Google Groups no longer supports new Usenet posts or subscriptions. Historical content remains viewable.
Dismiss

Apple II+, Raspberry Pi, and Kermit terminal column width

820 views
Skip to first unread message

Chris Torrence

unread,
Feb 9, 2014, 1:27:29 AM2/9/14
to
Hi all,

I'm trying to use my Apple II+ as a vt100 terminal to my Raspberry Pi. Since I've got a II+, I cannot use ProTERM, Z-Link, or Modem.MGR (the vt100 module requires a IIe). So I think my only real choice is Kermit-65. I'm using v. 3.87.

I have a Videx Ultraterm card in slot 3, an Apple SSC in slot 1, and a 16K language card in slot 0.

I've gotten really close to getting it to work. I was able to connect up to the RPi and log in, after realizing that I had to hit the right arrow key on the II+ to toggle upper/lowercase. I can't seem to get the shift key to do uppercase, even though I've got the shift key mod installed, but that isn't a big deal.

My real problem is that the terminal width seems to be stuck at 40 columns, even though the II+ is in 80-column mode. I *think* I've configured Kermit correctly, and over on the RPi side, if I do "stty size" it reports 80 columns by 24 lines. But even though the Ultraterm is in "80-column mode", only half the screen gets used because it just wraps at 40 columns. I have been able to get other programs to use the full 80 columns, so I know it's not a card problem.

Here are some of the Kermit settings:
TERMINAL: VT100, wrap-around on
LOCAL-ECHO: OFF
PARITY: NONE
DEVICE-DRIVER: SSC V1.15
KEYBOARD: 2P <-- indicates II+
DISPLAY: 80-COL slot=03
BAUD: 19200
FLOW: OFF
TIMER: OFF
SWAPKEYS-BS&DEL: ON
APPLICATION-MODE: ON
CURSOR-KEYS-VT100: OFF
TIMING: 11

Does anyone have any ideas for settings that I can change, either on the Kermit side, or on the Raspberry Pi side?

Thanks!
-Chris

Bill Buckels

unread,
Feb 9, 2014, 5:08:53 AM2/9/14
to
"Chris Torrence" <gort...@gmail.com> wrote:
>Does anyone have any ideas for settings that I can change, either on the
>Kermit side, or on the Raspberry Pi side?

Hi Chris,

If the Kermit you are using is written in Aztec C65 you could try using the
Aztec C configuration utility to set the peripheral properties.

The manual for the utility will tell you more.

C54 3.2b and CII both have a config utility.

Bill


Bill Buckels

unread,
Feb 9, 2014, 5:42:56 AM2/9/14
to
"Bill Buckels" <bbuc...@mts.net> wrote:
>If the Kermit you are using is written in Aztec C65 you could try using the
>Aztec C configuration utility to set the peripheral properties.

The config utility is available as noted below. Try running it aganst kermit
and see what happens.

Also you might want to start kermit from inside the Aztec C ProDOS Shell or
if its a DOS 3.3 kermit, inside the Aztec C DOS 3.3 Shell making sure you
have configured the shell to 89 columns beforehand.

Here's that post from November 2013... I think someone asked me at that time
how this is supposed to help anyone. It would be nice if it did help you.You
could always send me your kermit in PM and I'll see if it is in fact in
Aztec C... and let me know if its DOS 3.3 or ProDOS that you use if you do
that.

Bill

x--- snip ---x


----- Original Message -----
From: "Bill Buckels" <bbuc...@mts.net>
Newsgroups: comp.sys.apple2.programmer
Sent: Sunday, November 10, 2013 7:46 PM
Subject: Announcing the Resurrection of the Aztec C65 3.2b Utilities


Now Available for Download!

Read All About It:

This download is intended to be overlaid over the current AppleX
distribution for anyone who wants to build the programs, but can also simply
be downloaded, and used as-is. Executables and Disk Images are included. The
main topic is "retro" Serial Communications in Aztec C65, but included are
many others.

User Documentation:

http://www.aztecmuseum.ca/extras/AppleX_UTILITY.pdf

Programmer Notes:

http://www.aztecmuseum.ca/extras/AppleX_UTILITY_ReadMe.txt

Download it here:

http://www.aztecmuseum.ca/extras/AppleX_UTILITY.zip

This download also includes the Win32 Utility CiderTag which among other
tricks will strip the header from an Apple II Binary Program and use it to
create a new headless file copy with the header info as a file attribute
preservation tag acceptable to CiderPress.

CiderTag comes complete with source and a Win32 build environment for those
who like that sort of thing. Also, while I think of it, someone besides me
should probably double-dog test this program...

All the best,

Bill Buckels





Bill Buckels

unread,
Feb 9, 2014, 11:58:13 AM2/9/14
to
"Chris Torrence" <gort...@gmail.com> wrote:
>I think my only real choice is Kermit-65. I'm using v. 3.87.
Chris,

Ignore my previous messages about Aztec C and visit the following link for
VERSION 3.87 and rtfm: It is in pdf format and tells you how to configure
all:

http://www.columbia.edu/kermit/appleii.html

Your best bet is to dwnload everything:

http://www.columbia.edu/kermit/ftp/archives/appleii.zip

The historical Kermit software archive - the one that contains all the
Kermit programs and files from 1981 to August 2011 - is at Columbia
University.

This is written in Cross and has been since at least 1984 from what I found:

http://www.columbia.edu/kermit/ftp/e/mail.84b

I don't know why I thought someone had done an Aztec C Version. But then I
haven't thought about kermit since the '80's so I have already forgiven
myself.

I guess I would have taken a look if I needed to and read the f* manual.
Presumably that's why we write the things.

Bill


David Schmenk

unread,
Feb 9, 2014, 11:58:27 AM2/9/14
to
Chris-
You could also try ASCI Express with your II+. I've used it in the past. The only issue was the terminal description. You can find one here with the old school Apple and PC compatible term caps: http://stuff.mit.edu/afs/sipb/system/i386_nbsd1/os-1.4.1/usr/share/misc/termcap

Copy it to /etc/termcap

type: export TERM=apple-ae

It *should* work.

Dave...

Bill Buckels

unread,
Feb 9, 2014, 12:12:09 PM2/9/14
to
"David Schmenk" <dsch...@gmail.com> wrote:
>You could also try ASCI Express with your II+.

The Kermit that is on asimov comes with a program that sets 80 column in the
init file.

There's ahole load of communications programs there as well.

I'm going back to writing my manual, not that anyone reads them anymore.

Bill


Chris Torrence

unread,
Feb 9, 2014, 1:21:29 PM2/9/14
to
On Sunday, February 9, 2014 10:12:09 AM UTC-7, Bill Buckels wrote:
>
>
> The Kermit that is on asimov comes with a program that sets 80 column in the
>
> init file.
>
>
>
> There's ahole load of communications programs there as well.
>
>
>
> I'm going back to writing my manual, not that anyone reads them anymore.
>
>
>
> Bill

Bill,

In my original post, I mentioned that my Apple II+ is _already_ in 80-column mode. The characters are both upper and lowercase, and they are only taking up "half" the width of 40-column characters. I did download the entire Kermit source from Columbia, and I read the entire manual.

No matter what we may hope for, the Apple II community is a _small_ one. There are not that many of us that are willing to play around with 30-year old technology, and our numbers are shrinking. People come to comp.sys.apple2 because it is one of the only places where we can talk freely, respectfully exchange ideas, and have fun with the Apple platform.

Your work with the Aztec C Programming Language clearly shows that there are new things to be done with the Apple II. I think the Raspberry Pi is yet another way. My two daughters have been fascinated with the Apple II+ since I brought it out of the basement and put it on the desk next to the shiny MacBook.

If we work together, and help out newbies, maybe we can actually grow the community. At least we'll have fun trying.

Thanks for listening.
-Chris

Ivan X

unread,
Feb 9, 2014, 1:31:05 PM2/9/14
to
I no longer consider usable any non-vt100 derived termcaps. Too much Linux software appears to be hardcoded to expect a vt100-type emulator rather than being fully abstracted from the display codes. In my experience a term program has to provide actual vt100 emulation to be used effectively with the Pi.

D Finnigan

unread,
Feb 9, 2014, 1:39:22 PM2/9/14
to
Chris Torrence wrote:
> Hi all,
>
> I'm trying to use my Apple II+ as a vt100 terminal to my Raspberry Pi.
> Since I've got a II+, I cannot use ProTERM, Z-Link, or Modem.MGR (the
> vt100
> module requires a IIe). So I think my only real choice is Kermit-65. I'm
> using v. 3.87.
>
> I have a Videx Ultraterm card in slot 3, an Apple SSC in slot 1, and a 16K
> language card in slot 0.
>
> I've gotten really close to getting it to work. I was able to connect up
> to
> the RPi and log in, after realizing that I had to hit the right arrow key
> on the II+ to toggle upper/lowercase. I can't seem to get the shift key to
> do uppercase, even though I've got the shift key mod installed, but that
> isn't a big deal.

If you still can't get Kermit to go, try DCOM. Download here:

http://macgui.com/downloads/?file_id=2062

It's a VT100 emulator and I have used it to make a terminal to a Mac OS X
system over a direct serial link. It does work! It is also covered in
Chapter 10 of The New Apple II User's Guide. Setup is fairly
straightforward.

More info and features: http://macgui.com/usenet/?group=6&id=2661


--
]DF$
Apple II Book: http://macgui.com/newa2guide/
Usenet: http://macgui.com/usenet/ <-- get posts by email!
Apple II Web & Blog hosting: http://a2hq.com/

Chris Torrence

unread,
Feb 9, 2014, 3:32:29 PM2/9/14
to
Hi all,

Thanks for all of the suggestions. Here's what I've found:

Kermit: Connects fine, vt100 appears to work properly, but only uses 40 columns out of the 80 columns.

Ascii Express Pro 3.46b: Connects fine, uses all 80 columns, but doesn't handle the vt100 "color" escape codes, so things like lynx or news readers (like Ivan's a2news) look all messed up.

DCOM 3.21: Doesn't support VIDEX 80-column cards. Picking any of the other card options just results in garbage, so I can't tell if it is connecting.

I'm out of ideas with getting proper vt100 support on the Apple II+, with a Videx 80-column card...

-Chris

Egan Ford

unread,
Feb 9, 2014, 11:48:32 PM2/9/14
to
On 2/9/14, 1:32 PM, Chris Torrence wrote:
> I'm out of ideas with getting proper vt100 support on the Apple II+, with a Videx 80-column card...

Well there is one idea you could try. Write your own. To speed up
development you could use Aztec C or cc65. I do not think it'd take a
lot of effort.

Steve Nickolas

unread,
Feb 10, 2014, 6:25:30 AM2/10/14
to
My head exploded trying to just think about how to do it on a *PC* in
*Borland C*. And Borland has a lot of functionality in its libc to help.

-uso.

Bill Buckels

unread,
Feb 10, 2014, 6:29:34 AM2/10/14
to
"Egan Ford" <data...@gmail.com> wrote:
>Well there is one idea you could try. Write your own. To speed up
>development you could use Aztec C or cc65. I do not think it'd take a lot
>of effort.

Hi Egan,

"You can lead a man to slaughter but you can't make him think":)

I have skirted the issue to this point preferring to have the op take
another closer look. However that was the wrong approach considering he had
already taken as close a look as his tunnel allowed. My cyncism has blinded
me... I have no excuse. Only forgiveness is left for me:)

Also I admire your directness as always. Skirters sometimes get their skirts
torn off. My is completely at my feet now and my butt is freezing:)

I pointed him to the Aztec C Source and the manual (rtfm) very early in this
thread. Keep in mind that I am a net fisherman and not an angler and I don't
troll... I set a net and wait for the fish to come.

When he didn't take the bait I emailed him the fully functional Aztec C tty
terminal program in 80 column mode on a disk yesterday. But he didn't bother
with it until last evening.

The Atzec C tty's terminal module is written in 6502 and is included at the
end of this post.

The pure emulation for vt100 must also be considered:

Ivanx wrote: "I no longer consider usable any non-vt100 derived termcaps.
The Aztec C tty program supports only some of the features of the earlier
ADM-3A terminal. The ADM-3A does not support more advanced codes of the
later vt100.

Here is what Aztec C's tty supports out of the box. Completely self
explanatory and to some the comments are already included. Cover your little
eyes Yacky. You shouldn't oughta see this!" (Chopper):

The 80 column card firmware locations are accessed directly of course. I
have refrained from posting the serial IO module which is also in 6502.

All the best,

Bill

x--- snip ---x

*:ts=8
*
* APPLE //e TERMINAL
*
*

instxt <zpage.h>

HEAD EQU $65
TAIL EQU $67
STACK EQU $68
*
*
*
DATREG EQU $C088
STAREG EQU $C089
CMDREG EQU $C08A
CTLREG EQU $C08B
RMASK EQU $08
TMASK EQU $10
*
*
*
global slot_,2
*
cseg
public dumb_
dumb_ tsx
stx STACK
jsr INIT
; jsr outinit_
cli
LOOP jsr GETC
jsr PUTC
jmp LOOP
*
public EXIT
EXIT sei
sta R0
lda #0
sta R0+1
ldx slot_
lda #$0B
sta CMDREG,X
ldx STACK
txs
rts
*
INIT sei
ldx slot_
lda #$09 ;transmit, recieve enabled, irq on recv enabled
sta CMDREG,X
ldy #2
lda (SP),Y ;get baud rate
ora #$10 ;1 stop, 8 data, internal clock
sta CTLREG,X
lda DATREG,X
lda #<SERVICE
sta $3fe
lda #>SERVICE
sta $3ff
lda #0
sta HEAD
sta TAIL
rts
*
SERVICE tya
pha
txa
pha
ldx slot_
lda STAREG,X
and #RMASK
beq SKIP
lda DATREG,X
and #$7f
ldy HEAD
sta $200,y
inc HEAD
SKIP pla
tax
pla
tay
lda $45
rti
*
*
GETC ldx $C000
bpl GETC24
GETC20 cpx =$80 ;CTRL 2 _ -> exit
bne GETC21
jsr curs
lda $C010
lda =1
jmp EXIT
GETC21 cpx =$ff
bne GETC22
ldx =$88
bne GETC23
GETC22 cpx =$88
bne GETC23
ldx =$ff
GETC23 ldy slot_
lda STAREG,Y
and =TMASK
beq GETC24
sta $C010
txa
and =$7F
sta DATREG,Y
GETC24 rts
*
*
PUTC ldx TAIL
cpx HEAD
bne PUTC0
rts
PUTC0 ldy TAIL
lda $200,Y
inc TAIL
jmp outch
*
*
*
ALTCHRON EQU $C00F
ALTCHROFF EQU $C00E
ALTCHRRD EQU $C01E
PAGE2ON EQU $C055
PAGE2OFF EQU $C054
PAGE2RD EQU $C01C
COL80ON EQU $C00D
COL80OFF EQU $C00C
COL80RD EQU $C01F
STOR80ON EQU $C001
STOR80OFF EQU $C000
STOR80RD EQU $C018
SPKR EQU $C030
*
*
TP EQU $EA
BP EQU $EC
CHX EQU $FA
CHY EQU $FB
TMPY EQU $FC
TMPA EQU $FD
INESC EQU $FE
TMPC EQU $FF
*
cseg
public outinit_
outinit_
sta COL80ON
sta STOR80ON
sta ALTCHRON
lda #0
sta INESC
sta CHX
sta CHY
jsr clear
jmp curs
*
*
public putchar_
putchar_
lda #0
sta INESC
ldy #2
lda (SP),Y
cmp #$0a
bne outch
lda #$0d
jsr outch
lda #$0a
*
public outch
outch
jsr curs
lda INESC
beq outch1
jsr doesc
jmp curs
outch1
lda TMPA
cmp #$20
bcc outch2
jsr store
jsr advanc
jmp curs
outch2
jsr chk
jmp curs
*
*
chk
cmp #$1a ;^Z
bne chk1
jmp clear
chk1
cmp #$08 ;^H
bne chk2
dec CHX
bpl chkd
lda #79
sta CHX
dec CHY
bpl chkd
lda #0
sta CHY
rts
chk2
cmp #$0C ;^L
bne chk3
advanc
inc CHX
lda CHX
cmp #80
bne chkd
lda #0
sta CHX
inc CHY
lda CHY
cmp #24
bne chkd
jmp lf
chk3
cmp #$0A ;^J
bne chk4
inc CHY
lda CHY
cmp #24
bne chkd
dec CHY
jmp lf
chk4 cmp #$0B ;^K
bne chk5
dec CHY
bpl chkd
inc CHY
rts
chk5
cmp #$1B ;ESC
bne chk6
lda #1
sta INESC
rts
chk6
cmp #$0D ;^M
bne chk7
lda #$0
sta CHX
rts
chk7
cmp #$07 ;^G
bne chk8
jmp bell
chk8
cmp #$1E ;^^
bne chkd
lda #0
sta CHX
sta CHY
chkd
rts
*
*
doesc
tax
lda TMPA
cpx #$4
bne doesc1
sec
sbc #$20
sta CHX
lda TMPC
sta CHY
lda #0
sta INESC
rts
doesc1
cpx #$2
bne doesc2
sec
sbc #$20
sta TMPC
lda #$4
sta INESC
rts
doesc2
cmp #'=
bne doesc3
lda #$2
sta INESC
rts
doesc3
cmp #'T ;clear to EOL
bne doesc6
lda CHX
pha
doesc4
lda #$20
jsr store
lda CHX
cmp #79
beq doesc5
jsr advanc
jmp doesc4
doesc5
pla
sta CHX
jmp escdon
doesc6
cmp #'Q ;char insert
bne doesc9
lda CHX
sta TMPC
lda #79
sta CHX
doesc7
lda CHX
cmp TMPC
beq doesc8
dec CHX
jsr calc
lda CHX
lsr A
tay
lda (TP),Y
eor #$80
inc CHX
jsr store
dec CHX
bpl doesc7
doesc8
lda #$20
jsr store
jmp escdon
doesc9
cmp #'W ;char delete
bne doesc12
lda CHX
pha
doesc10
lda CHX
cmp #79
beq doesc11
inc CHX
jsr calc
lda CHX
lsr A
tay
lda (TP),Y
eor #$80
dec CHX
jsr store
inc CHX
bne doesc10
doesc11
lda #$20
jsr store
pla
sta CHX
jmp escdon
doesc12
cmp #'E ;line insert
bne doesc13
jsr inslin
jmp escdon
doesc13
cmp #'R ;line delete
bne doesc14
jsr dellin
jmp escdon
doesc14
cmp #$02
beq doesc15
cmp #$03
bne escdon
doesc15
lda #0
jmp EXIT
escdon
lda #0
sta INESC
rts
*
*
clear
ldx #47
clear1
lda STAB,X
sta TP+1
dex
lda STAB,X
sta TP
jsr clrlin
dex
bpl clear1
lda #0
sta CHX
sta CHY
rts
*
clrlin
sta PAGE2OFF
clrlin1
ldy #39
lda #$a0
clrlin2
sta (TP),Y
dey
bpl clrlin2
lda PAGE2RD
bmi clrlin3
sta PAGE2ON
jmp clrlin1
clrlin3
rts
*
store
pha
jsr calc
lda CHX
lsr A
tay
pla
eor #$80
sta (TP),Y
rts
*
lf
lda CHX
pha
lda #0
sta CHX
sta CHY
jsr dellin
pla
sta CHX
lda #23
sta CHY
jsr calc
rts
*
inslin
lda #0
sta CHX
lda CHY
sta TMPC
lda #23
sta CHY
inslin1
lda CHY
cmp TMPC
beq inslin2
jsr calc
lda TP
sta BP
lda TP+1
sta BP+1
dec CHY
jsr calc
jsr cpylin
jmp inslin1
inslin2
jsr calc
jsr clrlin
rts
*
dellin
lda #0
sta CHX
lda CHY
pha
dellin1
lda CHY
cmp #23
beq dellin2
jsr calc
lda TP
sta BP
lda TP+1
sta BP+1
inc CHY
jsr calc
jsr cpylin
jmp dellin1
dellin2
jsr calc
jsr clrlin
pla
sta CHY
rts
*
cpylin
sta PAGE2OFF
cpylin1
ldy #39
cpylin2
lda (TP),Y
sta (BP),Y
dey
bpl cpylin2
lda PAGE2RD
bmi cpylin3
sta PAGE2ON
jmp cpylin1
cpylin3
rts
*
curs
sty TMPY
sta TMPA
jsr calc
lda CHX
lsr A
tay
lda (TP),Y
cmp #$20
bcs curs1
ora #$40
curs1
eor #$80
cmp #$40
bcc curs2
cmp #$60
bcs curs2
and #$bf
curs2
sta (TP),Y
ldy TMPY
lda TMPA
rts
*
calc
lda CHY
asl A
tay
lda STAB,Y
sta TP
lda STAB+1,Y
sta TP+1
lda CHX
and #1
beq calc1
sta PAGE2OFF
rts
calc1
sta PAGE2ON
rts
*
bell
lda #$20
sta TMPC
bell1
lda #$2
jsr delay
sta SPKR
lda #$24
jsr delay
sta SPKR
dec TMPC
bne bell1
rts
*
delay
sec
delay1
pha
delay2
sbc #1
bne delay2
pla
sbc #1
bne delay1
rts
*
STAB fdb $400,$480,$500,$580,$600,$680,$700,$780
fdb $428,$4a8,$528,$5a8,$628,$6a8,$728,$7a8
fdb $450,$4d0,$550,$5d0,$650,$6d0,$750,$7d0
*








Bill Buckels

unread,
Feb 10, 2014, 6:45:35 AM2/10/14
to
"Steve Nickolas" <usot...@buric.co> wrote:
>My head exploded trying to just think about how to do it on a *PC* in
>*Borland C*. And Borland has a lot of functionality in its libc to help.

I have that same problem when I think about getting up in the morning:)

In your professional opinion, what functionality has borland got in its libc
that Aztec C doesn't in this regard? I haven't been a borland partner in
about a decade now, but through all their versions going back to my turbo c
1.0 I can't think of what this functionality might be.

Also, after your review of the existing Aztec C tty source code for the
Aztec C terminal, how does the Borland tty source code (if any) compare
specifically?

It has been 30 years since I wrote serial programs on the PC. Different UART
too. But it wasn't terribly hard for someone who knows what they are doing.

I also seem to remember that doing a term program amounts to mapping control
codes to screen hardware. This is straight forward.

We MS-DOS programmers worked the same way as we Apple II programmers. We
went to firmware and ROM calls.

Bill


Steve Nickolas

unread,
Feb 10, 2014, 7:25:34 AM2/10/14
to
On Mon, 10 Feb 2014, Bill Buckels wrote:

> "Steve Nickolas" <usot...@buric.co> wrote:
>> My head exploded trying to just think about how to do it on a *PC* in
>> *Borland C*. And Borland has a lot of functionality in its libc to help.
>
> I have that same problem when I think about getting up in the morning:)
>
> In your professional opinion, what functionality has borland got in its libc
> that Aztec C doesn't in this regard? I haven't been a borland partner in
> about a decade now, but through all their versions going back to my turbo c
> 1.0 I can't think of what this functionality might be.

Well, I don't know what Aztec has as far as console I/O, but the Borland
"conio.h" is very powerful. (It is also implemented by the DJGPP libc,
for 32-bit applications.) It has single functions that implement some of
the more powerful functionality of the VT-100. The trick is parsing the
VT-100 data stream (because it's a bit more complex of a state machine
than I'm used to coding - actually, it's an 8080-based computer, but
that's beside the point).

-uso.

D Finnigan

unread,
Feb 10, 2014, 3:57:08 PM2/10/14
to
I gave you the wrong version of DCOM. It's version 3.3 that has the Videx
driver for the Apple II Plus.

It's in these multipart Usenet posts, starting with DCOM vt100 emulator for
Apple, (1 of 5)
http://macgui.com/usenet/?group=7&author=James+Hayes

If you can't get it assembled, I can make the disk image for you. I couldn't
find any premade disk with version 3.3 on it.

Egan Ford

unread,
Feb 10, 2014, 5:09:05 PM2/10/14
to
On 2/10/14, 4:29 AM, Bill Buckels wrote:
> I emailed him the fully functional Aztec C tty
> terminal program in 80 column mode on a disk yesterday.

That's 90% of the work. Just needs VT100 support. That can be
harvested from the Kermit source or done by hand.

Egan Ford

unread,
Feb 10, 2014, 5:17:01 PM2/10/14
to
On 2/10/14, 4:25 AM, Steve Nickolas wrote:
> My head exploded trying to just think about how to do it on a *PC* in
> *Borland C*. And Borland has a lot of functionality in its libc to help.

I had written terminal emulation for UNIX back in the very early '90s.
IIRC, it wasn't that challenging. OTOH, UNIX does have great support
for serial communications. The work was really around reading in the
data stream/parsing it and capturing input and sending the correct codes.

Given how much open source C code there is that provides VT100 emulation
I'd have to wager that it wouldn't be too hard to port using Aztec C or
cc65 if there was a clear understanding of how to do serial
communications on the Apple II.

David Schmenk

unread,
Feb 10, 2014, 8:32:20 PM2/10/14
to
When it comes to the II+, much of the challenge involves more of the input, rather than the output. The lack of special keys makes operating newer applications a bit of a challenge. No function keys, meta key, lower case, and a host of characters that one would expect to have on a keyboard means special sequences to emulate them. That, IMHO, is the difficulty in using the II+ on Linux/Unix.

Dave...

Bill Buckels

unread,
Feb 10, 2014, 9:14:08 PM2/10/14
to
"David Schmenk" <dsch...@gmail.com> wrote:
>When it comes to the II+, much of the challenge involves more of the input,
>rather than the output. The lack of special keys makes operating newer
>applications a bit of a challenge. No function keys, meta key, lower case,
>and a host of characters that one would expect to have on a keyboard means
>special sequences to emulate them. That, IMHO, is the difficulty in using
>the II+ on Linux/Unix.

Hi Dave,

I would still like to know if the Aztec C Terminal works at all on the II+.

Can you test it? Or do you not have a II+?

The serial routines and the terminal routines are both in 6502 assembler.

The terminal routines are incomplete.

The serial routines cannot exceed 19200 baud rate.

But it works on the //e.

I want to get one of your cards before starting on the Pi. But in the
meantime it would even be nice to know how well or badly this terminal works
on a //e.

Bill


David Schmenk

unread,
Feb 10, 2014, 9:22:12 PM2/10/14
to
Bill-

I have a new *feature* I'm adding to A2Pi that could use a good II+ terminal. I'll give it a try later this week. If it works at 9600 baud, I would sure be happy.

Dave...

N.N. Thayer

unread,
Feb 13, 2014, 3:27:52 AM2/13/14
to
I wouldn't mind seeing an image of v3.3 uploaded somewhere - I was going to attempt using my own II+ as a vt100-type terminal at some point...

Incidentally, Chris, one other option for you might be Visiterm which is on Asimov. It uses hi-res to get something like 70 or 72 columns.

Chris Torrence

unread,
Feb 13, 2014, 12:21:58 PM2/13/14
to
Thanks for the suggestion for Visiterm. I'll try it tonight. Also, I created a disk image of DCOM from all of Jim Hayes' posts. I can put that up somewhere if you want. It was actually easy to do - I just fired up Virtual ][ on the Mac, created a DOS disk, and then copy/pasted the code straight into the Apple prompt. Just make sure to do DOS and not ProDOS because all of the DCOM file names are not ProDOS-compatible.

DCOM didn't end up working - I chose the Videx driver but it just gave garbage on the screen.

I think Ivan is correct - you need to have a "real" vt100 terminal. So far, Kermit has come the closest. It does vt100, I can connect to the RPi, but I only get 40 columns of text even in 80-column mode. I tried Bill's suggestion and did the exact same steps on an Apple //e, and it worked perfectly with Kermit. So I think everything is configured correctly on the RPi terminal side - for whatever reason Kermit is not playing nicely with my Videx Ultraterm card on the Apple II+.

Finally, I fired up the Aztec C Shell disk that Bill was kind enough to email to me. It's so weird to type "ls -l" on an Apple... Anyway, the shell worked fine on the II+, with 80-column text. But "tty" didn't do anything, even after setting the correct baud. Again, it's probably because of the terminal emulation.

I'll keep plugging away...

Cheers,
Chris

N.N. Thayer

unread,
Feb 13, 2014, 12:44:33 PM2/13/14
to
I wonder if a Videx Videoterm vs. Ultraterm would make a difference...

Chris Torrence

unread,
Feb 13, 2014, 1:25:33 PM2/13/14
to
On Thursday, February 13, 2014 10:44:33 AM UTC-7, N.N. Thayer wrote:
>
>
> I wonder if a Videx Videoterm vs. Ultraterm would make a difference...

Yeah, I'm trying to not think about that possibility, since I got a good deal on the Ultraterm on ebay...
-Chris

D Finnigan

unread,
Feb 14, 2014, 1:57:43 PM2/14/14
to
N.N. Thayer wrote:
>
> I wouldn't mind seeing an image of v3.3 uploaded somewhere - I was going
> to
> attempt using my own II+ as a vt100-type terminal at some point...

I made the disk image, and I can email it to you tomorrow. Send me a
reminder email if I forget.

It will also appear in the next Mac GUI Vault update, which will probably be
in March.

D Finnigan

unread,
Feb 16, 2014, 1:19:20 PM2/16/14
to
I pulled out the old Apple II Plus (if you're having trouble forming a
mental image of it, pull out your New Apple II User's Guide; it's the one
pictured therein) and put the Transwarp, Super Serial card, and Videx
Videoterm in it to try DCOM 3.3

I must say, the II Plus makes a surprisingly competent terminal emulator in
this configuration. As usual, I connected to an old World Power Macintosh G3
running Mac OS X 10.2.

The only downsides of the II Plus as a VT100 emulator is its lack of true
lowercase KB, and the fact that it has no up/down arrow keys.

The only minor downside of DCOM 3.3 is that it runs only at 1200 baud, so
it's somewhat slow, but not to the point where it's frustratingly slow.

David Schmenk

unread,
Feb 16, 2014, 3:16:43 PM2/16/14
to
I was able to connect with my II+ and AE view master and it worked just fine as a vt100 using Kermit65. I think that will be my II+ terminal program of choice, even with the somewhat goofy extra key sequences.

Dave...

Michael J. Mahon

unread,
Feb 16, 2014, 10:19:39 PM2/16/14
to
D Finnigan <dog...@macgui.com> wrote:
> I pulled out the old Apple II Plus (if you're having trouble forming a
> mental image of it, pull out your New Apple II User's Guide; it's the one
> pictured therein) and put the Transwarp, Super Serial card, and Videx
> Videoterm in it to try DCOM 3.3
>
> I must say, the II Plus makes a surprisingly competent terminal emulator in
> this configuration. As usual, I connected to an old World Power Macintosh G3
> running Mac OS X 10.2.
>
> The only downsides of the II Plus as a VT100 emulator is its lack of true
> lowercase KB, and the fact that it has no up/down arrow keys.
>
> The only minor downside of DCOM 3.3 is that it runs only at 1200 baud, so
> it's somewhat slow, but not to the point where it's frustratingly slow.

I'm curious--does DCOM use the shift-key mod, and does the accelerator
allow faster com speeds (or is it a configuration, not a processor speed
limitation)?
--
-michael - NadaNet 3.1 and AppleCrate II: http://home.comcast.net/~mjmahon

N.N. Thayer

unread,
Feb 16, 2014, 11:14:36 PM2/16/14
to
So, just to be clear - you *were* able to get DCOM 3.3 working properly with a Videoterm? No garbage, 80-col at 40-col width, etc.?

D Finnigan

unread,
Feb 18, 2014, 1:17:33 PM2/18/14
to
N.N. Thayer wrote:
>
> So, just to be clear - you *were* able to get DCOM 3.3 working properly
> with a Videoterm? No garbage, 80-col at 40-col width, etc.?

Oh yes! With the Videx Videoterm. It works. Full 80-columns.

I ran `top` and I could see the cursor moving around to update the process
list at 1200 baud.

I would have taken a picture to show the screen, but my camera was out of
reach that evening.

D Finnigan

unread,
Feb 18, 2014, 1:25:49 PM2/18/14
to
Michael J. Mahon wrote:
> D Finnigan <dog...@macgui.com> wrote:
>>
>> The only minor downside of DCOM 3.3 is that it runs only at 1200 baud, so
>> it's somewhat slow, but not to the point where it's frustratingly slow.
>
> I'm curious--does DCOM use the shift-key mod, and does the accelerator
> allow faster com speeds (or is it a configuration, not a processor speed
> limitation)?

Shift-key mod: yes. In fact, on the configuration screen, it asks if you are
using shift-key on PB2, or if you've got a true lowercase, replacement
keyboard.

(I actually have neither, so I lied and said I had shift-key.)

The communication speed of 1200 baud was enforced by the program, so I
presume that the Transwarp accelerator only accelerated the terminal
emulation, and not the transmission speed.

James Hayes, the programmer, probably did not have an accelerator for his
Apple system, and so was getting dropped characters at faster speeds.


Further reading:

- DCOM VT100 driver specifications:
http://macgui.com/usenet/?group=7&id=1187
Message has been deleted

Chris Torrence

unread,
Feb 18, 2014, 4:33:40 PM2/18/14
to
Just to follow on David's post. David was kind enough to email me the disk image of DCOM. With the Videx Ultraterm, I was able to hack the code and eventually get some characters to appear on the screen, in 80 columns, but it was a mess. The characters were only appearing in the top half , while the bottom half of the screen had cruft.

So, in conclusion, I think the Videx Ultraterm is just different enough from the Videoterm that none of these terminal programs will work properly. Kinda funny that with Kermit I get only the left half of the screen, with DCOM I only get the top half. And Ascii Express doesn't handle the vt100 codes.

So, does anyone have a Videx Videoterm that they want to sell? Or do an even swap for an Ultraterm?

-Chris

Michael J. Mahon

unread,
Feb 19, 2014, 4:42:52 AM2/19/14
to
D Finnigan <dog...@macgui.com> wrote:
> Michael J. Mahon wrote:
>> D Finnigan <dog...@macgui.com> wrote:
>>>
>>> The only minor downside of DCOM 3.3 is that it runs only at 1200 baud, so
>>> it's somewhat slow, but not to the point where it's frustratingly slow.
>>
>> I'm curious--does DCOM use the shift-key mod, and does the accelerator
>> allow faster com speeds (or is it a configuration, not a processor speed
>> limitation)?
>
> Shift-key mod: yes. In fact, on the configuration screen, it asks if you are
> using shift-key on PB2, or if you've got a true lowercase, replacement
> keyboard.
>
> (I actually have neither, so I lied and said I had shift-key.)

That could lead to lower-case-only operation.

> The communication speed of 1200 baud was enforced by the program, so I
> presume that the Transwarp accelerator only accelerated the terminal
> emulation, and not the transmission speed.
>
> James Hayes, the programmer, probably did not have an accelerator for his
> Apple system, and so was getting dropped characters at faster speeds.

Agreed--he was writing for a 1MHz machine, and so needed to limit the
configuration options to max 1200 baud.

But with a faster processor, the configuration options can be changed to
allow higher data rates. I'd guess that your com code runs about 4x faster,
you could hope for at least 4800 baud.

Of course, if the code does timing with (presumed 1MHz) loops, then it
probably cannot work correctly while accelerated.
>
> Further reading:
>
> - DCOM VT100 driver specifications:
> http://macgui.com/usenet/?group=7&id=1187



--

Michael J. Mahon

unread,
Feb 19, 2014, 4:57:43 AM2/19/14
to
Michael J. Mahon <mjm...@aol.com> wrote:
> D Finnigan <dog...@macgui.com> wrote:
>> Michael J. Mahon wrote:
>>> D Finnigan <dog...@macgui.com> wrote:
>>>>
>>>> The only minor downside of DCOM 3.3 is that it runs only at 1200 baud, so
>>>> it's somewhat slow, but not to the point where it's frustratingly slow.
>>>
>>> I'm curious--does DCOM use the shift-key mod, and does the accelerator
>>> allow faster com speeds (or is it a configuration, not a processor speed
>>> limitation)?
>>
>> Shift-key mod: yes. In fact, on the configuration screen, it asks if you are
>> using shift-key on PB2, or if you've got a true lowercase, replacement
>> keyboard.
>>
>> (I actually have neither, so I lied and said I had shift-key.)
>
> That could lead to lower-case-only operation.

The shift-key mod is on by default in later-model //e's. In earlier //e's,
there's a solder point on the main board that "installs" the SKM.

A problem with it is that if a pushbutton is ever plugged into PB2, and
that button is pressed at the same the SHIFT key is pressed, the +5v line
is shorted to ground!

I protect against this by installing a 1k resistor across the solder point
(or the cut bowtie on later models).

D Finnigan

unread,
Feb 19, 2014, 3:15:31 PM2/19/14
to
Michael J. Mahon wrote:
> Michael J. Mahon <mjm...@aol.com> wrote:
>> D Finnigan <dog...@macgui.com> wrote:
>>> Michael J. Mahon wrote:
>>>> D Finnigan <dog...@macgui.com> wrote:
>>>>>
>>>>> The only minor downside of DCOM 3.3 is that it runs only at 1200 baud,
>>>>> so
>>>>> it's somewhat slow, but not to the point where it's frustratingly
>>>>> slow.
>>>>
>>>> I'm curious--does DCOM use the shift-key mod, and does the accelerator
>>>> allow faster com speeds (or is it a configuration, not a processor
>>>> speed
>>>> limitation)?
>>>
>>> Shift-key mod: yes. In fact, on the configuration screen, it asks if you
>>> are
>>> using shift-key on PB2, or if you've got a true lowercase, replacement
>>> keyboard.
>>>
>>> (I actually have neither, so I lied and said I had shift-key.)
>>
>> That could lead to lower-case-only operation.
>
> The shift-key mod is on by default in later-model //e's. In earlier //e's,
> there's a solder point on the main board that "installs" the SKM.
>

Right you are, of course! IIe users should just choose the true upper/lower
case keyboard option when configuring DCOM, even though it won't disable the
built-in PB2 circuit that you mentioned.

Chris Torrence

unread,
Feb 26, 2014, 11:01:15 AM2/26/14
to
Hi all,

Just a followup. I received a genuine Videx Videoterm card from drkenb.blujay.com, with the Videx Soft Switch.

Both DCOM and Kermit now work properly in 80-column mode on my Apple II+. For DCOM, I did need to change that one line of code from Jim Hayes' post:
http://macgui.com/usenet/?author=James+Hayes&id=468&group=7
Here is the code change from his post:
730: PR#3:IN#0:PR#0: CALL 8217 : REM VIDEX USERS ONLY!!!

The only (minor) annoyance is that on Kermit, the shift-key mod seems to be ignored, so you have to hit the right-arrow key twice every time you want an uppercase character. Side note: To get the right-arrow key to work for upper/lowercase shift, you need to disable VT100 cursor keys when configuring Kermit.

DCOM works great with the shift-key mod. I actually created a ProDOS version of DCOM 3.3 if anyone is interested. It loads a little faster. Maybe I'll create a BeagleBros ProntoDOS version as well, to compare load times.

Thanks for all of the help!
-Chris

aiia...@gmail.com

unread,
Feb 26, 2014, 12:50:34 PM2/26/14
to
On Saturday, February 8, 2014 10:27:29 PM UTC-8, Chris Torrence wrote:
> Hi all,
>
>
>
> I'm trying to use my Apple II+ as a vt100 terminal

for sale or trade I have:

apple II+ accelerator
videx terminal cards (many different brands and models)
II+ RAM cards.


Rich

ro...@ronnysvedman.com

unread,
Mar 23, 2014, 4:58:34 PM3/23/14
to
On Sunday, February 9, 2014 7:31:05 PM UTC+1, Ivan X wrote:
> I no longer consider usable any non-vt100 derived termcaps. Too much Linux software appears to be hardcoded to expect a vt100-type emulator rather than being fully abstracted from the display codes. In my experience a term program has to provide actual vt100 emulation to be used effectively with the Pi.

One way of getting around that problem is to use gnu screen on the pi side. IT respects your terminfo/termcap very nicely and is in itself a real good vt100 emulator. I used that to comfortably use my Visual550 serial terminal under modern-day linux just a couple of years ago, and also to log in from my Atari ST, and modem.mgr on my unenhanced IIe. With enough perseverance you will make it work =) also, you used lynx on the pi, try elinks; it is the same style, but with frame support, it renders modern webpages better. I would not be surprised if it were a possibility to make elinks use mousetext linedrawing, haha =)

ro...@ronnysvedman.com

unread,
Mar 23, 2014, 5:02:41 PM3/23/14
to
On Sunday, February 9, 2014 9:32:29 PM UTC+1, Chris Torrence wrote:
> Hi all,
>
>
>
> Thanks for all of the suggestions. Here's what I've found:
>
>
>
> Kermit: Connects fine, vt100 appears to work properly, but only uses 40 columns out of the 80 columns.
>
>
>
> Ascii Express Pro 3.46b: Connects fine, uses all 80 columns, but doesn't handle the vt100 "color" escape codes, so things like lynx or news readers (like Ivan's a2news) look all messed up.
>
>
>
> DCOM 3.21: Doesn't support VIDEX 80-column cards. Picking any of the other card options just results in garbage, so I can't tell if it is connecting.
>
>
>
> I'm out of ideas with getting proper vt100 support on the Apple II+, with a Videx 80-column card...
>
>
>
> -Chris

elinks can be configured to use no color and no linedrawing characters. Also be sure to use TERM=vt100, not xterm or linux in the pi shell

Leandro Polimeno

unread,
Aug 12, 2014, 10:00:15 AM8/12/14
to
For the record, since the site A2CLOUD makes reference to this post.

As Kermit manual says, the swith 2-6 on SSC must be linked.

Regards,

Polimeno

Ivan X

unread,
Aug 12, 2014, 11:14:13 AM8/12/14
to
I think the issue with ASCII Express is that it doesn't do VT-100 emulation, only VT-52, hence the only partial emulation. (I could be wrong about this.) But the idea you had earlier of setting term to vt52 and then using Screen is a very interesting workaround I'll have to try sometime. That is, if I ever want to re-learn ASCII Express.

N.N. Thayer

unread,
Aug 8, 2015, 10:03:15 AM8/8/15
to
Thought I'd contribute.

I too have been able to get Kermit and DCOM 3.3 working on my II+ with a regular VideoTerm, enabling me to use the II+ as a serial console to a Linux machine. VT100 emulation works fine, though each terminal program has required some configuration tweaks of course.

I was also able to get ASCII Express working with VT52 emulation. Not a surprise, and not something I'll really use given the limitations of VT52, but I just figured I'd try it.

N.N. Thayer

unread,
Mar 2, 2017, 5:59:16 PM3/2/17
to
So here's a couple of things I've found, having fooled around some more with DCOM and Kermit:

DCOM's Videx driver isn't great. The display doesn't properly size itself for my Monitor ///. I have to shrink the vertical size a *lot* to get everything to fit - and by the time it does, a strange phenomenon is noticeable where display is increasingly stretched as you go from top to bottom. I did try the code change that you mentioned, but all that did was stop the whole process: DCOM told me to switch to an output device (the VideoTerm) that I had already switched to.

I found this later patch from the author ...

https://groups.google.com/forum/embed/#!topic/net.sources.bugs/YlfXy0VWUGk

... but the results from that one were even worse: it switched over to the VideoTerm, but the vertical sync was completely out of whack, and there were several garbage characters in addition to the normal DCOM intro message.

Also, my II+ lacks the shift-key mod, and I haven't found a way to change cases within DCOM, even using the standard Videx escape characters.

KERMIT's Videx driver, on the other hand, is top-notch. No issues at all, as good as anything I've seen with Magic Window II or any other professional-quality application. My issue here seems to be with its VT100 emulation: either it's incomplete (the Kermit manual basically says as much), or there's some setting I need to tweak. Anything that requires scrolling where lines hit the bottom of the screen and then shift upward does not work: the bottom line just repeats over itself. This includes IRC and chat clients.

It's a shame, because Kermit really does fit the bill in all other respects. The display is wonderful and the full character set is available for you to input, although turning off VT100 arrow keys can be a pain.

N.N. Thayer

unread,
Mar 4, 2017, 8:35:29 AM3/4/17
to
Hmm... some old USENET posts and scans of Washington Apple Pi show evidence of Kermit-65 v3.88 - but it would seem that version is lost to the ages. Even Columbia only has v3.87 available. :(

I also came across posts about NovaTerm for the C64 and other emulators that claimed to do VT100, but fell short. Apparently this was not an uncommon issue back in the day. DCOM seems more like the real deal - if only its Videx driver wasn't so awful!

6502en...@gmail.com

unread,
May 24, 2017, 6:49:33 AM5/24/17
to
While trying different setups for tcpser on the PI I found the following:

On an Apple II+ you can use ASCII Express Pro although there is no VT-100 emulation.

Apple II+
Videx VideoTerm in Slot 3
AE configured with SSC in Slot 2 and 4800 baud, lower case display with Videx VideoTerm

If you want to use upper case letters just press ESC and the key you want.

0 new messages