----- Original Message -----
| From: "John Wilson" <wilson at dbit.com>
| To: hecnet at Update.UU.SE
| Sent: Monday, 4 March, 2013 2:44:04 PM
| Subject: Re: [HECnet] Vt100 tester
|
| From: "Brian Schenkenberger, VAXman-" <system at TMESIS.COM>
|
| >Which tests???
|
| The worst one was with the letters going down both margins (part of
| set #1).
|
| >Are Eli Heffron still in biz?
|
| www.eli.com. I assume it's just "& son" now (Neal Heffron I think?
| he was cool). Their *awesome* dusty old shop is long gone (I was
| kicked
| out of there once in 6th or 7th grade) so I think it's
| mail-order-only,
| but they have plenty of stuff and their prices are reasonable.
| VT520s
| for less than my VT100 was!!!
Hey, looks like I can get a real DEC terminal for a reasonable price!
|
| >I've not tried these test on a plain ol' VT100. The oldest kit I've
| >used
| >was a VT220 and it passes all tests as expected.
|
| I'm sure the same is true for the author. Who'd think DEC would mess
| up
| their own compatibility? Plus the tests made no allowance for not
| having
| the AVO, which the 101 and plenty of 100s didn't.
|
| Anyway this is going to be OCD heaven -- more pointless details to
| obsess
| over! And I appreciate the warning to be on the lookout if I ever
| lose my
| mind and want to add VT220 emulation. I totally would have assumed
| it was
| just a matter of adding new stuff, not changing the old stuff.
|
| John Wilson
| D Bit
|
--
Cory Smelosky
http://gewt.net Personal stuff
http://gimme-sympathy.org Experiments
http://dev.gimme-sympathy.org Home experiments
Sorry for the side bar for those of you never been to Cambridge, MA -- but this is true DEC history ...
On Mon, Mar 4, 2013 at 2:44 PM, John Wilson <wilson at dbit.com> wrote:
www.eli.com. I assume it's just "& son" now (Neal Heffron I think?
he was cool). Their *awesome* dusty old shop is long gone
check out: http://www.eli.com/index.cfm?template=history
The old man was the first DEC reseller.
The old warehouse is still there. The son tried to get into the PC business in the late 80s and the shop that had all the cool old 'tronics in it closed, since the PC customers did not like the dirty and dark. Like John, other of us "nerdi" knights knew it as a holy place.
Clem
FYI: I bought my still working Tek 465B o'scope for $25 long ago (and it still has a DEC asset tag on the back ;-)
I still use to debug HW.
From: "Brian Schenkenberger, VAXman-" <system at TMESIS.COM>
Which tests???
The worst one was with the letters going down both margins (part of set #1).
Are Eli Heffron still in biz?
www.eli.com. I assume it's just "& son" now (Neal Heffron I think?
he was cool). Their *awesome* dusty old shop is long gone (I was kicked
out of there once in 6th or 7th grade) so I think it's mail-order-only,
but they have plenty of stuff and their prices are reasonable. VT520s
for less than my VT100 was!!!
I've not tried these test on a plain ol' VT100. The oldest kit I've used
was a VT220 and it passes all tests as expected.
I'm sure the same is true for the author. Who'd think DEC would mess up
their own compatibility? Plus the tests made no allowance for not having
the AVO, which the 101 and plenty of 100s didn't.
Anyway this is going to be OCD heaven -- more pointless details to obsess
over! And I appreciate the warning to be on the lookout if I ever lose my
mind and want to add VT220 emulation. I totally would have assumed it was
just a matter of adding new stuff, not changing the old stuff.
John Wilson
D Bit
From: "Jerome H. Fine" <jhfinedp3k at compsys.to>
At the risk of being extremely stupid, I would also appreciate
being able to run the same tests - just for the sake of knowing
what is being discussed. However, as a simple PDP-11 fellow,
I can't run VMS stuff and all I found was source for VMS.
Did you run the VTTEST code using a PDP-11? If so, which
operating system? Was that a real PDP-11 or E11?
It was Brian's great-escape.tmesis.com setup (which is awesome!!!).
For E11, I used the undocumented (because it's too easy to trip up and the
lack of DNS is annoying) TELNET command in the DOS version (only):
; great-escape.tmesis.com
; log in with VTTEST (no password) to test VT100 compatibility
net start ether-0 dhcp eth0:
def key f1 = "telnet 24.187.213.28"+chr$(13)
def key f2 = "VTTEST"+chr$(13)
So start E11 with this command file, then press F1 to connect, and F2
at the login prompt.
For the real VT100 and VT101, I used Linux. I don't have a getty on my
COM port so I did "telnet </dev/ttyS0 >/dev/ttyS0" and once that started up,
"stty ixon </dev/ttyS0" from another console so that XON/XOFF is honored
(lots of SUBs without that, and if you do it before starting telnet, the
ixon flag gets cleared when /dev/ttyS0 is reopened -- I don't use stty
enough to remember the switch that makes changes stick).
Then "open great-escape.tmesis.com" on the VT100 and you're off.
John Wilson
D Bit
"Jerome H. Fine" <jhfinedp3k at compsys.to> writes:
John Wilson wrote:
I wrote:
- if hidden "wrap" flag is set and in final column and [?7h in effect, <CRLF>
- if in final column, set "wrap" flag for next time (otherwise clear it)
- write character, advancing cursor unless in final column
The plot sickens ... I was starting to feel paranoid and wanted a second
opinion. So I replaced the UA9636/UA9639 driver/receiver chips in the VT101
that's been kicking around my basement for 10+ years waiting for me to get
around to that. Now it fails VTTEST too, but not the same way as the VT100!
(Characters only at the margins in the failing test, but with gaps.) The
132-col tests are all wrong but that's no surprise since the VT101 has no AVO.
Some quick experimentation shows that in the VT101, the rules are the same
as the VT100 ones above as long as the cursor doesn't leave the line where
a char was written to column 80 for the first time. But if it does (by any
means -- DCA, ESC M, <LF>, ESC [B etc.) then the "wrap" flag is cleared,
whether a printing character is displayed on the other line or not.
OK so DEC's *own* VT100 knockoff isn't even VT100-compatible. Nap time!
At the risk of being extremely stupid, I would also appreciate
being able to run the same tests - just for the sake of knowing
what is being discussed. However, as a simple PDP-11 fellow,
I can't run VMS stuff and all I found was source for VMS.
The target OS is immaterial. The output to the terminal or emulator to be
tested, however, is. You can run VTTEST by simply telnetting to my VTTEST
account at 'great-escape.tmesis.com.' Use the username: VTTEST There is
NO password needed as this is a captive account.
Did you run the VTTEST code using a PDP-11? If so, which
operating system? Was that a real PDP-11 or E11?
If a PDP-11 was used, is there a link available to VTTEST.SAV
(source as well would be even more appreciated if it is in
MACRO-11 or FORTRAN) so I can run the VTTEST code?
Jerome Fine
The VTTEST suite was written in C. Unless you have a C compiler for your
PDP-11s OS, your going to have to translate the test suite's source your-
self to Macro-11 or FORTRAN.
--
VAXman- A Bored Certified VMS Kernel Mode Hacker VAXman(at)TMESIS(dot)ORG
Well I speak to machines with the voice of humanity.
John Wilson <wilson at dbit.com> writes:
I wrote:
- if hidden "wrap" flag is set and in final column and [?7h in effect, <CRLF>
- if in final column, set "wrap" flag for next time (otherwise clear it)
- write character, advancing cursor unless in final column
The plot sickens ... I was starting to feel paranoid and wanted a second
opinion. So I replaced the UA9636/UA9639 driver/receiver chips in the VT101
that's been kicking around my basement for 10+ years waiting for me to get
around to that. Now it fails VTTEST too, but not the same way as the VT100!
(Characters only at the margins in the failing test, but with gaps.) The
132-col tests are all wrong but that's no surprise since the VT101 has no AVO.
Some quick experimentation shows that in the VT101, the rules are the same
as the VT100 ones above as long as the cursor doesn't leave the line where
a char was written to column 80 for the first time. But if it does (by any
means -- DCA, ESC M, <LF>, ESC [B etc.) then the "wrap" flag is cleared,
whether a printing character is displayed on the other line or not.
OK so DEC's *own* VT100 knockoff isn't even VT100-compatible. Nap time!
OK. You've convinced me to dust off my old VT100 and see what/how it does
with these VTTEST test. I have some other work to address first before my
international travel this week and VMS Bootcamp the week after but, if I'm
able to squeeze out a few precious moments, I'll look into it.
--
VAXman- A Bored Certified VMS Kernel Mode Hacker VAXman(at)TMESIS(dot)ORG
Well I speak to machines with the voice of humanity.
John Wilson <wilson at dbit.com> writes:
From: "Brian Schenkenberger, VAXman-" <system at TMESIS.COM>
Beware of the versions of VTTEST out there. Someone in VMS engineering had
tried to make it DECC compatible and horked it completely. I believe that
the version I have at the DECUS Library Compendium (DECUSlib.com) Freeware
CD version 4 is the only properly working version. Don't dowload any try
to use any of the later versions.
Not like I'm *looking* for trouble here but ... I was horrified by how badly
Ersatz-11(my PDP-11 emulator)'s VT100 emulation (which I've always thought was
good) was failing test #1 (Test of cursor movements), and then mystified when
I logged the output and saw it depending on behavior that I'm 100% sure I
tested on real VT100s centuries ago and convinced myself they don't work the
way the test expects. Then I got sidetracked with mounds of other fixes...
But today's exchange reminded me to unearth a real VT100 and schlep it out
within cable's reach of something which can telnet to great-escape.tmesis.com
and ... OK yes I have to investigate and fix my handling of setting tab stops
and of course I don't bother with smooth scroll at all (it's not reasonable
in text mode on the MDA/CGA/HGC/VGA) but other than that I was *delighted*
to see the real VT100-AA (with AVO) fail the tests in exactly the same way
as E11's VT100 emulation.
Which tests???
Am I insane, or do I have a counterfeit VT100 (would Eli Heffron steer me
wrong in ~1985?) or have these tests really not been tried on a plain ol'
VT100 in eons and there's bit rot no one has noticed? Or some fourth option
I'm not thinking of?
Are Eli Heffron still in biz?
I've not tried these test on a plain ol' VT100. The oldest kit I've used
was a VT220 and it passes all tests as expected.
--
VAXman- A Bored Certified VMS Kernel Mode Hacker VAXman(at)TMESIS(dot)ORG
Well I speak to machines with the voice of humanity.
>John Wilson wrote:
I wrote:
- if hidden "wrap" flag is set and in final column and [?7h in effect, <CRLF>
- if in final column, set "wrap" flag for next time (otherwise clear it)
- write character, advancing cursor unless in final column
The plot sickens ... I was starting to feel paranoid and wanted a second
opinion. So I replaced the UA9636/UA9639 driver/receiver chips in the VT101
that's been kicking around my basement for 10+ years waiting for me to get
around to that. Now it fails VTTEST too, but not the same way as the VT100!
(Characters only at the margins in the failing test, but with gaps.) The
132-col tests are all wrong but that's no surprise since the VT101 has no AVO.
Some quick experimentation shows that in the VT101, the rules are the same
as the VT100 ones above as long as the cursor doesn't leave the line where
a char was written to column 80 for the first time. But if it does (by any
means -- DCA, ESC M, <LF>, ESC [B etc.) then the "wrap" flag is cleared,
whether a printing character is displayed on the other line or not.
OK so DEC's *own* VT100 knockoff isn't even VT100-compatible. Nap time!
At the risk of being extremely stupid, I would also appreciate
being able to run the same tests - just for the sake of knowing
what is being discussed. However, as a simple PDP-11 fellow,
I can't run VMS stuff and all I found was source for VMS.
Did you run the VTTEST code using a PDP-11? If so, which
operating system? Was that a real PDP-11 or E11?
If a PDP-11 was used, is there a link available to VTTEST.SAV
(source as well would be even more appreciated if it is in
MACRO-11 or FORTRAN) so I can run the VTTEST code?
Jerome Fine
I wrote:
- if hidden "wrap" flag is set and in final column and [?7h in effect, <CRLF>
- if in final column, set "wrap" flag for next time (otherwise clear it)
- write character, advancing cursor unless in final column
The plot sickens ... I was starting to feel paranoid and wanted a second
opinion. So I replaced the UA9636/UA9639 driver/receiver chips in the VT101
that's been kicking around my basement for 10+ years waiting for me to get
around to that. Now it fails VTTEST too, but not the same way as the VT100!
(Characters only at the margins in the failing test, but with gaps.) The
132-col tests are all wrong but that's no surprise since the VT101 has no AVO.
Some quick experimentation shows that in the VT101, the rules are the same
as the VT100 ones above as long as the cursor doesn't leave the line where
a char was written to column 80 for the first time. But if it does (by any
means -- DCA, ESC M, <LF>, ESC [B etc.) then the "wrap" flag is cleared,
whether a printing character is displayed on the other line or not.
OK so DEC's *own* VT100 knockoff isn't even VT100-compatible. Nap time!
John Wilson
D Bit
From: Johnny Billquist <bqt at softjar.se>
Could you expand on how it fails?
Duh, sorry!! Here's the most egregious case:
Test of autowrap, mixing control and print characters.
The left/right margins should have letters in order:
L l
M m
mN n
O o
P p
Q q
qR r
S s
T t
U u
uV v
W w
X x
Y y
yZ z
Push <RETURN>
(This is a screen copy/paste from E11 but it looks identical on the real VT100.)
I've apparently overwritten the log file that I'd marked up with what the
problem was, but IIRC it had to do with moving the cursor around in between
writes to the last column of the screen. My testing eons ago showed that on
a real VT100, writing printing characters is this simple:
- if hidden "wrap" flag is set and in final column and [?7h in effect, <CRLF>
- if in final column, set "wrap" flag for next time (otherwise clear it)
- write character, advancing cursor unless in final column
(Plus the real VT100 has a bug if autowrap happens at the bottom of the
screen when the scroll region ends higher up, where the cursor does the right
thing but the wrapped character appears at column 80 instead of column 1 --
but that can't be intentional and it's fixed on the later terminals.)
ESC 7 / ESC 8 save/restore the "wrap" flag but other than that it survives
most things.
But the VTTEST output seems to expect that other things you do will clear
the flag. The source code I found (maybe not official?) isn't particularly
commented so it's hard to tell which parts are supposed to be clever tests
and which are generic screen writes.
John Wilson
D Bit
