11 Aug
2000
11 Aug
'00
7:17 a.m.
PALASM allows the user to generate his own test vectors for testing
purposes. I was not aware that the model 29 has a module that generates
these automatically. I'd be surprised if it really does it automatically.
I'd believe, however, that it compiles test vectors as part of the user's
design file. The output from the DOS-based PALASM is a JEDEC file which is
a standardized format for fuse maps for various programmable devices.
The fact that these devices were so costly is what makes them worth so much
less than nothing today. They required several updates per year back when
they were in supported service and, now that they're not, it's virtually
impossible to get software for them. If you're fortunate enough to have the
combination of device, adapter, internal firmware, and externally based
software to make them work they work about as well as anything, but because
of the divergence in the programmable device market since the model 29 was
brought out, it's too difficult to do what they (Data I/O) did because of
the way they did it. Their "next" step was to integrate the functions of
the model 29 and all its various accessories into a single unit produced the
"UniSite" which I only saw once in the hands of a major client and never
with a lesser one. The base unit for the Unisite cost $11k and, of course,
the name, UNISITE, was a misnomer, since it had a wide range of "sites" that
were ultimately needed if you had to support a wide range of devices of
types used after 1990.
That, of course, serves to show how difficult it is to have a single piece
of equipment support all the devices on the market. In the late 1980's
there were numerous ECL devices, including some that included TTL<=>ECL
device, TTL devices, CMOS devices, all in DIP's LCC's, PGA's. etc. It's
a
nightmare keeping up with packaging alone, let alone pin-driver requirements
for different technologies as well.
That's why I favored the lower-cost much more portable devices. I never
learned to like the style of adapters people made for the various packages,
though, since they were always too fragile to haul around in the briefcase
along with the programmer. I built my own, in some cases, and they were
even more fragile. <sigh>
Dick
----- Original Message -----
From: Joe <rigdonj(a)intellistar.net>
To: <classiccmp(a)classiccmp.org>
Sent: Friday, August 11, 2000 7:53 AM
Subject: Re: Palasm? Oops! Correction.
I was tired when I wrote the information below and I
made a serious
goof.
The part that plugs into the model 29 programmer and
that the PAL socket
adapter plugs into is the LogicPac and not the Design Adapter. The DA
(Design Adapter) plugs into the LogicPac in place of the PAL socket
adapter. After you use the DA to convert your logic equations to the JEDEC
fuse mappings, you remove the DA and replace it with the correct socket
adapter for your type of PAL and then command the programmer to burn the
fuse pattern. I don't think the DA is absolutely reguired, I think the
ABEL
software will also compile the logic equations into a
fuse mapping.
However
the DA will do tests on the PALs that the software
alone can't do.
FWIW the LogicPak works with the model 29A, 29B, 19 and 100A programmers
and orginally cost $1795. The LogicPak is REQUIRED in order to
read/program
any PALs.
Joe
At 10:05 PM 8/10/00 -0500, you wrote:
>I've missed part of this thread so I may make some wrong assumptions here
>but I'll throw in what I know about this stuff.
>
> DataI/O has both a software PALASM and a hardware PALASM. I have
both
>of them and the docs for them. They appear to do
the same thing and I
don't
>what the differences are but there probable are
some. I just got this
stuff
>recently and I haven't had time to learn all
the details about it. The
>software versions are called Abel and the was available in PCDOS, UNIX
and
>VAX VMS versions. I have the Abel PC DOS version
($895!) The hardware
>PALASM is called a PALASM Design Adapter and is PN 303-100 ($395). It's a
>unit that plugs into a DataI/O model 29 programmer and the unit with the
>socket for the PAL plugs into it. There are a number is different PAL
>socket units. Some of them only work with a few PALs but the one that I
>have is a DataI/O model 303A-011 and it seems to work with most PALs. I
>also have the manuals for it and the model 29 programmer. I also have a
>DataI/O UniPak EPROM socket adapter and LogicPak adapter and I THINK I
have
>the manuals for them. I also have the manual for
the 303A-002 P/T adapter
>(one of the PAL sockets) but not the hardware. You MUST have the manuals
to
>use the DataI/O stuff since the manuals contain
the codes that you have
to
>enter into the programmer in order to tell it what
kind of device you're
>using. I can make copies of what I have for the cost of copying and
postage
>if anyone needs them.
>
> Joe
>
>
>At 09:11 AM 8/10/00 -0600, Dick wrote:
>>Please see embedded renarks below.
>>
>>Dick
>>----- Original Message -----
>>From: John Honniball <John.Honniball(a)uwe.ac.uk>
>>To: <classiccmp(a)classiccmp.org>
>>Sent: Thursday, August 10, 2000 4:34 AM
>>Subject: Re: Palasm?
>>
>>
>>>
>>> On Wed, 9 Aug 2000 12:51:23 -0600 Richard Erlacher
>>> <richard(a)idcomm.com> wrote:
>>> > A plug in module for PALASM? What would give you that idea?
>>>
>>> Well, as I recall, that's what written on the module :-)
>>>
>>> > I've seen only
>>> > one place on a model 29 for anything to plug in, and that's the
>>programming
>>> > site. (there are several different ones)
>>>
>>> That's right, it's a huge plug-in module to go in the
>>> programming site socket. It has another plug-in module on
>>> top to accept the PALs that are being programmed.
>>>
>>It's been a while but the site does, in some cases have an optional
socket
>>adapter for differing packages. Unfortunately
this type of adapter was
>>needed in some case while other required an etirely different site to
>>accomodate a given package, i.e. you might need several site/adapter
>>combinations to handle all the different packages in which a given part
was
>>available.
>>>
>>> > The data was typically fed to the
>>> > Model 29 via a serial port, and in JEDEC form. The model 29
translated
>>the
>>> > jedec file into whatever signals were needed to program the device.
>>>
>>> As I have no manual, I can't tell how this thing was
>>> supposed to work.
>>>
>>There was software, in my experience, running on a PC, that told the
>>programmer what to do, specifically how to drive the device pins and how
to
>>process the JEDEC file. It also sent the
JEDEC file once that was set
up,
>>but these were separate steps. At the time I
seriously began using
>>field-programmable logic devices (once price came down) there were
>>sufficiently inexpensive programmers from second-rank independent
vendors
>>that the model 29 was unrealistic for someone
not programming in a
prduction
>>environment and whose needs involved a wide
range of devices.
>>
>>It's not unusual to see these model 29's at distributors' sites.
Perhaps
>>one of your local electronic distributors'
technicians can help you out.
>>The way to do that is to contact someone you know and tell them what you
>>want is to find whatever documentation there is on the model 29 and
>>examine/copy it. The guy at the will-call counter will probably be more
>>helpful than the phone-sales person.
>>
>>> > The last time I used a model 29 was in '85. They were already
pretty
old
> back then and wouldn't program the then
new Altera parts and Lattice
parts I
> was using. That was later fixed, but too
late for me. I bought a
cheapie
> that still works . . .
I've since acquired a Stag PPZ Universal Programmer.
6809-based machine with built-in CRT.
> Is there a feature that plugs in the intelligence to compile Palasm
files
> into the JEDEC? I'd be VERY surprised
to learn that.
I'll have to pop up into the loft and fetch this thing down
for a closer look (info in this e-mail is from memory).
I'll post model numbers, etc, in case anybody can identify
the module and tell me I've got it all wrong :-)
--
John Honniball
Email: John.Honniball(a)uwe.ac.uk
University of the West of England