8 Oct
2005
8 Oct
'05
9:50 a.m.
a scholar and a gentleman. I even considered building
my own, though not sure if MM made the artwork
available. Was put off by the fact that the board has
7 layers according to the article. If anyone is
knowledgeable enough of pcb production - is it
possible to make a 7 layer board w/o special
equipment/materials? Why did it even need 7 layers?
How many layers were did the IBM PC mobo have? I know
for a fact that the RE robot puter (80188) mobo had 2
sides (top and bottom, no sandwich).
--- cctech-bounces at classiccmp.org
<chenmel at earthlink.net> wrote:
welcome to the idea
On Fri, 7 Oct 2005 07:29:40 -0700 (PDT)
Chris M <chrism3667 at yahoo.com> wrote:
> dont have one, but want one. I hope youll make
disk
images
available.
I certainly will. A list member has extended a warm of me imaging my disks and putting them on his site.
More to follow
when they're all imaged and online.
> --- cctech-bounces at classiccmp.org
> <chenmel at earthlink.net> wrote:
> > Does anybody else on the list have an MPX-16
system?
> I recently
> > acquired one from a list member and am wondering
how
> many of these
> > systems still exist. I have a large
(relatively)
> collection of
> > diskettes with mine.
> >
> > The MPX-16, for those not familiar with it, was
> published as a project
> > in Steve Ciarcia's 'Circuit Cellar' column in
Byte
> magazine. It was a
> > three part 'construction' article and the
machine
> was sold by MicroMint
> > for a time. What I've heard is that about 500
> machines in total were
> > produced.
> >
> > It's an early 'IBM Compatible' in that they
designed
> it to be similar to
> > the PC, but only to a certain degree. It uses a
> serial console rather
> > than keyboard/display adapter, and it runs
CP/M-86
> and supposedly MS-DOS
> > though I don't have DOS diskettes for mine. It
has
> ISA slots and a
> > similar architecture to the IBM-PC, coming out
of
> that early era before
> > there were PC clones from the likes of Compaq.
> >
> > I'm curious of how many other MPX-16 systems
have
> survived to today.
> > There isn't a lot about it online. I can share
what
> information I have,
> > as I have manuals and docs with my system. I'm
> interested in hearing
> > from other people with this machine.
> >
> > Scott
>
>
>
>
> __________________________________
> Yahoo! Mail - PC Magazine Editors' Choice 2005
> http://mail.yahoo.com
__________________________________
Yahoo! Music Unlimited
Access over 1 million songs. Try it free.
http://music.yahoo.com/unlimited/
9 Oct
9 Oct
8:25 a.m.
a scholar and a gentleman. I even considered building
my own, though not sure if MM made the artwork
available. Was put off by the fact that the board has
7 layers according to the article. If anyone is
knowledgeable enough of pcb production - is it
possible to make a 7 layer board w/o special
I've never heard of anybody making more than double-sided PCBs at home (I
would love to be proved wrong!). But just about any PCB manufacturing
company could do it.
equipment/materials? Why did it even need 7 layers?
The advantages of more layes are :
You can route more tracks in the same area. A trivial example of this is
that you can normally only route one track between 2 adjacent IC pins of
a DIL pacakge on a particular layer. More layers -> more traces can be
fitted in.
More importantly you can have particular layes for power and ground
planes. That is, one layer will be almost all copper, with 'holes' round
pins you don't want to be connected to ground, and connected to those
pins you do wany grounded. Doing this will have a much lower impedance
(it's the inducatance that really matters, not the resistance, BTW) than
normal traces. You can also shield traced inside the board between ground
planes. And you can design traces as striplines over the ground plane
(and thus know their characteristic impedance) for high frequency work.
I doubt any 8088-based design needs you to go that far, but the dedicated
power and ground planes would probably be almost essential.
How many layers were did the IBM PC mobo have? I know
At a guess, it was a 6 layer board. Certainly 4 layers (signal traces on
the outside, intenral power and ground planes), I think there are hidden
signal traces too.
-tony
11 Oct
11 Oct
5:29 p.m.
I've never heard of anybody making more than
double-sided PCBs at
home (I would love to be proved wrong!).
I can't *prove* you wrong. But I've spoken with someone who claimed to
have done it, and when I started asking questions, talked a good enough
line that I found it believable. As she described it, you need
epoxy/fibreglass mix, which you cast a thin sheet of. You then sand it
smooth, copper-plate it, and etch. Slap on another coat of epoxy and
fibreglass, let it cure, sand it smooth, copper-plate, and etch.
Lather, rinse, repeat.
The hard part is of course quality control (and plated-through vias,
which you can get somewhere with by drilling holes before doing the
last copper-plating). And registration of the layers. But with
patience and attention to detail...after all, anyone doing this is
considering time to be worth a great deal less than money, or the job
would simply be shipped off to a commercial pcb fab house. :)
/~\ The ASCII der Mouse
\ / Ribbon Campaign
X Against HTML mouse at rodents.montreal.qc.ca
/ \ Email! 7D C8 61 52 5D E7 2D 39 4E F1 31 3E E8 B3 27 4B
12 Oct
12 Oct
11:25 a.m.
der Mouse wrote:
It's the sort of thing I can see our Mr Duell doing though.
Gordon.
I've never
heard of anybody making more than double-sided PCBs at
home (I would love to be proved wrong!).
I can't *prove* you wrong. But I've spoken with someone who claimed to
have done it, and when I started asking questions, talked a good enough
line that I found it believable. As she described it, you need
epoxy/fibreglass mix, which you cast a thin sheet of. You then sand it
smooth, copper-plate it, and etch. Slap on another coat of epoxy and
fibreglass, let it cure, sand it smooth, copper-plate, and etch.
Lather, rinse, repeat.
The hard part is of course quality control (and plated-through vias,
which you can get somewhere with by drilling holes before doing the
last copper-plating). And registration of the layers. But with
patience and attention to detail...after all, anyone doing this is
considering time to be worth a great deal less than money, or the job
would simply be shipped off to a commercial pcb fab house. :)
12:25 p.m.
der Mouse wrote:
It's the sort of thing I can see our Mr Duell doing though.
s/Mr/Dr/
:-)
You do realise that I wrote the >> bit above. I have never tried making
more than double-sided PCBs by hand. And for 1-off prototypes, where
typically I want to try something, get it working, then move on to the
next bit, I don't want a PCB anyway. It's possible to hand-wire stuff
well up into the low 100's mf MHz range if you are careful. The main
trick is to build everythign on a coppr ground plane (such as a pice of
copper-clad board, unetched), stick down little bits of stripboard,
strip side up, as component mounts, connect that which needs to be
grounded to the ground plane with the shortest possible wires, and fit
decoupling caps in the same way.
-tony
I've
never heard of anybody making more than double-sided PCBs at
home (I would love to be proved wrong!).
I can't *prove* you wrong. But I've spoken with someone who claimed to
have done it, and when I started asking questions, talked a good enough
line that I found it believable. As she described it, you need
epoxy/fibreglass mix, which you cast a thin sheet of. You then sand it
smooth, copper-plate it, and etch. Slap on another coat of epoxy and
fibreglass, let it cure, sand it smooth, copper-plate, and etch.
Lather, rinse, repeat.
The hard part is of course quality control (and plated-through vias,
which you can get somewhere with by drilling holes before doing the
last copper-plating). And registration of the layers. But with
patience and attention to detail...after all, anyone doing this is
considering time to be worth a great deal less than money, or the job
would simply be shipped off to a commercial pcb fab house. :)
14 Oct
14 Oct
8:17 a.m.
Tony Duell wrote:
s/Mr/Dr/
Oh, sorry
:-)
You do realise that I wrote the >> bit above. I have never tried making
No, I missed that bit.
more than double-sided PCBs by hand. And for 1-off
prototypes, where
typically I want to try something, get it working, then move on to the
next bit, I don't want a PCB anyway. It's possible to hand-wire stuff
well up into the low 100's mf MHz range if you are careful. The main
I bird's-nest pretty much all my experimental RF stuff, either on scrap
unetched circuit board or thin sheets of tinned steel.
Gordon.
9 Oct
9 Oct
10:46 a.m.
On Sat, 8 Oct 2005 06:50:21 -0700 (PDT)
Chris M <chrism3667 at yahoo.com> wrote:
a scholar and a gentleman. I even considered building
my own, though not sure if MM made the artwork
available. Was put off by the fact that the board has
7 layers according to the article. If anyone is
knowledgeable enough of pcb production - is it
possible to make a 7 layer board w/o special
equipment/materials? Why did it even need 7 layers?
How many layers were did the IBM PC mobo have? I know
for a fact that the RE robot puter (80188) mobo had 2
sides (top and bottom, no sandwich).
--- cctech-bounces at classiccmp.org
I remember wanting to build one, too. Now I have one.
I wonder if there would be enough interest and/or resources to produce a
new batch of the boards for it? It would be an expensive undertaking.
For one-off purposes, I suspect it would be cheaper and less work to
just make a wirewrap version. It wouldn't be that hard to do a good job
at that, using the existing layout from the PC board version as a
starting point, and laying in a good grounding/power infrastructure
before wire-wrapping the data paths. There aren't a lot of 'unobtanium'
chips in the MPX-16 design, and it is a nearly completely 'published'
work, the schematics spread over three Circuit Cellar columns in Byte,
also reprinted in Volume 4 of'Ciarcia's Circuit Cellar' published by
McGraw-Hill.
The MPX-16 fills an interesting gap in the history of hobbyist
computing. In a sense, it was a 'stab' at keeping the personal computer
a hobbyist enterprise. By publishing a 'kit' version with hobbyist
tweaks and differences (the MPX-16 has dip-switch settings to select 8"
floppies, for example) Ciarcia was trying to 'keep it real' for the
people accustomed to 'writing code with solder' as he once put it.
Another list member described it as 'the Circuit Cellar segment that
marked the end of
BYTE as a non PC hardware/software magazine.' (woodelf) Looking at all
the cool hardware projects that used to get published in Byte, I have to
agree. MicroMint didn't cease to exist, of course, and Ciarcia's
magazine Circuit Cellar INK has had a long run of truly great projects
and articles over the years.
Some news about my machine and the particular collection of diskettes
that came with it. There are many different copies of CP/M, it appears
whomever these disks belonged to either was involved in the porting to
the MPX-16 or just tracked it very closely. I brought the machine up
yesterday for the first time and it works great. Dual TM-100 disk
drives in the 'official' case makes it a most complete system. It boots
using the serial console as I haven't yet tried it with a video card. I
might never 'bring it to that point' as that just transforms it into a
kinda-PC/XT system.
I booted CP/M-86 on it successfully from the disk set marked 'Version
2.1.' There is .A86 source code for most of the CP/M on the three
diskette set. Not a lot of third party apps in the 69 diskettes in the
collection, but then some of the disks are marked 'BIOS source code' and
other low level system things. Seems the original owner was definitely
a Systems Programmer.
Then I thought I would try DOS so I took my stock (actually an original)
IBM PC-DOS 1.0 diskette and it booted perfectly. It boots much faster
than an IBM-PC, probably because I didn't observe it doing any kind of a
memory check. I duplicated the PC-DOS 1.0 diskette (shouldn't have even
_needed_ to use the original in this day and age) and then also booted
PC-DOS 1.1. 1.1 boots fine, too. Later versions wouldn't work,
obviously because they require double sided diskettes. (back in the day
I upgraded my father's PC-1 system to double sided diskettes when he
wanted to upgrade to DOS 2.1)
The biggest limitation in running DOS on the machine is that any app
that writes directly to video I/O or does any Video functionality at all
won't work. But it's mostly born to be a CP/M-86 system in any event.
There aren't a lot of apps on the large collection of disks. There is
some sort of a word processor, but when I brought it up it was obviously
designed for a particular terminal as it flooded the screen with a mess
and I couldn't get out of it at all without rebooting the machine.
There are some other small apps on one diskette with source code,
including a term emulator and what appears to be a low-level Diskette
editing utility.
All in all, the 69 diskette collection is a real jumble of things.
There are multiple copies of most of everything, including MicroMint
originals and 'working' copies. Several of the MicroMint originals have
'master' written where there should be a serial number penned in by hand
on the label. One diskette says 'CP/M-86 for the Micromint MPX-16
2/17/83' all written out in longhand like it might be the very first
release and/or even the original master. It's an exciting collection of
diskettes and after some organizing the images will be shared.
There are also a few 'PC' related diskettes. A preliminary demo version
(apparently bundled with PC World Magazine) of Microsoft Word on an
official Microsoft diskette. It works, with a 'demo' splash screen, on
a modern PC. There is a disk marked 'CP/M-86 1.0 for the IBM-PC' that
also boots properly on a modern PC (one of my ubiquetous Dell Optiplex
Pentium 3 machines- the one now with a 5-1/4" disk) There is a diskette
marked PC-DOS that boots to a diagnostic menu and won't proceed as I
don't have the 'diagnostic disk' to insert.
Well, enough about the MPX-16 and it's bundle of diskettes.
<chenmel at earthlink.net> wrote:
welcome to the idea
On Fri, 7 Oct 2005 07:29:40 -0700 (PDT)
Chris M <chrism3667 at yahoo.com> wrote:
> dont have one, but want one. I hope youll make
disk
images
available.
I certainly will. A list member has extended a warm of me imaging my disks and putting them on his
site.
More to follow
when they're all imaged and online.
> --- cctech-bounces at classiccmp.org
> <chenmel at earthlink.net> wrote:
> > Does anybody else on the list have an MPX-16
system?
> I recently
> > acquired one from a list member and am wondering
how
> many of these
> > systems still exist. I have a large
(relatively)
> collection of
> > diskettes with mine.
> >
> > The MPX-16, for those not familiar with it, was
> published as a project
> > in Steve Ciarcia's 'Circuit Cellar' column in
Byte
> magazine. It was a
> > three part 'construction' article and the
machine
> was sold by MicroMint
> > for a time. What I've heard is that about 500
> machines in total were
> > produced.
> >
> > It's an early 'IBM Compatible' in that they
designed
> it to be similar to
> > the PC, but only to a certain degree. It uses a
> serial console rather
> > than keyboard/display adapter, and it runs
CP/M-86
> and supposedly MS-DOS
> > though I don't have DOS diskettes for mine. It
has
> ISA slots and a
> > similar architecture to the IBM-PC, coming out
of
> that early era before
> > there were PC clones from the likes of Compaq.
> >
> > I'm curious of how many other MPX-16 systems
have
> survived to today.
> > There isn't a lot about it online. I can share
what
> information I have,
> > as I have manuals and docs with my system. I'm
> interested in hearing
> > from other people with this machine.
> >
> > Scott
>
>
>
>
> __________________________________
> Yahoo! Mail - PC Magazine Editors' Choice 2005
> http://mail.yahoo.com
__________________________________
Yahoo! Music Unlimited
Access over 1 million songs. Try it free.
http://music.yahoo.com/unlimited/
10 Oct
10 Oct
3:57 p.m.
I wonder if there would be enough interest and/or
resources to produce a
new batch of the boards for it?
At one company I worked for years ago, we made small
circuit boards for some thingamajig. If the artwork
was available (and MM is still around, I doubt they
refrain from releasing it provided they still had it)
I'd be inclined to give it a crack my self, save for
the fact that it has 7 layers. The Radio Electronics
robot board which appeared in about '88 would be an
easier project imo. Only 2 layers, and they put the
artwork right in the magazine. I imagine I'm going to
attempt it this winter. I e-mailed the company who
made the prototype and distributed it, and they
e-mailed me the FORTH interpreter rom image, but
unfortunately that's all they had. IIRC there was 3,
BIOS, BASIC, and FORTH.
It boots
using the serial console as I haven't yet tried it
with a video card. I
might never 'bring it to that point' as that just
transforms it into a
kinda-PC/XT system.
For what it's worth, it's exactly what I'd do if I
had one ;). I wouldn't be limited to that. I remember
he stated in the article that "in theory" a video card
should work. I recommend a PGA card - make that cards
- if you happen to have one (them). I don't, but I
have something similar, a dual card w/an onboard 80186
made by Vermont Microsystems. Anyone got the drivers
for it?
I booted CP/M-86 on it...the disk... .A86 source
code...CP/M on the three diskette set...third party
apps...69 diskettes...some of the disks...'BIOS
source >code'...low level
system things...
Dude, pleeease make those disks into images before
catastrophe befalls them :(
__________________________________
Yahoo! Music Unlimited
Access over 1 million songs. Try it free.
http://music.yahoo.com/unlimited/
6862
days inactive
6868
days old
7 comments
5 participants
participants (5)
-
ard＠p850ug1.demon.co.uk -
chenmel＠earthlink.net -
chrism3667＠yahoo.com -
gordonjcp＠gjcp.net -
mouse＠Rodents.Montreal.QC.CA