I have been wondering about how exactly the video screens at
Melbourne's main railway stations (Flinders/Spencer St., the loop,
North Melbourne, Richmond and Box Hill) operate; i.e., what sort of
hardware is used to generate the displays.
What I know is:
- the screens appear to be television-grade CRTs (some monochrome,
some colour)
- Text on the screens is proportional and kerned (i.e., the
bounding rectangles of the 'W' and 'A' in "GLEN WAVERLEY"
overlap), suggesting that it's somewhat more elaborate and
upmarket than character-cell-based systems (like the teletext-like
systems used for V/Line and airport info).
- from what I've heard, the system has been around since some time
between the mid-1970s and mid-1980s, which means that by those
standards it would have been particularly state of the art and
expensive. And it pretty much rules out microcomputer solutions
(it's unlikely to be a rack of Amiga 1000s, for example, let alone
a PC).
My guess is that it would be some kind of minicomputer (perhaps
controlled with the green-screen terminals seen in station offices)
connected to character generators/frame buffers (possibly designed for
TV titling or similar purposes), and routed to a number of CRTs in
some form not unlike PAL composite video. Does anybody know more
about this system?
(Please post replies to this group; my email address may not work)
Thanks,
gc
What I can tell you is they were first introduced when the first part of the
City Loop opened, in 1981, though I guess the ones at Box Hill came later.
The only enhancements I can recall since then was adding the number of
minutes until train departure, and the time and destination of the following
train. Also the screens on the track side of them were added shortly after
installation at aboveground stations because of sunlight problems.
At the same time that the loop opened, the signs at Richmond indicating
platform/mins for various destinations were installed, though they didn't
work very well at first.
Daniel
--
Daniel Bowen, Melbourne, Australia
dbo...@custard.REMOVE.net.au
Melbourne public transport FAQ http://www.custard.net.au/melbtrans/
>The only enhancements I can recall since then was adding the number of
>minutes until train departure, and the time and destination of the following
>train. Also the screens on the track side of them were added shortly after
The following train information was only added in the last year or two
(and is extremely useful and much appreciated). The screens appear to
be monitors only, so the hardware driving them may have changed
several times since they were installed. So the kerning may be a
recent addition too. Anyone remember if the text was kerned in the
80s?
Peter Shute
Regards,
Bill.
"Eric Vincent" <vin...@heritage.com> wrote in message
news:3bb04f1c$1...@news.alphalink.com.au...
> > Anyone remember if the text was kerned in the 80s?
>
> Yes, it's always been like it is now. I remember being very impressed at
the
> time the loop opened, that the Ws & As overlapped their bounding boxes. At
> the time the primitive home computers (like Microbee and Commodore 64) had
> really chunky text, and the screens at the stations were much clearer with
> better formed characters.
>
> Eric
>
> "PShute" <psh...@melbpc.org.au> wrote in message
> news:bcf0rtkuubtsl34dh...@4ax.com...
Game Cat <groups.2...@spamgourmet.com> wrote in message
news:406f676f.01092...@posting.google.com...
> Hello;
>
> I have been wondering about how exactly the video screens at
> Melbourne's main railway stations (Flinders/Spencer St., the loop,
> North Melbourne, Richmond and Box Hill) operate; i.e., what sort of
> hardware is used to generate the displays.
>
> What I know is:
>
> - the screens appear to be television-grade CRTs (some monochrome,
> some colour)
Yes they are.
> - Text on the screens is proportional and kerned (i.e., the
> bounding rectangles of the 'W' and 'A' in "GLEN WAVERLEY"
> overlap), suggesting that it's somewhat more elaborate and
> upmarket than character-cell-based systems (like the teletext-like
> systems used for V/Line and airport info).
Yes, great effort went into this at the time the system was designed by GEC
(in a loft above the Paxman diesel workshop in Artarmon NSW) to a
MURLA specification.
> - from what I've heard, the system has been around since some time
> between the mid-1970s and mid-1980s, which means that by those
> standards it would have been particularly state of the art and
> expensive. And it pretty much rules out microcomputer solutions
> (it's unlikely to be a rack of Amiga 1000s, for example, let alone
> a PC).
No, it was all microcomputer controlled from the beginning. Although
computers only reached homes and offices in the 80's, the first Intel micro
came out round 1974. Industrial users were quick to find uses for them. The
PIDS use a heirachical system of 8085 single-board computers. The software
was developed using an Intel MDS (blue box) with a 2Mhz 8080 CPU, 64Kb RAM
and two 1/2Mb 8" Floppy disc drives running the Intel ISIS operating system.
> My guess is that it would be some kind of minicomputer (perhaps
> controlled with the green-screen terminals seen in station offices)
> connected to character generators/frame buffers (possibly designed for
> TV titling or similar purposes), and routed to a number of CRTs in
> some form not unlike PAL composite video. Does anybody know more
> about this system?
The video driving boards came from Matrox, but there was no need for a
minicomputer.
The characters were not scanned, they were made by choosing the largest
available Letraset,
sticking each character on a piece of graph paper, tracing the outline,
ruling lines every eight
columns, getting a secretary to type the resulting 0's and 1's into the
source file as data
statements. Then a poofy MURLA architect, complete with bow-tie, flew up
from Melbourne
and told GEC to add or remove bits until the letters looked just right to
him. Perhaps it really
was worth all the fuss if people still think that they look good.
>
> (Please post replies to this group; my email address may not work)
One of the guys who may remember more than I do about this used to go on RTM
excursions
before becoming one of the pioneers of industrial digital systems in this
country. Haven't seen
him on any trips for years. He knows who I am talking about! He did build a
small 7 1/4" bush
railway for his kids (or himself?). There is a web page for his line at
http://btr.cjb.net so maybe
you could email him via the BTR website and see what he remembers.
>
> Thanks,
>
> gc
"Puti Unggoy" <exa...@example.com> wrote in message news:<3bb1d...@news3.prserv.net>...
> GC, Some answers below:
>
> Game Cat <groups.2...@spamgourmet.com> wrote in message
> news:406f676f.01092...@posting.google.com...
> > Hello;
> >
> > I have been wondering about how exactly the video screens at
> > Melbourne's main railway stations (Flinders/Spencer St., the loop,
> > North Melbourne, Richmond and Box Hill) operate; i.e., what sort of
> > hardware is used to generate the displays.
> >
> > What I know is:
> >
> > - the screens appear to be television-grade CRTs (some monochrome,
> > some colour)
>
> Yes they are.
>
> > - Text on the screens is proportional and kerned (i.e., the
> > bounding rectangles of the 'W' and 'A' in "GLEN WAVERLEY"
> > overlap), suggesting that it's somewhat more elaborate and
> > upmarket than character-cell-based systems (like the teletext-like
> > systems used for V/Line and airport info).
>
> Yes, great effort went into this at the time the system was designed by GEC
> (in a loft above the Paxman diesel workshop in Artarmon NSW) to a
> MURLA specification.
Has the technology developed been used anywhere other than Melbourne?
> No, it was all microcomputer controlled from the beginning. Although
> computers only reached homes and offices in the 80's, the first Intel micro
> came out round 1974. Industrial users were quick to find uses for them. The
> PIDS use a heirachical system of 8085 single-board computers. The software
> was developed using an Intel MDS (blue box) with a 2Mhz 8080 CPU, 64Kb RAM
> and two 1/2Mb 8" Floppy disc drives running the Intel ISIS operating system.
Interesting. (Was ISIS related to CP/M in any way?)
And how custom-made are the boxes at stations? Are they (then) commercially
available microcomputers with add-on cards, or were they custom-built from
scratch?
> > My guess is that it would be some kind of minicomputer (perhaps
> > controlled with the green-screen terminals seen in station offices)
> > connected to character generators/frame buffers (possibly designed for
> > TV titling or similar purposes), and routed to a number of CRTs in
> > some form not unlike PAL composite video. Does anybody know more
> > about this system?
>
> The video driving boards came from Matrox, but there was no need for a
> minicomputer.
>
> The characters were not scanned, they were made by choosing the largest
> available Letraset,
> sticking each character on a piece of graph paper, tracing the outline,
> ruling lines every eight
> columns, getting a secretary to type the resulting 0's and 1's into the
> source file as data
> statements. Then a poofy MURLA architect, complete with bow-tie, flew up
> from Melbourne
> and told GEC to add or remove bits until the letters looked just right to
> him. Perhaps it really
> was worth all the fuss if people still think that they look good.
data statements? It's not written in Microsoft BASIC-80 or something
like that, is it?... :-)
Cheers,
gc
A small system was later made for SRA for Sydney Terminal. Several non-PIDs
applications followed, the biggest were the controllers for the Telecom Time
Division Cross-Connect switches. These are in nearly every Telstra exchange
to handle data traffic and were still being built until recently when
Siemens who had taken over part of the GEC business stopped building systems
in Sydney.
One of the guys who worked on the project went onto to do something with an
automated underground railway for Olympic Dam. This was modeled in HO to
allow the control system to be checked.
> > No, it was all microcomputer controlled from the beginning. Although
> > computers only reached homes and offices in the 80's, the first Intel
micro
> > came out round 1974. Industrial users were quick to find uses for them.
The
> > PIDS use a heirachical system of 8085 single-board computers. The
software
> > was developed using an Intel MDS (blue box) with a 2Mhz 8080 CPU, 64Kb
RAM
> > and two 1/2Mb 8" Floppy disc drives running the Intel ISIS operating
system.
>
> Interesting. (Was ISIS related to CP/M in any way?)
It pre-dated CPM and was much easier to learn and use, more like MS-DOS.
However ISIS took 16K of RAM whereas CPM only needed 4K. Very important for
the development of home computers.
> And how custom-made are the boxes at stations? Are they (then)
commercially
> available microcomputers with add-on cards, or were they custom-built from
> scratch?
Intel single board computers were used. They are Multibus cards, about 12 x
6", other cards plug into a passive back plane. You bought a box with power
supply and added boards as needed. Many firms used to make Multibus cards.
You put the box in a 19" rack in a cubicle.
> > > My guess is that it would be some kind of minicomputer (perhaps
> > > controlled with the green-screen terminals seen in station offices)
> > > connected to character generators/frame buffers (possibly designed for
> > > TV titling or similar purposes), and routed to a number of CRTs in
> > > some form not unlike PAL composite video. Does anybody know more
> > > about this system?
> >
> > The video driving boards came from Matrox, but there was no need for a
> > minicomputer.
> >
> > The characters were not scanned, they were made by choosing the largest
> > available Letraset,
> > sticking each character on a piece of graph paper, tracing the outline,
> > ruling lines every eight
> > columns, getting a secretary to type the resulting 0's and 1's into the
> > source file as data
> > statements. Then a poofy MURLA architect, complete with bow-tie, flew up
> > from Melbourne
> > and told GEC to add or remove bits until the letters looked just right
to
> > him. Perhaps it really
> > was worth all the fuss if people still think that they look good.
>
> data statements? It's not written in Microsoft BASIC-80 or something
> like that, is it?... :-)
No, Basic was an interpreter and this stuff had to run fast on slow
processors so assembly language or PLM80 was the way. PLM80 compiled to
machine code that was almost as good a human writing assembler for small
jobs. PLM made smaller code than assembler once the job got too big for a
human to remain keep in his head.
During the development of the job more and more ISIS software became
available from Intel: even COBOL, Fortran etc. One package that showed up
was Basic-80 from some little firm called Microsoft. We used it as a slow
toy for such things as formatting documents.
Hope this helps.
>The characters were not scanned, they were made by choosing the largest
>available Letraset,
>sticking each character on a piece of graph paper, tracing the outline,
>ruling lines every eight
>columns, getting a secretary to type the resulting 0's and 1's into the
>source file as data
>statements. Then a poofy MURLA architect, complete with bow-tie, flew up
Could this be a breach of copyright?
Peter Shute
> >The characters were not scanned, they were made by choosing the largest
> >available Letraset,
> >sticking each character on a piece of graph paper, tracing the outline,
> >ruling lines every eight
> >columns, getting a secretary to type the resulting 0's and 1's into the
> >source file as data
> >statements. Then a poofy MURLA architect, complete with bow-tie, flew up
>
> Could this be a breach of copyright?
>
> Peter Shute
Probably not, I think that they are both just implementations of a
traditional font designed by someone else in the preceeding century. Anyway,
by the time the architect had added bits on here, cut bits off there, moved
it up or down etc. the two would be sufficiently different, and of course
sticky letters don't have steps on their sloping and curved edges.
By the way, do the colours in Melbourne still change when a train is
departing due to the magnetic field caused by the traction current?
Does anyone know (remember) what the screen resolution is?
These two questions are related as in the recent system in Sydney they don't
use CRT monitors on the platforms due to this magnetic effect making modern
high resolution screens unreadable. Hence the very expensive LCD panels.
> By the way, do the colours in Melbourne still change when a train is
> departing due to the magnetic field caused by the traction current?
Very much so! A funky green effect goes across the corners of the screen.
> Does anyone know (remember) what the screen resolution is?
>
> These two questions are related as in the recent system in Sydney they
don't
> use CRT monitors on the platforms due to this magnetic effect making
modern
> high resolution screens unreadable. Hence the very expensive LCD panels.
...which will most likely need to be replaced in a few years. As mentioned
on the ng many, many times, these screens are starting to burn out after
only 18 months.
I would have thought that plasma would be more appropriate for this
application.
--
Jeremy Lunn
Melbourne, Australia
Find me on Jabber today! Try my email address as my JID.
The Sydney screens are plasmas.
It's like the screen on a laptop by today's standards. Many new desktop
computers have them now too.
>
> Eric
>
> "Paul" <eh...@o.com> wrote in message
> news:9p6pap$mi1$1...@bugstomper.ihug.com.au...
Is it one computer per CRT, or one per set of CRTs?
And how do they interact with the rest of the system? Do they act as an
output device of sorts, to which a more central computer sends the characters
it wants displayed and the computer puts them on the screen, or do they
actively fetch data from a database and keep the display updated?
Do the actual boxes that do the rendering know anything about train timetables
or are they just "glass printers" of a sort?
Where does the data come from?
A few years ago I saw an odd glitch on one of the screens (at Flinders St.,
or possibly Richmond); it was a screen announcing a train with an unusual time
and destination (it may have been East Camberwell), and it had a footnote
about changing to a bus for East Kew/Deepdene (possibly a relic of the East
Kew railway line; though wasn't that ripped up in the 1940s?)
> No, Basic was an interpreter and this stuff had to run fast on slow
> processors so assembly language or PLM80 was the way. PLM80 compiled to
> machine code that was almost as good a human writing assembler for small
> jobs. PLM made smaller code than assembler once the job got too big for a
> human to remain keep in his head.
>
> During the development of the job more and more ISIS software became
> available from Intel: even COBOL, Fortran etc. One package that showed up
> was Basic-80 from some little firm called Microsoft. We used it as a slow
> toy for such things as formatting documents.
>
> Hope this helps.
Indeed it does; thanks.
gc
>> It's like the screen on a laptop by today's standards. Many new
>> desktop computers have them now too.
Not 'Plasma' screens. Back lit 'active matrix' LCD.
> Thanks for the answer, but what I want to know is why the screens that
> laptops use (and some desktops) are called Plasma screens. I thought
> they were either Active Matrix or Passive Matrix which both use
> transistors of some description to switch on the pixels which are
> illuminated by a backlight.
Yes,
> My understanding is that plasma is a glowing gas sort of thing. As far
> as I know (and I don't know much) laptop screens don't use glowing gas
> anymore, but those old red displays (not LEDs) from 80s laptops were
> some sort of glowy gas, and could therefore truly be called plasma
> displays.
LCDs are 'transmissive' devices. They do not emit light. Hence LCD
screens are now back lit buy an electrolumisent pannel. These things are
not very effcient, and most of the light output is blocked by the LCD
anyway. LCDs will always have brightness problems as the light source has
to come through them.
Plasma displays are active. The pixels generate light on their own. They
can be much brighter than an LCD.
Plasma displays however suffer from screen burn. The 20 year old 7
segment plasma display on my Microwave oven is now very difficult to read
some combinations - the segments that have been driven hard are now quite
dim.
The CityRail plasma screens are probably being driven quite hard to get
the high brightness required. This more than any thing is probably causing
the burn damage now evident on the older screens. Its very obvious at
Martin Place where the prototype was installed, as they changed the
software when going into 'production' and used a slightly different screen
layout. Martin Place has since been updated to be like the others, but the
old display arrangement is burned into the screens and quite visible.
Didn't this also happen recently at Box Hill after PRIDE2 was commissioned?
Paul
After the rail line was ripped up, there was a railway bus that took its
place for many years.
Daniel
--
Daniel Bowen, Melbourne, Australia
dbo...@custard.REMOVE.net.au
See how I destroyed my VCR: http://www.toxiccustard.com/vcr/
How does a railway bus (in this context) differ from a conventional public
transport bus?
gc
The East Camberwell to East Kew Victorian Railways Road Motor Service (route
902) that continued to 1990 was always known as a "railway bus". Same
applied to the Sandringham service (901 now part of route 600). Even the
separate route numbers acknowledged that they were "different".
Also the St Kilda-Brighton "Electric Street Railway" (closed 1959) was also
known as the "railway tram".
And the now privatised MMTB bus services were always "Tramways Buses".
Private buses were just buses. Maybe the Melbourne-Brighton buses that used
to ply Swanston Street might have been a bit more noticed by the general
public. And perhaps the Elsternwick bus that used to run from Batman Avenue,
possibly because it once used old MMTB halfcab Leyland OPS4s.
Paul in Melbourne
"Game Cat" <groups.2...@spamgourmet.com> wrote in message
news:406f676f.01100...@posting.google.com...
The 901 Sandringham to Black Rock/Cheltenham/Southland was a better example
(or at least, one I'm more familiar with than the East Kew bus). It was
co-ordinated with the trains, that is every train would be met by a bus, and
the buses would wait if the train was delayed.
The 901 was so successful that sometimes 2 or 3 buses would run to meet a
single train. It was merged with route 600 last decade (which also,
historically, met trains at St Kilda Station, often with 2-3 buses in peak
hours).
Daniel
--
Daniel Bowen, Melbourne, Australia
dbo...@custard.REMOVE.net.au
If that's so, are other things affected, eg computer disks, signalling, radio equipment? (mobile phones?)
I suppose new trains aren't allowed to emit that much interferance / EMF.
Otherwise, instead of paying $$$ for LCD and plasma, why not just shield the CRTs? There are lots of companies on the web that do this, dunno about pricing though.
Ben Reuter
(remove spam/nospam)
What is PRIDE2?
gc
Where did this run?
Wasn't there also a "railway bus" going from Sandringham to somewhere else,
possibly where a short-lived railway line used to run?
> And the now privatised MMTB bus services were always "Tramways Buses".
> Private buses were just buses. Maybe the Melbourne-Brighton buses that used
> to ply Swanston Street might have been a bit more noticed by the general
> public. And perhaps the Elsternwick bus that used to run from Batman Avenue,
> possibly because it once used old MMTB halfcab Leyland OPS4s.
So that's why the National buses run at proper hours (not just to 7pm
weekdays like most buses...)
gc
Basically where route 600 runs now, but only as far as Brighton.
See http://www.melbbuslink.com.au/mbl_600.shtml#MAP
One of the broad gauge trams from this route is at the Railway Museum in
Williamstown.
> Wasn't there also a "railway bus" going from Sandringham to somewhere
else,
> possibly where a short-lived railway line used to run?
Yes, the 901. See route 600 again - the section from Sandringham to
Beaumaris. It took over from a horse tram along that route, which is why it
still runs along a Tramway Parade in Beaumaris.
> > And the now privatised MMTB bus services were always "Tramways Buses".
> > Private buses were just buses. Maybe the Melbourne-Brighton buses that
used
> > to ply Swanston Street might have been a bit more noticed by the general
> > public. And perhaps the Elsternwick bus that used to run from Batman
Avenue,
> > possibly because it once used old MMTB halfcab Leyland OPS4s.
>
> So that's why the National buses run at proper hours (not just to 7pm
> weekdays like most buses...)
National Bus, and Melbourne Bus Link - that's correct. Merely historical
accident! (Though MBL's routes were run mostly by private company
"Melbourne-Brighton Buslines", apart from a few years owned by the Met at
one point.)
Daniel
--
Daniel Bowen, Melbourne, Australia
dbo...@custard.REMOVE.net.au
The present version of automated passenger information on stations, which
incoporates platform displays and digitised announcements.
>On Sat, 29 Sep 2001 17:08:40 +1100, "Paul" <eh...@o.com> wrote:
>>
>> "Puti Unggoy" <exa...@example.com> wrote in message
>> news:3bb56...@news3.prserv.net...
>>
>> > By the way, do the colours in Melbourne still change when a train is
>> > departing due to the magnetic field caused by the traction current?
>>
>> Very much so! A funky green effect goes across the corners of the screen.
>
>If that's so, are other things affected, eg computer disks, signalling, radio equipment? (mobile phones?)
>I suppose new trains aren't allowed to emit that much interferance / EMF.
Not much they can do about it due to the size of the traction motors.
AC drive motors would have different magnetic effects to DC motors.
David Johnson
trai...@ozemail.com.au
http://www.ozemail.com.au/~trainman/
------------------------------------
These comments are made in a private
capacity and do not represent the
official view of State Rail.
C.O.W.S. Page 11.
When the tramway ran from St Kilda Stn a connecting track was provided from
it to the rail line so that the trams could be transferred and serviced by
the VR. It was broad gauge after all.