Groups keyboard shortcuts have been updated
Dismiss
See shortcuts
MDS-II Keyboard Matrix Encoder
59 views
Skip to first unread message
Paul Birkel
Apr 7, 2021, 10:57:55 AM4/7/21
to intel-...@googlegroups.com
Roger has gracefully given me the opportunity to restore his damaged MDS-230
:-}.
Known issues for which I don't yet have a good resolution are:
- missing front panel
- missing zero switch-cover on the IPC
- missing keyboard
If anyone can help me with replacements for either the missing front panel
or zero switch-cover, that would be great :->. However ...
It's the keyboard that is my main concern assuming that the CRT subsystem
can be made operational.
Since I expect that the desired keyboard is probably mounted as a trophy on
someone's wall (and any one that might show up on eBay in the future will be
priced accordingly ...), I'm considering how to best go about creating a
functional (not cosmetic) replacement for use with the IOC. There are two
basic approaches that I think that I could take:
1) Use a MCU to translate an PS/2-type keyboard to the correct parallel
protocol (e.g., https://www.go4retro.com/products/ps2-encoder/ or
https://www.n4vlf.net/ps2.html). I believe that I have a couple of those
keyboards saved from my kids machines back in the era of supporting both
PS/2 and USB protocols.
2) More-or-less replicate the Intel encoding circuitry (8041/8071-based)
around a suitable raw keyboard / key-array recognizing that I wouldn't be
trying to replicate the exact upper/lower key labels/positions. "FUNC" and
"TPWR" would be some suitably repurposed keys/caps (which would also be the
case for a re-encoded PS/2 keyboard).
AFAICS the MDS keyboard is composed entirely of momentary switches, correct?
In either case it would be valuable to study the existing keyboard encoder
ROM content to ensure getting repeat, n-key rollover, and any specialized
behaviors correct. Otherwise I think that I can start from the tables in
"122029-001_Intellec_Series_II_CRT_and_Keyboard_Interface_Sep82.pdf" and
then adjust from there based on observed (mis)behavior as I bring up system
software (including, for example, CREDIT).
Martyn's video regarding the iPDS-100 keyboard
(https://www.youtube.com/watch?v=8CjR_TSPaXE) was helpful in seeing the FUNC
key use.
I've been discussing this topic a bit with Jack Mcmullen while studying his
Mini-MDS design. He suggested that the keyboard 8041/8071 code has already
been captured by someone -- possibly both the original as well as the update
specified in that Sep82 document where the "RPT" key was replaced by the
"FUNC" key, and repeat became the responsibility of the encoder algorithms.
He's not sure where that ROM code might be found and thought that either
Herb or Mark would be able to point me to it. A disassembly would be even
better, if one has been developed.
At the moment I'm leaning towards the MCU-based solution to adapt an
existing PS/2 keyboard to the requirements of the IOC & MDS software.
Code pointers? Comments?
Thank you,
paul
:-}.
Known issues for which I don't yet have a good resolution are:
- missing front panel
- missing zero switch-cover on the IPC
- missing keyboard
If anyone can help me with replacements for either the missing front panel
or zero switch-cover, that would be great :->. However ...
It's the keyboard that is my main concern assuming that the CRT subsystem
can be made operational.
Since I expect that the desired keyboard is probably mounted as a trophy on
someone's wall (and any one that might show up on eBay in the future will be
priced accordingly ...), I'm considering how to best go about creating a
functional (not cosmetic) replacement for use with the IOC. There are two
basic approaches that I think that I could take:
1) Use a MCU to translate an PS/2-type keyboard to the correct parallel
protocol (e.g., https://www.go4retro.com/products/ps2-encoder/ or
https://www.n4vlf.net/ps2.html). I believe that I have a couple of those
keyboards saved from my kids machines back in the era of supporting both
PS/2 and USB protocols.
2) More-or-less replicate the Intel encoding circuitry (8041/8071-based)
around a suitable raw keyboard / key-array recognizing that I wouldn't be
trying to replicate the exact upper/lower key labels/positions. "FUNC" and
"TPWR" would be some suitably repurposed keys/caps (which would also be the
case for a re-encoded PS/2 keyboard).
AFAICS the MDS keyboard is composed entirely of momentary switches, correct?
In either case it would be valuable to study the existing keyboard encoder
ROM content to ensure getting repeat, n-key rollover, and any specialized
behaviors correct. Otherwise I think that I can start from the tables in
"122029-001_Intellec_Series_II_CRT_and_Keyboard_Interface_Sep82.pdf" and
then adjust from there based on observed (mis)behavior as I bring up system
software (including, for example, CREDIT).
Martyn's video regarding the iPDS-100 keyboard
(https://www.youtube.com/watch?v=8CjR_TSPaXE) was helpful in seeing the FUNC
key use.
I've been discussing this topic a bit with Jack Mcmullen while studying his
Mini-MDS design. He suggested that the keyboard 8041/8071 code has already
been captured by someone -- possibly both the original as well as the update
specified in that Sep82 document where the "RPT" key was replaced by the
"FUNC" key, and repeat became the responsibility of the encoder algorithms.
He's not sure where that ROM code might be found and thought that either
Herb or Mark would be able to point me to it. A disassembly would be even
better, if one has been developed.
At the moment I'm leaning towards the MCU-based solution to adapt an
existing PS/2 keyboard to the requirements of the IOC & MDS software.
Code pointers? Comments?
Thank you,
paul
Jon Hales
Apr 7, 2021, 11:38:45 AM4/7/21
to intel-...@googlegroups.com
Paul
I'm interested to know about your mission to solve the missing keyboard issue - I'm in the same situation.
I'm fairly sure I saw an MDS keyboard on eBay.com a year or two ago, but I don't recall the price being asked.
When an MDS is booted in the absence of the keyboard, the serial interfaces on the IOC can be accessed via a serial terminal or terminal emulator software (Teraterm, Putty, etc). I was advised about this several years ago on the previous discussion system managed by Eric. I saved the messages in the attached note, which includes code to increase the baud rate.
Regarding the other items, I assume you are looking for the blue version of the front panel - is that correct?
MDS 22x/23x power switch
To add to your list of 'missing MDS parts', I'm looking for replacements for a couple of the main power switch units. I think these were originally made by Oslo. The part number I photographed was DLA3x27-2C9.
Best regards
Jon
--
You received this message because you are subscribed to the Google Groups "intel-devsys" group.
To unsubscribe from this group and stop receiving emails from it, send an email to intel-devsys...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/intel-devsys/02ab01d72bbe%2475534110%245ff9c330%24%40gmail.com.
Mark Fisher
Apr 7, 2021, 11:39:53 AM4/7/21
to intel-...@googlegroups.com
Hi Paul,
A couple of quick notes. On a Series II and III you can use the first serial port if no keyboard is connected. I think the speed is 110 Baud and you just need to hit the spacebar to get it started.
I can think of one key that is latched for sure. (TPWR) If I did not have a keyboard, I would probably go with the MCU option so you can use a IBM PC based keyboard (PS2, USB) and translate it to the appropriate output the MDS needs. I would try to emulate the updated MDS keyboard as it is more functional.
As for a front panel, you should at least cover the area if you have a lot of boards in the system to keep the heat management working the way it was intended. Maybe someone could 3D print one. I may be able to laser scan mine but really don't know what format is needed or desired to make it work.
Best Regards, Mark
Herb Johnson
Apr 7, 2021, 1:01:39 PM4/7/21
to intel-...@googlegroups.com, Paul Birkel
Well, Paul asked for comments. I thought about it, wrote too much, put
it aside. Here's my thoughts.
1) If the goal of Roger and Paul, is to restore to use an MDS-230, than
this business about a replacement keyboard is secondary. The point is to
use the MDS, not be in the MDS production business, or the keyboard
business. Right?
2) Given 1), then it become obvious that you need a *functional*
keyboard, to operate the thing, right? And that would be, an Arduino (or
your favorite microcontroller) that can a) talk to the MDS keyboard
interface and b) respond to some PS/2 keyboard and jam MDS-compatible
codes into the MDS keyboard interface.
If you want something better, you'll need to know the MDS interface and
codes *anyway*, so you'll have design criteria. So - just do that, make
the *functional* arduino/ps2 keyboard as you describe. It's easier to
just do-that, and then discard it later; than to not-do-that, design
something else that doesn't work, figure out why, then redesign it to
make it work.
Just my gut opinion.
3) Keyboards are a subculture in vintage computing. Paul, if that's
unfamiliar territory to you, pardon me but do some homework on the
matter. Why? Because you are now in the keyboard business. You'll be
looking at keyswitches, keycaps, frames for holding the keyswitches,
keyswitch scanning methods in hardware and software. You aren't alone,
and it's old work too.
There's not only a present-day keyboard design and construction
subculture, there's a *vintage* subculture. I know this, from selling
1980's Apple Macintosh and Apple II parts, every day. People buy old
keyboards for use on modern equipment; or harvest parts from them for
new keyboards. Every day.
4) If you decide to design a keyboard as "here's how Intel did it on the
MDS, let's replicate what they did", you still have problems. It will be
hard to get 1979 keyswitches, because they are already in demand AND
scarce. Keycaps? worse. Because as I said: it's a subculture, and it's
custom. So, not cheap.
5) But: there's a subculture in modern custom mechanical switch
keyboards too. Look *that* up if it's not familiar. There are kits one
can buy, PC boards and controllers and parts. One can populate them with
ancient keycaps and keyswitches, and make your own USB keyboard.
There's likely make-your-own ASCII parallel keyboards too, for vintage
computing use.
So why use someone else's make-your-own keyboard? a) it's a done deal,
you get the PC board and parts and codes and other mechanicals. You just
have to modify the codes and add an MDS interface. Now you can reuse
that Arduino design work, see?
b) Maybe, you'll save money? and time? if you just use their keyswitches
and keycaps, but roll your own PC board, modify their circuit design?
You'll find out the cost of custom keycaps.
but more likely c) - you can at least *learn* from their design and
parts choices. You can buy where they buy their keycaps and keyswitches.
You can be informed by their design choices.
6) So - given these points, that keyboards are real things with real
design and parts and construction issues. Given the kinds of interests,
just in *keyboards*. There's also interests, in Intel MDS systems. It's
likely other people want "an MDS compatible keyboard", but it's not that
easy to make one "cold".
So: are you making a one-off keyboard? or a project to make as many as
someone else might want? (shrug) I see techies who are content to make
one thing for themselves, and not because someone else is interested. I
see others who want to make many, or at least allow others to make many.
Their interest in other people's use vary. I've seen that here, in the
various iterations of the Multibus discussion lists, for other "widgets"
designed for use.
------------------------------------
So, those are my comments. Primarily, the matter of making a keyboard is
not a side issue to many techies and of course engineering
professionals. It's a real matter and a surprisingly active field of
interest, either vintage or modern.
It seems reasonable to make a PS/2-Arduino type keyboard for use with
the MDS - because you need *something* right away, and you need to
figure out how to make *that* work, to design something else.
But making the "something else" is a real-enough project, if you want it
to be a real project. Keyboard parts are not cheap, but cheaper if you
borrow someone else's keyboard or someone else's keyboard kit. Keyboard
designs are not trivial; but others have already designed them for you,
borrow from their works. If you want to entirely replicate an Intel MDS
keyboard, well, that's serious squared, good luck with that.
These are matters for you and Roger to determine. I suggest you take
advantage of prior and recent work; and the availability of old or new
parts-kits of keyswitches and keycaps. Good luck.
best regards, Herb
--
Herbert R. Johnson, New Jersey in the USA
http://www.retrotechnology.com OR .net
preserve, recover, restore 1970's computing
email: hjohnson AT retrotechnology DOT com
or try later herbjohnson AT comcast DOT net
it aside. Here's my thoughts.
1) If the goal of Roger and Paul, is to restore to use an MDS-230, than
this business about a replacement keyboard is secondary. The point is to
use the MDS, not be in the MDS production business, or the keyboard
business. Right?
2) Given 1), then it become obvious that you need a *functional*
keyboard, to operate the thing, right? And that would be, an Arduino (or
your favorite microcontroller) that can a) talk to the MDS keyboard
interface and b) respond to some PS/2 keyboard and jam MDS-compatible
codes into the MDS keyboard interface.
If you want something better, you'll need to know the MDS interface and
codes *anyway*, so you'll have design criteria. So - just do that, make
the *functional* arduino/ps2 keyboard as you describe. It's easier to
just do-that, and then discard it later; than to not-do-that, design
something else that doesn't work, figure out why, then redesign it to
make it work.
Just my gut opinion.
3) Keyboards are a subculture in vintage computing. Paul, if that's
unfamiliar territory to you, pardon me but do some homework on the
matter. Why? Because you are now in the keyboard business. You'll be
looking at keyswitches, keycaps, frames for holding the keyswitches,
keyswitch scanning methods in hardware and software. You aren't alone,
and it's old work too.
There's not only a present-day keyboard design and construction
subculture, there's a *vintage* subculture. I know this, from selling
1980's Apple Macintosh and Apple II parts, every day. People buy old
keyboards for use on modern equipment; or harvest parts from them for
new keyboards. Every day.
4) If you decide to design a keyboard as "here's how Intel did it on the
MDS, let's replicate what they did", you still have problems. It will be
hard to get 1979 keyswitches, because they are already in demand AND
scarce. Keycaps? worse. Because as I said: it's a subculture, and it's
custom. So, not cheap.
5) But: there's a subculture in modern custom mechanical switch
keyboards too. Look *that* up if it's not familiar. There are kits one
can buy, PC boards and controllers and parts. One can populate them with
ancient keycaps and keyswitches, and make your own USB keyboard.
There's likely make-your-own ASCII parallel keyboards too, for vintage
computing use.
So why use someone else's make-your-own keyboard? a) it's a done deal,
you get the PC board and parts and codes and other mechanicals. You just
have to modify the codes and add an MDS interface. Now you can reuse
that Arduino design work, see?
b) Maybe, you'll save money? and time? if you just use their keyswitches
and keycaps, but roll your own PC board, modify their circuit design?
You'll find out the cost of custom keycaps.
but more likely c) - you can at least *learn* from their design and
parts choices. You can buy where they buy their keycaps and keyswitches.
You can be informed by their design choices.
6) So - given these points, that keyboards are real things with real
design and parts and construction issues. Given the kinds of interests,
just in *keyboards*. There's also interests, in Intel MDS systems. It's
likely other people want "an MDS compatible keyboard", but it's not that
easy to make one "cold".
So: are you making a one-off keyboard? or a project to make as many as
someone else might want? (shrug) I see techies who are content to make
one thing for themselves, and not because someone else is interested. I
see others who want to make many, or at least allow others to make many.
Their interest in other people's use vary. I've seen that here, in the
various iterations of the Multibus discussion lists, for other "widgets"
designed for use.
------------------------------------
So, those are my comments. Primarily, the matter of making a keyboard is
not a side issue to many techies and of course engineering
professionals. It's a real matter and a surprisingly active field of
interest, either vintage or modern.
It seems reasonable to make a PS/2-Arduino type keyboard for use with
the MDS - because you need *something* right away, and you need to
figure out how to make *that* work, to design something else.
But making the "something else" is a real-enough project, if you want it
to be a real project. Keyboard parts are not cheap, but cheaper if you
borrow someone else's keyboard or someone else's keyboard kit. Keyboard
designs are not trivial; but others have already designed them for you,
borrow from their works. If you want to entirely replicate an Intel MDS
keyboard, well, that's serious squared, good luck with that.
These are matters for you and Roger to determine. I suggest you take
advantage of prior and recent work; and the availability of old or new
parts-kits of keyswitches and keycaps. Good luck.
best regards, Herb
Herbert R. Johnson, New Jersey in the USA
http://www.retrotechnology.com OR .net
preserve, recover, restore 1970's computing
email: hjohnson AT retrotechnology DOT com
or try later herbjohnson AT comcast DOT net
Herb Johnson
Apr 7, 2021, 2:17:14 PM4/7/21
to intel-...@googlegroups.com
For discussion and to locate, it would be useful if Paul and others
could photograph or sketch or extract diagrams, of the specific front
panel and IPC cover; and ideally the cables and connectors. If they are
to be replicated, then those would be the model. And who knows? One of
us (or a third party) may find some blue panel at a hamfest and be
surprised.
Seems to me that "laser scans" and "3D prints" are overkill. What I
recall of the parts is that they are metal panels which one could make
from aluminum stock from a hardware store (and application of tools).
This is 1980 technology after all. Good photos and measurements with a
ruler are readily done, such as by Mark. I'd take the measurements I can
get, to ones I cannot. Paul can cut a template from some material to
confirm fit.
The same applies to replicating the keyboard. Whatever the level of
replication, it's plausible to put it in a case of like color and like
keyboard arrangement. A photo and ruler measurements would be informative.
YEs, the Intel docs likely have specifications. But nothing is more
informative than "the real thing" and pointing to it. "oh, THAT keyboard
and not that OTHER keyboard". And third-parties who don't know Multibus
won't root through manuals.
I actually HAVE a keyboard of similar size in a similar case of similar
color. A lot of 1970 tech was in blue. Paul, if that keyboard is
desirable, I could pursue it with you privately but my apologies it
won't be cheap in cost or in shipping. A photo will inform you if you ask.
regards, Herb
On 4/7/2021 11:39 AM, Mark Fisher wrote:
> Hi Paul,
could photograph or sketch or extract diagrams, of the specific front
panel and IPC cover; and ideally the cables and connectors. If they are
to be replicated, then those would be the model. And who knows? One of
us (or a third party) may find some blue panel at a hamfest and be
surprised.
Seems to me that "laser scans" and "3D prints" are overkill. What I
recall of the parts is that they are metal panels which one could make
from aluminum stock from a hardware store (and application of tools).
This is 1980 technology after all. Good photos and measurements with a
ruler are readily done, such as by Mark. I'd take the measurements I can
get, to ones I cannot. Paul can cut a template from some material to
confirm fit.
The same applies to replicating the keyboard. Whatever the level of
replication, it's plausible to put it in a case of like color and like
keyboard arrangement. A photo and ruler measurements would be informative.
YEs, the Intel docs likely have specifications. But nothing is more
informative than "the real thing" and pointing to it. "oh, THAT keyboard
and not that OTHER keyboard". And third-parties who don't know Multibus
won't root through manuals.
I actually HAVE a keyboard of similar size in a similar case of similar
color. A lot of 1970 tech was in blue. Paul, if that keyboard is
desirable, I could pursue it with you privately but my apologies it
won't be cheap in cost or in shipping. A photo will inform you if you ask.
regards, Herb
On 4/7/2021 11:39 AM, Mark Fisher wrote:
> Hi Paul,
>
> As for a front panel, you should at least cover the area if you have a
> lot of boards in the system to keep the heat management working the way
> it was intended. Maybe someone could 3D print one. I may be able to
> laser scan mine but really don't know what format is needed or desired
> to make it work.
>
> Best Regards, Mark
>
> On Wed, Apr 7, 2021 at 9:57 AM Paul Birkel
>
> As for a front panel, you should at least cover the area if you have a
> lot of boards in the system to keep the heat management working the way
> it was intended. Maybe someone could 3D print one. I may be able to
> laser scan mine but really don't know what format is needed or desired
> to make it work.
>
> Best Regards, Mark
>
> On Wed, Apr 7, 2021 at 9:57 AM Paul Birkel
>
> Roger has gracefully given me the opportunity to restore his damaged
> MDS-230
> :-}.
>
> Known issues for which I don't yet have a good resolution are:
>
> - missing front panel
> - missing zero switch-cover on the IPC
> - missing keyboard
>
> If anyone can help me with replacements for either the missing front
> panel
> or zero switch-cover, that would be great :->.
> MDS-230
> :-}.
>
> Known issues for which I don't yet have a good resolution are:
>
> - missing front panel
> - missing zero switch-cover on the IPC
> - missing keyboard
>
> If anyone can help me with replacements for either the missing front
> panel
> or zero switch-cover, that would be great :->.
Paul Birkel
Apr 7, 2021, 3:28:17 PM4/7/21
to intel-...@googlegroups.com
Marc and John: Thank you very much for your observations.
Yes, I’ve read references to “just press the space key” and then use the serial interface, although none were as clear as your prior-thread attachment Jon. I earlier looked at “9800605-02B_MDS_Series_II_Boot_Monitor_Listing_May81.pdf” to see for myself what it all means. It’s pretty nicely documented source code and gives one some insight into the boot process that checks for the presence and activity of the various well-known ports and other system components that may be either uninstalled or not operating properly. I’ll be studying it further once I have something with which to play. So what I failed to clearly point out was the aspect of “IFF (if-and-only-if) CRT-working THEN nice-to-use-keyboard-input” precondition to bringing up baby :->. Definitely going a step at a time here, and getting serial-IO for the console running will be a great (and early) step forward.
OK, so TPWR is a latched on/off switch. Good to know.
Re: heat management – I hadn’t thought about that. Excellent warning/point. So I’ll need to do *something* about that missing front panel. For the moment there’s only going to be the IPC in there so not immediately an issue I expect.
Re: color scheme – yes the MDS-230 is in the blue livery, however I may be in Bondo territory for case repair in which case color-match for the missing front panel becomes just another mismatch. Any color will please me immensely :->. I have a sandy/cream MDS 201 expansion chassis (with front :->) that I’m considering putting under the MDS-230 so there’s not going to be a consistent color scheme anyway unless I resort to some repainting. At this point appearance is the very *least* of my concerns. (Not speaking for my spousal-unit in that regard, ahem.)
Re: mission – I’ll respond to Herb separately. However I’d simply like to use the CRT as Intel intended – assuming that It’s operational (or can be made that way). Still I’m at the other end of the shipping lane and who knows, the CRT neck might get broken or something else irrevocably damaged. So we’ll see. I know that Roger will do his dam’dest to stymie the Shipping Gorillas :->. So the keyboard mission is for functional-equivalence. Not beauty, nor art, nor stylishness – albeit all of those would be nice … but very non-essential. I’m not interested in getting into the “keeb” scene! Keyboards are to be heard (or not, as pleases your fancy; no clicky fetish here), not seen (at least in the modern’ish psychedelic sense). And, alas, I’m not a real touch typist so “feel” just isn’t much of a consideration.
The mission is not even to achieve the same key layout. Access to the ROM-code for the upgraded Intel MDS keyboard would make it easier to either implement a comparable encoder for some arbitrary existing key-layout (I have one possibility in hand, but on closer inspection this afternoon it isn’t one that I’m willing to break-and-rewire), or revise an existing PS/2-to-parallel solution to behave properly. At this point I’m heavily leaning towards the recoding of an existing MCU-based solution (https://www.go4retro.com/products/ps2-encoder/) but I’m completely open to criticism and concerns. I’m guessing that Jon (maybe others; who knows) might benefit from that approach since it seems to be pretty reusable. Heck, repaint your keyboard base to the matching MDS case-color …
I expect that my saved keyboards contain at least one that does PS/2. Right now the stash is a bit inaccessible but next week or so I’ll shift stuff around to see-what-I-can-see. Keyboards with PS/2 connectors on them look to be reasonably plentiful at low price-points on eBay. Some even “indecently” so :->. But no one is *ever* going to mistake one for a Genuine Intel MDS “keeb”.
To view this discussion on the web visit https://groups.google.com/d/msgid/intel-devsys/CAJ5BDDD%3DhNYB8feeeS7J0i3WgBsTbp6bGNge-LW5SNzFo_0dDw%40mail.gmail.com.
Roger Arrick
Apr 7, 2021, 3:32:28 PM4/7/21
to intel-...@googlegroups.com
I'm pretty sure my MDS II uses the serial port #2 for external console, not #1.
Baud can be jumpered but mine works at 2400-8N1
Roger
> <mailto:intel-devsys%2Bunsu...@googlegroups.com>.
> <https://groups.google.com/d/msgid/intel-devsys/CAJ5BDDD%3DhNYB8feeeS7J0i3WgBsTbp6bGNge-LW5SNzFo_0dDw%40mail.gmail.com?utm_medium=email&utm_source=footer>.
Baud can be jumpered but mine works at 2400-8N1
Roger
> To view this discussion on the web visit
> https://groups.google.com/d/msgid/intel-devsys/02ab01d72bbe%2475534110%245ff9c330%24%40gmail.com.
>
> https://groups.google.com/d/msgid/intel-devsys/02ab01d72bbe%2475534110%245ff9c330%24%40gmail.com.
>
> --
> You received this message because you are subscribed to the Google
> Groups "intel-devsys" group.
> To unsubscribe from this group and stop receiving emails from it, send
> an email to intel-devsys...@googlegroups.com
> <mailto:intel-devsys...@googlegroups.com>.
> You received this message because you are subscribed to the Google
> Groups "intel-devsys" group.
> To unsubscribe from this group and stop receiving emails from it, send
> an email to intel-devsys...@googlegroups.com
> To view this discussion on the web visit
> https://groups.google.com/d/msgid/intel-devsys/CAJ5BDDD%3DhNYB8feeeS7J0i3WgBsTbp6bGNge-LW5SNzFo_0dDw%40mail.gmail.com
> <https://groups.google.com/d/msgid/intel-devsys/CAJ5BDDD%3DhNYB8feeeS7J0i3WgBsTbp6bGNge-LW5SNzFo_0dDw%40mail.gmail.com?utm_medium=email&utm_source=footer>.
Paul Birkel
Apr 7, 2021, 3:51:03 PM4/7/21
to Herb Johnson, intel-...@googlegroups.com
[PAB] Thank you for the careful thinking and layout Herb. For clarity in my response I've marked my comments like this one. I hope that makes sense.
-----Original Message-----
From: Herb Johnson [mailto:hjoh...@retrotechnology.com]
Sent: Wednesday, April 07, 2021 1:01 PM
To: intel-...@googlegroups.com; Paul Birkel
Subject: Re: intel-devsys MDS-II Keyboard Matrix Encoder
2) Given 1), then it become obvious that you need a *functional*
keyboard, to operate the thing, right? And that would be, an Arduino (or
your favorite microcontroller) that can a) talk to the MDS keyboard
interface and b) respond to some PS/2 keyboard and jam MDS-compatible
codes into the MDS keyboard interface.
[PAB] "would" --> "could", otherwise ... exactly.
If you want something better, you'll need to know the MDS interface and
codes *anyway*, so you'll have design criteria. So - just do that, make
the *functional* arduino/ps2 keyboard as you describe. It's easier to
just do-that, and then discard it later; than to not-do-that, design
something else that doesn't work, figure out why, then redesign it to
make it work.
[PAB] If I understand your outlined process correctly, yes indeed.
Just my gut opinion.
3) Keyboards are a subculture in vintage computing. Paul, if that's
unfamiliar territory to you, pardon me but do some homework on the
matter. Why? Because you are now in the keyboard business. You'll be
looking at keyswitches, keycaps, frames for holding the keyswitches,
keyswitch scanning methods in hardware and software. You aren't alone,
and it's old work too.
[PAB] I'm not a "keeb", clicky, or cool-look freak (pardon the use of that term). No fetish. However Cherry-workalikes are cheap (China), a set of PC-layout keycaps are also cheap (China), and one could certainly replicate the Intel-specified circuitry -- but frankly I wouldn't think of going down this path unless the original 8041 ROM code was available. Hence my query. Building up something from scratch as regards the encoder and then the MDS-interface just isn't interesting to me.
There's not only a present-day keyboard design and construction
subculture, there's a *vintage* subculture. I know this, from selling
1980's Apple Macintosh and Apple II parts, every day. People buy old
keyboards for use on modern equipment; or harvest parts from them for
new keyboards. Every day.
[PAB] To each their own, granted. I stand in awe of them, and their fetish :->. I already have my own.
4) If you decide to design a keyboard as "here's how Intel did it on the
MDS, let's replicate what they did", you still have problems. It will be
hard to get 1979 keyswitches, because they are already in demand AND
scarce. Keycaps? worse. Because as I said: it's a subculture, and it's
custom. So, not cheap.
[PAB] See cheap Cherry-workalikes (above, and on eBay). I'd only build one, for me. The keyswitch matrix is clear and trivial in the schematic. The secret sauce is all in the 8041 ROM.
5) But: there's a subculture in modern custom mechanical switch
keyboards too. Look *that* up if it's not familiar. There are kits one
can buy, PC boards and controllers and parts. One can populate them with
ancient keycaps and keyswitches, and make your own USB keyboard.
There's likely make-your-own ASCII parallel keyboards too, for vintage
computing use.
[PAB] Indeed. I mentioned awe above. I judge not; just not the road that I intend to travel :->.
So why use someone else's make-your-own keyboard? a) it's a done deal,
you get the PC board and parts and codes and other mechanicals. You just
have to modify the codes and add an MDS interface. Now you can reuse
that Arduino design work, see?
b) Maybe, you'll save money? and time? if you just use their keyswitches
and keycaps, but roll your own PC board, modify their circuit design?
You'll find out the cost of custom keycaps.
[PAB] I've seen quotes for double-shot injection-molded keycaps. Serious bucks and a production run required (of course).
but more likely c) - you can at least *learn* from their design and
parts choices. You can buy where they buy their keycaps and keyswitches.
You can be informed by their design choices.
6) So - given these points, that keyboards are real things with real
design and parts and construction issues. Given the kinds of interests,
just in *keyboards*. There's also interests, in Intel MDS systems. It's
likely other people want "an MDS compatible keyboard", but it's not that
easy to make one "cold".
[PAB] Do I hear any volunteers :-}. Crickets ... we're not talking about a real market existing for these babies.
So: are you making a one-off keyboard? or a project to make as many as
someone else might want? (shrug) I see techies who are content to make
one thing for themselves, and not because someone else is interested. I
see others who want to make many, or at least allow others to make many.
Their interest in other people's use vary. I've seen that here, in the
various iterations of the Multibus discussion lists, for other "widgets"
designed for use.
------------------------------------
So, those are my comments. Primarily, the matter of making a keyboard is
not a side issue to many techies and of course engineering
professionals. It's a real matter and a surprisingly active field of
interest, either vintage or modern.
It seems reasonable to make a PS/2-Arduino type keyboard for use with
the MDS - because you need *something* right away, and you need to
figure out how to make *that* work, to design something else.
[PAB] Actually it's not "right away", it's "functional" that's the driver. For me.
But making the "something else" is a real-enough project, if you want it
to be a real project. Keyboard parts are not cheap, but cheaper if you
borrow someone else's keyboard or someone else's keyboard kit. Keyboard
designs are not trivial; but others have already designed them for you,
borrow from their works. If you want to entirely replicate an Intel MDS
keyboard, well, that's serious squared, good luck with that.
These are matters for you and Roger to determine. I suggest you take
advantage of prior and recent work; and the availability of old or new
parts-kits of keyswitches and keycaps. Good luck.
[PAB] Thank you. I'm keeping it as simple as possible, and no simpler. I'm a frugal guy.
-----Original Message-----
From: Herb Johnson [mailto:hjoh...@retrotechnology.com]
Sent: Wednesday, April 07, 2021 1:01 PM
To: intel-...@googlegroups.com; Paul Birkel
Subject: Re: intel-devsys MDS-II Keyboard Matrix Encoder
Well, Paul asked for comments. I thought about it, wrote too much, put
it aside. Here's my thoughts.
1) If the goal of Roger and Paul, is to restore to use an MDS-230, than
this business about a replacement keyboard is secondary. The point is to
use the MDS, not be in the MDS production business, or the keyboard
business. Right?
[PAB] Well this is a Paul-alone affair; Roger is my supplier. Given how hard it is to find these things I feel that I owe him a debt of gratitude. You are correct, Use The MDS! I'm not interested in going into the keyboard business. A solution that might be useful to others is preferable, but in the very end it's my time and I'm the only customer. I do know how keyboards get separated from their monitors/computers/equipment and head in different directions. So this is not an uncommon problem; just the MDS-specific particulars are new (AFAICS).
it aside. Here's my thoughts.
1) If the goal of Roger and Paul, is to restore to use an MDS-230, than
this business about a replacement keyboard is secondary. The point is to
use the MDS, not be in the MDS production business, or the keyboard
business. Right?
2) Given 1), then it become obvious that you need a *functional*
keyboard, to operate the thing, right? And that would be, an Arduino (or
your favorite microcontroller) that can a) talk to the MDS keyboard
interface and b) respond to some PS/2 keyboard and jam MDS-compatible
codes into the MDS keyboard interface.
If you want something better, you'll need to know the MDS interface and
codes *anyway*, so you'll have design criteria. So - just do that, make
the *functional* arduino/ps2 keyboard as you describe. It's easier to
just do-that, and then discard it later; than to not-do-that, design
something else that doesn't work, figure out why, then redesign it to
make it work.
Just my gut opinion.
3) Keyboards are a subculture in vintage computing. Paul, if that's
unfamiliar territory to you, pardon me but do some homework on the
matter. Why? Because you are now in the keyboard business. You'll be
looking at keyswitches, keycaps, frames for holding the keyswitches,
keyswitch scanning methods in hardware and software. You aren't alone,
and it's old work too.
There's not only a present-day keyboard design and construction
subculture, there's a *vintage* subculture. I know this, from selling
1980's Apple Macintosh and Apple II parts, every day. People buy old
keyboards for use on modern equipment; or harvest parts from them for
new keyboards. Every day.
4) If you decide to design a keyboard as "here's how Intel did it on the
MDS, let's replicate what they did", you still have problems. It will be
hard to get 1979 keyswitches, because they are already in demand AND
scarce. Keycaps? worse. Because as I said: it's a subculture, and it's
custom. So, not cheap.
5) But: there's a subculture in modern custom mechanical switch
keyboards too. Look *that* up if it's not familiar. There are kits one
can buy, PC boards and controllers and parts. One can populate them with
ancient keycaps and keyswitches, and make your own USB keyboard.
There's likely make-your-own ASCII parallel keyboards too, for vintage
computing use.
So why use someone else's make-your-own keyboard? a) it's a done deal,
you get the PC board and parts and codes and other mechanicals. You just
have to modify the codes and add an MDS interface. Now you can reuse
that Arduino design work, see?
b) Maybe, you'll save money? and time? if you just use their keyswitches
and keycaps, but roll your own PC board, modify their circuit design?
You'll find out the cost of custom keycaps.
but more likely c) - you can at least *learn* from their design and
parts choices. You can buy where they buy their keycaps and keyswitches.
You can be informed by their design choices.
6) So - given these points, that keyboards are real things with real
design and parts and construction issues. Given the kinds of interests,
just in *keyboards*. There's also interests, in Intel MDS systems. It's
likely other people want "an MDS compatible keyboard", but it's not that
easy to make one "cold".
So: are you making a one-off keyboard? or a project to make as many as
someone else might want? (shrug) I see techies who are content to make
one thing for themselves, and not because someone else is interested. I
see others who want to make many, or at least allow others to make many.
Their interest in other people's use vary. I've seen that here, in the
various iterations of the Multibus discussion lists, for other "widgets"
designed for use.
------------------------------------
So, those are my comments. Primarily, the matter of making a keyboard is
not a side issue to many techies and of course engineering
professionals. It's a real matter and a surprisingly active field of
interest, either vintage or modern.
It seems reasonable to make a PS/2-Arduino type keyboard for use with
the MDS - because you need *something* right away, and you need to
figure out how to make *that* work, to design something else.
But making the "something else" is a real-enough project, if you want it
to be a real project. Keyboard parts are not cheap, but cheaper if you
borrow someone else's keyboard or someone else's keyboard kit. Keyboard
designs are not trivial; but others have already designed them for you,
borrow from their works. If you want to entirely replicate an Intel MDS
keyboard, well, that's serious squared, good luck with that.
These are matters for you and Roger to determine. I suggest you take
advantage of prior and recent work; and the availability of old or new
parts-kits of keyswitches and keycaps. Good luck.
Paul Birkel
Apr 7, 2021, 4:13:53 PM4/7/21
to intel-...@googlegroups.com
More good points Herb.
Speaking only for myself, my highest and best crafts-material remains wood, glue, sandpaper, sanding sealer, and paint. And sweat-equity. I know that those materials aren't de rigueur with some/most. But I'm just a hobbyist and those materials do have an honorable history :->.
After examining the MDS-230, and considering the design of the MDS 201 front panel (four-slot chassis, blank), unless a real front panel shows up I'll probably craft a functional substitute for myself (starting from a template, yes). Use aluminum inserts for the two tabs on the bottom edge (at least as seen on the MDS 201). I need to do the same thing for a missing bezel to go with a PDP-11/40 front panel, and one for an Intersystems DPS-1 (no front panel in that case; another future project to substitute for that). Where to all these things go? To the Land of Lost Parts ...
Your 1970's keyboard in blue sounds comparable to one that I have (and am equally unlikely to give up :->). I haven't dug it out because this is not the project for it; I'd have to add the five navigation cluster keys, and both FUNC and TPWR. And the existing keycaps would probably be wrong anyway. But I will be refreshing my memory soon as to its particulars. A nice looker from one of those "surplus" outlets that were all over Byte, et al. advertisements. Parallel interface. I've always loved those shades of blue (seems like there were a number in that era; I imagine spawned by IBM. A fun read (or gaze): https://www.ibm.com/design/language/color/
-----Original Message-----
From: intel-...@googlegroups.com [mailto:intel-...@googlegroups.com] On Behalf Of Herb Johnson
Sent: Wednesday, April 07, 2021 2:16 PM
To: intel-...@googlegroups.com
Subject: Re: intel-devsys MDS-II Keyboard Matrix Encoder
Speaking only for myself, my highest and best crafts-material remains wood, glue, sandpaper, sanding sealer, and paint. And sweat-equity. I know that those materials aren't de rigueur with some/most. But I'm just a hobbyist and those materials do have an honorable history :->.
After examining the MDS-230, and considering the design of the MDS 201 front panel (four-slot chassis, blank), unless a real front panel shows up I'll probably craft a functional substitute for myself (starting from a template, yes). Use aluminum inserts for the two tabs on the bottom edge (at least as seen on the MDS 201). I need to do the same thing for a missing bezel to go with a PDP-11/40 front panel, and one for an Intersystems DPS-1 (no front panel in that case; another future project to substitute for that). Where to all these things go? To the Land of Lost Parts ...
Your 1970's keyboard in blue sounds comparable to one that I have (and am equally unlikely to give up :->). I haven't dug it out because this is not the project for it; I'd have to add the five navigation cluster keys, and both FUNC and TPWR. And the existing keycaps would probably be wrong anyway. But I will be refreshing my memory soon as to its particulars. A nice looker from one of those "surplus" outlets that were all over Byte, et al. advertisements. Parallel interface. I've always loved those shades of blue (seems like there were a number in that era; I imagine spawned by IBM. A fun read (or gaze): https://www.ibm.com/design/language/color/
-----Original Message-----
From: intel-...@googlegroups.com [mailto:intel-...@googlegroups.com] On Behalf Of Herb Johnson
Sent: Wednesday, April 07, 2021 2:16 PM
To: intel-...@googlegroups.com
Subject: Re: intel-devsys MDS-II Keyboard Matrix Encoder
--
You received this message because you are subscribed to the Google Groups "intel-devsys" group.
To unsubscribe from this group and stop receiving emails from it, send an email to intel-devsys...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/intel-devsys/8d1eb747-d4fa-a171-16d5-c6884cf53e7e%40retrotechnology.com.
You received this message because you are subscribed to the Google Groups "intel-devsys" group.
To unsubscribe from this group and stop receiving emails from it, send an email to intel-devsys...@googlegroups.com.
Paul Birkel
Apr 7, 2021, 4:29:29 PM4/7/21
to intel-...@googlegroups.com
Well, just to confuse me I see that 9800544-04D_Intellec_Series_II_Schematics_Mar80.pdf terms them "Serial Channel #0" and "Serial Channel #1" Serial Channel #0 is the one with the added TTY interface.
According to 9800605-02B_MDS_Series_II_Boot_Monitor_Listing_May81.pdf (page 18 of 100):
; CONSOLE IS EITHER SERIAL CRT OR TTY
So that would be Serial Channel #0. However I gather that at various times & places folks have patched their Monitor and this may be one common patch-point? That said, there are variations between the various processor boards, and I'm quoting for the MDS II IPC. However the IPB seems to be the same.
-----Original Message-----
From: intel-...@googlegroups.com [mailto:intel-...@googlegroups.com] On Behalf Of Roger Arrick
Sent: Wednesday, April 07, 2021 3:32 PM
To: intel-...@googlegroups.com
Subject: Re: intel-devsys MDS-II Keyboard Matrix Encoder
To unsubscribe from this group and stop receiving emails from it, send an email to intel-devsys...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/intel-devsys/606E08C1.6040509%40Arrick.com.
According to 9800605-02B_MDS_Series_II_Boot_Monitor_Listing_May81.pdf (page 18 of 100):
; CONSOLE IS EITHER SERIAL CRT OR TTY
So that would be Serial Channel #0. However I gather that at various times & places folks have patched their Monitor and this may be one common patch-point? That said, there are variations between the various processor boards, and I'm quoting for the MDS II IPC. However the IPB seems to be the same.
-----Original Message-----
From: intel-...@googlegroups.com [mailto:intel-...@googlegroups.com] On Behalf Of Roger Arrick
Sent: Wednesday, April 07, 2021 3:32 PM
To: intel-...@googlegroups.com
Subject: Re: intel-devsys MDS-II Keyboard Matrix Encoder
To view this discussion on the web visit https://groups.google.com/d/msgid/intel-devsys/606E08C1.6040509%40Arrick.com.
Bill Beech (NJ7P)
Apr 7, 2021, 4:41:47 PM4/7/21
to intel-...@googlegroups.com
Guys,
I would like to get copies of the 8741/2 binary images as you work
through some of this keyboard work. Also, thought of replacing the 8741
in the standard PC/XT/AT/PS2 keyboard with one to make it an MDS keyboard?
I have a disassembler for the 8741/2 here. I would be glad to create
source code for the binary files.
Bill
I would like to get copies of the 8741/2 binary images as you work
through some of this keyboard work. Also, thought of replacing the 8741
in the standard PC/XT/AT/PS2 keyboard with one to make it an MDS keyboard?
I have a disassembler for the 8741/2 here. I would be glad to create
source code for the binary files.
Bill
Jack Mcmullen
Apr 7, 2021, 6:52:28 PM4/7/21
to intel-...@googlegroups.com
If you modify the BIOS prom the CRT can still be the "display" and the keyboard can be a serial port. This can then allow a typical IBM keyboard using a serial converter to fulfill that hole and still look nice! The BIOS has several human interfaces it looks for, a jump out of the IOC keyboard code to the serial input code should do it. Now to my knowledge none of the ISIS software directly accesses the serial ports, with the exception of possibly KERMIT
Jack
-----Original Message-----
From: Jon Hales <jonh...@gmail.com>
To: intel-...@googlegroups.com
Sent: Wed, Apr 7, 2021 8:38 am
Subject: Re: intel-devsys MDS-II Keyboard Matrix Encoder
From: Jon Hales <jonh...@gmail.com>
To: intel-...@googlegroups.com
Sent: Wed, Apr 7, 2021 8:38 am
Subject: Re: intel-devsys MDS-II Keyboard Matrix Encoder
To view this discussion on the web visit https://groups.google.com/d/msgid/intel-devsys/CANPXPJkWWeayRYB-REi9EQ%3D%2BRA%3DzHpX7JPkpqsuAwcbDOVKQGw%40mail.gmail.com.
Jack Mcmullen
Apr 7, 2021, 7:05:47 PM4/7/21
to intel-...@googlegroups.com
As I told Paul I'm pretty sure the IPDS keyboard uses the same zCherry switches and have the same key tops in most cases are very close to the MDS. Although the IPDS is scanned via a PIO chip, theres a good possibility the MDS keyboard 8741 and a keyboard PCB clone could be replicate for the 225 keyboard.
Jack
--
You received this message because you are subscribed to the Google Groups "intel-devsys" group.
You received this message because you are subscribed to the Google Groups "intel-devsys" group.
To unsubscribe from this group and stop receiving emails from it, send an email to intel-devsys+unsub...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/intel-devsys/46cafcae-b51f-7e79-b08e-88610f7f560f%40nj7p.info.
Roger Arrick
Apr 7, 2021, 7:53:56 PM4/7/21
to intel-...@googlegroups.com
I just confirmed that on my MDS II, you can use serial channel #2 as the console
for the internal monitor ROM all the way to ISIS-II 4.1. This port is DTE so
cross 2 and 3, and also needs CTS jumpered - normally I use RTS or DTR.
But not channel #1/TTY port.
1 and 2 are how they are labeled on the back.
I seriously doubt this monitor ROM has been hacked.
_______________________________________________
Roger Arrick
Ro...@Arrick.com
Tyler, TX
for the internal monitor ROM all the way to ISIS-II 4.1. This port is DTE so
cross 2 and 3, and also needs CTS jumpered - normally I use RTS or DTR.
But not channel #1/TTY port.
1 and 2 are how they are labeled on the back.
I seriously doubt this monitor ROM has been hacked.
_______________________________________________
Roger Arrick
Ro...@Arrick.com
Tyler, TX
Jack Mcmullen
Apr 7, 2021, 7:57:40 PM4/7/21
to intel-...@googlegroups.com
Serial Channel 1 (TTY) is what Rich's USB adapter board plugged into
>> send an email to intel-devsys+unsub...@googlegroups.com
>> <mailto:intel-devsys%2Bunsu...@googlegroups.com>.
>> To view this discussion on the web visit
>> https://groups.google.com/d/msgid/intel-devsys/02ab01d72bbe%2475534110%245ff9c330%24%40gmail.com.
>>
>> --
>> You received this message because you are subscribed to the Google
>> Groups "intel-devsys" group.
>> To unsubscribe from this group and stop receiving emails from it, send
>> an email to intel-devsys+unsub...@googlegroups.com
>> <mailto:intel-devsys+unsub...@googlegroups.com>.
>> <mailto:intel-devsys+unsub...@googlegroups.com>.
>> To view this discussion on the web visit
>> https://groups.google.com/d/msgid/intel-devsys/CAJ5BDDD%3DhNYB8feeeS7J0i3WgBsTbp6bGNge-LW5SNzFo_0dDw%40mail.gmail.com
>> <https://groups.google.com/d/msgid/intel-devsys/CAJ5BDDD%3DhNYB8feeeS7J0i3WgBsTbp6bGNge-LW5SNzFo_0dDw%40mail.gmail.com?utm_medium=email&utm_source=footer>.
>
--
You received this message because you are subscribed to the Google Groups "intel-devsys" group.
You received this message because you are subscribed to the Google Groups "intel-devsys" group.
To unsubscribe from this group and stop receiving emails from it, send an email to intel-devsys+unsub...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/intel-devsys/606E4609.7010204%40Arrick.com.
Roger Arrick
Apr 7, 2021, 8:10:09 PM4/7/21
to intel-...@googlegroups.com
I'm no MDS hero, what Paul is working on is parts of a system I bought 2 yrs ago
that was improperly packaged and destroyed. I felt bad for the lady who had
recently lost her husband who was the owner of this stuff and didn't even ask
for a refund.
It's mostly the case that was destroyed, the individual components seem semi-ok
- power supply, card cage, IOC board, CPU board (with missing switch cap), DD
FDC board set, even the CRT is intact. I'm confident this could be made into a
working system. The DD FDC board set tested out in another machine.
I was holding these parts as replacements to keep my MDS II system running in
the future but it makes more sense to let them go to someone willing to put in
the work to make a working system. I'm all about preservation so this meets my
altruistic goals. And Paul paid for it too, so I'm no saint. :) There are so
few of us interested in these systems - just a handful out of 7,000,000,000
people - so we should treat each other as gold, we're much rarer :)
As for the missing panels, I'd suggest MDF. Maybe you can find someone with a
CNC and good at Fusion360.
As for the paint - GOOD LUCK - nothing off-the shelf matches.
It seems to be close to a Sherwin Williams Polane-T Nitro Blue, common of the era.
_______________________________________________
Roger Arrick
Ro...@Arrick.com
Tyler, TX
Paul Birkel wrote:
that was improperly packaged and destroyed. I felt bad for the lady who had
recently lost her husband who was the owner of this stuff and didn't even ask
for a refund.
It's mostly the case that was destroyed, the individual components seem semi-ok
- power supply, card cage, IOC board, CPU board (with missing switch cap), DD
FDC board set, even the CRT is intact. I'm confident this could be made into a
working system. The DD FDC board set tested out in another machine.
I was holding these parts as replacements to keep my MDS II system running in
the future but it makes more sense to let them go to someone willing to put in
the work to make a working system. I'm all about preservation so this meets my
altruistic goals. And Paul paid for it too, so I'm no saint. :) There are so
few of us interested in these systems - just a handful out of 7,000,000,000
people - so we should treat each other as gold, we're much rarer :)
As for the missing panels, I'd suggest MDF. Maybe you can find someone with a
CNC and good at Fusion360.
As for the paint - GOOD LUCK - nothing off-the shelf matches.
It seems to be close to a Sherwin Williams Polane-T Nitro Blue, common of the era.
_______________________________________________
Roger Arrick
Ro...@Arrick.com
Tyler, TX
Paul Birkel wrote:
Jack Mcmullen
Apr 7, 2021, 8:26:39 PM4/7/21
to intel-...@googlegroups.com
-----Original Message-----
From: Roger Arrick <Ro...@Arrick.com>
To: intel-...@googlegroups.com
Sent: Wed, Apr 7, 2021 5:10 pm
Subject: Re: intel-devsys MDS-II Keyboard Matrix Encoder
From: Roger Arrick <Ro...@Arrick.com>
To: intel-...@googlegroups.com
Sent: Wed, Apr 7, 2021 5:10 pm
Subject: Re: intel-devsys MDS-II Keyboard Matrix Encoder
--
You received this message because you are subscribed to the Google Groups "intel-devsys" group.
To unsubscribe from this group and stop receiving emails from it, send an email to intel-devsys+unsub...@googlegroups.com.
You received this message because you are subscribed to the Google Groups "intel-devsys" group.
To unsubscribe from this group and stop receiving emails from it, send an email to intel-devsys+unsub...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/intel-devsys/606E49DA.5010306%40Arrick.com.
Paul Birkel
Apr 8, 2021, 2:28:31 AM4/8/21
to intel-...@googlegroups.com
Ah, it had occurred to me that the labeling on the rear panel might be different but I neglected to check. That's a (bit) confusing inconsistency on the part of Intel but I can understand that the zero-indexed approach to numbering isn't necessarily user-friendly for some audiences. I have much to learn here.
To unsubscribe from this group and stop receiving emails from it, send an email to intel-devsys...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/intel-devsys/606E4609.7010204%40Arrick.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/intel-devsys/606E4609.7010204%40Arrick.com.
Paul Birkel
Apr 8, 2021, 2:34:48 AM4/8/21
to intel-...@googlegroups.com
I’ll definitely check it out. It doesn’t sound exorbitant. My familiarity with “auto parts stores” are the usual consumer outlets. The DC region here tends to be a dead zone for industrial-type activities, but surely someone is supplying all of those body shops that I see around :->!
Do they give you the match specifications in a form that can be used by other suppliers (e.g., a Pantone ID)? It would be great to get that documented and in-the-record here.
To unsubscribe from this group and stop receiving emails from it, send an email to intel-devsys...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/intel-devsys/805082743.668878.1617841597534%40mail.yahoo.com.
Jack Mcmullen
Apr 8, 2021, 3:06:04 AM4/8/21
to intel-...@googlegroups.com
Hey Paul.
First off, forget Autozone, ORieleys, Quest, and all other "part" stores. What you need and I used is an Auto Paint Supplier. They supply body shops/collision repair etc. Stop in one and ask
I was able to give them the 1929 paint codes and with their computer and samples of what paint was on 2 pieces they matched it perfectly. I bought several (like 8, 4 of each color) rattle cans same number touch up bottles. They have two the new formulas on file and gave me a copy when I moved so I can get it made up here in Washington
To view this discussion on the web visit https://groups.google.com/d/msgid/intel-devsys/03d601d72c41%24587862e0%24096928a0%24%40gmail.com.
Eric Smith
Apr 8, 2021, 3:14:33 AM4/8/21
to intel-devsys
On Wed, Apr 7, 2021 at 8:57 AM Paul Birkel <pbi...@gmail.com> wrote:
- missing zero switch-cover on the IPC
The interrupt buttons are C&K Digitast, which are widely available, or equivalents. Those come in narrow and wide variants, and I think Intel used the narrow variant. I suspect it's C&K p/n SER-L-BK-RD-7.62-AU-OA. Three years ago, Digi-Key was selling them for $7.45 each, but the price could have gone up. An equivalent part is E-Switch p/n 5501MBLKRED, which Digi-Key listed fro $4.28.
The C&K switch covers _might_ be available separately, but I'd either replace the whole switch, or replace the cover with one from a new switch.
I have NOT actually purchased any of these, so I'm not 100% certain that it's an exact replacement.
Of course, you won't get them with the "0", so you'll have to somehow mark that yourself.
Eric Smith
Apr 8, 2021, 3:20:56 AM4/8/21
to intel-devsys
On Wed, Apr 7, 2021 at 8:57 AM Paul Birkel <pbi...@gmail.com> wrote:
A disassembly would be even better, if one has been developed.
Jack Mcmullen
Apr 8, 2021, 3:23:28 AM4/8/21
to intel-...@googlegroups.com
If I am wrong with "Cherry" Somewhere in my mind that has made a home. I've been wrong before.
-----Original Message-----
From: Eric Smith <spac...@gmail.com>
To: intel-devsys <intel-...@googlegroups.com>
Sent: Thu, Apr 8, 2021 12:14 am
Subject: Re: intel-devsys MDS-II Keyboard Matrix Encoder
From: Eric Smith <spac...@gmail.com>
To: intel-devsys <intel-...@googlegroups.com>
Sent: Thu, Apr 8, 2021 12:14 am
Subject: Re: intel-devsys MDS-II Keyboard Matrix Encoder
--
You received this message because you are subscribed to the Google Groups "intel-devsys" group.
To unsubscribe from this group and stop receiving emails from it, send an email to intel-devsys...@googlegroups.com.
To view this discussion on the web visit
You received this message because you are subscribed to the Google Groups "intel-devsys" group.
To unsubscribe from this group and stop receiving emails from it, send an email to intel-devsys...@googlegroups.com.
To view this discussion on the web visit
Eric Smith
Apr 8, 2021, 3:35:51 AM4/8/21
to intel-devsys
On Thu, Apr 8, 2021 at 1:23 AM 'Jack Mcmullen' via intel-devsys <intel-...@googlegroups.com> wrote:
If I am wrong with "Cherry" Somewhere in my mind that has made a home. I've been wrong before.
The keyboard might use Cherry. I'm talking about the interrupt buttons on the IPB & IPC, that poke out through the front panel. Those aren't Cherry, unless Cherry made a clone of the C&K Digitast.
Jack Mcmullen
Apr 8, 2021, 3:43:22 AM4/8/21
to intel-...@googlegroups.com
Nice!!! Thanks Eric!!
-----Original Message-----
From: Eric Smith <spac...@gmail.com>
To: intel-devsys <intel-...@googlegroups.com>
Sent: Thu, Apr 8, 2021 12:20 am
Subject: Re: intel-devsys MDS-II Keyboard Matrix Encoder
From: Eric Smith <spac...@gmail.com>
To: intel-devsys <intel-...@googlegroups.com>
Sent: Thu, Apr 8, 2021 12:20 am
Subject: Re: intel-devsys MDS-II Keyboard Matrix Encoder
--
You received this message because you are subscribed to the Google Groups "intel-devsys" group.
To unsubscribe from this group and stop receiving emails from it, send an email to intel-devsys...@googlegroups.com.
To view this discussion on the web visit
You received this message because you are subscribed to the Google Groups "intel-devsys" group.
To unsubscribe from this group and stop receiving emails from it, send an email to intel-devsys...@googlegroups.com.
To view this discussion on the web visit
Paul Birkel
Apr 8, 2021, 3:47:43 AM4/8/21
to intel-...@googlegroups.com
Bill:
Unfortunately the initial precondition here is getting someone to open up their keyboard. IAW the upgrade documentation (convert "RPT" to "FUNC") there was a supplied ROM chip-set for the IOC and a replacement 8041A / 8741A for the keyboard. So I infer that the keyboards were manufactured with socketed, rather than soldered-in, encoders. I wouldn't think that a field-upgrade would require a desoldering tool, but maybe my expectations are completely wrong? Need to inspect one. Of course a desoldering operation is possible, but it's not my keyboard to which surgery would need to be applied in that case. Someone else would need to perform that operation and take that risk.
Assuming sockets (could someone please check theirs?) the next step would be to read out the ROM. That's dependent on either the availably of a PROM programmer supporting that chip-family or willingness to breadboard a MCU into a suitable circuit. I expect that something similar to the circuit documented at https://www.sbprojects.net/projects/8049spy/index.php ("8049 Spy") will do. The only "big deal" there is applying +12Vdc to the EA (External Access) pin 7. "Reading a byte from ROM on an 8048 is only a matter of taking its EA line high up to 12V, keep RESET low and apply a 12 bit address. Then you must release the RESET (making it high), and the ROM byte will be presented on the DB port of the processor."
The programming (and thus reading) algorithm is a bit different for the 8741A. EA (pin 7) is driven to 23Vdc (higher). And TEST 0 (pin 1) is driven to +5Vdc to enable "verify mode". I'm quoting from the Intel datasheet "8041A/8641A/8741A Universal Peripheral Interface 8-bit Microcomputer" (AFN-00188B). We'd want to experiment with at least an 8471A first to verify the published algorithm.
Hacking the keyboard is an interesting idea but I suspect that not all of the necessary state information would get passed over the PS/2 protocol/cable (e.g., TPWR, but that's a problem either way) but it's certainly a consideration. However given that we'd still need an MCU to go from PS/2 protocol to the correct parallel interface for the IOC I'm thinking that it would be better to keep all of the hacking in one place (the off-board MCU). Of course we could simply rip out PS/2 support entirely from the keyboard and cable the parallel interface directly. But at that point all we're really accomplishing is reuse of a shell, a set of non-ideal keys, and the existing matrix wiring. Still need to rewrite the 8071A software.
So at least for me that's more pain and risk without any clear gain (e.g., greater fidelity to the MDS keyboard layout); it eliminates a "dongle" between the keyboard and the MDS back-panel but I don't see that having one is a real problem. I'd just Velcro it to the back panel (with a short MDS-side cable), or perhaps one could literally incorporate the back panel plug into its shell as a direct-attach dongle and have the PS/2 DIN connector off the side so as to limit the added depth and make the DIN connector more easily accessible.
Or one could put it inside the case and expose a DIN connector -- but I'm definitely not in favor of that approach since it potentially amounts to a destructive/irreversible modification of the MDS for the DIN-penetration. However perhaps clearance is such that the J1 Keyboard connector could be replaced by a plate that mounts the DIN connector without any impact on existing metal-work (and there is sufficient back-clearance for the connector WRT the IOC board), in which case I prefer that approach over having a dongle. (Now add a PS/2 mouse that feeds the MCU to emulate the arrow-keys for some real anachronism :-.>.)
Keeping the MCU protocol/encoding conversion separate from the keyboard allows folks to use whatever PS/2 keyboard look/feel strikes their fancy. And swap it when their fancy changes, or the keyboard fails.
Thank you for the disassembly offer; I hope that we get to call-in that favor someday :->!
But, so far I'm hearing that no one has already extracted the 8041A / 8741A ROM? Bill Mcmullen thought that someone already had.
Is anyone willing to take a peek inside their keyboard to see which IC they have and whether it is socketed? That would be a start.
Unfortunately the initial precondition here is getting someone to open up their keyboard. IAW the upgrade documentation (convert "RPT" to "FUNC") there was a supplied ROM chip-set for the IOC and a replacement 8041A / 8741A for the keyboard. So I infer that the keyboards were manufactured with socketed, rather than soldered-in, encoders. I wouldn't think that a field-upgrade would require a desoldering tool, but maybe my expectations are completely wrong? Need to inspect one. Of course a desoldering operation is possible, but it's not my keyboard to which surgery would need to be applied in that case. Someone else would need to perform that operation and take that risk.
Assuming sockets (could someone please check theirs?) the next step would be to read out the ROM. That's dependent on either the availably of a PROM programmer supporting that chip-family or willingness to breadboard a MCU into a suitable circuit. I expect that something similar to the circuit documented at https://www.sbprojects.net/projects/8049spy/index.php ("8049 Spy") will do. The only "big deal" there is applying +12Vdc to the EA (External Access) pin 7. "Reading a byte from ROM on an 8048 is only a matter of taking its EA line high up to 12V, keep RESET low and apply a 12 bit address. Then you must release the RESET (making it high), and the ROM byte will be presented on the DB port of the processor."
The programming (and thus reading) algorithm is a bit different for the 8741A. EA (pin 7) is driven to 23Vdc (higher). And TEST 0 (pin 1) is driven to +5Vdc to enable "verify mode". I'm quoting from the Intel datasheet "8041A/8641A/8741A Universal Peripheral Interface 8-bit Microcomputer" (AFN-00188B). We'd want to experiment with at least an 8471A first to verify the published algorithm.
Hacking the keyboard is an interesting idea but I suspect that not all of the necessary state information would get passed over the PS/2 protocol/cable (e.g., TPWR, but that's a problem either way) but it's certainly a consideration. However given that we'd still need an MCU to go from PS/2 protocol to the correct parallel interface for the IOC I'm thinking that it would be better to keep all of the hacking in one place (the off-board MCU). Of course we could simply rip out PS/2 support entirely from the keyboard and cable the parallel interface directly. But at that point all we're really accomplishing is reuse of a shell, a set of non-ideal keys, and the existing matrix wiring. Still need to rewrite the 8071A software.
So at least for me that's more pain and risk without any clear gain (e.g., greater fidelity to the MDS keyboard layout); it eliminates a "dongle" between the keyboard and the MDS back-panel but I don't see that having one is a real problem. I'd just Velcro it to the back panel (with a short MDS-side cable), or perhaps one could literally incorporate the back panel plug into its shell as a direct-attach dongle and have the PS/2 DIN connector off the side so as to limit the added depth and make the DIN connector more easily accessible.
Or one could put it inside the case and expose a DIN connector -- but I'm definitely not in favor of that approach since it potentially amounts to a destructive/irreversible modification of the MDS for the DIN-penetration. However perhaps clearance is such that the J1 Keyboard connector could be replaced by a plate that mounts the DIN connector without any impact on existing metal-work (and there is sufficient back-clearance for the connector WRT the IOC board), in which case I prefer that approach over having a dongle. (Now add a PS/2 mouse that feeds the MCU to emulate the arrow-keys for some real anachronism :-.>.)
Keeping the MCU protocol/encoding conversion separate from the keyboard allows folks to use whatever PS/2 keyboard look/feel strikes their fancy. And swap it when their fancy changes, or the keyboard fails.
Thank you for the disassembly offer; I hope that we get to call-in that favor someday :->!
But, so far I'm hearing that no one has already extracted the 8041A / 8741A ROM? Bill Mcmullen thought that someone already had.
Is anyone willing to take a peek inside their keyboard to see which IC they have and whether it is socketed? That would be a start.
--
You received this message because you are subscribed to the Google Groups "intel-devsys" group.
To unsubscribe from this group and stop receiving emails from it, send an email to intel-devsys...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/intel-devsys/46cafcae-b51f-7e79-b08e-88610f7f560f%40nj7p.info.
You received this message because you are subscribed to the Google Groups "intel-devsys" group.
To unsubscribe from this group and stop receiving emails from it, send an email to intel-devsys...@googlegroups.com.
Paul Birkel
Apr 8, 2021, 3:52:52 AM4/8/21
to intel-...@googlegroups.com
Jack: I had wondered about that as a possibility, but was first-pass thinking through avoiding modifying the boot/monitor software. Your proposal still requires a MCU for protocol conversion from PS/2 (or perhaps then USB) to RS-232, if not for encoding-conversion. How do TPWR and FUNC work through the serial channel?
To view this discussion on the web visit https://groups.google.com/d/msgid/intel-devsys/1359339909.627743.1617835942390%40mail.yahoo.com.
Paul Birkel
Apr 8, 2021, 3:58:16 AM4/8/21
to intel-...@googlegroups.com
Bingo!
You say “Presently this project has partially reverse-engineered source code for the unenhanced keyboard 8741A, part number 9100101, in the kbd directory, and for the enhanced keyboard 8741A, part number 104675-001, in the kbd-enhanced directory.” In what sense(s) is it partial and what remains to be accomplished?
From: intel-...@googlegroups.com [mailto:intel-...@googlegroups.com] On Behalf Of Eric Smith
Sent: Thursday, April 08, 2021 3:21 AM
To: intel-devsys
Subject: Re: intel-devsys MDS-II Keyboard Matrix Encoder
On Wed, Apr 7, 2021 at 8:57 AM Paul Birkel <pbi...@gmail.com> wrote:
--
You received this message because you are subscribed to the Google Groups "intel-devsys" group.
To unsubscribe from this group and stop receiving emails from it, send an email to intel-devsys...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/intel-devsys/CAFrGgTQg9sbd8rWrapjOH%2B7gy7z-M__q8PbETimndsHn9rwUFA%40mail.gmail.com.
Paul Birkel
Apr 8, 2021, 4:10:19 AM4/8/21
to intel-...@googlegroups.com
Jack: “Cherry” would be the keyboard; these are the switches on the IPB/IPC edge for the interrupts.
Eric: _great_ information, thank you.
Digi-Key: https://www.digikey.com/en/products/detail/c-k/SERL-BK-RD-7-62-AU-OA/417769 $8.27 (336 in stock)
Mouser: https://www.mouser.com/ProductDetail/CK/SERLBKRD762AUOA?qs=CJN2HVZRpZ0PuJ6pLavs2A%3D%3D $6.22 (205 in stock)
Digi-Key: https://www.digikey.com/en/products/detail/e-switch/5501MBLKRED/101666 $5.75 (1,212 in stock)
Mouser: https://www.mouser.com/ProductDetail/E-Switch/5501MBLKREDGRN?qs=f57gQzlyLiqtQ3Pb7mKRgA%3D%3D $4.59 (347 in stock)
Note that in the latter case Mouser only stocks RED with a green LED. Not suitable in this case. But they do supply a datasheet, whereas Digi-key doesn’t.
I’ll compare the product sheet dimensions to the IPC when it arrives. IMO “zero” can reasonably be inferred from the progression. At least for my purposes.
To view this discussion on the web visit https://groups.google.com/d/msgid/intel-devsys/805528308.759139.1617866605293%40mail.yahoo.com.
Herb Johnson
Apr 8, 2021, 10:40:02 AM4/8/21
to intel-...@googlegroups.com
I dithered about whether to post the following documentation trail or
not. But everyone else has posted their details. I may as well follow
the crowd.
So: Web search tells me about the keyboard and its operation.
https://github.com/brouhaha/mds-fw
Would seem to answer a lot of questions, specifically some 8041 codes.
I'm not sure if these are IOC 8041's or keyboard 8041's. But: focus. The
goal is not to replicate an 8041.
https://www.retrotechnology.com/restore/mon80_proms.html
I have some info on my site, and I point to other sites. Like Al
Kossow's who has these docs:
http://bitsavers.informatik.uni-stuttgart.de/pdf/intel/MDS2/firmware/
http://bitsavers.informatik.uni-stuttgart.de/pdf/intel/MDS2/9800544-04D_Intellec_Series_II_Schematics_Mar80.pdf
http://bitsavers.informatik.uni-stuttgart.de/pdf/intel/MDS2/122029-001_Intellec_Series_II_CRT_and_Keyboard_Interface_Sep82.pdf
http://bitsavers.informatik.uni-stuttgart.de/pdf/intel/MDS2/MDS_Troubleshooting_Manual.pdf
the schematic document 9800544-04D, PDF page 40, schematic sheet 5 of
12, shows J1 "Keyboard inputs" as a connector with 8 data/ inputs (to
the IOC), KB_PRESENT/ input; READ KB/ output, STATUS/DATA output,
KB_RESET/ output (from the IOC). Outputs are 7432's ORs (apparently
not open-collector). Inputs are to chps AB0 AB1 74LS240 inputs.
Keyboard schematic same document PDF page 56. Someone familiar with an
8041 can interpret inputs/outputs but it would appear that the IOC can
read two "locations" of the 8041, reset it, and perform a read by
strobing READ_KB/.
Point being: the keyboard hardware interface is not complicated. It's
known. Good.
Now: about how the keyboard *works*. Read the documents!
I didn't study the protocols among the documents; I'm not writing the
code for the replacement. The protocols can't be awful. It's a
*keyboard*! How complicated can codes from a *keyboard* be? The
operation appears to be documented. The ROM in the IOC appears to be
captured and likely disassembled. The programs running in the MDS are
documented and likely disassembled. and surely, someone has one of these
puppies operating - Ask. Them.
So how much further detail is needed? Does someone actually HAVE TO
desolder a 8041 from an ancient keyboard just to figure out what that
keyboard *does*? And- maybe "someone" already did that. Do some homework.
Back to Paul's objectives.
So Paul's scheme seems plausible. Obtain that PS/2 parallel keyboard
controller he pointed out. Make sure it works! A row of LEDs will be
sufficient, maybe a TTL latch.
Then futz with its code to provide two bytes of data on demand (data,
status), also accept a reset. Or add a hardware latch and some
gate-logic to create that "status" byte. OK? This is not hard.
Next. Get basic A-Z keyboard operation going. If the MDS is working,
Paul can operate it from the serial port and treat the keyboard as an
I/O device, debug in that fashion. Otherwise, futz up a breadboard and
some LED's.
Then program into the Arduino, the features of the Intel keyboard as
documented or as actual owners can demonstrate. If you get it wrong,
*reprogram* and repeat.
I've said ZERO about the PS/2 keyboard. That's correct.
..and that's that.
The hard part won't be decoding the Intel keyboard, making something
work. The hard part will be *restoring the Intel system shipped to Paul
to work at all*. That's where Paul will really need help, and parts,
and luck. And. Homework.
But you know, Paul was wrong about "anyone interested in an Intel
keyboard? (Cricket-chirps) " recent correspondence in this list says
otherwise. My spider-sense about keyboards - based on my recent years of
selling people keyswitches and keycaps - tells me that five people would
LOVE to build, or buy, or at least look at, a faux Intel MDS keyboard.
And fifty people would at least read a Web page about it.
It can be a thing, if someone wants to do that. Does it need an 8041? I
dunno, not necessarily. Depends on your "interests".
People are entitled to have their interests, and one person's interests
is another's useful (or useless) detail. I've helped people restore
their S-100 systems in the past, and now their vintage Apple Macs today,
and even their keyboards. Details? It's all good, someone somewhere
needs that detail.
Regards, Herb
not. But everyone else has posted their details. I may as well follow
the crowd.
So: Web search tells me about the keyboard and its operation.
https://github.com/brouhaha/mds-fw
Would seem to answer a lot of questions, specifically some 8041 codes.
I'm not sure if these are IOC 8041's or keyboard 8041's. But: focus. The
goal is not to replicate an 8041.
https://www.retrotechnology.com/restore/mon80_proms.html
I have some info on my site, and I point to other sites. Like Al
Kossow's who has these docs:
http://bitsavers.informatik.uni-stuttgart.de/pdf/intel/MDS2/firmware/
http://bitsavers.informatik.uni-stuttgart.de/pdf/intel/MDS2/9800544-04D_Intellec_Series_II_Schematics_Mar80.pdf
http://bitsavers.informatik.uni-stuttgart.de/pdf/intel/MDS2/122029-001_Intellec_Series_II_CRT_and_Keyboard_Interface_Sep82.pdf
http://bitsavers.informatik.uni-stuttgart.de/pdf/intel/MDS2/MDS_Troubleshooting_Manual.pdf
the schematic document 9800544-04D, PDF page 40, schematic sheet 5 of
12, shows J1 "Keyboard inputs" as a connector with 8 data/ inputs (to
the IOC), KB_PRESENT/ input; READ KB/ output, STATUS/DATA output,
KB_RESET/ output (from the IOC). Outputs are 7432's ORs (apparently
not open-collector). Inputs are to chps AB0 AB1 74LS240 inputs.
Keyboard schematic same document PDF page 56. Someone familiar with an
8041 can interpret inputs/outputs but it would appear that the IOC can
read two "locations" of the 8041, reset it, and perform a read by
strobing READ_KB/.
Point being: the keyboard hardware interface is not complicated. It's
known. Good.
Now: about how the keyboard *works*. Read the documents!
I didn't study the protocols among the documents; I'm not writing the
code for the replacement. The protocols can't be awful. It's a
*keyboard*! How complicated can codes from a *keyboard* be? The
operation appears to be documented. The ROM in the IOC appears to be
captured and likely disassembled. The programs running in the MDS are
documented and likely disassembled. and surely, someone has one of these
puppies operating - Ask. Them.
So how much further detail is needed? Does someone actually HAVE TO
desolder a 8041 from an ancient keyboard just to figure out what that
keyboard *does*? And- maybe "someone" already did that. Do some homework.
Back to Paul's objectives.
So Paul's scheme seems plausible. Obtain that PS/2 parallel keyboard
controller he pointed out. Make sure it works! A row of LEDs will be
sufficient, maybe a TTL latch.
Then futz with its code to provide two bytes of data on demand (data,
status), also accept a reset. Or add a hardware latch and some
gate-logic to create that "status" byte. OK? This is not hard.
Next. Get basic A-Z keyboard operation going. If the MDS is working,
Paul can operate it from the serial port and treat the keyboard as an
I/O device, debug in that fashion. Otherwise, futz up a breadboard and
some LED's.
Then program into the Arduino, the features of the Intel keyboard as
documented or as actual owners can demonstrate. If you get it wrong,
*reprogram* and repeat.
I've said ZERO about the PS/2 keyboard. That's correct.
..and that's that.
The hard part won't be decoding the Intel keyboard, making something
work. The hard part will be *restoring the Intel system shipped to Paul
to work at all*. That's where Paul will really need help, and parts,
and luck. And. Homework.
But you know, Paul was wrong about "anyone interested in an Intel
keyboard? (Cricket-chirps) " recent correspondence in this list says
otherwise. My spider-sense about keyboards - based on my recent years of
selling people keyswitches and keycaps - tells me that five people would
LOVE to build, or buy, or at least look at, a faux Intel MDS keyboard.
And fifty people would at least read a Web page about it.
It can be a thing, if someone wants to do that. Does it need an 8041? I
dunno, not necessarily. Depends on your "interests".
People are entitled to have their interests, and one person's interests
is another's useful (or useless) detail. I've helped people restore
their S-100 systems in the past, and now their vintage Apple Macs today,
and even their keyboards. Details? It's all good, someone somewhere
needs that detail.
Regards, Herb
Roger Arrick
Apr 8, 2021, 11:37:19 AM4/8/21
to intel-...@googlegroups.com
Take any microcontroller with USB host and capture the keyboard codes then send
them to its RS232 port. Wire that one TX line over to the MDS serial port #2 RX
line and you'll be in business.
Roger
Paul Birkel wrote:
> Jack: “Cherry” would be the keyboard; these are the switches on the
> IPB/IPC edge for the interrupts.
>
>
>
> Eric: _/great/_ information, thank you.
> [mailto:intel-...@googlegroups.com]
> *Sent:* Thursday, April 08, 2021 3:23 AM
> *To:* intel-...@googlegroups.com
> *Subject:* Re: intel-devsys MDS-II Keyboard Matrix Encoder
> <mailto:intel-devsys...@googlegroups.com>.
> <https://groups.google.com/d/msgid/intel-devsys/805528308.759139.1617866605293%40mail.yahoo.com?utm_medium=email&utm_source=footer>.
> <https://groups.google.com/d/msgid/intel-devsys/040901d72c4e%24afffd810%240fff8830%24%40gmail.com?utm_medium=email&utm_source=footer>.
them to its RS232 port. Wire that one TX line over to the MDS serial port #2 RX
line and you'll be in business.
Roger
Paul Birkel wrote:
> Jack: “Cherry” would be the keyboard; these are the switches on the
> IPB/IPC edge for the interrupts.
>
>
>
>
>
>
> Digi-Key:
> https://www.digikey.com/en/products/detail/c-k/SERL-BK-RD-7-62-AU-OA/417769
> $8.27 (336 in stock)
>
> Mouser:
> https://www.mouser.com/ProductDetail/CK/SERLBKRD762AUOA?qs=CJN2HVZRpZ0PuJ6pLavs2A%3D%3D
> $6.22 (205 in stock)
>
>
>
> Digi-Key:
> https://www.digikey.com/en/products/detail/e-switch/5501MBLKRED/101666
> $5.75 (1,212 in stock)
>
> Mouser:
> https://www.mouser.com/ProductDetail/E-Switch/5501MBLKREDGRN?qs=f57gQzlyLiqtQ3Pb7mKRgA%3D%3D
> $4.59 (347 in stock)
>
>
>
> Note that in the latter case Mouser only stocks RED with a green LED.
> Not suitable in this case. But they do supply a datasheet, whereas
> Digi-key doesn’t.
>
>
>
> I’ll compare the product sheet dimensions to the IPC when it arrives.
> IMO “zero” can reasonably be inferred from the progression. At least
> for my purposes.
>
>
>
> *From:* 'Jack Mcmullen' via intel-devsys
>
>
> Digi-Key:
> https://www.digikey.com/en/products/detail/c-k/SERL-BK-RD-7-62-AU-OA/417769
> $8.27 (336 in stock)
>
> Mouser:
> https://www.mouser.com/ProductDetail/CK/SERLBKRD762AUOA?qs=CJN2HVZRpZ0PuJ6pLavs2A%3D%3D
> $6.22 (205 in stock)
>
>
>
> Digi-Key:
> https://www.digikey.com/en/products/detail/e-switch/5501MBLKRED/101666
> $5.75 (1,212 in stock)
>
> Mouser:
> https://www.mouser.com/ProductDetail/E-Switch/5501MBLKREDGRN?qs=f57gQzlyLiqtQ3Pb7mKRgA%3D%3D
> $4.59 (347 in stock)
>
>
>
> Note that in the latter case Mouser only stocks RED with a green LED.
> Not suitable in this case. But they do supply a datasheet, whereas
> Digi-key doesn’t.
>
>
>
> I’ll compare the product sheet dimensions to the IPC when it arrives.
> IMO “zero” can reasonably be inferred from the progression. At least
> for my purposes.
>
>
>
> [mailto:intel-...@googlegroups.com]
> *Sent:* Thursday, April 08, 2021 3:23 AM
> *To:* intel-...@googlegroups.com
> *Subject:* Re: intel-devsys MDS-II Keyboard Matrix Encoder
> To view this discussion on the web visit
>
> https://groups.google.com/d/msgid/intel-devsys/CAFrGgTQtW9bBfVhO42mDwUNSa1S7y5GGzdHLb-abG67kGptrOQ%40mail.gmail.com
> <https://groups.google.com/d/msgid/intel-devsys/CAFrGgTQtW9bBfVhO42mDwUNSa1S7y5GGzdHLb-abG67kGptrOQ%40mail.gmail.com?utm_medium=email&utm_source=footer>
>
> https://groups.google.com/d/msgid/intel-devsys/CAFrGgTQtW9bBfVhO42mDwUNSa1S7y5GGzdHLb-abG67kGptrOQ%40mail.gmail.com
>
>
> .
>
> --
> You received this message because you are subscribed to the Google
> Groups "intel-devsys" group.
> To unsubscribe from this group and stop receiving emails from it, send
> an email to intel-devsys...@googlegroups.com
> <mailto:intel-devsys...@googlegroups.com>.
>
> .
>
> --
> You received this message because you are subscribed to the Google
> Groups "intel-devsys" group.
> To unsubscribe from this group and stop receiving emails from it, send
> an email to intel-devsys...@googlegroups.com
> To view this discussion on the web visit
> https://groups.google.com/d/msgid/intel-devsys/805528308.759139.1617866605293%40mail.yahoo.com
> <https://groups.google.com/d/msgid/intel-devsys/805528308.759139.1617866605293%40mail.yahoo.com?utm_medium=email&utm_source=footer>.
>
> --
> You received this message because you are subscribed to the Google
> Groups "intel-devsys" group.
> To unsubscribe from this group and stop receiving emails from it, send
> an email to intel-devsys...@googlegroups.com
> <mailto:intel-devsys...@googlegroups.com>.
> --
> You received this message because you are subscribed to the Google
> Groups "intel-devsys" group.
> To unsubscribe from this group and stop receiving emails from it, send
> an email to intel-devsys...@googlegroups.com
> To view this discussion on the web visit
> https://groups.google.com/d/msgid/intel-devsys/040901d72c4e%24afffd810%240fff8830%24%40gmail.com
> <https://groups.google.com/d/msgid/intel-devsys/040901d72c4e%24afffd810%240fff8830%24%40gmail.com?utm_medium=email&utm_source=footer>.
Bill Beech (NJ7P)
Apr 8, 2021, 3:06:01 PM4/8/21
to intel-...@googlegroups.com
The PC interface used a shift register to make 8-bit parallel ASCII from the DIN keyboard....
Bill
To view this discussion on the web visit https://groups.google.com/d/msgid/intel-devsys/03f601d72c4c%24408a96c0%24c19fc440%24%40gmail.com.
Bill Beech (NJ7P)
Apr 8, 2021, 3:11:13 PM4/8/21
to intel-...@googlegroups.com
Not ASCII, but keyboard codes...
To view this discussion on the web visit https://groups.google.com/d/msgid/intel-devsys/e85874fb-b1a5-132d-591a-446c5afab495%40nj7p.info.
Jack Mcmullen
Apr 8, 2021, 4:05:29 PM4/8/21
to intel-...@googlegroups.com
My Bad, sorry
-----Original Message-----
From: Eric Smith <spac...@gmail.com>
To: intel-devsys <intel-...@googlegroups.com>
Sent: Thu, Apr 8, 2021 12:35 am
Subject: Re: intel-devsys MDS-II Keyboard Matrix Encoder
From: Eric Smith <spac...@gmail.com>
To: intel-devsys <intel-...@googlegroups.com>
Sent: Thu, Apr 8, 2021 12:35 am
Subject: Re: intel-devsys MDS-II Keyboard Matrix Encoder
--
You received this message because you are subscribed to the Google Groups "intel-devsys" group.
To unsubscribe from this group and stop receiving emails from it, send an email to intel-devsys...@googlegroups.com.
To view this discussion on the web visit
You received this message because you are subscribed to the Google Groups "intel-devsys" group.
To unsubscribe from this group and stop receiving emails from it, send an email to intel-devsys...@googlegroups.com.
To view this discussion on the web visit
Jack Mcmullen
Apr 8, 2021, 4:19:48 PM4/8/21
to intel-...@googlegroups.com
Morning Paul,
There are PS2 to Serial adapters
Now IBM like keyboards to MDS will have differences in keys. Credit and Aedit will need "MAC" files to map a PS2 ket to an editor function. Rest of ISIS is normal ASCII.
I;ve forgotten what TPWR and FUNC do in ISIS, please remind
To view this discussion on the web visit https://groups.google.com/d/msgid/intel-devsys/03f601d72c4c%24408a96c0%24c19fc440%24%40gmail.com.
Roger Arrick
Apr 8, 2021, 5:03:37 PM4/8/21
to intel-...@googlegroups.com
This adapter is BS.
There's no conversion going on there.
Here's a pinout:
https://www.pccables.com/products/00361.html
And it's DE-9 not DB-9. lol
Roger
'Jack Mcmullen' via intel-devsys wrote:
> Morning Paul,
> There are PS2 to Serial adapters
> https://www.amazon.com/CablesOnline-MiniDin6-Female-Adapter-AD-M01-2/dp/B00HGI5JPA
>
> Now IBM like keyboards to MDS will have differences in keys. Credit and
> Aedit will need "MAC" files to map a PS2 ket to an editor function. Rest
> of ISIS is normal ASCII.
>
> I;ve forgotten what TPWR and FUNC do in ISIS, please remind
>
>
> -----Original Message-----
> From: Paul Birkel <pbi...@gmail.com>
> To: intel-...@googlegroups.com
> Sent: Thu, Apr 8, 2021 12:53 am
> Subject: RE: intel-devsys MDS-II Keyboard Matrix Encoder
>
> Jack: I had wondered about that as a possibility, but was first-pass
> thinking through avoiding modifying the boot/monitor software. Your
> proposal still requires a MCU for protocol conversion from PS/2 (or
> perhaps then USB) to RS-232, if not for encoding-conversion. How do
> TPWR and FUNC work through the serial channel?
>
> *From:* 'Jack Mcmullen' via intel-devsys
> [mailto:intel-...@googlegroups.com]
> *Sent:* Wednesday, April 07, 2021 6:52 PM
> *To:* intel-...@googlegroups.com
> *Subject:* Re: intel-devsys MDS-II Keyboard Matrix Encoder
> <mailto:intel-devsys%2Bunsu...@googlegroups.com>.
> <https://groups.google.com/d/msgid/intel-devsys/CANPXPJkWWeayRYB-REi9EQ%3D%2BRA%3DzHpX7JPkpqsuAwcbDOVKQGw%40mail.gmail.com?utm_medium=email&utm_source=footer>.
> <https://groups.google.com/d/msgid/intel-devsys/1359339909.627743.1617835942390%40mail.yahoo.com?utm_medium=email&utm_source=footer>.
> <https://groups.google.com/d/msgid/intel-devsys/03f601d72c4c%24408a96c0%24c19fc440%24%40gmail.com?utm_medium=email&utm_source=footer>.
> <https://groups.google.com/d/msgid/intel-devsys/1661480240.32767.1617913185569%40mail.yahoo.com?utm_medium=email&utm_source=footer>.
There's no conversion going on there.
Here's a pinout:
https://www.pccables.com/products/00361.html
And it's DE-9 not DB-9. lol
Roger
'Jack Mcmullen' via intel-devsys wrote:
> Morning Paul,
> There are PS2 to Serial adapters
> https://www.amazon.com/CablesOnline-MiniDin6-Female-Adapter-AD-M01-2/dp/B00HGI5JPA
>
> Now IBM like keyboards to MDS will have differences in keys. Credit and
> Aedit will need "MAC" files to map a PS2 ket to an editor function. Rest
> of ISIS is normal ASCII.
>
> I;ve forgotten what TPWR and FUNC do in ISIS, please remind
>
>
> -----Original Message-----
> From: Paul Birkel <pbi...@gmail.com>
> To: intel-...@googlegroups.com
> Sent: Thu, Apr 8, 2021 12:53 am
> Subject: RE: intel-devsys MDS-II Keyboard Matrix Encoder
>
> Jack: I had wondered about that as a possibility, but was first-pass
> thinking through avoiding modifying the boot/monitor software. Your
> proposal still requires a MCU for protocol conversion from PS/2 (or
> perhaps then USB) to RS-232, if not for encoding-conversion. How do
> TPWR and FUNC work through the serial channel?
>
> [mailto:intel-...@googlegroups.com]
> *Sent:* Wednesday, April 07, 2021 6:52 PM
> *To:* intel-...@googlegroups.com
> *Subject:* Re: intel-devsys MDS-II Keyboard Matrix Encoder
> To view this discussion on the web visit
> https://groups.google.com/d/msgid/intel-devsys/02ab01d72bbe%2475534110%245ff9c330%24%40gmail.com.
>
> --
> You received this message because you are subscribed to the Google
> Groups "intel-devsys" group.
> To unsubscribe from this group and stop receiving emails from it, send
> an email to intel-devsys...@googlegroups.com
> <mailto:intel-devsys...@googlegroups.com>.
> https://groups.google.com/d/msgid/intel-devsys/02ab01d72bbe%2475534110%245ff9c330%24%40gmail.com.
>
> --
> You received this message because you are subscribed to the Google
> Groups "intel-devsys" group.
> To unsubscribe from this group and stop receiving emails from it, send
> an email to intel-devsys...@googlegroups.com
> To view this discussion on the web visit
> https://groups.google.com/d/msgid/intel-devsys/CANPXPJkWWeayRYB-REi9EQ%3D%2BRA%3DzHpX7JPkpqsuAwcbDOVKQGw%40mail.gmail.com
> <https://groups.google.com/d/msgid/intel-devsys/CANPXPJkWWeayRYB-REi9EQ%3D%2BRA%3DzHpX7JPkpqsuAwcbDOVKQGw%40mail.gmail.com?utm_medium=email&utm_source=footer>.
> --
> You received this message because you are subscribed to the Google
> Groups "intel-devsys" group.
> To unsubscribe from this group and stop receiving emails from it, send
> an email to intel-devsys...@googlegroups.com
> <mailto:intel-devsys...@googlegroups.com>.
> You received this message because you are subscribed to the Google
> Groups "intel-devsys" group.
> To unsubscribe from this group and stop receiving emails from it, send
> an email to intel-devsys...@googlegroups.com
> To view this discussion on the web visit
> https://groups.google.com/d/msgid/intel-devsys/1359339909.627743.1617835942390%40mail.yahoo.com
> <https://groups.google.com/d/msgid/intel-devsys/1359339909.627743.1617835942390%40mail.yahoo.com?utm_medium=email&utm_source=footer>.
> --
> You received this message because you are subscribed to the Google
> Groups "intel-devsys" group.
> To unsubscribe from this group and stop receiving emails from it, send
> an email to intel-devsys...@googlegroups.com
> <mailto:intel-devsys...@googlegroups.com>.
> You received this message because you are subscribed to the Google
> Groups "intel-devsys" group.
> To unsubscribe from this group and stop receiving emails from it, send
> an email to intel-devsys...@googlegroups.com
> To view this discussion on the web visit
> https://groups.google.com/d/msgid/intel-devsys/03f601d72c4c%24408a96c0%24c19fc440%24%40gmail.com
> <https://groups.google.com/d/msgid/intel-devsys/03f601d72c4c%24408a96c0%24c19fc440%24%40gmail.com?utm_medium=email&utm_source=footer>.
>
> --
> You received this message because you are subscribed to the Google
> Groups "intel-devsys" group.
> To unsubscribe from this group and stop receiving emails from it, send
> an email to intel-devsys...@googlegroups.com
> <mailto:intel-devsys...@googlegroups.com>.
> --
> You received this message because you are subscribed to the Google
> Groups "intel-devsys" group.
> To unsubscribe from this group and stop receiving emails from it, send
> an email to intel-devsys...@googlegroups.com
> To view this discussion on the web visit
> https://groups.google.com/d/msgid/intel-devsys/1661480240.32767.1617913185569%40mail.yahoo.com
> <https://groups.google.com/d/msgid/intel-devsys/1661480240.32767.1617913185569%40mail.yahoo.com?utm_medium=email&utm_source=footer>.
Jack Mcmullen
Apr 8, 2021, 5:13:55 PM4/8/21
to intel-...@googlegroups.com
One nice thing about the self contained "processor" is, in the case of the 87xx series, is the instruction set is easily understoor and a dedicated scanning keyboard setup is very straight forward. With the schematic functional understanding of key down to code out will expose itself quickly. I am not sure if the 225 keyboard output is a ASCII code or a key code dealt with by the IOC.
One focus should be having the integrated CRT as display and an external keyboard on serial #2. This will require a mod to the IPC firmware to allow this but as I remember that code not impossible
Jaxk
To view this discussion on the web visit https://groups.google.com/d/msgid/intel-devsys/03fb01d72c4d%2401a56740%2404f035c0%24%40gmail.com.
Jack Mcmullen
Apr 8, 2021, 5:32:34 PM4/8/21
to intel-...@googlegroups.com
Sorry lost context somewhere in the banter
To view this discussion on the web visit https://groups.google.com/d/msgid/intel-devsys/040901d72c4e%24afffd810%240fff8830%24%40gmail.com.
Jack Mcmullen
Apr 8, 2021, 5:53:59 PM4/8/21
to intel-...@googlegroups.com
Yes but I was showing a qick search showing that their available. Functionality to our needs requires a review of whats available versus what it does. I was not saying this is "the solution"
-----Original Message-----
From: Roger Arrick <Ro...@Arrick.com>
From: Roger Arrick <Ro...@Arrick.com>
> send an email to intel-devsys+unsub...@googlegroups.com
> <mailto:intel-devsys%2Bunsu...@googlegroups.com>.
> To view this discussion on the web visit
> https://groups.google.com/d/msgid/intel-devsys/02ab01d72bbe%2475534110%245ff9c330%24%40gmail.com.
>
> --
> You received this message because you are subscribed to the Google
> Groups "intel-devsys" group.
> To unsubscribe from this group and stop receiving emails from it, send
> an email to intel-devsys+unsub...@googlegroups.com
> <mailto:intel-devsys+unsub...@googlegroups.com>.
> <mailto:intel-devsys+unsub...@googlegroups.com>.
> To view this discussion on the web visit
> https://groups.google.com/d/msgid/intel-devsys/CANPXPJkWWeayRYB-REi9EQ%3D%2BRA%3DzHpX7JPkpqsuAwcbDOVKQGw%40mail.gmail.com
> <https://groups.google.com/d/msgid/intel-devsys/CANPXPJkWWeayRYB-REi9EQ%3D%2BRA%3DzHpX7JPkpqsuAwcbDOVKQGw%40mail.gmail.com?utm_medium=email&utm_source=footer>.
> --
> You received this message because you are subscribed to the Google
> Groups "intel-devsys" group.
> To unsubscribe from this group and stop receiving emails from it, send
> an email to intel-devsys+unsub...@googlegroups.com
> <mailto:intel-devsys+unsub...@googlegroups.com>.
> <mailto:intel-devsys+unsub...@googlegroups.com>.
> To view this discussion on the web visit
> https://groups.google.com/d/msgid/intel-devsys/1359339909.627743.1617835942390%40mail.yahoo.com
> <https://groups.google.com/d/msgid/intel-devsys/1359339909.627743.1617835942390%40mail.yahoo.com?utm_medium=email&utm_source=footer>.
> --
> You received this message because you are subscribed to the Google
> Groups "intel-devsys" group.
> To unsubscribe from this group and stop receiving emails from it, send
> an email to intel-devsys+unsub...@googlegroups.com
> <mailto:intel-devsys+unsub...@googlegroups.com>.
> <mailto:intel-devsys+unsub...@googlegroups.com>.
> To view this discussion on the web visit
> https://groups.google.com/d/msgid/intel-devsys/03f601d72c4c%24408a96c0%24c19fc440%24%40gmail.com
> <https://groups.google.com/d/msgid/intel-devsys/03f601d72c4c%24408a96c0%24c19fc440%24%40gmail.com?utm_medium=email&utm_source=footer>.
>
> --
> You received this message because you are subscribed to the Google
> Groups "intel-devsys" group.
> To unsubscribe from this group and stop receiving emails from it, send
> an email to intel-devsys+unsub...@googlegroups.com
> <mailto:intel-devsys+unsub...@googlegroups.com>.
> <mailto:intel-devsys+unsub...@googlegroups.com>.
--
You received this message because you are subscribed to the Google Groups "intel-devsys" group.
To unsubscribe from this group and stop receiving emails from it, send an email to intel-devsys+unsub...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/intel-devsys/606F6FA0.8080707%40Arrick.com.
Paul Birkel
Apr 9, 2021, 2:55:46 AM4/9/21
to intel-...@googlegroups.com
A busy thread yesterday! For my purposes I don’t (objectively) wish to shanghai half of a serial channel in order to interface a keyboard. I intend, if I can get the MDS-230 CRT interface working properly, to attach a substitute keyboard to the existing IOC keyboard interface. To my mind that’s the elegant approach, even if it simply leverages and remaps an existing PS/2-class keyboard.
At the moment only Jon seems to be interested in that approach. I don’t see “a market opportunity” based on the discussion here. I could be wrong. Regardless if/when I prove out a capability I’ll report back.
Jon: Do you have a preference for where the FUNC key would be mapped? Possibly modern-Alt, however it seems to me that modern-Esc would perhaps be a better choice given that its use in CREDIT seems to be to manipulate the screen as-a-whole rather than affect key-behavior as is the case for modern-Alt. Or maybe F1? I have no familiarity with where FUNC affects ISIS software so guidance in this regard would be appreciated.
Eric contributed the key missing information that I was seeking (keyboard encoder ROM and disassembly). Again, thank you Eric; the comments there clear up several points for me. The behavior of TPWR (“Typewriter”, who knew …) as a upper case lock is now quite clear; I can map the IBM-style “CAPS LOCK” to that, then – although it’s been pointed out that TPWR is physically a locking switch, not momentary. I was concerned that the keyboard encoder behavior included a finite state machine convolving history with current scan code(s), but I don’t as yet see one. I can see where the initial disassembly would benefit from better labels and additional comments; I’ll see about working on that.
Anyway, “no plan survives contact with the enemy” so … we’ll see how matters proceed once I inspect the MDS-230 and get it operational using a plain serial terminal. Maybe an M33 ASR if I’m feeling particularly in need of slow self-flagellation!
To unsubscribe from this group and stop receiving emails from it, send an email to intel-devsys...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/intel-devsys/900092157.84655.1617918835257%40mail.yahoo.com.
Herb Johnson
Apr 9, 2021, 11:12:45 AM4/9/21
to intel-...@googlegroups.com, Paul Birkel
On 4/9/2021 2:56 AM, Paul Birkel wrote:
> I intend, if I can get the MDS-230 CRT interface working properly, to
> attach a substitute keyboard to the existing IOC keyboard interface. To
> my mind that’s the elegant approach, even if it simply leverages and
> remaps an existing PS/2-class keyboard.
>
> At the moment only Jon seems to be interested in that approach. I don’t
> see “a market opportunity” based on the discussion here. I could be
> wrong. Regardless if/when I prove out a capability I’ll report back.
Well, you know, Paul: I already argued for your approach. And you
> I intend, if I can get the MDS-230 CRT interface working properly, to
> attach a substitute keyboard to the existing IOC keyboard interface. To
> my mind that’s the elegant approach, even if it simply leverages and
> remaps an existing PS/2-class keyboard.
>
> At the moment only Jon seems to be interested in that approach. I don’t
> see “a market opportunity” based on the discussion here. I could be
> wrong. Regardless if/when I prove out a capability I’ll report back.
already had that in mind, down to a specific $20 microcontroller. I read
the Web page about it, and the documentation. I read the IOC
documentation, looked at the keyboard interface. It. will. do. just. fine.
And: that *was* the market opportunity, that's why it and many many
other microcontrollers like it for keyboards, exist. And: there's
already means of creating keyboards from scratch, if one wishes to do
such things.
I saw no point, in arguing with engineers, why there was no engineering
to be done here. Or why futzing with serial ports was barking up the
wrong tree. Or that questions on how keyboards and software and hardware
works, were already determined and documented, and-heres-the-Web-links.
And Paul: you already responded that you knew about keyswitches and
keycaps, but it wasn't a "fetish" for you.
So I wrote, and then deleted unsent, further posts of that sort.
And about the "fetish" of keyboard design. Paul - calling an interest a
"fetish", does not encourage further discussion. In this case, that
level of design and detail is simply unnecessary, and that's that.
There's no shame in desiring one kind of keyboard over another. Or
having certain purposes for some layout or capability. Keyboards are an
entire industry; I assure you it's a considerable vintage computing
interest. Is that a "fetish"? or just another kind of detail, another
kind of part and feature?
What is necessary, as you point out Paul, is to make the MDS system run
again. That presumably will be in discussion, once you get the parts and
see what can be made of them. Some of that discussion already occurred:
colors of paint, types of toggles and pushbuttons. Getting stuff to work
can be tough; making stuff is I think easier. It's all different kinds
of fun for other people.
I don't wanna spoil people's fun. But repairs set their own priorities.
Things work as they should, or don't. They don't care about one's "fun".
I find that repairs are a means of focusing one's attentions outside of
one's interests and preferences. Vintage computing repairs and parts,
are my daily business and activity. Read my fine Web pages for details.
Attributed to Mark Twain but actually said prior: "Thunder is good;
thunder is impressive. But it's lightning that does the work." Paul,
Keep us posted on the repairs, good luck with them.
regards, Herb Johnson
Jack Mcmullen
Apr 9, 2021, 12:31:02 PM4/9/21
to intel-...@googlegroups.com
I think we've covered about all the options for a keyboard. Your orig idea using a Rasberry or other single chip processor still may be the leader. I guess you'll know when you get there.
To view this discussion on the web visit https://groups.google.com/d/msgid/intel-devsys/065c01d72d0d%2471eaff90%2455c0feb0%24%40gmail.com.
Eric Smith
Apr 9, 2021, 2:00:46 PM4/9/21
to intel-devsys
You say “Presently this project has partially reverse-engineered source code for the unenhanced keyboard 8741A, part number 9100101, in the kbd directory, and for the enhanced keyboard 8741A, part number 104675-001, in the kbd-enhanced directory.” In what sense(s) is it partial and what remains to be accomplished?
It's only partially reverse-engineered because it isn't a good approximation to real "source code". It's just a dissembly with a few comments.
It is, however, verified to cross-assenble ti a bit-identical output.
Paul Birkel
Apr 9, 2021, 2:38:21 PM4/9/21
to intel-...@googlegroups.com
Excellent. Thanks.
From: intel-...@googlegroups.com [mailto:intel-...@googlegroups.com] On Behalf Of Eric Smith
Sent: Friday, April 09, 2021 2:01 PM
To: intel-devsys
Subject: Re: intel-devsys MDS-II Keyboard Matrix Encoder
You say “Presently this project has partially reverse-engineered source code for the unenhanced keyboard 8741A, part number 9100101, in the kbd directory, and for the enhanced keyboard 8741A, part number 104675-001, in the kbd-enhanced directory.” In what sense(s) is it partial and what remains to be accomplished?
It's only partially reverse-engineered because it isn't a good approximation to real "source code". It's just a dissembly with a few comments.
It is, however, verified to cross-assenble to a bit-identical output.
--
You received this message because you are subscribed to the Google Groups "intel-devsys" group.
To unsubscribe from this group and stop receiving emails from it, send an email to intel-devsys...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/intel-devsys/CAFrGgTQXo_id0TWrw%2BoKi12ap5fXt3dbZ4epJJU3NEsYrstk0g%40mail.gmail.com.
forjack842
Apr 9, 2021, 6:07:16 PM4/9/21
to intel-...@googlegroups.com
Sent via the Samsung Galaxy A11, an AT&T 4G LTE smartphone
To view this discussion on the web visit https://groups.google.com/d/msgid/intel-devsys/07b401d72d6f%2498ea2320%24cabe6960%24%40gmail.com.
Roger Arrick
Apr 11, 2021, 12:00:06 PM4/11/21
to intel-...@googlegroups.com
Does anyone here do series III and IV work?
I don't and probably never will.
I have 3 binders with info about system III/IV/PDS.
2 of the binders are blue intel.
1st binder (white)
intel ;comments Dec 86 (no #)
Series III/IV troubleshooting Q3/1986 (122240-009)
2nd binder (blue intel)
Series IV OS and programming 121753-002
Series IV overview 2.5 121752002
Series IV installation/checkout 121757-003
3rd binder (blue intel)
IPDS giant binder full
162606-002 ?
10 sections - CREDIT, DEBUG, PGM REF, DUAL PROCESSOR, PROM PGMING, etc
A-E reference sections
Index
EMV51 emulator
Trying to decide what to do with these.
Will probably ebay them.
Roger in TX
I don't and probably never will.
I have 3 binders with info about system III/IV/PDS.
2 of the binders are blue intel.
1st binder (white)
intel ;comments Dec 86 (no #)
Series III/IV troubleshooting Q3/1986 (122240-009)
2nd binder (blue intel)
Series IV OS and programming 121753-002
Series IV overview 2.5 121752002
Series IV installation/checkout 121757-003
3rd binder (blue intel)
IPDS giant binder full
162606-002 ?
10 sections - CREDIT, DEBUG, PGM REF, DUAL PROCESSOR, PROM PGMING, etc
A-E reference sections
Index
EMV51 emulator
Trying to decide what to do with these.
Will probably ebay them.
Roger in TX
mark.p...@btinternet.com
Apr 11, 2021, 12:58:24 PM4/11/21
to intel-...@googlegroups.com
Roger
I appreciate it may be a challenge, but Is there any chance you could scan some of them before selling so that they can be added to the shared document repository?
For reference, none of these specific versions are currently in the document repository, although some different versions are available as noted below
Series III/IV troubleshooting Q3/1986 (122240-009) - no version in the repository
Series IV OS and programming 121753-002 - earlier version 001 and later version 004 in the repository
Series IV overview 2.5 121752-002 - earlier version 001 in the repository
Series IV installation/checkout 121757-003 - earlier version 002 and later version 005 in the repository
162606-002 - earlier version 001 and later versions 003 & 004 in the repository
If you cannot scan all of them, having the troubleshooting guide and possibly the later overview would be good
Mark
I appreciate it may be a challenge, but Is there any chance you could scan some of them before selling so that they can be added to the shared document repository?
For reference, none of these specific versions are currently in the document repository, although some different versions are available as noted below
Series III/IV troubleshooting Q3/1986 (122240-009) - no version in the repository
Series IV OS and programming 121753-002 - earlier version 001 and later version 004 in the repository
Series IV overview 2.5 121752-002 - earlier version 001 in the repository
Series IV installation/checkout 121757-003 - earlier version 002 and later version 005 in the repository
162606-002 - earlier version 001 and later versions 003 & 004 in the repository
If you cannot scan all of them, having the troubleshooting guide and possibly the later overview would be good
Mark
--
You received this message because you are subscribed to the Google Groups "intel-devsys" group.
To unsubscribe from this group and stop receiving emails from it, send an email to intel-devsys...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/intel-devsys/60731D02.2090600%40Arrick.com.
You received this message because you are subscribed to the Google Groups "intel-devsys" group.
To unsubscribe from this group and stop receiving emails from it, send an email to intel-devsys...@googlegroups.com.
Herb Johnson
Apr 11, 2021, 2:30:08 PM4/11/21
to intel-...@googlegroups.com
Since these are documents, without advertizing, they can be posted "book
rate" and shipped to someone to be scanned, or to be purchased. I've had
no problems, other than delay, in items sent postal "book rate". I think
a likely cost is a few tens of dollars. Keep the package square is my
suggestion, not long.
We all do some eBay buying or selling, so I'd just say in my own
opinion, the eBay value of these documents is in the few tens of
dollars, and may not sell. But anything can happen. The binders may be
of interest to someone. In general: it's a seller's market over the last
year.
Otherwise this is a matter between Roger and whomever wants these for
scanning or ownership, unless third parties care to donate to cover
shipping them around. Just for my making noise on this matter, I'd send
$10 to the appropriate party, for these to be scanned, if they ask me
for it. "money where my mouth is".
regards, Herb
On 4/11/2021 12:58 PM, mark.pm.ogden via intel-devsys wrote:
> Roger
> I appreciate it may be a challenge, but Is there any chance you could scan some of them before selling so that they can be added to the shared document repository?
>
> For reference, none of these specific versions are currently in the document repository, although some different versions are available as noted below
>
rate" and shipped to someone to be scanned, or to be purchased. I've had
no problems, other than delay, in items sent postal "book rate". I think
a likely cost is a few tens of dollars. Keep the package square is my
suggestion, not long.
We all do some eBay buying or selling, so I'd just say in my own
opinion, the eBay value of these documents is in the few tens of
dollars, and may not sell. But anything can happen. The binders may be
of interest to someone. In general: it's a seller's market over the last
year.
Otherwise this is a matter between Roger and whomever wants these for
scanning or ownership, unless third parties care to donate to cover
shipping them around. Just for my making noise on this matter, I'd send
$10 to the appropriate party, for these to be scanned, if they ask me
for it. "money where my mouth is".
regards, Herb
On 4/11/2021 12:58 PM, mark.pm.ogden via intel-devsys wrote:
> Roger
> I appreciate it may be a challenge, but Is there any chance you could scan some of them before selling so that they can be added to the shared document repository?
>
> For reference, none of these specific versions are currently in the document repository, although some different versions are available as noted below
>
> -----Original Message-----
> From: Roger Arrick
> From: Roger Arrick
> Does anyone here do series III and IV work?
> I don't and probably never will.
>
> I have 3 binders with info about system III/IV/PDS.
> 2 of the binders are blue intel.
>
> I don't and probably never will.
>
> I have 3 binders with info about system III/IV/PDS.
> 2 of the binders are blue intel.
>
> Trying to decide what to do with these.
> Will probably ebay them.
>
> Roger in TX
--
> Will probably ebay them.
>
> Roger in TX
--
Jack Mcmullen
Apr 11, 2021, 4:34:56 PM4/11/21
to intel-...@googlegroups.com
I'm interested in the Volume 3-IPDS stuff as I have 2 of there. Or would be willing to chip in $25 for a full scanned in copy in say PDF. Easier to store for me.
Jack
--
You received this message because you are subscribed to the Google Groups "intel-devsys" group.
To unsubscribe from this group and stop receiving emails from it, send an email to intel-devsys+unsub...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/intel-devsys/069e1660-52fb-787d-0fe1-017ec82c20ba%40retrotechnology.com.
Bill Beech (NJ7P)
Apr 11, 2021, 4:42:17 PM4/11/21
to intel-...@googlegroups.com
Roger,
I second Mark's request. I would be glad to scan the documents, and
then return them to you, if you don't want to mess with it. Other
versions of a document can be helpful.
And Herb's comment about book rate is also my findings.
Bill
I second Mark's request. I would be glad to scan the documents, and
then return them to you, if you don't want to mess with it. Other
versions of a document can be helpful.
And Herb's comment about book rate is also my findings.
Bill
Reply all
Reply to author
Forward
0 new messages