classiccmp.org] Namens Tony Duell
Verzonden: zondag 21 juni 2009 22:30
Aan: cctalk at classiccmp.org
Onderwerp: Re: How to lose most of an an entire collection in one shot
devices. I
Me too, and for the speed a HP 9000/300 ;-)
I like the view of a HP 125 ET-head but it is not a very handy machine.
The 9000's series 200 or 300 are much more useful.
Rik
I really like the 125 for several reasons. First, it has an HPIB
I wonder why you like the HP125 and not the HP120, given that
they're almost the same machine electronically. All the
points you make apply to the 120 as well.
interface for disks, etc. That interface also
allowed it to be a
controller for calibration instruments. It also used an easy to
create
Sure, but most HP machines have HPIB :-). The HP120 doesn't
properly support the HPIB port in the BIOS (IIRC there's no
BIOS function to read from an HPIB device) so you pretty much
have to talk to the 9914 directly. Not a big problem, but if
you get it into the wrong state you may find you can no
longer talk to the disk drives.
For talking to instruments, I prefer to use an HP9000/200
machine with a second HPIB port (HP98624). Firstly there's
much better standard software support for the HPIB interface,
and secondly, if one of the instruments 'hangs the bus'
(likely in my case as I might well be repairing said
instrument), I can still talk to the disks.
disk format. Also handy.
> When I said I didn't like the design of the HP120, I
didn't mean the
> 'looks ' of the machine. I meant the
electronic design --
a separate
terminal
processor communicating with the application processor
through a little 'mailbox' and a very strange video circuit.
I worked for two companies which made terminal / PC combo suppose I have a weakness for them. The 125
behaved as a decent
terminal on an HP 3000, and ran as a CP/M computer simultaneously,
allowing a switch between disparate processes that the standard PC
world wouldn't see for many years. The "strange" circuitry made it
easy for the user to alternate processes, and allowed communication
between
It's not so much that, but rather the actual design of the
video circuit.
It uses that National 8350 chip (actually a differently
mask-programmed one for a different screen format). That's a
strange chip, it interrupts the Z80 at the start of each
character line, and fills an 80 byte shift register with characters.
And then there's the attributesm which are set on a character
line basis.
Basically, every character is either normal or enhanced
(decided by bit 7 of the chracter code). You can only have
one type of enhancement in a given line. So you could have
normal and underlined characters in the same line, but you
can't have normal, inverse, and underlined characters all in
the same line.
Another problem is that AFAIK there's no official way for a
CP/M program to set the parameters of the serial ports (not
even the 'printer' port).
The serial chips are on the terminal processor bus, there's
no direct way for the application processor to talk to them.
There's an undocumented way to run code on the terminal
processor (in that at least one HP application does that),
but I can't find out how to do it.
them. It's appropriate to have complex
hardware and
software to make
the use of the machine easier. At least,
that's my theory. The HP
125
Sure, but I don't think the HP120 falls into that category.
was both pretty and easy to use, as an HPIB
controller, an HP
terminal, and as a CP/M machine.
I'd much rather have an HP9000/200....
-tony