Colin,
Might you have a copy of the basic compiler or basic
disk for the 5360?
i have most of the other languages for the sys 36 but
need a copy of basic.
tape or 8 in diskette is ok.
Bill
Message: 4
Subject: IBM 5360 FREE to good hom
To: cctalk(a)classiccmp.org
From: "Colin Eby" <ceby2(a)csc.com>
Date: Sun, 17 Aug 2003 21:42:34 -0400
Reply-To: cctech(a)classiccmp.org
Okay --
One more time.... Can I persuade any of you nice folks
on this list to
come
cart off my IBM?
Colin
----------------------------------------------------------------------------------------
Thank you, Rich ! It would be very useful for me. The version of
CP/NET that I'm using is 1.1, but the 1.0 is a good choice too,
I think (with the documentation). Yesterday I proceed to extract
and compile the RS232 version of SNIOS. Actually, the CP/NET
requester do a full load, but cause of one parameter
(always$retries, if I remember well) appears to hang itself and
don't returns to CP/M. The problem can be too in the particular CCP
that comes with CP/NET, and this can fail in the reload after the
load of CP/NET main block, but this is a matter to investigate.
Once I solve it, I must locate what SIO is using CP/NET to try
to establish communications.
Next step would be to check what SIO or Serial Ports are defined
under ALTAIRZ80 with its IO Device or Port Identifiers.
After this, it must be modified the SNIOS to select one free and
safe IO port.
And finally comes the SIMH matter... Map one of the simulated SIOs
with one real RS232 can be relatively easy under Linux/Unix, but not
under Windoze (perhaps with one IP to RS232 mapper I've used sometimes)
and this make me think in one interprocessor solution (that links
two copies of SIMH running at same time) applied in the HP2100
simulator. This would suppose to link the simulated RS232 of a couple
of SIMH with one chain mechanism already implemented in the HP2100
sim.
But, of course, the funny thing would be to connect one Altos 586
with MP/M with one PeeCee with SIMH running ;-)
Thanks again. If you want to send me the CP/NET stuff by mail, this is
my address:
Sergio Pedraja
Calle San Luis 4, 1
39010 Santander
Spain
I pay the shipping costs. But, of course, the email is more economic
:-)
Cheers
Sergio
----- Mensaje Original -----
Remitente: "Rich Beaudry" <r_beaudry(a)hotmail.com>
Fecha: Jueves, Agosto 21, 2003 5:22 am
Asunto: CP/NET
> Sergio,
>
> >I am trying to obtain one copy of CP/NET or MPM/NET.
> >In .dsk format suitable to be used with SIMH simulators.
>
> <snip>
>
> >I can't locate two binaries, MAIL.COM and SERVER.RSP
>
> I have a disk (8", SSSD) labeled "CP/NET 1.0". I also have a
> photocopy of
> the User's Guide for CP/NET (3rd Printing, 1981). The files you
> are looking
> for are not there, but perhaps you have a different version of
> CP/NET. Mine
> appears to be complete, judging from a quick look through the
> manual. There
> are two .COM files for electronic mail (SNDMAIL.COM and
> RCVMAIL.COM), and
> there are also "server" versions of these files. Also, the only
> differenceI can see for the server side is a change in
> NETWRKIF.RSP. There are full
> directions for altering the server and client setups in the manual.
>
> >In the good side, I could generate the NETWRKIF.RSP
> >file using the indications in the CP/NET manual.
>
> My manual also has complete directions. I haven't tried them out,
> however....
>
> >Of course, this is only a first step. If all goes well we
> >couldn't have yet one CP/NET working environment.
> >Here is where I'm lost. What kind of "network" devices
> >used CP/NET ? In the manual comes one example to
> >customize one SNIOS.RSP in the cliente to use one
> >Serial Port, but I assume the existence of others. I think
> >it would be great to allow the simulation of one
> >MP/M-CP/NET-CP/M network with a couple
> >(or more) of SIMH AltairZ80's running at same time.
>
> My manual also only mentions serial devices (in fact, it even suggests
> sample 7-bit and 8-bit serial protocols, and message formats).
> According to
> the manual, NETWRKIF and SNIOS would have to be altered for a new
> device.It would most likely be very difficult to alter them to
> work with new
> networking hardware. If SIMH connects the emulated computer's
> serial ports
> to the real PC ports, then perhaps it would be best to go with two
> computers, hooked by serial cables, one running MP/M and CP/NET
> ("master"),and the other running CP/M and CP/NET ("slave")
>
> Hope this helps, and if you want, I can supply a PC-format disk
> with all
> files from my CP/NET 1.0 disk. I could also supply a
> scan/photocopy of the
> manual as well....
>
> Rich B.
There is another command to run. To get rid of the partitions you will need
to use FDISK.
Once you get rid of the three partitions and reformat it as a single
partition you should be OK.
I would use DOS 6.22, because that is what I have.
Paxton
Astoria, OR
Sergio,
>I am trying to obtain one copy of CP/NET or MPM/NET.
>In .dsk format suitable to be used with SIMH simulators.
<snip>
>I can't locate two binaries, MAIL.COM and SERVER.RSP
I have a disk (8", SSSD) labeled "CP/NET 1.0". I also have a photocopy of
the User's Guide for CP/NET (3rd Printing, 1981). The files you are looking
for are not there, but perhaps you have a different version of CP/NET. Mine
appears to be complete, judging from a quick look through the manual. There
are two .COM files for electronic mail (SNDMAIL.COM and RCVMAIL.COM), and
there are also "server" versions of these files. Also, the only difference
I can see for the server side is a change in NETWRKIF.RSP. There are full
directions for altering the server and client setups in the manual.
>In the good side, I could generate the NETWRKIF.RSP
>file using the indications in the CP/NET manual.
My manual also has complete directions. I haven't tried them out,
however....
>Of course, this is only a first step. If all goes well we
>couldn't have yet one CP/NET working environment.
>Here is where I'm lost. What kind of "network" devices
>used CP/NET ? In the manual comes one example to
>customize one SNIOS.RSP in the cliente to use one
>Serial Port, but I assume the existence of others. I think
>it would be great to allow the simulation of one
>MP/M-CP/NET-CP/M network with a couple
>(or more) of SIMH AltairZ80's running at same time.
My manual also only mentions serial devices (in fact, it even suggests
sample 7-bit and 8-bit serial protocols, and message formats). According to
the manual, NETWRKIF and SNIOS would have to be altered for a new device.
It would most likely be very difficult to alter them to work with new
networking hardware. If SIMH connects the emulated computer's serial ports
to the real PC ports, then perhaps it would be best to go with two
computers, hooked by serial cables, one running MP/M and CP/NET ("master"),
and the other running CP/M and CP/NET ("slave")
Hope this helps, and if you want, I can supply a PC-format disk with all
files from my CP/NET 1.0 disk. I could also supply a scan/photocopy of the
manual as well....
Rich B.
>From: Federico <nuvola66(a)katamail.com>
>
>Fabio Finotti wrote:
>I would like project a circuit to emulate a old 8 inch floppy disk
>(tandon tm848 e).
>So i would like to know wath kind of signal the controller send to
>floppy drive and the reponse
>of floppy to controller.
>
>Thank you.
>
Hi
You should look at a data sheet for something like
a WD1793 or nec765. This will help in understanding
normal formats. Of course, you do not have to use these.
There are a number of control and status signals
to deal with head load, motor control, stepping, index
detection and door closed. There are usually to data
lines, one for read and one for write ( with the exception
early 8" that had built-in clock/data separators.
Typically, the clock and data are encoded with one
of several methods. FM, MFM and M2FM are a few. These
are methods of keeping the clocking with the data.
On floppies, it is necessary to have regular transitions
of the write signal ( no long strings of constant levels ).
This is because most early read heads used coils and
required edges to be detected ( coil heads are differentiators ).
New hard disk heads may use techniques that see the actual
strength of field.
Most soft sectored formats use illegal clock/data sequences
to make the beginnings of sectors( by illegal I mean that
they can't exist in the data portion ). These are usually called
address marks.
Like I said, look at the data sheets. They explain these things.
Dwight
I was going to do this offline, but I have had a couple of requests for
this info, so I will put it out to the list. I do not have the manuals
scanned, so I will excerpt from them. If any of you reading this have
one of the systems (or part of) that ends in an 8, meaning an 8 MB drive,
I would especially like to hear from you in order to try to see if it is
the electronics or the actual drive that I have that is bad. I have 3
working 30 MB systems, and I think I will be getting rid of them soon, as
I have moved the software over to a PC.
Here goes the explanation on the DSD 880 series from my perspective.
Data Systems Design (Qualogy) made two disk subsystems for Digital Equipment
Corporation (DEC) computers. It is possible they made other models, and
other adapters for Prime or DG, but I don't know anything about them.
The subsystems were external, in a 19 inch wide case, about 24 inches deep,
and about 6 inches high. They used a ribbon cable, 26 pin to connect to
the backplane adapter. The adapter had diagnostics upon bootstrap and
a bootstrap prom on it. The models were DSD 880xnn, where the X was either
an S or a D, meaning a single sided 8" floppy drive or a double sided 8"
floppy drive. The nn was either 8 or 20 or 30 for the megabytes. The
floppy emulated an RX02 or the double sided emulated an RX03 with a patch
to the OS. The winchester emulated an RL01 (8 MB with some waste), a short
RL02 (10MB) with a couple of megabytes of bad blocks, or 2 or 3 RL02s with
a patch to the OS for 3 RL02's. I am also doing this mostly in context of
RT-11. The manuals I have are dated 1982 and 1984, so that gives you some
idea of the age.
The bootstrap will boot either the floppy or the hard drive. There were
two different formatter/controller cards, models 8840 and 8841. There were
a few different adapter/interface cards, 8832 and 8836 for q-bus and 8830
for unibus. The winchesters had head and spindle locks located in access
doors underneath, for shipping. There was a modification to the 8836 to
support 22 bit addressing. The 8832/8836 are dual wide boards, while the
8830 is a quad board. The configuration of the 8832 and 8836 are in two
jumper blocks at location F4 and D3 on the card. These control the location
of the RL CSR, the boot address/enable, the floppy CSR, dma burst length,
the RL and floppy vector addresses and the priority levels. The unibus
version has similar features in 5 locations. The units shipped with a
diagnostic floppy that had a diagnostic monitor, a floppy and winchester
exerciser, a diagnostic program for the winchester in RL emulation and
a scan program for bad blocks.
The hyperdiagnostics panel is similar between the 8 and 20/30 drives,
except the mode 0 stuff is different. I will start with mode 1 stuff,
where the panel has mode and class (left,right)
Mode 1: floppy disk format stuff
class 0 format double density
class 1 format single density
class 2 set media to double density
class 3 set media to single density
class 4 set media double density and scan
class 5 set media single density and scan
Mode 2: system tests
class 0 floppy disk exerciser and write
class 1 ditto, but no write
class 2 fixed disk exerciser
class 3 floppy/fixed disk exerciser, write floppy
class 4 single pass of class 3
class 5 single pass of class 3 without write
class 6 floppy/fixed disk exerciser, no fixed/floppy write
class 7 fixed disk write enable, use this before fixed write tests
Mode 3: more hardware tests
class 0 controller switch and indicator test
class 1 general controller tests, alu,mem, crc, pll
class 2 alu and serdes test
class 3 memory test
class 4 crc test
class 5 pll test
class 6 display microcode version
mode 4: floppy disk alignment stuff.
pretty specific, must have sa800/sa850 floppy maint manual
mode 5: read/write tests
class 0 single pass sequential scan floppy
class 1 butterfly seek floppy
class 2 butterfly read floppy headers
class 3 sequential write/read floppy
class 4 sequential scan hard drive
class 5 butterfly seek hard drive
class 6 sequential write/read hard drive
class 7 fixed disk write enable ( do before 6)
Mode 6 and 7 are for offline backup and restore, except 7 - 7
which is to set the floppy-type flag on the winchester bat track map
to tell the system if it is a single or double sided floppy, or no floppy.
Finally there are 40 or so error codes that can pop up in the diagnostic
display.
Hope this satisfies some of the curiosity.
Joe Heck
Okay --
One more time.... Can I persuade any of you nice folks on this list to come
cart off my IBM?
Colin
----------------------------------------------------------------------------------------
This is a PRIVATE message. If you are not the intended recipient, please
delete without copying and kindly advise us by e-mail of the mistake in
delivery. NOTE: Regardless of content, this e-mail shall not operate to
bind CSC to any order or other contract unless pursuant to explicit written
agreement or government initiative expressly permitting the use of e-mail
for such purpose.
----------------------------------------------------------------------------------------
Microsoft wins because of the thing he mentioned about
development support. They support developers.
Developers love Linux and support it, but it is a pain
to develop for Linux compared to microsoft.
The business case he makes about absorbing and running
microsoft applications is probably right too, but I
doubt that Microsoft will allow it to happen. There
are
too many "no reverse engineer" minefields in the EULA
now to allow it to happen anyway.
As he said, rolling out thousands of units during the
implementation of a massive application needs really
solid
support and tools.
I don't think that the fact that everyone can roll
their own development is necessarily a plus for Linux.
Having the mass that an IBM or Microsoft lends with
their market position to making it possible to have a
single product from top to bottom to use is way
cheaper
than the current Linux way. You have Redhat and SUSE,
but they still have the problem that if you make a
commercial Linux App you have to have a RH7.1 RH7.2
RH8 and RH9 all with different nuances (and not minor
ones) to support the market, where with Microsoft you
usually need only one, or can buy a package to handle
the differences from someone.
I have read on mailing list the your message that speaks about floppy disk
and controller :
>Read/Write
>signals,
>normally using a 26LS31 driver and 26LS32 receiver.
>But in a sense they're analogue in _time_. In write mode, the >drive
>writes a flux transition to the disk for every pulse on the >write data
>line. In read mode, the drive sends a pulse on the read data >line for
>every flux transition on the disk. There are some restrictions >on
>allowable frequencies of transisitions, of course.
etc. etc.
I am looking for documentation on the floppy disk interface.I would like
to use a floppy disk in order to save data but I don't know how format
them. Can you help me? ?
Fabio
