Chips & Technologies 82C611 "Micro Channel Interface"

[David L. Beem]

As promised, I want to start bringing up topics covering several chips
from different manufacturers that did a good deal of the work to
interface to the microchannel bus. With the development of at least
one new microchannel adapter being considered we need as much data as
possible. Although I see some of these parts still in surplus
inventory, many of the manufacturers have of course purged them by
now.
Louis did two topics covering a selection of the chips on a
single day back five years ago. There were no (public) replies, which
is disheartening. Hopefully there is more interest now (but harder to
find the parts) for MCA development nowadays on the newsgroup. I will
provide some examples of the adapters that use particular interface
chips & anyone is certainly welcome to respond about an adapter I
don't know of.
Ideally would be datasheets or other information on how the chips
work. The ADFs themselves can provide many clues. I'm leading in with
a chip I only have found on one real adapter, without an ADF (at least
for what I now find): The recent topic of the Seattle Telecom & Data
PSX/386/486 I have posted about.
A search tonight revealed a document by National Semiconductors
for an adapter they apparently registered, but never developed. It
also uses a Chips & Technologies 82C611 (beware of webpages covering
the Opti 82C611, which is a VLB IDE interface chip wholy unrelated to
our needs) & reveals a great deal about how the chip operates. For
those of you interested, the document now resides at
http://www.gilanet.com/David/wcd00a37.pdf
Another publication I am looking for is "CHIPS 82C611, 82C612
Micro CHIPS: Micro Channel Interface Parts (1988)". From what I
remember "C&T" has now gone out of business or been absorbed into
another company (Intel?). As you can see in the title, there is a
small chip line from C&T, with the 82C611 somewhat at the low-end.
The next post will cover the C&T 82C612, which is on at least two
adapters I have found. After that it is on to the 82C614 (but I
haven't found any adapters that use it) & the apparently older C&T
82C574 & 82C575 that I have found on a couple of adapters apiece.
Other manufacturers come later.
Read up on the National Semicondutor document; I'm going to snip
some information from it for successive posts on this thread. The
pinouts of the 82C611 in the NS implementation will be compared to
what I show with a voltmeter on the PSX card. If I don't find an ADF
for the PSX card then I very well may have to try writing one, with
the National Semiconductor paper as my only example of an ADF for the
82C611.

So the C&T 82C611 is used on the following adapters:
STD PSX/386/486, Adapter ID 003Ah, No ADF found
NS Dual UART adapter, Adapter ID 6E6Dh, No adapter manufactured?
Others?

David
Put my name in front of my domain

[John K. Mes]

Post by David L. Beem
STD PSX/386/486, Adapter ID 003Ah, No ADF found
NS Dual UART adapter, Adapter ID 6E6Dh, No adapter manufactured?

Hi David,

So, what you're saying is if there is the expertise & desire, a custom,
new adapter could be built?

But if the examples are a guide, the only types of interface the 82c611
would excel at are data collection and other serial streaming data type
interfaces?

A very cool possibility, to be sure!

-John the Intrigued

[David L. Beem]

Hi John,

Post by John K. Mes
So, what you're saying is if there is the expertise & desire, a
custom, new adapter could be built?

There is the desire & the expertise is being gained. Without the chips
you will hear about over the next week a new microchannel adapter is
possible, but could be a little harder. Let's just say there is an adapter
or two being formulated right now.

Post by John K. Mes
But if the examples are a guide, the only types of interface the
82c611 would excel at are data collection and other serial
streaming data type interfaces?

The 82C611 is definitely a lower level in the C&T line, but still
capable of quite a bit. Consider that it is on the PSX adapter, a full
computer-on-a-card upgrade most likely targeted at the Model 50 & 60,
interfacing a whole 386SX or 486SLC2 CPU subsystem with NCR SCSI to the MCA
bus. In the National Semiconductors design there are still are a few
discrete logic chips & a PAL, besides the 82C611, to interface two serial
ports (NS ignores that there is also a bi-di parallel port on the 16552 too)
to the MCA bus, but contrast that to the IBM Dual Async card, with 22 out of
32 chips just to do the same thing!

Post by John K. Mes
A very cool possibility, to be sure!

Wait till you hear about it's cooler brothers. Also cool is that the
C&T chips seem to be in stock in some locations, which many others we go
over won't have the same benefit. William has already let on to a couple
others, but we will cover them all.
David
Put my name in front of my domain

[William R. Walsh]

Hi!

Post by David L. Beem
As promised, I want to start bringing up topics covering several chips
from different manufacturers that did a good deal of the work to
interface to the microchannel bus. With the development of at least
one new microchannel adapter being considered we need as much data as
possible. Although I see some of these parts still in surplus
inventory, many of the manufacturers have of course purged them by
now.

I believe that Rochester Electronics (I believe Louis made some reference to
them on his pages, but I can't find it now...) carries at least the SMC MCA
interface chip. I don't know if they'd deal with a small order, but it might
be worth finding out.

Post by David L. Beem
Another publication I am looking for is "CHIPS 82C611, 82C612
Micro CHIPS: Micro Channel Interface Parts (1988)". From what I
remember "C&T" has now gone out of business or been absorbed into
another company (Intel?).

Yes, C&T was absorbed by Intel...at least their video graphics line was. I
fear, however, that Intel may have done the same thing to C&T that they did
with Corollary. There doesn't seem to be much information on their products
and I've never gotten anything from Intel for the one Acer notebook I have
here with C&T video.

Post by David L. Beem
Other manufacturers come later.

I found tonight while rummaging around in what's left of my basement several
Madge TR adapters. These seem to use an NCR 86C05A for the MCA bus
interface. I believe the Madge Smart Ringnode series can busmaster, so these
chips might be promising as well. Dunno how you'd find information on them,
but NCR has traditionally been very good to me when I've needed info or help
with one of their products...although you won't find much on their website.

Post by David L. Beem
Read up on the National Semicondutor document; I'm going to snip
some information from it for successive posts on this thread. The
pinouts of the 82C611 in the NS implementation will be compared to
what I show with a voltmeter on the PSX card. If I don't find an ADF
for the PSX card then I very well may have to try writing one, with
the National Semiconductor paper as my only example of an ADF for the
82C611.

I'll give it a looksee, but not tonight. Gotta work tomorrow and it's gonna
be a big bad day.

William

[David L. Beem]

Hi William,

Post by William R. Walsh
I believe that Rochester Electronics (I believe Louis made some
reference to them on his pages, but I can't find it now...) carries
at least the SMC MCA interface chip. I don't know if they'd
deal with a small order, but it might be worth finding out.

The SMC MCA interface chip sits where we want the rest of them: Ryan
located a datasheet that covers the chip exactly as needed. I fear for the
others that time isn't on our side & we will end up reverse engineering ADFs
& existing designs to find out how the chips work & what they do. Even now I
am unsure of the difference in capabilities between the 82C611 & 82C612. The
SMC chip will be covered later (maybe even last, because it has
documentation).

Post by William R. Walsh
Yes, C&T was absorbed by Intel...at least their video graphics line
was. I fear, however, that Intel may have done the same thing to
C&T that they did with Corollary. There doesn't seem to be much
information on their products and I've never gotten anything from
Intel for the one Acer notebook I have here with C&T video.

C&T caught Intel's attention because they were one of the first to
combine Intel peripheral chips into chipsets. Chipsets are big money now &
we are talking about the same with our little interface: Less chips on the
card, not necessarily cost effective when replacing discrete logic, but
maybe saving some headaches. Little hope, but I gave the document name in
case someone knows of it out there.

Post by William R. Walsh
I found tonight while rummaging around in what's left of my basement
several Madge TR adapters. These seem to use an NCR 86C05A
for the MCA bus interface. I believe the Madge Smart Ringnode series
can busmaster, so these chips might be promising as well. Dunno how
you'd find information on them, but NCR has traditionally been very
good to me when I've needed info or help with one of their products...
although you won't find much on their website.

One type of the Madge Ringnode uses the 86C05, but not another.
Interested to hear what the differences in ADFs or adapters are (besides
them both having the red tab)? Stay tuned.
The NCR 86C05 is indeed busmaster-capable, but so is the C&T 82C614.
There is also a lesser NCR MCA interface chip, the 86C01 (and the 86C71,
although I'm not sure where it places & haven't located an adapter using
it), but that will be covered later. All in due time.
Another bit of trivia is that NCR used a competing MCA interface chip,
the Altera EPB2001, on one of their own Token Ring adapters! The Altera
EPB2001 has to be the most heavily used microchannel interface chip out
there, while I can find nothing that uses the other chip of the series, the
EPB2002. Both seem to be out of production now (the EPB2002 went first),
even though Altera is still in business.

Post by William R. Walsh
I'll give it a looksee, but not tonight. Gotta work tomorrow and it's
gonna be a big bad day.

It's a fascinating document. An excellent lead-in for a topic that
seems to be catching more interest than it did for Louis five years ago. He
also found a few others (including some Intel chips) I still need to
research a bit more.
David
Put my name in front of my domain

[David L. Beem]

Hi William,

...Even now I am unsure of the difference in capabilities between
the 82C611 & 82C612...

Thanks to Elar's goldmine of datasheets they have been defined:

"The 82C611 is optimized for memory and I/O interfaces such as those on
multi-function cards. It does not support the DMA arbitration and handshake
signals."

82C612 definition on that thread. Is this too fast of pace or do I need
to slow it down?
David
Put my name in front of my domain

[Elar Saar]

Meanwhile, I've found some relevant datasheets for 82C614, 82C574,
EP2001, EP2002. Also, not really useful two page datasheets for
82C611/612, 82C576. All these files are on my home server.

http://icl486.pointclark.net/~pdf/CT82C614.pdf
http://icl486.pointclark.net/~pdf/CT82C574.pdf
http://icl486.pointclark.net/~pdf/EPB2001.pdf
http://icl486.pointclark.net/~pdf/EPB2002.pdf
http://icl486.pointclark.net/~pdf/CT82C576.pdf
http://icl486.pointclark.net/~pdf/CT82C611.pdf

Elar

[David L. Beem]

Hi Elar,

Post by Elar Saar
Meanwhile, I've found some relevant datasheets for 82C614, 82C574,
EP2001, EP2002. Also, not really useful two page datasheets for
82C611/612, 82C576. All these files are on my home server.
http://icl486.pointclark.net/~pdf/CT82C614.pdf
http://icl486.pointclark.net/~pdf/CT82C574.pdf
http://icl486.pointclark.net/~pdf/EPB2001.pdf
http://icl486.pointclark.net/~pdf/EPB2002.pdf
http://icl486.pointclark.net/~pdf/CT82C576.pdf
http://icl486.pointclark.net/~pdf/CT82C611.pdf

Wow, the motherlode. Thank You, Thank You, Thank You. Do you mind if I
host them from a webserver? There might be some heavy traffic downloading
them from your server otherwise.
David
Put my name in front of my domain

[Elar Saar]

Post by David L. Beem
Wow, the motherlode. Thank You, Thank You, Thank You. Do you mind if I
host them from a webserver?

Not at all.

Post by David L. Beem
There might be some heavy traffic downloading
them from your server otherwise.

There's not much traffic for the first day:

Loading Image...

Elar

[David L. Beem]

Hi Elar,

Post by Elar Saar

Do you mind if I host them from a webserver?

Not at all.

Thank You, although for me in the States the copyright notice on the
82C614 is a little troubling: "You may not reproduce, transmit, transcribe,
store in a retrieval system, or translate into any language or computer
language, in any form or by any means, electronic, mechanical, optical,
chemical, manual, or otherwise, any part of this publication without the
express written permission of Chips and Technologies". So you can't even put
hardcopy in a file cabinet ("retrieval system")? Whoops, the copyright
notice was "part of" the document I just copied here. Nonetheless:

www.gilanet.com/David/datasheets

David
Put my name in front of my domain

[William R. Walsh]

Hi!

I've had a look at that National Semiconductor sheet so far--interesting
reading, but somewhat over my head in spots...I'm still thinking about it.

In the meantime, would you mind if I made a copy of the files on "this end"?

William

[David L. Beem]

Hi William,

Post by William R. Walsh
I've had a look at that National Semiconductor sheet so far--interesting
reading, but somewhat over my head in spots...I'm still thinking about it.

The NSC document is intended to promote use of their (among many other
manufacturers) 16C552 chip. For me it almost seems like a senior project or
college class (although mine never got very deep into hardware) in style.
Just in how they show the address decoding is a very worthwhile study.

Post by William R. Walsh
In the meantime, would you mind if I made a copy of the files on "this end"?

You've seen the copyright notices, so at your own risk. Together they
give a pretty good idea of what most of the C&T chips do, with pinouts. The
Altera line will be covered too, even though those seem to be out of stock &
need a programmer to burn in resource decoding & Adapter ID.
There is a correction on your end I requested via e-mail a couple weeks
back with no response. On your links page
(http://www.walshcomptech.com/ps2/complinks.htm), my domain is spelled
right, but the link itself is not. I've not gone prime time yet, but getting
ready to do so.
David
Put my name in front of my domain

[William R. Walsh]

Hi!

Post by David L. Beem
There is a correction on your end I requested via e-mail a couple weeks
back with no response. On your links page
(http://www.walshcomptech.com/ps2/complinks.htm), my domain is spelled
right, but the link itself is not. I've not gone prime time yet, but getting
ready to do so.

I sent a message with a reply, so I'm not sure what could have happened to
it. In any event, I didn't catch the transposition of characters in the
domain name until just minutes ago, and until I see if I still have my
WS_FTP Pro installer file (or if I have to register their latest version) I
can't do much about the error.

I'll keep the copyright info in mind when and if I decide to post the files
you have gathered. I've been through the mill once with this, but I still
think the people who contacted me were not entirely within their rights to
do so. (I did take the "offending" material down, however.)

William

[deo]

Post by William R. Walsh
Hi!

... until I see if I still have my

Post by William R. Walsh
WS_FTP Pro installer file (or if I have to register their latest version) I
can't do much...

http://filezilla.sourceforge.net/