Twisted and Sad (or hard drive emulation)
Anthony Marchini
an old hard drive for the TRS80. Let me rephrase, Is the command sequence
between the TRS80 and the old hard drive box relatively simple?
I know, what's the point?
Well, I like my TRS80 machines, they still run (Mod III and IV by the way),
and I know that the mechanical hardware isn't going to last much longer.
I have a parallel I/O board in a PC which I can connect directly to the
processor bus of the Mod III, I could add a PAL or two for anything time
critical, I was just wondering what kind of command sequences are used by
the OS to control the hard disk. Is it just two difficult to even try?
I was going to write a DOS application that managed, say a 5 meg file, and
emulate the bus transactions necessary to move info back and forth out of
the TRS80.
Now from my limited memory of the TRS80 Mod III architecture, the Hard drive
mechanism would have direct access the TRS80 bus I doubt if there is
anything in the bootstrap ROM that actually accessed the external bus, so I
am guessing that the old hard drive would simply suspend operation of the
TRS 80 and write directly to the memory of the system, loading in the new
bootstrap and OS. Is that correct?
Tony
Jack C Empson Jr
The Model III uses a version of LDOS and boots from a floppy. As for connecting it
to the computer there is a interface board in the case to convert a mfm drive to
the bus. The original drive was a 5meg but I do know that a good ole ST225 (20 meg)
will work with some jumper changes on the interface board.
Sylvan Butler wrote:
> On Mon, 8 Nov 1999 12:55:03 -0500, Anthony Marchini <tonym@ce_comet.net> wrote:
> >Now from my limited memory of the TRS80 Mod III architecture, the Hard drive
> >mechanism would have direct access the TRS80 bus I doubt if there is
> >anything in the bootstrap ROM that actually accessed the external bus, so I
> >am guessing that the old hard drive would simply suspend operation of the
> >TRS 80 and write directly to the memory of the system, loading in the new
> >bootstrap and OS. Is that correct?
>
> well I only ever saw a handful of M3 systems which had hard disks. Of
> those, only one would boot from the hard disk and that one had a special
> boot rom installed (it was in a lab and the other systems also had a
> special boot rom, but to boot across the "network" not from a HD).
>
> sdb
>
> --
> | Sylvan Butler | Not speaking for Hewlett-Packard | sbutler-boi.hp.com |
> | Watch out for my e-mail address. Thank UCE. #### change ^ to @ #### |
> They that can give up essential liberty to obtain a little temporary
> safety deserve neither liberty nor safety. --Benjamin Franklin, 1759
> "Don't Tread On Me!"
--
**************************************************
* Jack Empson Instructional Support Technician *
* State University of New York @ Buffalo *
* 10 Norton Buffalo, New York 14260 *
* Voice 716.645.3518 Fax 716.645.3198 *
**************************************************
Dan French
original owner made. You have about 5 seconds to choose the device that you want
to boot to (hard drive / floppy), after that (as I recall) it would try to boot off
of the hard drive. Unfortunately, I did something to the system (crashed the hard
drive?) and it's been in storage ever since. It was cool while it worked.
Dan French
e-mail has spam check on it, correct to send a e-mail message.
Anthony Marchini
impression that it connected directly to the processor bus.
My memory is so hazy now.
Bummer
Tony
Jack C Empson Jr wrote in message <38277EBD...@acsu.buffalo.edu>...
>
>The Model III uses a version of LDOS and boots from a floppy. As for
connecting it
>to the computer there is a interface board in the case to convert a mfm
drive to
>the bus. The original drive was a 5meg but I do know that a good ole ST225
(20 meg)
>
>will work with some jumper changes on the interface board.
>
>
>
>
>> On Mon, 8 Nov 1999 12:55:03 -0500, Anthony Marchini <tonym@ce_comet.net>
wrote:
>> >Now from my limited memory of the TRS80 Mod III architecture, the Hard
drive
>> >mechanism would have direct access the TRS80 bus I doubt if there is
>> >anything in the bootstrap ROM that actually accessed the external bus,
so I
>> >am guessing that the old hard drive would simply suspend operation of
the
>> >TRS 80 and write directly to the memory of the system, loading in the
new
>> >bootstrap and OS. Is that correct?
>>
>> well I only ever saw a handful of M3 systems which had hard disks. Of
>> those, only one would boot from the hard disk and that one had a special
>> boot rom installed (it was in a lab and the other systems also had a
>> special boot rom, but to boot across the "network" not from a HD).
>>
>> sdb
>>
Sylvan Butler
Skipp asks that you dont spam him thanks
: I was wondering if there was a simple means of commanding a PC to emulated
: an old hard drive for the TRS80. Let me rephrase, Is the command sequence
: between the TRS80 and the old hard drive box relatively simple?
: I know, what's the point?
It's fun, that's what's the point. You might ask my friend Jim at
http://www.strappe.com He built and ran the "Yet another bbs" for many
years. It was a model 1, lnw 80 and a pc as a ram disk. He might take a
moment to answer you, but he's quite busy as of late.
cheers
skipp
: Well, I like my TRS80 machines, they still run (Mod III and IV by the way),
: and I know that the mechanical hardware isn't going to last much longer.
: I have a parallel I/O board in a PC which I can connect directly to the
: processor bus of the Mod III, I could add a PAL or two for anything time
: critical, I was just wondering what kind of command sequences are used by
: the OS to control the hard disk. Is it just two difficult to even try?
: I was going to write a DOS application that managed, say a 5 meg file, and
: emulate the bus transactions necessary to move info back and forth out of
: the TRS80.
: Now from my limited memory of the TRS80 Mod III architecture, the Hard drive
: mechanism would have direct access the TRS80 bus I doubt if there is
: anything in the bootstrap ROM that actually accessed the external bus, so I
: am guessing that the old hard drive would simply suspend operation of the
: TRS 80 and write directly to the memory of the system, loading in the new
: bootstrap and OS. Is that correct?
: Tony
Lord Apollyon
<tonym@ce_comet.net> wrote:
> Now from my limited memory of the TRS80 Mod III architecture, the Hard drive
> mechanism would have direct access the TRS80 bus I doubt if there is
> anything in the bootstrap ROM that actually accessed the external bus, so I
> am guessing that the old hard drive would simply suspend operation of the
> TRS 80 and write directly to the memory of the system, loading in the new
> bootstrap and OS. Is that correct?
Well, seeing as you have access to the entire LDOS source tree, and you
could in theory make a driver to do what you see fit with... you could
come up with just about any "HD translation" method you liked, and make a
driver for LDOS to use it. [This is I think the BEST part of Tim Mann's
(via Ray Soltoff I guess) legacy to the TRS80 community. That code will
let us hack our way to keep those old machines running, no matter what
funky interfaces we attach to them. :)
=Rob=
--
The reply-to-address will expire on Midnight 1-December-1999.
Spammers: You will lose your network access. Guaranteed.
96 domains, 353 web-accounts, and 555 dialup ISP accounts flushed.
Ward Donald Griffiths III
>
> The hard drive connected to the Floppy controller??? I was under the
> impression that it connected directly to the processor bus.
> My memory is so hazy now.
> Bummer
> Tony
> Jack C Empson Jr wrote in message <38277EBD...@acsu.buffalo.edu>...
> >
> >The Model III uses a version of LDOS and boots from a floppy. As for
> connecting it
> >to the computer there is a interface board in the case to convert a mfm
> drive to
> >the bus. The original drive was a 5meg but I do know that a good ole ST225
> (20 meg)
> >
> >will work with some jumper changes on the interface board.
No, the hard drive doesn't connect to the floppy controller, it's just
that the Mod 3 (and most Mod 4) can't boot directly from the HD, so it
boots from a floppy that then transfers control to the HD. Just like
you do with a PC that's had its OS hosed by a virus or normal DOS/Win
operation.
--
Ward Griffiths
Anthony Marchini
the top of anyones head, how many bytes may be available in the Mod III rom?
I suppose, If I make the interface, I only need a few bytes to coordinate
byte wide transfers over a parallel interface. I could do it in TTL for a
foot and a half.
That sounds like the ticket.
tony
Ward Donald Griffiths III wrote in message <3827F96C...@home.com>...
Jan Vanden Bossche (aka the TyRannoSaurus)
Ward Donald Griffiths III <wdg...@home.com> wrote in article
<3827F96C...@home.com>...
> Anthony Marchini wrote:
>
> No, the hard drive doesn't connect to the floppy controller, it's just
> that the Mod 3 (and most Mod 4) can't boot directly from the HD, so it
The 4P can, can't it ?
> boots from a floppy that then transfers control to the HD. Just like
> you do with a PC that's had its OS hosed by a virus or normal DOS/Win
> operation.
But still, it's a good idea. Even if you have to boot from a TRS-80 floppy,
if then you could get to the PC's harddisk, it would be a great plus. There
are plenty of old PC's around that nobody uses. Why not use them as a model
4's HD ?
Let me propose one modification: hook the I/O bus to the PC's standard
parallel port. They run ZIP-drives, scanners and SCSI interfaces off that
port, why not a humble model 4 access program ? The only hardware needed
then, would be a special cable.
> Ward Griffiths
Greetings from the TyRannoSaurus
Jan-80
Ward Donald Griffiths III
> Ward Donald Griffiths III <wdg...@home.com> wrote in article
> <3827F96C...@home.com>...
> > that the Mod 3 (and most Mod 4) can't boot directly from the HD, so it
>
> The 4P can, can't it ?
The key word was "most Mod 4" and that includes at least half of the 4P
systems and all desktop 4s prior to the 4D to the best of my knowledge
if they haven't had ROM upgrades. And it was always best to have a
boot floppy anyway in my experience. (And the discussion started with
Mod 3, not Mod 4).
>
> > boots from a floppy that then transfers control to the HD. Just like
> > you do with a PC that's had its OS hosed by a virus or normal DOS/Win
> > operation.
>
> But still, it's a good idea. Even if you have to boot from a TRS-80 floppy,
> if then you could get to the PC's harddisk, it would be a great plus. There
> are plenty of old PC's around that nobody uses. Why not use them as a model
> 4's HD ?
Nothing wrong with that at all, I will cheerfully use anything along
those
lines, as my 4P's hard disk's board is toast. I'll admit to preferring
that the technology extend to Linux PCs to make it easier to share
between Tim Mann's xtrs and the "real thing".
>
> Let me propose one modification: hook the I/O bus to the PC's standard
> parallel port. They run ZIP-drives, scanners and SCSI interfaces off that
> port, why not a humble model 4 access program ? The only hardware needed
> then, would be a special cable.
Sounds great to me. I'll admit that my skills are not up to it, my
hardware and Z-80 assembly skills are rusty and in fact I was never
anything more than a CSR at an RSCC in my so-called prime, most of my
little hardware hacking back then was with Color Computers. (Mostly
following DBK's instructions and designs).
> Greetings from the TyRannoSaurus
> Jan-80
--
Ward Griffiths
Jack C Empson Jr
"Jan Vanden Bossche (aka the TyRannoSaurus)" wrote:
>
> Let me propose one modification: hook the I/O bus to the PC's standard
> parallel port. They run ZIP-drives, scanners and SCSI interfaces off that
> port, why not a humble model 4 access program ? The only hardware needed
> then, would be a special cable.
I don't think this can be done. A zip drive uses a bi-directional port.
Most older computers including some pc's use a unidirectional printer port.
lamar...@wgcr.org
>
> "Jan Vanden Bossche (aka the TyRannoSaurus)" wrote:
>
> >
> > Let me propose one modification: hook the I/O bus to the PC's standard
> > parallel port. They run ZIP-drives, scanners and SCSI interfaces off that
> > port, why not a humble model 4 access program ? The only hardware needed
> > then, would be a special cable.
>
> I don't think this can be done. A zip drive uses a bi-directional port.
> Most older computers including some pc's use a unidirectional printer port.
It could be done even on a unidirectional port by using the four status
bits for input lines (don't laugh -- the IOmega ZIP parport driver does
this exact thing if a bidirectional parport isn't available).
The way these parport devices work is by using a special chip that is a
'bus replicator' -- in effect, the parport is hooked up to a chip that
lets commands given over the parport drive a complete, fully functional
ISA bus. Then, either an IDE, floppy, or other controller chip is
attached via the replicated ISA bus. Nasty stuff. Look at the ftape or
ppa driver source code for Linux for an idea of what kind of code you're
dealing with.
HOWEVER, that is the exact thing that would have to be done here. The
parport would be interfaced to a circuit that would take the data sent
over the parport and translate it to address and data transitions (the
50 pin Model 3/4/2/12/16/6000 IO expansion bus is a complete Z80 IO bus,
with 8 address lines, 8 data lines, ioread and iowrite signals, etc --
and the timing of those signals is critical). A microcontroller is
about the only way to do it. With a microcontroller taking high level
commands from the parport (such as 'INPUT 80' (for input from port 80),
etc). ALSO, interrupts have to be handled. Again, this exact thing is
already being done in most parport devices -- just, it's an ISA bus and
not a TRS-80 bus.
(NOTE: the external hard drive controllers for the 12 Meg and later
external hard drives for the Models 2/12/16/16B/6000 used an interface
card in the CPU that mirrored the Model 3/4 expansion bus -- the 8 Meg
controller was different, however, and used a different card, IIRC.)
It certainly CAN be done -- but it would be non-trivial.
--
Lamar Owen
WGCR Internet Radio
1 Peter 4:11
lamar...@wgcr.org
>
> About 12 years ago -- I got a Model 4 that has a special "boot rom" that the
> original owner made. You have about 5 seconds to choose the device that you want
> to boot to (hard drive / floppy), after that (as I recall) it would try to boot off
> of the hard drive. Unfortunately, I did something to the system (crashed the hard
> drive?) and it's been in storage ever since. It was cool while it worked.
I had a model 4 with the MAD Software XDROM and HBUILD6 installed -- and
it booted flawlessly from HD. MAD Software is still in business --
http://madsoft.lonestar.org for more info on their excellent products.
I sold said mod 4 (with a whole slew of software on it, plus Pro MC,
Pro-MRAS, a pristine copy of The Source, hires graphics board, 320K RAM
modification, and a bunch of other stuff) in 1992 before I got married
-- I have since stayed with Unix machines (first, a Tandy 6000, then a
pair of AT&T 3B1's, then an Apollo DN3500 network of three workstations,
then a 386sx-20 running SoftLandingSystems distribution of Linux 0.13,
---- and now up to a Pentium MMX 200 running RedHat Linux 6.1.). Having
a Linux machine is nice -- I can run Tim Mann's top-o-line xtrs and
execute all my old Z80 code..... ;-)
Ward Donald Griffiths III
>
> "Jan Vanden Bossche (aka the TyRannoSaurus)" wrote:
>
> >
> > Let me propose one modification: hook the I/O bus to the PC's standard
> > port, why not a humble model 4 access program ? The only hardware needed
> > then, would be a special cable.
>
> I don't think this can be done. A zip drive uses a bi-directional port.
> Most older computers including some pc's use a unidirectional printer port.
That's I/O bus, not printer port. Although I recall that some input
periperals (joysticks?) were connected to TRS-80 printer ports once
upon a time.
--
Ward Griffiths
"Men will never be free until the last king is strangled with
the entrails of the last priest." Denis Diderot
Frank Durda IV
: Now that would be fun, patch the rom, I could do that with no problem. Off
: the top of anyones head, how many bytes may be available in the Mod III rom?
Without rewriting modules, how many bytes are unused in the more common
US versions of the "C" ROM? Easy: Four bytes are free, two are actually
accessible (two are masked by the printer status port memory-map mirroring).
There are some small regions of two to six bytes available in the "A"
ROM - about 16 bytes total as I recall - and essentially no free bytes in
the "B" ROM.
That's why I had to re-write just about all of the "C" ROM to come up with
the space for the "XROM" boot ROMs that allow booting directly from MFM and
SCSI hard disks and such. (You get a much better keyboard driver out of the
deal as well.)
If you undertake such a project, re-writing the keyboard driver is where you
will recover most of your space. You basically can't touch the cassette
driver because it is timing critical, AND some applications decided to just
jump into the middle of those functions and use some routines they found
handy in there that are unrelated to cassette I/O operations. Move things
around at all and stuff breaks. It took a lot of work to find/fix all of
those issues. The Model III ROM sets have far more cases of jumps to
fixed locations in each other than the Model 4 ROM sets, although all the
"official" entry locations in the Model III and 4 ROMs are the same.
Frank Durda IV - only this address works:|"How do I know?
<uhclem.dec99%nemesis.lonestar.org> | I wrote the code, that's how."
This Anti-spam address expires Dec. 31st |
Copr. 1999, ask before reprinting.
Frank Durda IV
: The key word was "most Mod 4" and that includes at least half of the 4P
: systems and all desktop 4s prior to the 4D to the best of my knowledge
: if they haven't had ROM upgrades. And it was always best to have a
: boot floppy anyway in my experience. (And the discussion started with
: Mod 3, not Mod 4).
For the record, a complete Model 4 ROM set will drop-in to a Model III and
run. It tries to program some ports that don't exist on the Model III,
but the result is harmless.
As to Model 4 hard disk boot-cable ROMs, see:
http://madsoft.lonestar.org/catalog/xrom.html
for a beast that does exist, does what you want, and is still sold.
Frank Durda IV - only this address works:|"I picked up a Magic 8-Ball the
<uhclem.dec99%nemesis.lonestar.org> | other day and it said 'Outlook
| not so good'. I said 'Sure,
This Anti-spam address expires Dec. 31st | but Microsoft still ships it.'"
Frank Durda IV
: Of those, only one would boot from the hard disk and that one had a special
: boot rom installed (it was in a lab and the other systems also had a
: special boot rom, but to boot across the "network" not from a HD).
You are describing the Network 4 system, which had a hard-disk-based teacher
station (aka a server), and student stations (aka clients). It ran a
badly-hacked copy of TRSDOS 1.3 with a hard disk driver grafted in. Because
Network 4 ran in Model III mode, existing Model III systems could also be
used as teacher stations. In both platforms, the "C" ROM was replaced with
one containing a hard disk boot loader.
Frank Durda IV - only this address works:|Sign on back of a Cable TV truck:
<uhclem.dec99%nemesis.lonestar.org> |"This vehicle makes frequent stops."
| (written in the dirt underneath:)
This Anti-spam address expires Dec. 31st |"Just like our cable service."
Lord Apollyon
uhclem...@nemesis.lonestar.org (Frank Durda IV) wrote:
> That's why I had to re-write just about all of the "C" ROM to come up with
> the space for the "XROM" boot ROMs that allow booting directly from MFM and
> SCSI hard disks and such. (You get a much better keyboard driver out of the
> deal as well.)
This begs an obvious question: Is a fully assemblable set of sources to
build the ROMs available somewhere, or did you do this the old fashioned
way? (Disassembly).
I've not looked too hard for this, but I'm just curious to know if the ROM
sources are floating on any of the software archives.
=R=
--
The reply-to-address will expire on Midnight 1-December-1999.
Spammers: You will lose your network access. Guaranteed.
96 domains, 354 web-accounts, and 559 dialup ISP accounts flushed.
Ken Walker
I would think that if anyone had them, it would be Frank. He wrote alot
(if not all) of them. I know he wrote the 4P boot rom (FDIV).
Let us know if you find them.
Ken (*:
Dan French
-- Yes (I do)
-- Yes (I will, if I can)
If pir...@microsoft.com contacts me, I will disavow any knowledge.
Dan French
remove spam food to send e-mail.
Jan Vanden Bossche (aka the TyRannoSaurus)
Jack C Empson Jr <emp...@acsu.buffalo.edu> wrote in article
<3829D75C...@acsu.buffalo.edu>...
>
> "Jan Vanden Bossche (aka the TyRannoSaurus)" wrote:
> >
> > Let me propose one modification: hook the I/O bus to the PC's standard
> > parallel port. They run ZIP-drives, scanners and SCSI interfaces off
> > that port, why not a humble model 4 access program ? The only hardware
> > needed then, would be a special cable.
>
> I don't think this can be done. A zip drive uses a bi-directional port.
> Most older computers including some pc's use a unidirectional printer
> port.
Fine. Then search for a PC that DOES have a bidirectional parallel port.
Plenty around ...
And still, I don't think that's completely true. I can run a ZIP-drive off
an original IBM PC, so what do you mean by "most older ... pc's" ? How
much older do you want ? Aren't the 4 returnlines used ("ack", "busy",
"paper out" or something ?) when there is a standard parallele port ?
> * Jack Empson Instructional Support Technician *
Greetings from the TyRannoSaurus
Jan-80
Lord Apollyon
> That's funny, I asked the same question about 2 months ago. (*:
>
> I would think that if anyone had them, it would be Frank. He wrote alot
> (if not all) of them. I know he wrote the 4P boot rom (FDIV).
Hrm. So I take Frank's silence on this issue to mean he does have them,
however he is unable to (legally) or unwilling to make them available, and
he just doesn't want to be in the position to say "no" to a dozen
squalling infants begging him for their bottle of warm ROM sources
formula? :)
=Rob=
Jack C Empson Jr
"Jan Vanden Bossche (aka the TyRannoSaurus)" wrote:
>
> And still, I don't think that's completely true. I can run a ZIP-drive off
> an original IBM PC, so what do you mean by "most older ... pc's" ? How
> much older do you want ? Aren't the 4 returnlines used ("ack", "busy",
> "paper out" or something ?) when there is a standard parallele port ?
The older printer ports as in the MIII and 4 will only send information out
the data lines.
A zip drive also sends data back to the computer via the same lines.
--
**************************************************
* Jack Empson Instructional Support Technician *
* State University of New York @ Buffalo *
* 10 Norton Buffalo, New York 14260 *
* Voice 716.645.3518 Fax 716.645.3198 *
**************************************************
186000 mps it's not just a good idea, it's the law!
Frank Durda IV
: This begs an obvious question: Is a fully assemblable set of sources to
: build the ROMs available somewhere, or did you do this the old fashioned
: way? (Disassembly).
I have source code for some items. Naturally I have source code for the
things I wrote from scratch, own and still sell.
Frank Durda IV - only this address works:|"The Knights who say "LETNi"
<uhclem.dec99%nemesis.lonestar.org> | demand... A SEGMENT REGISTER!!!"
|"A what?"
This Anti-spam address expires Dec. 31st |"LETNi! LETNi! LETNi!" - 1983
Frank Durda IV
: Hrm. So I take Frank's silence on this issue to mean he does have them,
Read into it what you want, but consider that some people have to walk away
from the computer every now and then and can't monitor their hobby group
constantly. Then read other post on this subject.
Ben Yates
The port for another computer of mine (circa 1980) made by the computer
manufacturer (there were clones that I'm told don't) does both input and
output, but asynchronously (not at the same time) [OK - It's a TI!]
Does the zip drive use standard handshaking? Or does it assume that anything
it sends to the PC will be captured. Like a "Ready" line?
Jack C Empson Jr wrote in message <382CB0C8...@acsu.buffalo.edu>...
Jan Vanden Bossche (aka the TyRannoSaurus)
Jack C Empson Jr <emp...@acsu.buffalo.edu> wrote in article
<382CB0C8...@acsu.buffalo.edu>...
>
>
> "Jan Vanden Bossche (aka the TyRannoSaurus)" wrote:
>
> > And still, I don't think that's completely true. I can run a ZIP-drive
> > off an original IBM PC, so what do you mean by "most older ... pc's" ?
> > How much older do you want ? Aren't the 4 returnlines used ("ack",
> > "busy", "paper out" or something ?) when there is a standard parallel
> > port ?
>
> The older printer ports as in the MIII and 4 will only send information
> out the data lines.
> A zip drive also sends data back to the computer via the same lines.
Yes, but I wasn't talking about hooking a ZIP-drive to a model III parallel
port. Let's get back on track here: it was about using a PC as a HD for a
model III or 4. How do we hook it up ?
I suggested that we use the I/O bus on the TRS-80 side, and the parallel
port on the PC side, with the argument that ZIP drives could be attached to
(virtually) any (IBM truly compatible/clone) PC, so it shouldn't be too
much of a problem.
If we figure out how Iomega does it on a standard IBM PC, somebody could do
it for attaching a TRS-80 to a PC //port, only this time, with the TRS-80
as the master in the setup.
> * Jack Empson Instructional Support Technician *
Greetings from the TyRannoSaurus
Jan-80
Shanti
>I suggested that we use the I/O bus on the TRS-80 side, and the parallel
>port on the PC side, with the argument that ZIP drives could be attached to
>(virtually) any (IBM truly compatible/clone) PC, so it shouldn't be too
>much of a problem.
>
>If we figure out how Iomega does it on a standard IBM PC, somebody could do
>it for attaching a TRS-80 to a PC //port, only this time, with the TRS-80
>as the master in the setup.
>
Jeff Vavasour's emulator uses the printer ports of both the PC and Model I
to transfer TRS-80 disks to PC image files. It is only a six wire interface
cable and works rather well. Perhaps that would be a good starting point.
All the cable construction information is in the model1-e.zip archive. It
should not be too difficult to adapt this scheme for I/O bus use on a Model
III.
Amardeep
Lord Apollyon
uhclem...@nemesis.lonestar.org (Frank Durda IV) wrote:
> I have source code for some items. Naturally I have source code for the
> things I wrote from scratch, own and still sell.
Is it possible to share the sources for the items you've not written? The
goal is really one of documentation, and the possibility to extend the ROM
functionality to perhaps boot off of HDs using funky-assed interfacing
schemes that would scare even the most grizzled hardware hacker.
In article <FL47A...@nemesis.lonestar.org>,
uhclem...@nemesis.lonestar.org (Frank Durda IV) wrote:
> Read into it what you want, but consider that some people have to walk away
> from the computer every now and then and can't monitor their hobby group
> constantly. Then read other post on this subject.
Sorry, Frank. I was trying to be cheeky, but probably failed dismally.
The squalling infants part at the end of the paragraph was trying to
indicate some "poking fun" more than a serious jab.
I don't think there's a person on this newsgroup who would or COULD fault
what you've contributed to the TRS-80 community [both real and emulated]
over the years. So don't take what I said the wrong way.
If you cannot or will not share source code, I'm sure your reasons are
good ones. It *IS* just 14k or so of object code. With all of the
disassembling I've had to do on modern machines, 14k is like a toy. If
most of us TRS-80 hackers really had to do it, I'm sure we could.
Can you share the sources you have of the Model 1, 3, or 4 ROMs with the
TRS-80 community?
Jan Vanden Bossche (aka the TyRannoSaurus)
Shanti <sha...@nospam4me.com> wrote in article
<BQjX3.7754$wG3.4...@typhoon.nyroc.rr.com>...
> >
> > I suggested that we use the I/O bus on the TRS-80 side, and the
> > parallel port on the PC side [..]
>
> Jeff Vavasour's emulator uses the printer ports of both the PC and
> Model I to transfer TRS-80 disks to PC image files. It is only a
> six wire interface cable and works rather well.
Never heard of it. But then again, I never read README-files :)
> Perhaps that would be a good starting point.
Yes, but would parallel-to-parallel be fast enogh to drive a harddisk ?
> All the cable construction information is in the model1-e.zip
> archive. It should not be too difficult to adapt this scheme
> for I/O bus use on a Model III.
Yep.
Gentlemen, is there anyone out there, who's up to the challenge ?
> Amardeep
Lord Apollyon
Vanden Bossche (aka the TyRannoSaurus)" <Ja...@ping.be.nospamplease>
wrote:
> Yes, but would parallel-to-parallel be fast enogh to drive a harddisk ?
Well, if you connected to the TRS-80's IN/OUT lines rather than the memory
bus subsystem, you could be as slow as you liked [the transfers would be
driven by the TRS80-side driver].
If you were opting to connect parallel to the TRS80's memory bus, ouch.
You have hard real-time requirements to satisfy, or crashorama.
I'm assuming you're just after an interconnect scheme whereby you can
connect a TRS-80 to a PC, and have the PC act as a "hard drive server" to
the TRS80. The "hard disk" could be an image file, or the entire hard
disk if you really wanted. But the interconnect speed wouldn't be a big
issue - the PC would buffer between the two, and it'd run as fast as the
little Z80 could suck or blow the data.
> Gentlemen, is there anyone out there, who's up to the challenge ?
This shouldn't be TOO hard. You'll need PC-side and TRS80-side drivers,
and the cable. That Vavasour technique sounded like the least painful.
Here's an excerpt from his manual: [full (c) and accreditation remain with
Jeff Vavasour, though his Model I emulator and files were placed in the
public domain].
Enjoy!
=R=
3.2.1 Constructing the Cable
To start, you will need to purchase a 25-pin male D-type connector (also
known as a standard PC printer connector), a 34-pin female edge-card connector
(the standard counterpart to the Model I printer port), and some ribbon cable
(you will need to use only six wires). These may be obtained at many
electronics and computer stores. If a store you try does not have all the
necessary parts, they may be able to direct you to another supplier.
If you look carefully, you should see that the pins on the 25-pin connector
are numbered, running left to right as 1 to 13 on the top row, and 14 to 25
on the bottom row, when looking at the pin side. In contrast the front of
the edge-card connector (I'll call it the 34-pin from now on) is numbered
as follows:
1 3 5 7 9 11 13 15 17 19 21 23 25 27 29 31 33
__ __ __ __ __ __ __ __ __ __ __ __ __ __ __ __ __
__ __ __ __ __ __ __ __ __ __ __ __ __ __ __ __ __
2 4 6 8 10 12 14 16 18 20 22 24 26 28 30 32 34
This is the view from the side with the hole for the Model I's card edge to
slide in.
You will need to wire the following pins together:
Wire # from 34-pin to 25-pin
------ ----------- ---------
1 11 ----------> 13
2 13 ----------> 12
3 15 ----------> 11
4 17 ----------> 15
5 20 ----------> 25
6 21 ----------> 2
If you can, check them with a resistance meter afterward to ensure they
are in good contact and connected properly.
3.2.2 Testing the cable
You may test the cable by connecting it to both computers, loading BASIC on
each, and then enter "POKE 14312,x" on the Model I followed by "PRINT
INP(&H379)"* on the PC. Repeat this where "x" using x=0, 16, 0, 32, 0, 64, 0,
128. The number from the "PRINT" command on the PC should change each time
(though it will not match the "x" value you used). If it fails to change for
some of the values of "x", double check the first four lines in the wiring
chart above. If none of them change also check the fifth wire.
The sixth line is tested by entering "OUT &H378,x"* on your PC followed by
"PRINT PEEK(14312)" on your Model I where x=0 and 255. x=0 should give a 127
on the Model I, while x=255 should give a 255. If this fails to happen,
then wires 5 and/or 6 are incorrect.
This cable construction has been tested and is known to be safe if properly
constructed. Be careful though. I assume no liability for any damage that
may result.
3.2.3 Nonstandard parallel ports on your PC
This section is for users whose PC's parallel port is not the standard 25-pin
connector.
-------------------------------------------------------------------------------
*On PS/2 or PS/2-compatible machines, you will find that you will have to
use &H3BC and &H3BD instead of &H378 and &H379
You will need to identify the following pins on your parallel port: BUSY,
PAPER EMPTY, SELECT, ERROR, D7, and GROUND. Service personel for your PC
should be able to provide this information. Then use the following wiring
chart:
Wire # from 34-pin to PC
------ ----------- -----
1 11 -----> SELECT
2 13 -----> PE
3 15 -----> BUSY
4 17 -----> ERROR
5 20 -----> GND
6 21 -----> D7
Ken Walker
Ken
Ward Griffiths
The bulk of the Mod 1/3/4 ROM set is copyrighted by Microsoft.
There is lots of evidence that they are not into sharing. Starting
with a letter from Bill Gates to the hobbyist community back before
the Mod 1 showed up concerning Altair BASIC. (And the Halloween
documents tell us that they don't like people sharing things that
Microsoft was never involved with).
--
Ward Griffiths wdg...@home.com http://members.home.net/wdg3rd/
"It is not merely that I dislike, distrust and disbelieve anyone who
seeks political office. I would extend privacy rights even to
politicians were it not for two countervailing circumstances. First,
they themselves violate privacy rights wholesale. They regulate
virtually everything that peaceful people can do behind closed doors,
from taking drugs to having sex. It is elitist hypocrisy for them to
demand the privacy rights that they routinely deny to ordinary people.
If a politician wishes me to respect his personal life, then he needs
to respect mine." Wendy McElroy
Anthony Marchini
at what price?
Tony
Frank Durda IV wrote in message ...
>Lord Apollyon (spamba...@paypc.com) wrote:
>: This begs an obvious question: Is a fully assemblable set of sources to
>: build the ROMs available somewhere, or did you do this the old fashioned
>: way? (Disassembly).
>
Anthony Marchini
www.boondog.com
and connecting it to the I/O bus of the TRS 80.
As I stare at my M3 schematics I notice that the printer port is full
bidirectional. The BUSY,OUTPAPER , UNIT SELECT and FAULT lines interspace
with 4 other lines that are just connected to the edge connector at 19,29,30
and 32.
Originally (and i was wrong) I thought the hard drives of the Mod III were
able to usurp control of the Mod III bus, holding it in a reset state or
something.
I then hoped that the Mod III bios (if you can call it that) somehow looked
at the I/O bus for a "BIOS extension" (i know... I have been under the
influence of these other machines for too long) meaning, some kind of port
value that would show the existance of a certain product.
Now, i am on another tac.... The floppy connector. Does anyone remember if
you can boot from any drive or not? I forget, and the MOD III is in storage.
I bring it up because I Am thinking that I could hook and interface onto the
external floppy port and when the system boots up, emulate a floppy
drive....which would be nuts, but probably pretty easy for a quick PIC
microcontroller (that would probably end up being more powerful than the Z80
i want to interface to) to emulate the signals to the floppy port necessary
for the MOD III to think that a floppy was there and to load in a bootstrap,
which would then...........
access the parallel port , because it would be easier than communicating
throught the floppy port and then the program on the PC Emulating a hard
drive...
So in essance... a solid state boot floppy sector to launch a parallel port
(or I/O port) boot program that would load an operating system through an
existing PC and no moving parts.... excepting the PC.
I am one of those people who always sees the hard way of doing things.
Tony
Lord Apollyon wrote in message ...
Lord Apollyon
> Starting
> with a letter from Bill Gates to the hobbyist community back before
> the Mod 1 showed up concerning Altair BASIC. (And the Halloween
> documents tell us that they don't like people sharing things that
> Microsoft was never involved with).
Wow. Gates has been a PowerPrick from birth. I'm impressed.
=R=
Frank Durda IV
: The bulk of the Mod 1/3/4 ROM set is copyrighted by Microsoft.
No it isn't. None of the Model I Level 1, Model III or 4 ROM is copyrighted
by Microsoft.
Anthony Marchini
The machine boots into basic if you hold done break right? How bout...
Type in a quick basic program, an easy to remember one...
100 for X% = 1 to N (however many bytes)
105 j % = peek(input port address- possibly printer input port)
110 poke X% + (safe address), j%
115 poke (output address...possibly printer strobe),255
120 print "THATS "; X% ' slight delay
125 next X%
130 poke 16526,LSB(safe address)
135 poke 16527,MSB(safe address)
140 X = USR(0) 'bombs away
Now the parallel port is hooked to a parallel port of the PC , possibly not
unlike the one from www.boondog.com but maybe even the parallel port. The
trick is catching the printer strobe...signalling the next byte. A tight
loop on an input line should garantee that.
Now the PC program is running first. It has the first byte loaded up on the
parallel output line...does a pc parallel port strobe the outputs? I can't
remember now. In any case, you then hit run on the TRS 80 and it loads in
the first byte , stores it, strobes the parallel port...prints to the screen
(a little delay). The pc does a tight scan for the printer strobe....and
outputs the next byte and so on. Then we jump to it.
This program then loads in a small operating system, clears a few key
locations necessary for bootstrap startup and vectors to the normal DOS
bootup.
Now all I need is a hacked DOS like LDOS , that has the I/O changed to just
the I/O Port , (in this case I think a port built to the spec of the I/o
Interface would have to be done up). Then I just need Jeff Vavasour or
Mathew Reed to loan (ok..ill give you money) the source code for floppy
emulation and wammo. PC plays floppy to a real TRS80.. Then its just a
matter of changing the emulation params to pretend its a hard drive as
well...or a darn big floppy.
The cool thing about this, is that a person with a MOD III without floppys
could then experience the thrill of running the machine with floppys. I
wonder how many of those are sitting in a pile of dust waiting to be played
with.
My problem, the easy way to go would be to connect the pc to the printer
port of the TRS80, which would put it out of action. This is a problem if
the wanted to print from it.
The hard way, come up with a latching port scheme and connect to the I/O
port. This would require another poke...to configure external I/O ...but it
is doable. I was quickly toying with the idea to solder to the 45 kit from
boondog.com. I keep bringing that site up because I was intrigued by the
cost of the kit and I have been itching for a reason to get one.
So who wants to rain on my parade now?
Anthony Marchini
P.S. What is the procedure for a /DCT driver??? that would besomething to
load. The vector for the OS would be the biggy too i suppose.
Lord Apollyon wrote in message ...
>In article <Ym0Y3.436$9F4....@newsfeed.slurp.net>, "Anthony Marchini"
><marchini@pe_nn.c1o1m> wrote:
>
>You get the King Hardware Hack Skank 1999 award. :) Good lord. All of
>this just to avoid having to bootstrap from floppy? If your floppies are
>decent shape, why not just make a /DCT driver for LDOS to use the parallel
>port as the interface, at least in your first "cut" of the system. I
>mean, there's a fair bit of work to get that going properly.
>
Anthony Marchini
Lord Apollyon
uhclem...@nemesis.lonestar.org (Frank Durda IV) wrote:
Frank, if the copyright isn't held by Microsoft, is it held by Tandy? Is
it legal to share the sources to these? Are you willing to share the
source code you have to the ROMs? [Obviously, you will probably not wish
to share the sources to the ROMs you sell]
=R=
--
The reply-to-address will expire on Midnight 1-December-1999.
Spammers: You will lose your network access. Guaranteed.
96 domains, 356 web-accounts, and 560 dialup ISP accounts flushed.
Lord Apollyon
<marchini@pe_nn.c1o1m> wrote:
> Originally (and i was wrong) I thought the hard drives of the Mod III were
> able to usurp control of the Mod III bus, holding it in a reset state or
> something.
I think with the Model III's, they had to use a boot floppy. (If this is
wrong, someone will rush to correct me and crack my head with a ruler).
> Now, i am on another tac.... The floppy connector. Does anyone remember if
> you can boot from any drive or not? I forget, and the MOD III is in storage.
I'm pretty sure it's Drive 0 only.
> I bring it up because I Am thinking that I could hook and interface onto the
> external floppy port and when the system boots up, emulate a floppy
> drive....which would be nuts, but probably pretty easy for a quick PIC
> microcontroller (that would probably end up being more powerful than the Z80
> i want to interface to) to emulate the signals to the floppy port necessary
> for the MOD III to think that a floppy was there and to load in a bootstrap,
> which would then...........
> access the parallel port , because it would be easier than communicating
> throught the floppy port and then the program on the PC Emulating a hard
> drive...
You get the King Hardware Hack Skank 1999 award. :) Good lord. All of
this just to avoid having to bootstrap from floppy? If your floppies are
decent shape, why not just make a /DCT driver for LDOS to use the parallel
port as the interface, at least in your first "cut" of the system. I
mean, there's a fair bit of work to get that going properly.
If you're savvy with a PIC and hardware, then maybe a patch to the Model
III ROM to bootstrap via this parallel port scheme could be in order.
Keep in mind, even though it's tight in there, you could slim down some
routines and use ultra-lean skanko-bootstrap parallel schemes [just to
snag the first 256 or so bytes to get a more complex loader into RAM,
where you can do the super-freeky loader stuff]. For leanness of code, it
might be a "serial" transmission scheme that uses just one line of the
parallel port to snag the bootstrap code from the PC-side - whatever skank
it would take to make the z80-side code as small as possible - who cares
how large the pc-side stuff is - it'll be tiny anyway by comparison.
Did those serial-networked Model III's boot over the net, or did they
require a special boot floppy as well? If they booted over the net, then
you could use that net protocol for booting into the hella-funky HD
driver.
> So in essance... a solid state boot floppy sector to launch a parallel port
> (or I/O port) boot program that would load an operating system through an
> existing PC and no moving parts.... excepting the PC.
I think emulating the floppy stuff would be a significant project unto
itself - I think you might be able to do different skank a bit more
easily.
> I am one of those people who always sees the hard way of doing things.
No, I understand perfectly. One doesn't mind expending a bit of effort so
that we can expend zero effort to make use of something repeatedly. Works
for me. :)
Lord Apollyon
<marchini@pe_nn.c1o1m> wrote:
> Wait, I got IT!
> The machine boots into basic if you hold done break right? How bout...
> Type in a quick basic program, an easy to remember one...
>
> 100 for X% = 1 to N (however many bytes)
[...]
Egads, that's not only skanky, it's tedious! A special boot floppy or a
slightly patched ROM would be MUCH better. I think those ROM chips were
socketed, so the challenge would be to find EPROMs that are signal and
form-factor compatible.
> Then I just need Jeff Vavasour or
> Mathew Reed to loan (ok..ill give you money) the source code for floppy
> emulation and wammo. PC plays floppy to a real TRS80..
Silly goat. You've forgotten Tim Mann's "xtrs" emulator, which has full
source code available. His emulator fully supports the WDC controller
commands and can read "protected" floppies. Tsk tsk.
> The cool thing about this, is that a person with a MOD III without floppys
> could then experience the thrill of running the machine with floppys. I
> wonder how many of those are sitting in a pile of dust waiting to be played
> with.
Yes, you could work around dead floppies with this approach. Though if
you're that hard-up to get your TRS-80 working, I daresay making a ROM
patch would probably be 1) easier, 2) alot cheaper, 3) more painless to
use.
> My problem, the easy way to go would be to connect the pc to the printer
> port of the TRS80, which would put it out of action. This is a problem if
> the wanted to print from it.
*Laugh* he wants his cake, and eat it too! :)
> P.S. What is the procedure for a /DCT driver??? that would besomething to
> load. The vector for the OS would be the biggy too i suppose.
/DCT would be post OS load... But I think if you weren't afraid of some
hackery, you could get a small LDOS loader that'd use your driver to load
the rest of its ass into RAM and complete the bootstrap process. [In
effect, replace the floppy/dct with yours, internally to LDOS, that is].
There are REAMS of excellent documentation and source code [yes, it's to
LDOS 6, but it will be VERY very instructive for LDOS 5 as well] on Tim
Mann's site.
Do yourself a favour and check it out!
=R=
Ward Griffiths
>
> Ward Griffiths (wdg...@home.com) wrote:
> : The bulk of the Mod 1/3/4 ROM set is copyrighted by Microsoft.
>
> No it isn't. None of the Model I Level 1, Model III or 4 ROM is copyrighted
> by Microsoft.
I know that the Mod 1/3 Level 1 BASIC wasn't by Microsoft at all.
The Level 2 (Mod 1/3/4) was (Hell, you know at least as well as I
do) written by MS and while the specific release was copyrighted
by Tandy/RS, the long-term rights were held by MS -- much like
magazine publication is copyrighted by the magazine publisher for
"first serial publication" but the author (nowadays) retains the
right to further use unless contractually surrendered (very common
backalong, but nowadays writers sometimes _read_ contracts and the
laws that apply to them).
Ward Griffiths
>
> Ward Griffiths (wdg...@home.com) wrote:
> : The bulk of the Mod 1/3/4 ROM set is copyrighted by Microsoft.
>
> No it isn't. None of the Model I Level 1, Model III or 4 ROM is copyrighted
> by Microsoft.
And by "the bulk of", Level 1 BASIC was 4k; Level 2 was 12k,
Model 3/4 ROM BASIC was 14k. And Level 1 was not a regular part
of the ROM set, it was usually sort of alone.
Ken Walker
Can you give us a copy of the ROM sources (Tandy)?
Lord Apollyon
> I know that the Mod 1/3 Level 1 BASIC wasn't by Microsoft at all.
> The Level 2 (Mod 1/3/4) was (Hell, you know at least as well as I
> do) written by MS and while the specific release was copyrighted
> by Tandy/RS, the long-term rights were held by MS -- much like
> magazine publication is copyrighted by the magazine publisher for
> "first serial publication" but the author (nowadays) retains the
> right to further use unless contractually surrendered (very common
> backalong, but nowadays writers sometimes _read_ contracts and the
> laws that apply to them).
I've since located the Model 1 Level II ROM sources and they appear to be
an 8080-assembler BASIC core [with occasional DC.B's to add the z80
specific mnemonics], with a TRS-80 specific "prologue" tacked on the front
of it. The source credits indicate "BILL GATES, PAUL ALLEN, and MONTE
DAVIDOFF". Monte evidently wrote the entire mathematics section of
BASIC. The root (c) of the BASIC was listed as 1975, with lots of ALTAIR
references, however, the version number of the BASIC was "4.46".
I suspect M$ got lots of mileage out of that 8080 code base, and just
added machine-specific personality prologues to it at the low addresses.
The assembler format is some ultra-ancient predecessor of a MACRO-11esque
[any old VAX programmers out there? :) ] x-assembler that ran on a PDP-10
[according to the code notes].
Doing some random object code to source code spot checks, it's either ROM
A or ROM B. [It has "RADIO SHACK LEVEL II BASIC" rather than "R/S L2
BASIC"], and the first 200 bytes and the final 200 match perfectly to ROM
images I possess.
I have plenty of x-assemblers that could assemble this, but the directives
and pseudo-ops would need translation.
Once I have it assembling easily, I could difference the ROM with the
Model III's and create commented source that would build Model III ROMs.
How many Model III ROM versions were there? I have 3 ROM versions for the
Model 1, but only 1 for the Model III. Were the ModelIII/A files
different from the on-chip ROM of a Model 3?
Frank Durda IV
: The assembler format is some ultra-ancient predecessor of a MACRO-11esque
: [any old VAX programmers out there? :) ] x-assembler that ran on a PDP-10
: [according to the code notes].
Real genuine Microsoft BASIC up through the Model 100 and Model 4 disk BASIC
(not Model 4 ROM BASIC) was assembled on a DECsystem-20 using the MACRO-20
assembler. (I liked TOPS-20, really miss it.) Using its macro/conditional
language, MACRO-20 was able to take a single source tree and generate output
for incompatible hardware platforms including different CPU architectures.
8080/6502 and others were produced by this thing. Microsoft had at least two
DEC-20s, one was called "Central Heating Plant" and the other "Miss Piggy".
(Well named as behind the "E" box of a DEC-20, all that ECL circuitry heated
things up nicely...) The mnemonics MS invented for this "language" show the
8080 origins of MS-BASIC, but that didn't stop it generating output for 6502s,
8080s, even TI994a systems all from the same source tree.
This meant good and bad things happened, like a bug fix some other vendor
entered for some 6502 platform would end up in our 8085 ROMs (usually without
MS ever telling us), and some bug introduced by making a change for another
vendor found its way into our trees as well or vice versa.
The MS royalty issue was the reason Tandy went their own way with the later
Model III ROMs. By 1982, no royalty payments were made for Model III platforms
to Microsoft at all because Tandy felt all the code contained in the platform
was Tandys. A variant of that "MS-free" ROM set was used in the Model 4/4D/4P
systems, with all of this ROM code completely assembled at Tandy using ALDS
and before that, a cross-assembler running on a Tandem non-stop system, later
on a VAX-11/780 system, and finally the work was moved to a BSD 4.1 port of
ALDS and finally done on real TRSDOS systems using ALDS.
: Doing some random object code to source code spot checks, it's either ROM
: A or ROM B. [It has "RADIO SHACK LEVEL II BASIC" rather than "R/S L2
: BASIC"], and the first 200 bytes and the final 200 match perfectly to ROM
: images I possess.
For entry point compatibility that is no surprise. Similar lengthy sequences
can be found in PC-compatible BIOS ROMs simply because they have to be that
way to be compatible, even though the code bases were independently developed.
As an example, Tandy had to change all of their BIOS ROMs to enter and exit
subroutines in the Tandy 1000 BIOS with a sequence other than the traditional
PUSH AX/PUSH BX/PUSH CX/PUSH DX/PUSH DI/PUSH SI and the matching POP SI/POP DI/
POP DX/POP CX/POP BX/POP AX because IBM happened to use this same push/pop
order (it is logical and what Intel recommends in their books) and the US
Customs department took the frequent occurrence of this sequence in comparing
Tandy ROMs against true-blue IBM BIOS ROMs to be copyright infringement of
the real genuine IBM ROMs, so they started confiscating Tandys ROMs that
were fabbed in Japan which were destined for the Tandy 1000 systems. Clearly
an unintentional similarity, but detected by using simplistic comparisons
of bytes. Vector tables and hardware initialization tables cause similar
roblems because they pretty much have to be the same bytes in the same order
and the same location.
: How many Model III ROM versions were there? I have 3 ROM versions for the
: Model 1, but only 1 for the Model III. Were the ModelIII/A files
: different from the on-chip ROM of a Model 3?
Easily over two dozen different versions were released by Tandy to
manufacturing of Model III and Model 4 ROMsets. For starters, there was US,
German, French, US with Network III, US with Network 4, US with various bug
fixes, repeats of German, French, Model 4 initialization cores for all three
languages, revised releases of that, with Network III, with Network 4,
stir and repeat. Lots and lots of releases. (The Model III TRSDOS memtest
program lists three different checksums for just one ROM as valid but there
were many others.) The "C" ROM was the most frequently-changed ROM, on
purpose.
Lord Apollyon
uhclem...@nemesis.lonestar.org (Frank Durda IV) wrote:
> Real genuine Microsoft BASIC up through the Model 100 and Model 4 disk BASIC
> (not Model 4 ROM BASIC) was assembled on a DECsystem-20 using the MACRO-20
> assembler. (I liked TOPS-20, really miss it.)
Heheh, this is typical. Nearly everyone I know who used TOPS-20 has
nothing but praise for it. How much of VMS was like TOPS-20? I enjoyed
VMS, and VAX assembler very much, actually. I was stunned at how
high-level the instructions were [some could do binary->ASCII conversions
with some formatting, polynomial evaluation, CRC-calculations of a buffer,
etc. And then the writeable control store, where a suitably priv'ed piece
of code could augment the instruction set... [not just via an
unimplemented exception vector either]. Most sexy.
> (Well named as behind the "E" box of a DEC-20, all that ECL circuitry heated
> things up nicely...) The mnemonics MS invented for this "language" show the
> 8080 origins of MS-BASIC, but that didn't stop it generating output for 6502s,
> 8080s, even TI994a systems all from the same source tree.
Sounds optimal perhaps from a porting speed issue, but non-optimal for
performance... What it sounds like (and somewhat looks like) is
abstracted assembly language via the 8080. Going "up" to a Z80, it's
probably not so bad. But going "down" to a 6502, it's not going to be as
fast as it could be. I know the TI-99/4a's CPU was pretty trick - some
folks used to call it "very FORTH-like".
> The MS royalty issue was the reason Tandy went their own way with the later
> Model III ROMs. By 1982, no royalty payments were made for Model III
> platforms to Microsoft at all because Tandy felt all the code contained
in the
> platform was Tandys.
Interesting, but was it M$-free by then? Id est, did Tandy re-implement
the ROM code from scratch in a clean-room development environment, with
the requisite entry points being specified of course?
> ALDS and finally done on real TRSDOS systems using ALDS.
What does ALDS stand for? Was it a Tandy-only piece of software, or was
it commonly available? [And does a copy exist in one of the net-archives
for the TRS80?]
> Easily over two dozen different versions were released by Tandy to
> manufacturing of Model III and Model 4 ROMsets. For starters, there was US,
> German, French, US with Network III, US with Network 4, US with various bug
> fixes, repeats of German, French, Model 4 initialization cores for all three
> languages, revised releases of that, with Network III, with Network 4,
> stir and repeat.
Eesh, what a mess. Actually, I'm amazed they didn't break anything with
all of that. All it would take is for a routine entry point to shift by
one, crash-o-rama, baybee. I guess this would be why the changes were
kept to ROM C, seeing as most of the entry points used by app-writers were
low-address routines/vectors.
So, Frank... will you share any of these Model 3 ROM sources with us?
=Rob=
--
The reply-to-address will expire on Midnight 1-December-1999.
Spammers: You will lose your network access. Guaranteed.
97 domains, 359 web-accounts, and 561 dialup ISP accounts flushed.
Tim Mann
> the ROM code from scratch in a clean-room development environment, with
> the requisite entry points being specified of course?
You could do a byte-for-byte comparison of a Model I and Model III ROM
and draw your own conclusions. It looks to me like the I/O drivers were
completely rewritten, but the core Basic interpreter was nearly unchanged.
Only a few scattered bytes differ between addresses 0x709 and 0x2B85,
and much between 0x2B86 and 0x2FFF is unchanged as well. However, the
code below 0x709 is very different, and the 2K of code above 0x2FFF is
all new.
I don't claim to know why this was done or what the legal implications
were in this particular case, since I don't know any details of the
agreements between Tandy and Microsoft, and I am definitely not a lawyer.
Tim Mann <ma...@pa.dec.com> Compaq Systems Research Center
http://www.research.compaq.com/SRC/personal/Tim_Mann/
Frank Durda IV
: Sounds optimal perhaps from a porting speed issue, but non-optimal for
: performance... What it sounds like (and somewhat looks like) is
: abstracted assembly language via the 8080. Going "up" to a Z80, it's
: probably not so bad.
It was estimated at the time that 15-20% additional speed could have been
gained by deliberately using Z-80 instructions rather than sticking to
pure-8080 opcodes (getting to use ld hl,de instead of mov h,d and mov l,e
at 6 clocks versus 8 on a Z80 is one example) and using the same algorithms
which they did unless they had to interface with the OS in some way, such as
the LS-DOS/TRSDOS 6 IY pointer to flags. They used subroutines for anything
that had to touch local CPU opcodes like the IX and IY registers of the Z-80.
I recoded several of the functions in Model 4 disk BASIC in real Z-80 using
tighter algorithms and code, which sped up test programs by over 30%.
(MS Model 4 Disk BASIC was also produced from the same MACRO-20/8080ized
source files, so there is fat in there.) Unfortunately, one of the
modifications has a bug in it that I never got around to identifying, so that
work is not really of much use. The original goal was to speed it up and
re-write over 12% of the work (at the time, that amount of unique code was
considered enough to avoid copyright issues for the entire work). I forget
whether I reached that goal of percentage re-write or not.
I can say that from studying the original MS source code common to the Model
100 and Model 4, I can tell you that Paul was the math brains in MS BASIC,
not Bill. Either that or Paul remembered to put his name in the comment block
at the start of more modules than Bill did. Most of the code clearly has
two different coding styles.
: > The MS royalty issue was the reason Tandy went their own way with the later
: > Model III ROMs. By 1982, no royalty payments were made for Model III
: > platforms to Microsoft at all because Tandy felt all the code contained
: > in the platform was Tandys.
: Interesting, but was it M$-free by then?
In Tandys mind, yes.
: Id est, did Tandy re-implement : the ROM code from scratch in a clean-room
: development environment, with the requisite entry points being specified
: of course?
Since Tandy defined the entry points to MS, no big deal (The BIOS layer in
the "A" ROM and all of the "C" ROM was always a Tandy creation). As to a
"clean-room", the idea of using that as a way to absolutely protect yourself
against cloning lawsuits wasn't to come along for another couple of years,
so the answer is no.
In any event, Tandy walking away from Microsoft on the Model III never appeared
to be an issue (Tandy had a period when telling Microsoft to drop-dead was
a daily occurrence), and Microsoft never pushed the point until the Model 4
came along, where they demanded that Tandy buy Disk BASIC from MS for the
Model 4 *IF* Tandy ever expected to see BASCOM, Multiplan, MS-Fortran or any
other MS products for that platform. This is the very sort of "tying" (forcing
the purchase of one unrelated product in order to be able to purchase other
products) that has gotten MS into trouble fifteen years later. I still don't
understand why everyone was so surprised about the antics of Windows '95
and '98 "bundling" and "tying" as the company had been doing it for years,
only they used to just spell the technique differently: "Blackmail".
Of course, when Tandy didn't want XENIX (we were busily putting the finishing
touches on UNOS for the Model 16), Microsoft sweetened the pot by saying
they would let Tandy have XENIX Multiplan and XENIX BASIC royalty-free, so
that Tandy made almost pure-profit on those two applications.
: > ALDS and finally done on real TRSDOS systems using ALDS.
: What does ALDS stand for?
Assembly Language Development System. The Assembler internals were taken
almost line for line from an article someone found in a magazine. I still
have the "source paper" for that assembler around here somewhere.
: Was it a Tandy-only piece of software, or was
: it commonly available?
Tandy sold versions of ALDS for the Model II, III and 4 and it had an utility
for moving files between these platforms via the serial port (ALTRAN). There
was also a port of the ALDS assembler that ran on our VAX under BSD 4.x which
shared many of the same assembler features but was written in C. ALDS had
some goofy pseudo-ops, like one for generating roman numeral strings.
: [And does a copy exist in one of the net-archives : for the TRS80?]
No idea.
: So, Frank... will you share any of these Model 3 ROM sources with us?
There are issues that have to be worked out on some items. I will not go into
what they are in a public forum, period. I will say that impatience will
likely make it not happen for a long long time, so I ask all to stop waving
the large red flag. When items appear on the web site, they are available.
Ward Griffiths
I know _I_ never had sources. I was never anything than a tech
support guy in the RSCC in Los Angeles (who kept getting called
from anywhere East late in the day for Xenix problems when the
stores in other time zones had closed). On copyright stati, I'll
take Frank's word, he was there in Texas, I spent only a week and
a bit there one time when they decided to train RSCC CSRs in what
they'd been doing for(seemed-like)ever, though I did learn a lot
about CP/M 3.0 that was interesting.
Frank, I was unaware of the pseudo-clean-room status of the later
Z-80 BASICs. Thank you, as I had assumed that MS held nearly
full title on those. I am _glad_ to be corrected. Now about the
Color Computer BASIC interpreters ...
And will somebody give me a status on the Scripsit-16 source
code? I know John Esak _used_ to have it, but my emails to him
bounce lately. And I'd really like to see it run with Linux.
Lord Apollyon
> gained by deliberately using Z-80 instructions rather than sticking to
> pure-8080 opcodes
I'm not at all surprised. The Z80 had many nifty features and
enhancements over the 8080. And abstracting via the 8080 ensured you
never used any of them, and in many instances, used 2 (or more)
instructions, where one could have done the job.
> Of course, when Tandy didn't want XENIX (we were busily putting the finishing
> touches on UNOS for the Model 16), Microsoft sweetened the pot by saying
> they would let Tandy have XENIX Multiplan and XENIX BASIC royalty-free, so
> that Tandy made almost pure-profit on those two applications.
My goodness. They've (M$oft) been doing this crap for their entire
existence, by the looks of things. I was just a kid back then, so the
"politics" of all of this were either not apparent to me, or not even
within my awareness map.
> Assembly Language Development System. The Assembler internals were taken
> almost line for line from an article someone found in a magazine. I still
> have the "source paper" for that assembler around here somewhere.
Fascinating.... it's amazing how long-lived some things are. I know *ASMH
(IBM 370 assembler) was the model for many personal computer assemblers
(like ORCA/M for the Apple ][)... a monster of an assembler it was even.
> Tandy sold versions of ALDS for the Model II, III and 4 and it had an utility
> for moving files between these platforms via the serial port (ALTRAN). There
> was also a port of the ALDS assembler that ran on our VAX under BSD 4.x which
> shared many of the same assembler features but was written in C. ALDS had
> some goofy pseudo-ops, like one for generating roman numeral strings.
OOOKAY. Most bizarre. Creeping featurism, eh?
> There are issues that have to be worked out on some items. I will not go into
> what they are in a public forum, period.
OK, OK. Though I never understand why "gagged" people can't ever say why
they're gagged. :) It's been a perennial mystery to me.
> I will say that impatience will
> likely make it not happen for a long long time, so I ask all to stop waving
> the large red flag. When items appear on the web site, they are available.
Fair enough. Heheh, it was the lack of a reply to several direct
questions which got us all excited. :) Please don't take it as disrespect
- we're just a passionate lot, that's all. Really, Frank, I'm sure the
issues which preclude your sharing them are serious, so it'd be remiss of
us to make "demands" on you. We as a community are quite a great deal
indebted to you; do not think we are not mindful of this.
Frank Durda IV
: And will somebody give me a status on the Scripsit-16 source
: code? I know John Esak _used_ to have it, but my emails to him
: bounce lately. And I'd really like to see it run with Linux.
Around 1988, Tandy sold a license (or substantial rights) to Scripsit to
an Washington DC-based company, who did ports for IBM mainframes and other
heavy iron environments. Kevin Dack, the creator of Scripsit, left Tandy
shortly after that to consult and later work for this company. Kevins
unexpected departure that summer forced me into the Deskmate arena to complete
work on the Deskmate sampled-music and sound programs he had started.
It was never entirely clear to me if Tandy sold all rights to Scripsit, or
just a license to do ports for non-competing platforms (plus use of the name)
to the other company. Tandy had aleady pretty-much abandoned it, apart
from releasing Scripsit PC and then not actually advertising or marketing it
at all.
Superscriptsit, was an unrelated product that came from a guy named Sam
Soloman/Solomon (can't remember which). He didn't work for Tandy. Tandy
did do support and maintenance on this code, particularly for the Model 4
version. It had lots of bugs and Sam didn't believe in re-assembly; he did
all fixes - no matter how massive - with patches.
Ward Griffiths
>
> Ward Griffiths (wdg...@home.com) wrote:
> : And will somebody give me a status on the Scripsit-16 source
> : code? I know John Esak _used_ to have it, but my emails to him
> : bounce lately. And I'd really like to see it run with Linux.
>
> Around 1988, Tandy sold a license (or substantial rights) to Scripsit to
> an Washington DC-based company, who did ports for IBM mainframes and other
> heavy iron environments. Kevin Dack, the creator of Scripsit, left Tandy
> shortly after that to consult and later work for this company. Kevins
> unexpected departure that summer forced me into the Deskmate arena to complete
> work on the Deskmate sampled-music and sound programs he had started.
>
> It was never entirely clear to me if Tandy sold all rights to Scripsit, or
> just a license to do ports for non-competing platforms (plus use of the name)
> to the other company. Tandy had aleady pretty-much abandoned it, apart
> from releasing Scripsit PC and then not actually advertising or marketing it
> at all.
The Washington DC based company (I forget the name) was basically
John Esak who was for many years the top filePro guru on the Tandy
Professional/Tangent SIG on Compuserve. Several times I offered
help in doing a port to the AT&T Unix PC, but he declined since by
then the Unix PC was as dead an issue as the Tandy 6000.
> Superscriptsit, was an unrelated product that came from a guy named Sam
> Soloman/Solomon (can't remember which). He didn't work for Tandy. Tandy
> did do support and maintenance on this code, particularly for the Model 4
> version. It had lots of bugs and Sam didn't believe in re-assembly; he did
> all fixes - no matter how massive - with patches.
Probably my least favorite product that ever bore the Scripsit
name, and I'm including the cassette-based products. I was
always having to try to salvage scrambled documents (with widely
varying degrees of success). I usually used those as
opportunities to recommend an Express Order of Prosoft's
AllWrite product, which is what I used on my own 1/3/4 systems.
(And occasionally under LS-DOS on my Model 16s).
Model 4 SuperScripsit eventually achieved some degree of stability.
That can not be said for the Model 3 version.
Neil Morrison
16K version of 8080 basic. When Tandy wanted a 12K, they hacked bits of
the 16K out to get the size down, but left in the 'hooks' so the disk
could have the full version. They weren't even sure this would work, and
were happy when it did. (They even had a tape version of the 16K -
remember 'Level III Basic' on tape?)
BGates said in the interview that he might release the source, but AFAIK
he never did. I guess we will have to pry it out of his cold dead hands.
RANT WARNING !!!
Why is that patents have so short a life, so the inventors of the Xerox,
and television basically got nothing, whereas copyrights are so long
that we still have to worry about dead code, out of print books, and
forgotten music from dead composers? Now Sony (I think it is?) is trying
to do a sneaky deal with Congress to extend copyrights even longer,
after two previous extensions, so they can maintain their monopoly and
screw even more money out of us for CDs! Must be something to do with
money, right?
RANT ENDS.
NM
Mark P. Fishman
> Model 4 SuperScripsit eventually achieved some degree of stability.
> That can not be said for the Model 3 version.
Really? Are you sure you don't mean Scripsit Pro (Model 4 only)?
That was quite reliable. My understanding is that while Model
III SuperScripsit could be used most of the time, Model 4
SuperScripsit was almost bound to crash on heavily edited
documents.
Cheers -- Mark F.
--
reply to: mfishman at alum dot em eye tee dot ee dee ewe
Lord Apollyon
> RANT WARNING !!!
> Why is that patents have so short a life, so the inventors of the Xerox,
> and television basically got nothing, whereas copyrights are so long
> that we still have to worry about dead code, out of print books, and
> forgotten music from dead composers? Now Sony (I think it is?) is trying
> to do a sneaky deal with Congress to extend copyrights even longer,
> after two previous extensions, so they can maintain their monopoly and
> screw even more money out of us for CDs! Must be something to do with
> money, right?
> RANT ENDS.
Actually, the CD standard is shared with Philips. If you wish to have
that "CD" logo, which depicts that your device is fully conformant with
the specifications, you pay your $5000 [per product] to Philips. If you
want the complete Red, Yellow, Orange, and Blue books for the standards,
yes, that's right, another $5,000 goes to Eindhoven Netherlands.
Actually, if you want to hear a really nasty tale about the evil
application of copyright....
There's a small privately printed little book by the Inklings [a literary
group consisting of the 'great Philologists' of the early 20th century [EV
Gardner, Tolkien, CS Lewis, etc].
60 copies were known to be printed. All of the copyright holders are
dead. Only 8 or 9 libraries have copies, most of these are in closed
stacks [we need to check your stool samples and make sure you're worthy to
even touch this book sorta garbage].
It turns out the British Public Library in London has a copy. And so when
I was in London back in 1987, I and a friend who were very much into
ancient languages [Latin, Anglo-Saxon, etc].... went to view this book.
It was in a special area of the library, where all of your pens, bags, and
the like are seized and held ransom by the library staff nazis.
You were forbidden to use the photocopier on more than 2 pages, and there
were fuckwads there to make sure you didn't commit the sin of exceeding
your 2 page limit. No amount of reasoning, pleading, begging, or trying
to appeal to the rationale of preserving and making available to more
people these luminaries' works.
For them, keeping them locked away and rotting unread is preferable.
As it turned out, the book is only 60 pages. My friend and I copied the
old fashioned way: by hand with pencils on paper scraps. [We joked it was
like the old monk-scribes of yore] - it was slow going in some of them
because Anglo-Saxon uses some odd letters which are hard to scribble
legibly. [eth and thorn].
Anyway, this is way off-topic for the newsgroup, but a very nasty example
of the "prevent people from being able to access an old and forgotten work
at all costs, so long as copyright propriety is maintained [SPIT!]". In
my example, there was clearly ZERO economic value to the copyright. A
collection of 30 songs and poems in languages very few people can even
read isn't likely to top the NY Times Bestseller charts.
The attitude of the librarians was astounding.
=R=
--
The reply-to-address will expire on Midnight 1-January-2000.
Spammers: You will lose your network access. Guaranteed.
99 domains, 360 web-accounts, and 561 dialup ISP accounts flushed.
Sylvan Butler
>Model 4 SuperScripsit eventually achieved some degree of stability.
>That can not be said for the Model 3 version.
Scripsit, the original Model I version, is what got me started with
small, efficient text editors. It was also my first "big" reverse
engineering project. I hope that binder of annotated disassembly is
still around somplace...
SuperScripsit was horrendous.
When I needed more word processing power than Scripsit, SuperScripsit was
not an option. Instead I switched to Zorlof which later became LeScript,
IIRC... That was an amazingly powerful package for the time/platform.
sdb
--
| Sylvan Butler | Not speaking for Hewlett-Packard | sbutler-boi.hp.com |
| Watch out for my e-mail address. Thank UCE. #### change ^ to @ #### |
They that can give up essential liberty to obtain a little temporary
safety deserve neither liberty nor safety. --Benjamin Franklin, 1759
"Don't Tread On Me!"
Jim K
ago to see if they'd moved the program to MSDOS but they seem to have
faded. I liked the /// for italics adn *** for bold -- neat idea.
On 22 Nov 1999 20:28:37 GMT,
nospam+_n...@hpb13799z.zboi.hpz.com.invalid (Sylvan Butler)
wrote:
>
>When I needed more word processing power than Scripsit, SuperScripsit was
>not an option. Instead I switched to Zorlof which later became LeScript,
>IIRC... That was an amazingly powerful package for the time/platform.
>
>sdb
Jim Kajpust - Personal Responsibility - Personal Freedoms
Michigan - www.concentric.net/~jkajpust
Ward Griffiths
>
> I loved LeScript. I tried to track down the company a couple of years
> ago to see if they'd moved the program to MSDOS but they seem to have
> faded. I liked the /// for italics adn *** for bold -- neat idea.
Le Script was OK, but I went with AllWrite because I'd gotten
into nroff with Xenix, and AllWrite was very similar --with a
quick search and/or destroy replacing semi-colons with periods
and vice versa, the files were over 90% compatible. That and
the editor being separate from the formatter -- I didn't like
WYSIWYG then and I still don't, I use vi to edit the little
HTML that I do.
Mark P. Fishman
> I loved LeScript. I tried to track down the company a couple of years
> ago to see if they'd moved the program to MSDOS but they seem to have
> faded. I liked the /// for italics adn *** for bold -- neat idea.
Yes, they had MS-DOS versions. I have a 1.80 and a 2.0 for
MS-DOS (IBM-compatible, really). They faded somewhere around
there.
2.0 was rather different from the 1.x series: changes to
internal file format, and other "enhancements". The changes
were supposed to improve performance.
Anitek does seem to have been into direct hardware accesses.
It's harder to run LeScript on the emulators than some other
programs.
Jan Vanden Bossche (aka the TyRannoSaurus)
Jim K <jkaj...@concentric.net> wrote in article
<Oxw6OC5KFiShY+...@4ax.com>...
> I loved LeScript. I tried to track down the company a couple of years
> ago to see if they'd moved the program to MSDOS
Lescript for DOS did exist. I think I had a copy, once. It was described in
80-micro, somewhere in 87, I think.
I still use Lescript 1.4 or 1.7, on all my TRS-80's. If you want to do
crazy things with your matrix printer, you can't beat its direct coding, or
macro-facility. Still love it.
And 1.7 could do collumns.
> but they seem to have
> faded. I liked the /// for italics adn *** for bold -- neat idea.
Yes. It was color-coding on a IBM color screen.
> On 22 Nov 1999 20:28:37 GMT, (Sylvan Butler) wrote:
>
> > Instead I switched to Zorlof which later became LeScript,
> > IIRC... That was an amazingly powerful package for the
> > time/platform.
> >sdb
>
> Jim Kajpust - Personal Responsibility - Personal Freedoms
> Michigan - www.concentric.net/~jkajpust
Greetings from the TyRannoSaurus
Jan-80
Jan Vanden Bossche (aka the TyRannoSaurus)
Mark P. Fishman <mfis...@ll.mit.edu> wrote in article
<Pine.LNX.4.05.99112...@shmulin.llan.ll.mit.edu>...
> On 23 Nov 1999, Jim K wrote:
>
> > I loved LeScript.
I still do.
> Yes, they had MS-DOS versions. I have a 1.80 and a 2.0 for
> MS-DOS (IBM-compatible, really). They faded somewhere around
> there.
>
> 2.0 was rather different from the 1.x series: changes to
> internal file format, and other "enhancements". The changes
> were supposed to improve performance.
But didn't...
2.0 could work with an alternate (IBM) character set or an Hi-Res board to
give borders etc.
> Anitek does seem to have been into direct hardware accesses.
> It's harder to run LeScript on the emulators than some other
> programs.
Virtually impossible on Jeff V's.
Lescript 1.4 and higher recognised a model 4 running in model III mode and
switched to the 80x24 screen and back when exiting. Neat.
After running Lescript 1.4 on a 4, you had to set the system back into
standard high speed: SYSTEM (FAST).
> Cheers -- Mark F.
Kenneth Brody
[...]
> And will somebody give me a status on the Scripsit-16 source
> code? I know John Esak _used_ to have it, but my emails to him
> bounce lately. And I'd really like to see it run with Linux.
John still has/owns the Scripsit code, but he told me getting him "to
compile it on Linux for profit, would be like trying to get Howard Stern
to be nice to people... not going to happen in this life time."
--
+---------+----------------------------------+-----------------------------+
| Kenneth | kenb...@bestweb.net | "The opinions expressed |
| J. | | herein are not necessarily |
| Brody | http://www.bestweb.net/~kenbrody | those of fP Technologies." |
+---------+----------------------------------+-----------------------------+
GCS (ver 3.12) d- s+++: a C++$(+++) ULAVHSC^++++$ P+>+++ L+(++) E-(---)
W++ N+ o+ K(---) w@ M@ V- PS++(+) PE@ Y+ PGP-(+) t+ R@ tv+() b+
DI+(++++) D---() G e* h---- r+++ y?
Sylvan Butler
>Ward Griffiths wrote:
>[...]
>> And will somebody give me a status on the Scripsit-16 source
>> code? I know John Esak _used_ to have it, but my emails to him
>> bounce lately. And I'd really like to see it run with Linux.
>[...]
>
>John still has/owns the Scripsit code, but he told me getting him "to
>compile it on Linux for profit, would be like trying to get Howard Stern
>to be nice to people... not going to happen in this life time."
So does he just hate Linux?
Does he make any money from Scripsit anymore?
If not, he could consider releasing it under GNU as open source.
Mark P. Fishman
> John still has/owns the Scripsit[-16] code, but he told me
> getting him "to compile it on Linux for profit, would be
> like trying to get Howard Stern to be nice to people... not
> going to happen in this life time."
I won't even suggest he do it for free, but maybe he'd be
willing to let someone else try?
Lord Apollyon
Vanden Bossche (aka the TyRannoSaurus)" <Ja...@ping.be.nospamplease>
wrote:
> > Anitek does seem to have been into direct hardware accesses.
> > It's harder to run LeScript on the emulators than some other
> > programs.
>
> Virtually impossible on Jeff V's.
>
> Lescript 1.4 and higher recognised a model 4 running in model III mode and
> switched to the 80x24 screen and back when exiting. Neat.
How about Tim Mann's or Dave Keil's? Colour me silly perhaps, but I
consider those to be the "state of the art" in TRS80 emulator
compatibility.
sola...@don'tmesswithtexas.net
>My goodness. They've (M$oft) been doing this crap for their entire
>existence, by the looks of things. I was just a kid back then, so the
>"politics" of all of this were either not apparent to me, or not even
>within my awareness map.
It generally wasn't obvious to end-users what was going on, since the
"bundling" was a little different that what M$ has been doing lately. In
this case, for example, telling Tandy that "if you don't buy X from us for
your new TRS-80 model, we won't port Y and Z to your new model either"
wasn't something that the end user would really know about since X, Y, and
Z were still sold by Tandy as separate products.
-----------------------------------------------------------------------------
BREAK UP MICROSOFT!!
(Preferably with a wrecking ball and some dynamite.)
---------------------------------------------------------------------
solarfox@DON'TMESSWITHtexas.net (Gary Akins jr.)
http://lonestar.texas.net/~solarfox
---------------------------------------------------------------------
Jan Vanden Bossche (aka the TyRannoSaurus)
Sylvan Butler wrote
> >
> >John still has/owns the Scripsit code, but he told me getting him "to
> >compile it on Linux for profit, would be like trying to get Howard Stern
> >to be nice to people... not going to happen in this life time."
>
> So does he just hate Linux?
> Does he make any money from Scripsit anymore?
>
> If not, he could consider releasing it under GNU as open source.
If that exists on Linux, would be THE reason to start using Linux on my
systems.
> sdb
Kenneth Brody
>
> Frank Durda IV wrote:
[...]
> > It was never entirely clear to me if Tandy sold all rights to Scripsit, or
> > just a license to do ports for non-competing platforms (plus use of the name)
> > to the other company. Tandy had aleady pretty-much abandoned it, apart
> > from releasing Scripsit PC and then not actually advertising or marketing it
> > at all.
>
> The Washington DC based company (I forget the name) was basically
> John Esak who was for many years the top filePro guru on the Tandy
> Professional/Tangent SIG on Compuserve. Several times I offered
> help in doing a port to the AT&T Unix PC, but he declined since by
> then the Unix PC was as dead an issue as the Tandy 6000.
(Shameless plug follows...)
He was a vice president for The small Computer Company around that time,
though I don't know if he was still with "small" at the time that he bought
the rights to Scripsit. The small Computer Company wrote most of the
Profile series of software for Tandy computers [though the model 1 version
and I believe one of the model 3 versions were made by someone else] as
well as filePro and filePro Plus for non-Tandy's. In case anyone is
interested, the company is still around, but has moved to Indianapolis
and is now called "fP Technologies, Inc." Their web site is at
http://www.fileproplus.com
and
http://www.fptechonologies.com
They can also be reached at "sup...@fileproplus.com"
Ward Griffiths
> He was a vice president for The small Computer Company around that time,
> though I don't know if he was still with "small" at the time that he bought
> the rights to Scripsit. The small Computer Company wrote most of the
> Profile series of software for Tandy computers [though the model 1 version
> and I believe one of the model 3 versions were made by someone else] as
> well as filePro and filePro Plus for non-Tandy's. In case anyone is
> interested, the company is still around, but has moved to Indianapolis
> and is now called "fP Technologies, Inc." Their web site is at
"Profile" for the 1/3 (and Color Profile) were unrelated to
Profile II, Profile Plus and Profile 16, which were indeed by
the small Computer Company -- for the non-Tandy Unix market
and for enhancements that Tandy didn't license, filePro was
the name. I've got Profile 16 for my T6k and a DOS version of
filePro that isn't presently in use (since DOS isn't presently
in use, my TL/2 hasn't been plugged in for 2-3 years).
That's another product I'd like to see for Linux. We've got
DB2, Oracle, Sybase and other "state of the art" professional
database development systems. Gimme filePro. At least gimme
a reasonable price on filePro. I _hate_ SQL and all of its
heirs and assigns.
But Scripsit-16 first. I would rather write than file.
Mike Yetsko
files were 'EFC') and it was done on a Model 1, and then converted to
the Model II when that came out. (I think the very first Model II in
the
northeast US went to the author. Which promptly blew up, which is
another story in itself!)
Anyway, ProFile for the Model II was a hit. And pieces of it were
actually
pulled and sold as other products, like menu generators. Eventually
the buyer at Tandy almost had to bully Howie to 'convert' his software
BACK to the Model 1 type platform (including the Model 3/4) as a
ProFile Plus upgrade to the original ProFile. It was done, and sales
were fantastic!
Which is no wonder. ProFile+ from sCC was probably one of the best
software packages written in it's day. It was fast, easy to use, and
more reliable than any other piece of software I've ever played with
on a Model II. (Except my own, of course!)
Mike
Kenneth Brody <kenb...@bestweb.net> wrote in message
news:38441E63...@bestweb.net...
> Ward Griffiths wrote:
> >
> > Frank Durda IV wrote:
> [...]
> > > It was never entirely clear to me if Tandy sold all rights to
Scripsit, or
> > > just a license to do ports for non-competing platforms (plus use
of the name)
> > > to the other company. Tandy had aleady pretty-much abandoned it,
apart
> > > from releasing Scripsit PC and then not actually advertising or
marketing it
> > > at all.
> >
> > The Washington DC based company (I forget the name) was basically
> > John Esak who was for many years the top filePro guru on the Tandy
> > Professional/Tangent SIG on Compuserve. Several times I offered
> > help in doing a port to the AT&T Unix PC, but he declined since by
> > then the Unix PC was as dead an issue as the Tandy 6000.
> [...]
>
> (Shameless plug follows...)
>
> He was a vice president for The small Computer Company around that
time,
> though I don't know if he was still with "small" at the time that he
bought
> the rights to Scripsit. The small Computer Company wrote most of the
> Profile series of software for Tandy computers [though the model 1
version
> and I believe one of the model 3 versions were made by someone else]
as
> well as filePro and filePro Plus for non-Tandy's. In case anyone is
> interested, the company is still around, but has moved to Indianapolis
> and is now called "fP Technologies, Inc." Their web site is at
>
Kenneth Brody
[...]
> That's another product I'd like to see for Linux. We've got
> DB2, Oracle, Sybase and other "state of the art" professional
> database development systems. Gimme filePro. At least gimme
> a reasonable price on filePro. I _hate_ SQL and all of its
> heirs and assigns.
Yes, Virginia, there is a Santa Claus...
Contact <sa...@fileproplus.com> or (800)847-4740 and ask them for the
prices for the Linux version of filePro.
--
+---------+----------------------------------+-----------------------------+
Kenneth Brody
>
> The original ProFile was called The Electric File Clerk (that's why the
> files were 'EFC') and it was done on a Model 1, and then converted to
> the Model II when that came out. (I think the very first Model II in
> the
> northeast US went to the author. Which promptly blew up, which is
> another story in itself!)
As we used to say... "It's A small World".
I thought the first EFC was on the Model II, so I asked Howie about it.
He told me that, in fact, the Model I was the first version, though he
used it mostly as a tool for some jobs he was hired for. He also told
me about how he got (one of?) the first Model II's in New York, and when
it died, a Radio Shack repairman, Mike Yetsko, came out to fix it.
And, apparently, you were the root cause of the only Level 1 ("possible
hardware damage") bug in Profile known. :-)
[...]
> Which is no wonder. ProFile+ from sCC was probably one of the best
> software packages written in it's day. It was fast, easy to use, and
> more reliable than any other piece of software I've ever played with
> on a Model II. (Except my own, of course!)
[...]
"In its day"? It still is!
Mike Yetsko
itself. His first Model II blew up in days. And the computer store
snowballed him, so he got on the phone to Penny in Ft Worth.
I ended up being told to drive out to Long Island to meet Howie
at the computer center and fix his machine. (There's a 'Tandy
story' to that too!) I did, but it blew up again a couple days later.
That wasn't Howies fault. His Model II had a really wierd problem.
A bad connection on the socket for the video card caused data
bit 4 not to be reliable. As a result, the machine could turn on with
the video driver not initialized correctly, causing current to
surge and blow this $12 driver transistor on the monitor board.
While the socket was a rare problem, that driver blowing wasn't.
I ended up not replacing the boards, but keeping that transistor in
stock. Tandy ended up changing the Model II to a new video
monitor that had a 'free run' capability that wouldn't self destruct.
Thing is, it was intermittent as hell. I ended up driving up to Howies
on my own time to fix it again. I swapped his computer out eventually
and then found the intermittent socket by putting the machine in my
bathroom at home and 'steaming the place up'.
The hardware destruction for the model 1? You mean Howie
figuring out how to destroy the stepper motor on some drives?
Tell Howie I said hi! And by the way, I'm divorced now. Long story.
I'll be trying to get a more direct route to Howie.
Mike
Kenneth Brody <kenb...@bestweb.net> wrote in message
news:3845BA6B...@bestweb.net...
Kenneth Brody
[...]
> A bad connection on the socket for the video card caused data
> bit 4 not to be reliable. As a result, the machine could turn on with
> the video driver not initialized correctly, causing current to
> surge and blow this $12 driver transistor on the monitor board.
> While the socket was a rare problem, that driver blowing wasn't.
> I ended up not replacing the boards, but keeping that transistor in
> stock. Tandy ended up changing the Model II to a new video
> monitor that had a 'free run' capability that wouldn't self destruct.
> The hardware destruction for the model 1? You mean Howie
> figuring out how to destroy the stepper motor on some drives?
No, it was a model II. You showed him how the bank-siwtched memory worked
to allow direct access to the video memory, so EFC could draw directly to
the screen which was orders of magnitude faster than the BIOS calls. Well,
you never told him to disable interrupts while doing this, and when Tandy
added print spool capabilities (the spooler ran in the bank-switched
memory) the system would lock up if the spooler was running and the clock
tick happened to be at the time that the video memory was switched in.
(It didn't know the spooler memory was switched out, and attempted to
use video memory as executable code.)
Well, while the video memory was switched in, the video driver was
turned off, causing the same problem as you found in the bad socket
connection. The fly-back transformer (I think that's what it was)
would overload and blow out something in the video.
Burning out stepper motors is a no-brainer in comparison.
> Tell Howie I said hi! And by the way, I'm divorced now. Long story.
> I'll be trying to get a more direct route to Howie.
If you want, I'll pass along your e-mail address to him.
[...]
--
+---------+----------------------------------+-----------------------------+
Mike Yetsko
any newsgroup, but I joined the filepro mail list.
As the the video... What happened is the sync had a narrow
'bandwidth'. Get out that, and current went sky high as things
tried to stabilize. The most common failure was the Darlington
transisitor. And that was good! Consider it a 'fuse' for
protection of the expensive flyback!
Mike
Mike Yetsko
an interrupt tick? OUCH!!
Hmm, I think by then I was in other areas, not so heavy into Model
II stuff.
Mike
Cortland Richmond
Didnt' teh VM-5 also have a "flaming monitor" problem?
I still have a mdoel 4, a 4P, TRSDOS Scripsit and a copy of the DOS
versionof Scripsit my Mom was using.
Cortland
Here's to the Good Old Days: Thank GOD they're gone!
Mike Yetsko wrote:
> Yeah, well, that trip to Howies brownstone in NYC was a story in
> itself. His first Model II blew up in days. And the computer store
> snowballed him, so he got on the phone to Penny in Ft Worth.
> I ended up being told to drive out to Long Island to meet Howie
> at the computer center and fix his machine. (There's a 'Tandy
> story' to that too!) I did, but it blew up again a couple days later.
>
> That wasn't Howies fault. His Model II had a really wierd problem.