Appletalk on a IIgs with phoneNet connection to Win2K?
Tim Haynes
I have a IIgs that is currently using Appletalk over a Localtalk connection
(actually phoneNet connectors that I scrounged up). One of my phoneNet
connectors is kinda weird - it has a 9pin serial connection (perfect for a
Windows COM port) that is plugged into an adapter to bring it back to the
Mac printer port (geez, I'm out of the loop! What are those connectors
called again? DIN? DB9? Grrr... memory fading...).
Anyway, I got to thinking - Win2K Server has Appletalk services that you can
run. I'm wondering if I can plug this phoneNet cord into COM1 on my Win2K
machine and transfer files to my IIgs that way. Anybody tried such a trick?
I suspect I will need a driver of some kind to recognize the phoneNet
connector on Win2K, but can find no such beast (for free) on the 'net.
Anybody have any ideas or comments?
Cheers,
Tim
John Oyler
The serial port on a windows box is rs232... incompatible with localtalk
or rs422. It might be bad to try this, wouldn't want to fry anything.
The db9 phonenet is for the orginal mac, I believe, which didn't have the
miniDIN 8 plug, like the later macs do. Also, if you have a Localtalk PC
card for the windows box, it would indeed use this, as opposed to the mini8.
Quite possibly a few older printers of various makes with localtalk options
may have used this, I'm unsure. Now, if only I knew what that hdb15 phonenet
adapter was, that I found last year...
John
Gary Gouriluk
network.
Gary
Byron Desnoyers Winmill
Wintel machines. There are differences in signaling, which are important for
AppleTalk networking. That said, your PhoneNet adapter is probably for the
early Macintosh computers (which used DB9). It is also worth noting that many
localtalk drivers (at least in the Unix world) appear ethernet only.
Byron.
Tim Haynes (trha...@yahoo.SPAMBGONE.com) wrote:
: Hey gang,
william strutts
printer port and the modem port. Up to the Mac Plus,
they used a non-ADB keyboard. The Mac SE introduced
the ADB keyboard and that is what you might be thinking
of.
--
--
William R. Strutts - wrstr...@home.com - Whatever!
C'est moi! http://www.facelink.com/wrstrutts
Just hacking away...
"John Oyler" <jo...@mediaone.net> wrote in message
news:_bR07.6608$Lc3.1...@typhoon.jacksonville.mediaone.net...
william strutts
mouse port. It was pre-ADB and didn't employ the
ADB keyboard until the Mac SE which was deployed
after the IIgs introduced the newer style ports. I used
to have a Mac Plus and it had the mini-8 connectors.
--
--
William R. Strutts - wrstr...@home.com - Whatever!
C'est moi! http://www.facelink.com/wrstrutts
Just hacking away...
"Byron Desnoyers Winmill" <wbde...@acs5.acs.ucalgary.ca> wrote in message
news:9i0nnv$3a4$1...@nserve1.acs.ucalgary.ca...
Marsha
>Hey gang,
>
>Anyway, I got to thinking - Win2K Server has Appletalk services that you can
>run. I'm wondering if I can plug this phoneNet cord into COM1 on my Win2K
>machine and transfer files to my IIgs that way
As far as I know the Win NT/2K only supports Appletalk over ethernet or via a LocalTalk card for the PC.
-----= Posted via Newsfeeds.Com, Uncensored Usenet News =-----
http://www.newsfeeds.com - The #1 Newsgroup Service in the World!
-----== Over 80,000 Newsgroups - 16 Different Servers! =-----
Roger Johnstone
strutts" <wrstr...@home.com> wrote:
> The only DB-9 connector on the original Mac was the
> mouse port. It was pre-ADB and didn't employ the
> ADB keyboard until the Mac SE which was deployed
> after the IIgs introduced the newer style ports. I used
> to have a Mac Plus and it had the mini-8 connectors.
>
> --
> William R. Strutts - wrstr...@home.com - Whatever!
>
> C'est moi! http://www.facelink.com/wrstrutts
The original Mac and the Mac 512K both used DE-9 sockets for the serial
ports. They were not wired the same as the DE-9 serial sockets on a PC. The
mouse port on the Mac 128K, 512K and Plus also used a DE-9 connector. As you
said, the mini DIN-8 serial connectors were introduced with the Mac Plus and
the mini DIN-4 ADB with the IIgs and later the Mac SE and II.
--
Roger Johnstone, Invercargill, New Zealand
Apple II - Future Cop:LAPD - Warcraft II
http://homepage.mac.com/rojaws
______________________________________________________________________
"The Weekenders" by Doug Langdale
Lor: You see, you can justify anything with a little effort.
Phil Beesley
wrote:
> "Tim Haynes" <trha...@yahoo.SPAMBGONE.com> wrote:
> >Hey gang,
> >
> >Anyway, I got to thinking - Win2K Server has Appletalk services that you can
> >run. I'm wondering if I can plug this phoneNet cord into COM1 on my Win2K
> >machine and transfer files to my IIgs that way
>
> As far as I know the Win NT/2K only supports Appletalk over ethernet or via a
> LocalTalk card for the PC.
Correct. With Win2K, the situation is even worse because the only
LocalTalk card on Microsoft's Hardware Compatibility List is a
COPS/Daystar LT200 which is less readily available than other models.
Generally ISA adapters are deprecated for use with W2K server so the
easiest way to go is to use some sort of ethernet/LocalTalk bridge.
I have been experimenting with a IIgs and various other LocalTalk-only
systems connected to W2K server via an SE/30 running LocalTalk Bridge
2.1 software. Systems tested so far:
Mac Plus, system 6.0.4 & 6.0.8 -- connects to server OK as user and
guest;
Mac 512, workstation 1.1 client -- server visible in Chooser but an
"incompatible AppleShare version" dialog is displayed on login;
IIgs, system 6.0.1 -- server is not visible in AppleShare control panel.
I hope that this is a configuration error on my part and I'll check
further at the weekend. In the meantime, can anyone tell me which
AppleShare versions are supported by the IIgs system.
Phil Beesley
David Empson
> The original Mac used the mini-8 connectors for the
> printer port and the modem port.
Nope. The original Mac 128K through to the Mac 512KE used a DE-9 female
connector for the printer and modem ports (as did the original
LaserWriter). There is even a diagram of it in Inside Mac, Volume 3
(which I have right here). The pinout is quite different to a PC.
The Mac Plus was the first Mac model to use the Mini-Din-8 connector for
the serial ports. (It is documented in Inside Mac, Volume 4.)
I don't know what the Lisa used for its serial ports, but it wouldn't
surprise me if it was the same DE-9 female.
Byron Desnoyers Winmill
: The only DB-9 connector on the original Mac was the
: mouse port. It was pre-ADB and didn't employ the
: ADB keyboard until the Mac SE which was deployed
: after the IIgs introduced the newer style ports. I used
: to have a Mac Plus and it had the mini-8 connectors.
Nope. The original Mac used DB9 for the mouse, printer and modem ports.
AFAIK, Macs used RS-422 since day one.
Byron.
Matthew Russotto
David Empson <dem...@actrix.gen.nz> wrote:
>
>I don't know what the Lisa used for its serial ports, but it wouldn't
>surprise me if it was the same DE-9 female.
The Lisa used DB25 (female, IIRC). Remember it was a much larger
machine. One port was RS-232, the other RS-422.
--
Matthew T. Russotto russ...@pond.com
"Extremism in defense of liberty is no vice, and moderation in pursuit
of justice is no virtue."
Rev. Roy Miller
> In article <1ew3gjj.imykd1fehnc2N%dem...@actrix.gen.nz>,
> David Empson <dem...@actrix.gen.nz> wrote:
> >
> >I don't know what the Lisa used for its serial ports, but it wouldn't
> >surprise me if it was the same DE-9 female.
>
> The Lisa used DB25 (female, IIRC). Remember it was a much larger
> machine. One port was RS-232, the other RS-422.
So, could a Lisa, with the right driver, use the RS-422 port as an
AppleTalk port??
I'd love to have my Lisa AppleTalked to my Mac/II network.
Roy
Matthew Russotto
Rev. Roy Miller <mil...@inetnebr.com> wrote:
>Matthew Russotto wrote:
>
>> In article <1ew3gjj.imykd1fehnc2N%dem...@actrix.gen.nz>,
>> David Empson <dem...@actrix.gen.nz> wrote:
>> >
>> >I don't know what the Lisa used for its serial ports, but it wouldn't
>> >surprise me if it was the same DE-9 female.
>>
>> The Lisa used DB25 (female, IIRC). Remember it was a much larger
>> machine. One port was RS-232, the other RS-422.
>
>So, could a Lisa, with the right driver, use the RS-422 port as an
>AppleTalk port??
Certainly, as it did so under MacWorks. I don't recall if there was
ever a way to use it under Lisa 7/7. If you know how to write a
driver for Lisa 7/7..
(note this is the Lisa 2. The late and unlamented Lisa 1 with the
Twiggy drives is another matter)
>I'd love to have my Lisa AppleTalked to my Mac/II network.
william strutts
I had assumed, wrongly, that the Mac Plus only introduced
the larger RAM and the SCSI adapter.
--
--
William R. Strutts - wrstr...@home.com - Whatever!
C'est moi! http://www.facelink.com/wrstrutts
Just hacking away...
"Roger Johnstone" <roj...@mac.com> wrote in message
news:9i18v4$hn5$1...@lust.ihug.co.nz...
william strutts
Plus and not a Fat Mac so I don't recall. I should have
qualified that better. I had wrongly assumed that the Mac
Plus only introduced the SIMM memory and the SCSI
port.
--
--
William R. Strutts - wrstr...@home.com - Whatever!
C'est moi! http://www.facelink.com/wrstrutts
Just hacking away...
"David Empson" <dem...@actrix.gen.nz> wrote in message
news:1ew3gjj.imykd1fehnc2N%dem...@actrix.gen.nz...
John Oyler
> AppleTalk port??
Roy,
I am almost certain the Lisa could do appletalk. Butchering a phonenet
dongle might be necessary, but would be trivial. The lisa is 68k with what?
1 meg of ram? That's more than the first macs, so it looks like it's
powerful enough to do it. Question is, is there enough info floating around
about the lisa, and the proper development tools to compile the code? Would
be fun though, can't wait til I have a Lisa.
Other computers that are apparently localtalk capable include...
Various SGI boxen, which have rs422 in either db25 or de9...
Atari Falcon, and other atari's with LAN ports... (no drivers or appletalk
stack available, though there are works in progress)
The PC (obvious to most of us, I think, requires one of several Ltalk cards
for the isa or pci bus)
Apple II (the GS has it built in, the IIe requires a workstation card)
Amiga 2/3/4000 (requires the emplant card, seemingly only works in
emulation, another card, the Doubletalk, apparently can do file/print
sharing from workbench)
Other than the obvious, I can only confirm that the SGI can actually do it,
using a particular version of netatalk. I found instructions on the web
awhile back. Also, finally got my localtalk pc card working a few weeks ago,
though it's not quite all there yet. Just wish it didn't conflict with my
arcnet card.
*off-topic* On another note, would you happen to know where a person could
find the pinouts for the Lisa expansion slots?
Thanks,
John
> I'd love to have my Lisa AppleTalked to my Mac/II network.
Me too. But first things first... I must find a Lisa!!! ;-)
> Roy
Rev. Roy Miller
> > So, could a Lisa, with the right driver, use the RS-422 port as an
> > AppleTalk port??
>
> Roy,
> I am almost certain the Lisa could do appletalk. Butchering a phonenet
> dongle might be necessary, but would be trivial. The lisa is 68k with what?
> 1 meg of ram? That's more than the first macs, so it looks like it's
> powerful enough to do it. Question is, is there enough info floating around
> about the lisa, and the proper development tools to compile the code? Would
> be fun though, can't wait til I have a Lisa.
Lisa was developed, as was the later Mac, using Pascal (and the Pascal system
Assembler.) It's a 68000 at 5 mhz, my Lisa has one meg, though a friend might be
able to help me get it up to it's limit of 2 megs of RAM (I never understood
limiting a 24 bit address bus chip to 1/4 or less of it's address space....)
>
>
> Other computers that are apparently localtalk capable include...
>
> Various SGI boxen, which have rs422 in either db25 or de9...
> Atari Falcon, and other atari's with LAN ports... (no drivers or appletalk
> stack available, though there are works in progress)
> The PC (obvious to most of us, I think, requires one of several Ltalk cards
> for the isa or pci bus)
I have an ISA card and software - but it's limited to a 25 mhz machine... I'm
planning to stick it in an old IBM PS/1 486/25 and run PC-DOS on it.
>
> Apple II (the GS has it built in, the IIe requires a workstation card)
> Amiga 2/3/4000 (requires the emplant card, seemingly only works in
> emulation, another card, the Doubletalk, apparently can do file/print
> sharing from workbench)
>
> Other than the obvious, I can only confirm that the SGI can actually do it,
> using a particular version of netatalk. I found instructions on the web
> awhile back. Also, finally got my localtalk pc card working a few weeks ago,
> though it's not quite all there yet. Just wish it didn't conflict with my
> arcnet card.
>
> *off-topic* On another note, would you happen to know where a person could
> find the pinouts for the Lisa expansion slots?
On eBay there is usually someone trying to sell full schematics for the Lisa,
would that cover it?
Roy
Michael J. Mahon
>I have an ISA card and software - but it's limited to a 25 mhz machine... I'm
>planning to stick it in an old IBM PS/1 486/25 and run PC-DOS on it.
This seems a curious requirement, since ISA bus speed is independent
of system clock. Is this requirement the result of the driver software
using some kind of timing loop, or what? Could it be patched for faster
processors?
-michael
Email: mjm...@aol.com
Home page: http://members.aol.com/MJMahon/
Roy Miller
> Roy Miller wrote (regarding AppleTalk on a PC):
>
> >I have an ISA card and software - but it's limited to a 25 mhz machine... I'm
> >planning to stick it in an old IBM PS/1 486/25 and run PC-DOS on it.
>
> This seems a curious requirement, since ISA bus speed is independent
> of system clock. Is this requirement the result of the driver software
> using some kind of timing loop, or what? Could it be patched for faster
> processors?
I don't know Michael. The package is in my garage, and I'm about to head for bed,
but I'll check. What I do remember is the manual saying that when used with 386
machines, it might now work at speeds greater than 25 mhz. I'm assuming that it's
a bus speed and not the cpu, so that a 486 (which, evidently, didn't exist when
the card and software were created) shouldn't be a problem. I'll post the results
here.
Roy
Tim Haynes
there were so many subtle technical issues involved, or that so many people
would understand them still!
I guess I'll have to live with my Appletalk to SE/30 connection for my
networking needs.
Thanks again!
Tim
Marsha
Now that I've used a different newserver I can see your first post I have the following suggestion...
Get a localtalk to ethernet bridge like a GatorBox, that should allow the GS to see an Appleshared volume on your W2K server.
( I've only played with this setup under LINUX )
Marsha
Marsha
Now that I've used a different newserver I can see your first post I have the following suggestion...
Mister Boffo
<trha...@yahoo.SPAMBGONE.com> wrote:
>Hey gang,
>
>I have a IIgs that is currently using Appletalk over a
>Localtalk connection (actually phoneNet connectors that I
>scrounged up). One of my phoneNet connectors is kinda weird -
>it has a 9pin serial connection (perfect for a Windows COM
>port) that is plugged into an adapter to bring it back to the
>Mac printer port (geez, I'm out of the loop! What are those
>connectors called again? DIN? DB9? Grrr... memory fading...).
The Mac connector is RS-422, the PC is RS-232.
However, this does not mean a connection is impossible. The
actual network is merely two wires run in parallel from machine
to machine, in a "Bus" architecture. That fat little box
contains a transformer and a bunch of noise filtering components
to interface the bus with the serial in/out ports.
The main difference between the cheaper Phonenet and expensive
Appletalk boxes is what is known as "termination". The ends of
the network need to have a resistor connecting the signal to
ground. The resistor dampens noise and signal reflections in
the wire.
Appletalk is self-terminating. By default both ports on an
unused box are already terminated. The network cable, when
inserted into the box, physically presses back on a switch,
and the switch disconnects the terminating resistor for that
port. Pull out the cable, and the resistor reconnects. Very
slick. Probably why Apple always charged $75 and up for that
seemingly useless little box.
Phonenet requires manual termination. You have to stick a
little phone plug into the spare port, and the plug is wired
with a resistor. Now, it WILL work without termination, but
likely only with short runs between 2-6 machines in the same
room. If you plan to span the network 1000 feet or so,
termination is very important.
And would ya know it, I've got a disassembled Appletalk dongle
right here. Wiring looks like this inside:
T1
BUS Wire 1 --+------+--- ----------+--------- Serial Red
| | ) ## ( |
Term R2 | ) ## ( `--- R4 -- Serial Blu
| | ) ## (
Term Sw | ) ## (
| | ) ## (
BUS Wire 2---+----- | -- ## --- N/C
| ) ## (
| ) ## (
R1 ) ## (
| ) ## ( .--- R3 -- Serial Wht
| ) ## ( |
BUS Ground----------+--- ----------+--------- Serial Grn
|
|
`-----------+-- R5 --+-------- Serial GND
| |
`-- C1 --'
(noise filter)
Okay, I don't know what the serial wire colors represent but
it seems fairly easy to figure out. The main thing you need to
know about RS422 is that you might say the sending line has
its own ground, and so does the receiving line. It's usually
represented as RX+ RX- and TX+ TX-
In order for you to send a signal you have to send from RX+
through the transformer and out to RX-. Similarly to receive
a signal the transformer must connect to TX+ and TX-.
What they have done is tie BOTH signal lines together. So
basically the input is automatically receiving whatever is
output to the network.
I would not be suprised to see that R3 is dampening RX+ and
R4 is dampening RX- to keep it from being overloaded by the
local signal transmissions on that same machine.
Conversion to a PC serial is a problem since the power in
the serial port typically flows like this:
Sending from PC1 to PC2 Sending from PC2 to PC1
TX -->>>-- RX TX ------- RX
PC1 PC2 PC1 PC2
GND --<<<-- GND GND -->>>-- GND
RX ------- TX RX --<<<-- TX
Ground is multidirectional, balancing the current flow of
the various signal lines traveling in either direction.
One option is the following. With this, Serial TX flows to
local ground, completing the circuit. Meanwhile when a
network signal is received, power flows from ground to
Serial RX, keeping the polarity the same:
T1
BUS Wire 1 --+------+--- ----------+--------- Serial TX
| | ) ## ( |
Term-R2 | ) ## ( `--- R4 -- N/C
| | ) ## (
Switch | ) ## (
| | ) ## (
BUS Wire 2---+----- | -- ## -------------------- GND
| ) ## (
| ) ## (
R1 ) ## (
| ) ## ( .--- R3 -- N/C
| ) ## ( |
BUS Shield----------+--- ----------+--------- Serial RX
|
|
`----------+-- R5 --+------ Serial Shield
| |
`-- C1 --'
(noise filter)
I don't really know if this would work. Can you think of any
better possible wiring schemes?
-------------------------------------------------------------
PHONE-NET box:
Hey, wouldn't ya know it, I got one of those too. :)
.-- R3 -- Ser Org
T1 |
BUS 1 -----+-------------- -------+-------- Ser Brn
| ) ## (
| ) ## (
| ) ## (
BUS 2 ---- | ---------+--- -------+-------- Ser Red
| | |
+---. +---. `-- R4 -- Ser Blk
Filter | | | | Filter
for R1 C1 R2 C2 for
BUS 1 | | | | BUS 2
| | | |
`---+------+---+------------------ Ser Shield
Hmmm, looks a little familiar. Same transformer layout. Only
real difference is the filtering. I would have to guess that
the extra filtering is needed because these cables are not
shielded at all, being plain phone cord.
On another front, I have found that Phonenet and Appletalk
are so similar that you can just wire the two together
without any harm. Take an Appletalk BUS cable and hack it
off. Take a phonenet cord and hack it off. Connect the two
outer wires of the phonenet cord and attach them to the bus
cable wires. Tape it together, and plug it in.
-Mr. Boffo
Tim Haynes
Wow! You have an astonishing wealth of technical information at your
fingertips! Unfortunately, my skill with a soldering iron is limited to
repairing cracked traces and not much else. :( But thanks for taking some
much time to give me such detailed info! If I ever get a few free days,
maybe I'll take up the challenge.
Thanks again!
Tim
Wayne Stewart
<big snip>
> On another front, I have found that Phonenet and Appletalk
> are so similar that you can just wire the two together
> without any harm. Take an Appletalk BUS cable and hack it
> off. Take a phonenet cord and hack it off. Connect the two
> outer wires of the phonenet cord and attach them to the bus
> cable wires. Tape it together, and plug it in.
They did make adapters like that, I have a couple of them.
Wayne
David Wilson
And here are the serial connections from page A-4 of "Inside Appletalk"
Second Edition:
Wire Sig MiniDIN-8 DE-9
-------------------------------------
Red TxD- pin 3 pin 5
Blu RxD- pin 5 pin 9
Wht RxD+ pin 8 pin 8
Grn TxD+ pin 6 pin 4
GND Case Shield Shield
>In order for you to send a signal you have to send from RX+
>through the transformer and out to RX-. Similarly to receive
>a signal the transformer must connect to TX+ and TX-.
The +/- signals are differential. When TxD+ is +3volts the TxD-
will be -3volts giving 6 volts between them. When the signal changes
from logical 1 to logical zero the polarities of both lines change.
--
David Wilson School of IT & CS, Uni of Wollongong, Australia
Mister Boffo
Ah-HAH! So nice to see that my electronics theory is actually valid. I
was also right about the resistors being there to reduce the local
current going to RxD+ and RxD-! :)
As I think about it some more, in a direct serial link, RxD+/- is
normally connected straight to TxD+/- of another computer. So RxD+/-
should be fully capable of accepting whatever current TxD+/- puts out.
It is possible that R3 and R4 don't "protect" RxD+/- but instead are
there to make sure some power flows through the transformer from TxD+
to TxD-. Without R3 and R4, all the output of TxD+/- could be sucked
up by RxD+/- without doing anything in T1.
Resistor R1 is still a bit of a question mark for me, but I do believe
you can also construct a filter out of an inductor and a resistor. So
in this case R1 and the "secondary" coil of T1 are working together to
form a second noise filter for the Appletalk network bus, in addition
to the R5/C1 noise filter...
I'm crossposting this to sci.electronics.basics, in hopes of getting
some input from an electronics expert on this. :)
-Mr. Boffo
Matthew Russotto
Mister Boffo <mister...@hotmail.com> wrote:
>
>It is possible that R3 and R4 don't "protect" RxD+/- but instead are
>there to make sure some power flows through the transformer from TxD+
>to TxD-. Without R3 and R4, all the output of TxD+/- could be sucked
>up by RxD+/- without doing anything in T1.
I think it might be a bit more subtle than that. The probable reason you
need a connection from your own transmitter to your own receiver is
for collision detection. With a straight connection, your own signal
would overwhelm any signal from the bus, leaving you unable to detect
collisions.
>Resistor R1 is still a bit of a question mark for me, but I do believe
>you can also construct a filter out of an inductor and a resistor. So
>in this case R1 and the "secondary" coil of T1 are working together to
>form a second noise filter for the Appletalk network bus, in addition
>to the R5/C1 noise filter...
I think R1 is probably to prevent external high frequency noise from inducing
a signal on the bus.
What bothers me is the connection from the bus ground to serial
ground. With that, if your computers are at different ground levels,
your Appletalk network can damage them. What's the point of isolating
things with a transformer if you are going to directly connect
grounds?
Note that PhoneNet, being unshielded, doesn't have a bus ground.
Mister Boffo
Russotto) wrote:
>In article <3b4c6986....@news.mw.centurytel.net>,
>Mister Boffo <mister...@hotmail.com> wrote:
>>
>>It is possible that R3 and R4 don't "protect" RxD+/- but instead are
>>there to make sure some power flows through the transformer from TxD+
>>to TxD-. Without R3 and R4, all the output of TxD+/- could be sucked
>>up by RxD+/- without doing anything in T1.
>
>I think it might be a bit more subtle than that. The probable reason you
>need a connection from your own transmitter to your own receiver is
>for collision detection. With a straight connection, your own signal
>would overwhelm any signal from the bus, leaving you unable to detect
>collisions.
You may have missed the earlier part of this discussion. The receiver
and transmitter are connected to each other since that's the only way
for the bidirectional Appletalk bus to connect to both of them at
once.
However, yes, I do think I see what you mean. You can check for errors
by sending out data, and then immediately checking if you can read it
back on the incoming side okay. If the data comes back garbled then
there was probably a collision.
Overall it's a rather weird way to use your serial port. :)
Honestly, I don't think there is anything at all different between the
serial and modem ports on a Macintosh if this is all there is to it.
Appletalk appears to involve no additional hardware beyond the normal
serial port, and AppleTalk may be a completely software-based network.
Apple always claimed the printer port was for AppleTalk, but I don't
see any objective reason why it matters which port get used. Perhaps
Apple used more expensive serial hardware for the printer port, so
that it could signal much faster than the modem port? Anybody know
what the max speed is of the modem port vs printer port?
-Mr. Boffo
Phil Beesley
<mister...@hotmail.com> wrote:
The AppleTalk and modem ports are identical to each other on all Apple
computers that have both ports ;-) There are some PowerBook models, for
instance, that have only one port and a software switch must be used to
switch between "printer" and "modem" modes.
Different models of computer have a different maximum speed for the
ports (but both ports will work at the same speed). Standard AppleTalk
over LocalTalk cabling operates at about 230.4 kbps. A couple of
vendors (Radius, Sum Microsystems) implemented LocalTalk accelerators
but these are unusual and largely incompatible with Apple's standard.
Phil
Matthew Russotto
Mister Boffo <mister...@hotmail.com> wrote:
>On Wed, 11 Jul 2001 15:41:26 GMT, russ...@wanda.pond.com (Matthew
>Russotto) wrote:
>
>However, yes, I do think I see what you mean. You can check for errors
>by sending out data, and then immediately checking if you can read it
>back on the incoming side okay. If the data comes back garbled then
>there was probably a collision.
>
>Overall it's a rather weird way to use your serial port. :)
>
>Honestly, I don't think there is anything at all different between the
>serial and modem ports on a Macintosh if this is all there is to it.
>Appletalk appears to involve no additional hardware beyond the normal
>serial port, and AppleTalk may be a completely software-based network.
Appletalk requires RS-422 drivers and a serial controller capable of SDLC
communication at 230Kbps (I think using NRZi encoding, but it's been
way too long since I traced this). A simple asynchronous RS-232 port won't
work. But you're right -- there is no hardware difference between the
serial and modem ports on the Macintosh.
>Apple always claimed the printer port was for AppleTalk, but I don't
>see any objective reason why it matters which port get used. Perhaps
>Apple used more expensive serial hardware for the printer port, so
>that it could signal much faster than the modem port? Anybody know
>what the max speed is of the modem port vs printer port?
For the Lisa, only one of the ports was RS-422, so you had to use that one.
For the Mac, the difference had to do with the floppy disk driver. It
disabled interrupts, which could cause lost characters. This didn't
matter much for AppleTalk, which is expecting that sort of thing. But
it causes problems with modems. So the floppy driver polls the modem
port periodically while interrupts are disabled. The printer port is
left to its own devices.
David Empson
> In article <3b507efb....@news.mw.centurytel.net>,
> Mister Boffo <mister...@hotmail.com> wrote:
> >Honestly, I don't think there is anything at all different between the
> >serial and modem ports on a Macintosh if this is all there is to it.
> >Appletalk appears to involve no additional hardware beyond the normal
> >serial port, and AppleTalk may be a completely software-based network.
>
> Appletalk requires RS-422 drivers and a serial controller capable of SDLC
> communication at 230Kbps (I think using NRZi encoding, but it's been
> way too long since I traced this).
It is FM encoding, but I forget whether it is FM0 or FM1.
The RS-422 transmit driver chip must have an enable input (controlled by
the RTS signal from the Zilog SCC on the Mac and IIgs), so that the
transmitter can be tri-stated when the computer isn't sending data.
(The transmitter is permanently enabled when the port is being used for
normal asynchronous communication.)
There are also some specific requirements of the LocalTalk protocol
which impose further restrictions on the SDLC implementation (to do with
the synchronization mechanism, mainly). A Zilog SCC or compatible
device is OK, but some other devices are not, without additional
hardware.
> A simple asynchronous RS-232 port won't work. But you're right -- there is no
> hardware difference between the serial and modem ports on the Macintosh.
That's "printer" and modem ports.
> >Apple always claimed the printer port was for AppleTalk, but I don't
> >see any objective reason why it matters which port get used. Perhaps
> >Apple used more expensive serial hardware for the printer port, so
> >that it could signal much faster than the modem port? Anybody know
> >what the max speed is of the modem port vs printer port?
>
> For the Lisa, only one of the ports was RS-422, so you had to use that one.
> For the Mac, the difference had to do with the floppy disk driver. It
> disabled interrupts, which could cause lost characters. This didn't
> matter much for AppleTalk, which is expecting that sort of thing. But
> it causes problems with modems. So the floppy driver polls the modem
> port periodically while interrupts are disabled. The printer port is
> left to its own devices.
The early AppleTalk implementation (prior to Open Transport) probably
also disables interrupts for the duration of a packet being sent or
received (as does the IIgs), so floppy activity and AppleTalk will be
mutually exclusive.
On later Mac models, Open Transport can presumably use DMA for
LocalTalk, so there should be such a performance it.
Matthew Russotto
David Empson <dem...@actrix.gen.nz> wrote:
>Matthew Russotto <russ...@wanda.pond.com> wrote:
>
>> In article <3b507efb....@news.mw.centurytel.net>,
>> Mister Boffo <mister...@hotmail.com> wrote:
>> >Honestly, I don't think there is anything at all different between the
>> >serial and modem ports on a Macintosh if this is all there is to it.
>> >Appletalk appears to involve no additional hardware beyond the normal
>> >serial port, and AppleTalk may be a completely software-based network.
>>
>> Appletalk requires RS-422 drivers and a serial controller capable of SDLC
>> communication at 230Kbps (I think using NRZi encoding, but it's been
>> way too long since I traced this).
>
>It is FM encoding, but I forget whether it is FM0 or FM1.
FM0; I just checked the log from booting the vMac Mac emulator.
There's a wealth of crap^Wtreasure lying around on my hard drive...