WANTED: SETSIO File
Mr Emmanuel Roche, France
who has the last version of SETSIO for the PCW?
About 20 years ago, I studied the SETSIO utility provided with one of
the very first Amstrad PCW-8256 sold in France. As usual, I dumped it
on paper, then wrote down the various sections of code: ASCII string,
inline addresses, byte, word, table, etc.
Recently, I needed to do some Z-80 DART programming. So, I refound my
old dump. Problem: I did not transfer my PCW files. They are still on
3" floppies.
So, I asked my correspondents if anyone had a copy of SETSIO, and got
one.
So, I launched my disassembler, entered the break points that I had
written down 20 years ago, then started editing the resulting MAC
file.
Problem: the longer I was commenting the file, the more curious the
program was looking. Normally, my disassembler is quite good at
recreating intelligent code, especially after I do all the work by
hand. Alerted, I left WordStar and examined the COM file and my paper
dump.
Haaa!... I had disassembled SETSIO #93, based on my dump of SETSIO
#117! No wonder, then, that the code looked strange!
So, this raises several questions, since I got one of the first
Amstrad PCW-8256 sold in France.
1) What is the last known version of SETSIO? Obviously, it must be
more than #117.
2) When will someone save it (and the other utilities) on a Web site?
24 years is a lot of time, for microcomputers. I doubt that Amstrad
will prosecute you and put you to jail for saving it.
3) As far as I know, Amstrad and Locomotive are dead. Maybe it is time
to release the source code of their utilities?
4) Is anybody out there, interested in the PCW? In France, the PCW was
almost a flop: only 50,000 were sold. There is no, as far as I know,
any PCW club in France. Hence this message in English.
5) 20 years ago, the standard version of CP/M Plus for the PCW was
Version 1.4. Was there later versions, are they available, and, if
so, what did they correct?
Yours Sincerely,
Mr. Emmanuel Roche, France
Brian Watson
> 2) When will someone save it (and the other utilities) on a Web site?
> 24 years is a lot of time, for microcomputers. I doubt that Amstrad
> will prosecute you and put you to jail for saving it.
>
> 3) As far as I know, Amstrad and Locomotive are dead. Maybe it is time
> to release the source code of their utilities?
Neither Amstrad nor Locomotive software items are out of copyright, but
Amstrad have been very generous with letting their copyrighted material be
used by enthusiasts.
As far as I know, Locomotive Software items are still sold and supported by
SD Micros (Steve Denson). It might be an idea to contact him through the FAQ
to see what he has to say about it.
--
Brian Watson (formerly of Protext Software)
(www.imagebus.co.uk/shop)
Zbigniew Niedzwiedz
<roch...@laposte.net> wrote:
>Recently, I needed to do some Z-80 DART programming. So, I refound my
>old dump. Problem: I did not transfer my PCW files. They are still on
>3" floppies.
Probably you could connect your 3" drive (remember to set jumper to "B")
to PC and use www.cpcmania.com/cpcdiskxp/cpcdiskxp.htm to make images
and try to extract files from them. I did that successfully with two
models of 3" drives (Hitachi SSDD and DSDD) but don't know if any of
them were used in PCW 8256.
I have some 3" diskettes with unknown content (probably used on Amstrad
machines) and two images of them has string "SETSIO #117" inside (some
CPC images has also "SETSIO #93" string inside). If you are interested I
can send them to you.
--
zo...@lu.onet.pl - Zbigniew Nied�wied� - GSM/SMS:+48601263351
http://www.omega.computer.lublin.pl GaduGadu #589387
Suma inteligencji na planecie jest sta�a. Populacja ro�nie.
John Elliott
: old dump. Problem: I did not transfer my PCW files. They are still on
: 3" floppies.
I don't think SETSIO would help you there. It doesn't program the DART; it
makes CD SA INIT and CD SA BAUD calls, and the actual DART-programming
happens in the BIOS.
: 1) What is the last known version of SETSIO? Obviously, it must be
: more than #117.
I have #221. The differences appear to be minor, related to having the
program detect and run under Spectrum +3 CP/M. #093 comes from CPC CP/M.
: 5) 20 years ago, the standard version of CP/M Plus for the PCW was
: Version 1.4. Was there later versions, are they available, and, if
: so, what did they correct?
The latest version I have is 1.15 / 2.15 (the 2.x versions are for the
9512 and 9512+; 1.x versions for others). The main changes include support
for third-party device drivers, which I have documented at
<http://www.seasip.demon.co.uk/Cpm/amsfid.html> and 3.5" floppy drives.
You should be able to get 1.15 or 2.15 boot floppies from SD Microsystems.
--
John Elliott
Thinks: This is what a nice clean life leads to. Hmm, why did I ever lead one?
-- Bluebottle, in the Goon Show
Mr Emmanuel Roche, France
Many thanks for your answer.
> Neither Amstrad nor Locomotive software items are out of copyright, but
> Amstrad have been very generous with letting their copyrighted material be
> used by enthusiasts.
Well... If copyright is 75 years long, like in the USA (and getting
longer each time Disney productions are going to become public
domain), then we can wait 4 human generations before seeing what will
remain of this stuff. Anyway, I am French and "Copyright", as its name
implies, is an English invention. (During the Middle Ages, Catholic
copists copied the Bible during 1,500+ years without any "copyright".
It seems that it was the Protestants, who were selling their
translations into popular languages of the Bible, who invented the
Copyright. It will be interesting to see if the Americans will manage
to impose the Copyright idea to the Chinese, who have a 4,000-years
long civilisation with an almost endless list of inventions, yet, in
4,000 years, no Chinese ever invented the Copyright.)
> As far as I know, Locomotive Software items are still sold and supported by
> SD Micros (Steve Denson). It might be an idea to contact him through the FAQ
> to see what he has to say about it.
Well, over the last 20 years, I regularly wrote to Locomotive
Software, asking to buy the Mallard BASIC-86 for CP/M-86 (single user)
and Mallard BASIC-86 for Concurrent CP/M (multi user). Never got an
answer. It is written black on white on all the Mallard BASIC manuals
I have. Needless to say, when someone warned me that Steve Denson
bought back what remained from Locomotive, I wrote him. No answer, as
usual. If you have those 2 rarities, I hope that you will not wait 75
years before releasing them...
Mr Emmanuel Roche, France
Many thanks for your message.
> Probably you could connect your 3" drive (remember to set jumper to "B")
> to PC and usewww.cpcmania.com/cpcdiskxp/cpcdiskxp.htmto make images
> and try to extract files from them. I did that successfully with two
> models of 3" drives (Hitachi SSDD and DSDD) but don't know if any of
> them were used in PCW 8256.
Ha? Interesting. Never heard about it. Will investigate it more
Sunday.
> I have some 3" diskettes with unknown content (probably used on Amstrad
> machines) and two images of them has string "SETSIO #117" inside (some
> CPC images has also "SETSIO #93" string inside). If you are interested I
> can send them to you.
Many thanks, but someone else send it (#117) to me. It seems that #93
is the CPC version.
Mr Emmanuel Roche, France
Many thanks for your message.
> I don't think SETSIO would help you there. It doesn't program the DART; it
> makes CD SA INIT and CD SA BAUD calls, and the actual DART-programming
> happens in the BIOS.
Yes, this is what I noticed this morning. No OUT mnemonics.
> : 1) What is the last known version of SETSIO? Obviously, it must be
> : more than #117.
>
> I have #221. The differences appear to be minor, related to having the
> program detect and run under Spectrum +3 CP/M. #093 comes from CPC CP/M.
If I have well understood you, this would mean that the last port of
the Amstrad utilities was to the Spectrum? Were the other Amstrad
utilities also updated?
> : 5) 20 years ago, the standard version of CP/M Plus for the PCW was
> : Version 1.4. Was there later versions, are they available, and, if
> : so, what did they correct?
>
> The latest version I have is 1.15 / 2.15 (the 2.x versions are for the
> 9512 and 9512+; 1.x versions for others). The main changes include support
> for third-party device drivers, which I have documented at
> <http://www.seasip.demon.co.uk/Cpm/amsfid.html> and 3.5" floppy drives.
>
> You should be able to get 1.15 or 2.15 boot floppies from SD Microsystems.
Well, SD never answered my previous letters...
Since I am the only one in my region of France with an Amstrad
PCW-8256, at least I am sure never to meet this version.
It is curious that your version is 1.15, since mine, 20 years older,
is 1.4... What is its date?
Let see, the last time I learned to count,
1.4 -- 1985
1.5
1.6
1.7
1.8
1.9
1.10
1.11
1.12
1.13
1.14
1.15
So, there were 12 important enough modifications to CP/M Plus to
warrent a new version number...
Is anybody out there
who know what were those important changes?
James Coupe
>implies, is an English invention.
Why is this even slightly relevant? Copyright laws exist, even in
France. Who invented them hundreds of years ago has nothing to do with
anything.
--
James Coupe
PGP Key: 0x5D623D5D YOU ARE IN ERROR.
EBD690ECD7A1FB457CA2 NO-ONE IS SCREAMING.
13D7E668C3695D623D5D THANK YOU FOR YOUR COOPERATION.
John Elliott
: the Amstrad utilities was to the Spectrum? Were the other Amstrad
: utilities also updated?
All the Locomotive utilities in +3 CP/M are numbered #221.
In PCW CP/M 1.14, the utilities are numbered #218 (except for PAPER.COM,
#115). Oddly, SETSIO #218 and SETSIO #117 are identical except for the
version number.
The only bits of PCW CP/M 1.15 that I have are the EMS files, so I don't
know what SETSIO came with them.
: It is curious that your version is 1.15, since mine, 20 years older,
: is 1.4... What is its date?
Early 1990s? 1993, I think.
: So, there were 12 important enough modifications to CP/M Plus to
: warrent a new version number...
More, because they go back at least as far as 1.2.
: Is anybody out there
: who know what were those important changes?
At this distance in time, I suspect you'd have to disassemble and diff all
the EMS files. Which would be a job and a half.
Mr Emmanuel Roche, France
> Why is this even slightly relevant? Copyright laws exist, even in
> France. Who invented them hundreds of years ago has nothing
> to do with anything.
Copyright laws exist, but they are particular to the history in
general, and religion in particular, of the country were they are
enforced. That's why each country has different copyright laws. They
are artificial inventions: you will never see an animal in Nature
invent or enforce copyright, yet almost everything has been invented
by Nature during the last 15 Billions years. (The oldest laws in
France deal with water for peasants, and are coming straight from the
Romans! Notice, too, that the Romans never invented the copyright...)
Finally, many years ago, I found 2 paragraphs on this subject that
resonated so much with me that I retyped them:
SOFTVAL.WS4
-----------
- "The Value of Software" (post-scriptum of an article)
H.T. Gordon
DDJ, No.19, October 1977, p.42
(Retyped by Emmanuel ROCHE.)
While I wish to make the software in this paper freely available
without restriction to individual users, I retain full copyright
for any commercial use. My motivation is not greed (my fee would
be trivial and usually waived) but curiosity. Some time ago, I
sent a binary-to-decimal conversion program to MOS Technology,
stating that it was to be in the public domain. Recently, I
asked them whether they had used it in their PET. The reply was
that they did not plan to release any information on its
programs. Puzzling, since anyone who buys a PET will be able to
read its ROM. I am an admirer of their 6502 and KIM designs, and
I sympathize with their desire to avoid hassles. Still, this
gave me food for thought. Software is a trickier thing than the
books and magazine articles for which copyright law was written,
and in which changing a few words won't affect value. A
seemingly minor software revision may so enhance value that its
original will (and should) vanish forever. Like it or not, every
programmer is part of a collective mind, and progress demands
that he educate and be educated by others. There is no precise
answer to the question: what is the value of a piece of
software, and who owns that part of the value? In the software-
cost controversy argued in DDJ by Tom Pittman and others, I
strongly favor keeping all costs at the bare minimum and legal
conflicts at zero, since I am primarily a user. Program costs
are not easy to calculate. In my present "package", OPLEGL took
a lot of time, BYTNUM much less because I was only modifying
Larry Fish's logic, NUMBYT was even easier because I knew what
needed to be done, and SIMBUG was child's play. Still, all these
things were simultaneously working themselves out of my mind
and, without the basic (originally more grandiose) concept of a
scanning-debugger, I would never have bothered with OPLEGL. This
concept was inspired by Jim Butterfield's screening-out of the
64 "easy" illegals in his relocator program. I was receptive to
this because of earlier problems caused by an illegal opcode.
So, in a way, SIMBUG is a costly program.
The real value of anything is not in its cost, but in its
utility. It may cost a lot to produce a white elephant, but the
result is still worthless. In our social system, utility tends
to be measured in the marketplace, in dollars. This has worked
well for hardware, and for any software that can be inextricably
linked to hardware. Unattached software can be valued in dollars
only to the extent that users can be compelled either to buy it
or do without. But it is so readily diffusible and copiable that
many users will not buy it, except at a price little above the
cost (in money and time) of copying. Copyrighted printed matter
is now photocopied illegally by individuals with complete
impunity, because enforcement is not practical. Where law fails,
we cannot expect too much of ethics. So, the utility (in the
economic sense) of even the most useful "software" is low. One
could say the same of rainfall or sunlight!
EOF
Axel Berger
>During the Middle Ages, Catholic copists copied the Bible during 1,500+
>years without any "copyright".
Saint Columba fled to Iona and christianized Scotland because he had
copied someone else's psalter. That was around 560 AD.
Mr Emmanuel Roche, France
Many thanks for your message.
> Saint Columba fled to Iona and christianized Scotland because he had
> copied someone else's psalter. That was around 560 AD.
Well, I never heard about him, so had a quick search. You are right:
http://en.wikipedia.org/wiki/Columba
but I notice that a psalter is far from being a complete Bible:
http://en.wikipedia.org/wiki/Psalter
"Psalters were amongst the most popular types of illuminated
manuscripts" so, one wonders if Columba did not stole, rather than
copy, a beautiful psalter? This would explain why several people died.
All that for a book with pictures! (But it was "state-of-the-art" at
the time...)
Brian Watson
>
> Many thanks for your answer.
>
>> Neither Amstrad nor Locomotive software items are out of copyright,
>> but Amstrad have been very generous with letting their copyrighted
>> material be used by enthusiasts.
>
> Well... If copyright is 75 years long, like in the USA (and getting
> longer each time Disney productions are going to become public
> domain), then we can wait 4 human generations before seeing what will
> remain of this stuff.
That's right. Life's like that sometimes.
> Anyway, I am French and "Copyright", as its name
> implies, is an English invention.
Possibly, but YOUR country has signed up to it and there are a VERY large
number of creative Frenchmen who are grateful for the protection it gives to
their work.
> (During the Middle Ages, Catholic
> copists copied the Bible during 1,500+ years without any "copyright".
But they were VERY protective of it being reproduced without changes being
made to the text.
> It seems that it was the Protestants, who were selling their
> translations into popular languages of the Bible, who invented the
> Copyright.
Rubbish.
> It will be interesting to see if the Americans will manage
> to impose the Copyright idea to the Chinese, who have a 4,000-years
> long civilisation with an almost endless list of inventions, yet, in
> 4,000 years, no Chinese ever invented the Copyright.)
China is getting wise to the fact that bootlegging ADVANCED technology is
harder than making cheaper versions of plastic toys etc.
China has things to sell that are VERY original in concept and signing up to
reciprocal arrangements is making more and more sense to them, as a culture.
>> As far as I know, Locomotive Software items are still sold and
>> supported by SD Micros (Steve Denson). It might be an idea to
>> contact him through the FAQ to see what he has to say about it.
>
> Well, over the last 20 years, I regularly wrote to Locomotive
> Software, asking to buy the Mallard BASIC-86 for CP/M-86 (single user)
> and Mallard BASIC-86 for Concurrent CP/M (multi user). Never got an
> answer.
Possibly because Locomotive had, by that time, stopped supporting the
software. That does not place it in the public domain, does it. Just because
someone won't give you, or sell you, something doesn't authorise you to
steal it.
> It is written black on white on all the Mallard BASIC manuals
> I have. Needless to say, when someone warned me that Steve Denson
> bought back what remained from Locomotive, I wrote him. No answer, as
> usual. If you have those 2 rarities, I hope that you will not wait 75
> years before releasing them...
I respect copyright because I have several things worth protecting by it.
In fact, as well as the ex-Arnor Protext programs, I write plays for a
living and have already taken action against a couple of groups who staged
my work without permission or paying the neccesary fees. In both cases I
obtained far greater recompense than had they done what they should have
done to start with.
I bought the copyrights to the 8bit Protext programs and will sue the ass
off anyone who sells them for profit, whether they are a company or some
jerk operating out of his bedroom.
Doubt me? Try it. I have VERY good legal insurance.
I HAVE said they can be freely distributed and I am going to make the code
available for hobbyists to play with.
That is at MY discretion, not yours (or theirs).
--
Brian
"Fight like the Devil, die like a gentleman."
www.imagebus.co.uk/shop
Brian Watson
>
>> Why is this even slightly relevant? Copyright laws exist, even in
>> France. Who invented them hundreds of years ago has nothing
>> to do with anything.
>
> Copyright laws exist, but they are particular to the history in
> general, and religion in particular, of the country were they are
> enforced.
You are doing a great job of alienating people who can help you.
Brian Watson
There is another poster in another Group that I take who argues like you.
He looks up every statement made that contradicts his rather foolish point
of view and ends up with nothing more than the equivalent of one word on his
side against a full page of verified and validated statements against him.
I strongly suggest you stop arguing like an idiot.
John Elliott
: At this distance in time, I suspect you'd have to disassemble and diff all
: the EMS files. Which would be a job and a half.
OK, I'm a glutton for punishment. Here are the differences I've spotted so
far between all the PCW EMS files I have:
Between 1.2 and 1.4:
* Change to the dot-matrix printer font: The " and ' characters are
chisel-like, not curly.
* The serial port driver can now be driven in interrupt mode.
* The floppy driver uses a semaphore to signal completion, not a
simple on/off flag.
* There is a kernel panic (RST 30h) which is invoked if the kernel
runs out of memory or a process descriptor gets corrupted.
Between 1.4 and 2.1:
* Support added for the PCW9512 and its daisywheel printer.
* Memory layout changed, so that external device drivers can be
loaded (but it doesn't actually load them yet).
* New escape code in the terminal emulator: ESC @ (insert character)
Between 2.1 and 2.9:
* Now supports the PCW9512 sheet feeder.
* Will now load external device drivers (*.FID)
* The "Drive is A:" message moves four characters to the left (!)
* New ESC 4 1 / ESC 4 0 sequences to disable / enable screen
scrolling.
Between 2.9 and 2.11:
* Support added for second-generation PCWs: 9256, 9512+.
* Support added for booting the LocoScript installer (insert the boot
floppy and press SHIFT+EXTRA+RELAY. Only applies to
second-generation PCWs).
* Allows the floppy drives to be stepped at different rates.
Between 1.4 and 1.12:
* Minor version numbers synchronised between 1.x (dot-matrix PCWs)
and 2.x (daisywheel PCWs). This probably means some 2.x versions
were skipped.
* The dot-matrix printer driver allows redefinition of arbitrary
characters, using jumps at FFF9h and FFFCh
Between 2.11 and 1.12/2.12:
* Support for the standalone Centronics port accessory.
Between 1.12/2.12 and 1.14/2.14:
* Support for loading drive parameters from *.FIB files.
* Improved Centronics port detection.
* If the memory test reports 256k, reset the memory read/write
register and try again.
* SHIFT+EXTRA+RELAY now works on first-generation PCWs as well.
Between 1.14/2.14 and 1.15/2.15:
* A single change: When doing a SHIFT+EXTRA+RELAY reboot, takes all
devices out of interrupt mode so that they don't interrupt at an
awkward moment.
If anyone's got PCW EMS files from versions I haven't, ie:
1.0 2.0
1.1 2.2-2.8
1.3 2.10
1.5-1.11 2.13
1.13 2.14
I'll happily look at them and find out what's different about them.
(One amusing fact learned from all this disassembly: PCW CP/M, and
LocoScript, and LocoLink, all contain a kernel with extremely similar
functions to those provided by the CPC firmware. It's got KL ADD FRAME FLY
and KL DEL FRAME FLY, for example, and two speeds of ticker. It's even got
KL TIME SET and KL TIME PLEASE, despite the fact that nothing in CP/M calls
them.)
Jacob Nevins
> OK, I'm a glutton for punishment. Here are the differences I've spotted so
>far between all the PCW EMS files I have:
Interesting, thanks!
>Between 1.4 and 2.1:
[...]
> * New escape code in the terminal emulator: ESC @ (insert character)
Well, I never knew that. I'd have loved that as a counterpart to ESC N
(delete character) back in my BASIC days.
John Elliott
: Well, I never knew that. I'd have loved that as a counterpart to ESC N
: (delete character) back in my BASIC days.
I think it came in as a side-effect of right-to-left text support in the
underlying screen driver.
Thanks to a couple of helpful contributors, I've been able to take a look
at three more versions: 1.1, 1.7H and 1.11. From these, I reckon:
* Between 1.1 and 1.2: Support for localisation. The keystrokes for Retry,
Ignore, and Cancel are localised, as are the default screen and printer
language.
* Between 1.4 and 1.7H: Internal memory layout changed to support *.FID
drivers. ESC @ escape code added.
* 1.11 is in step with 2.11, as I'd have expected.
I also have some guesses:
* That version numbers were synchronised at 2.9. Version 1.7H looks slightly
older, to my way of thinking, than 2.1, which would give the following
equivalences:
1.7 :: 2.0
1.8 :: 2.1
1.9 :: 2.9
* That there were separate 'H' and non-'H' versions. Version 1.7H looks
older than 2.1, but it supports loading drivers, which 2.1 doesn't. That
suggests to me that there was a 1.7 (no H) that couldn't load drivers,
and perhaps a 2.1H that could.
I've put a page at <http://www.seasip.demon.co.uk/Cpm/xbiosint.html> with
all my findings and guesses so far on this subject.