[Arfa Daily]

Booked in by shop as a 'Bingo calling unit'. It has a large 5 digit 7 seg
display made from discrete LEDs. Has two 5 digit BCD encoded switch units to
set start and end number. Has jack input to which is connected a hand-held
push button unit. Chip date codes indicate that this thing is from 1981. It
looks 'professionally' hand built, if you see what I mean. It has stickers
on with a company name, but that could be largely meaningless. Two problems
reported. First that it had a problem with the LEDs, and secondly that it
had "lost its memory".

LED problem was straightforward enough. One segment of one digit was not
lighting due to one LED in the series chain of five which made up the
segment, being open circuit. A replacement LED restored that.

Now, the second problem. For playing bingo, the start number would be set to
"1", and the end number to "90". All ok so far, and if this is done, it does
indeed generate random numbers in the range 1 to 90, one per button push.

The next requirement is that the unit is going to need to know for any game
in progress, which of the numbers between 1 and 90 that it has already
produced so that they are not duplicated, and this is where it all goes tits
up.

After a while, numbers *do* start to duplicate. If you set the start and end
numbers to say "1" and "10", it might take 15 button pushes to get all ten
numbers out, the other five being one or more times dups.

So I have a good look at what's in it. 2 x 4027 dual J-K, 3 x 4011 quad
NAND, 2 x 4081 quad AND,
1 x 4040 12 stage counter, 1 x 4020 14 stage counter, 1 x 555 timer and
last but not least, 1 x MK50395 decoder / driver. Now I don't know if I'm
just being dumb-arsed here, or having a senior moment, but I'm buggered if I
can see any chips in that line up that could form a memory and comparator
for potentially 99999 numbers. Or even for 90 numbers come to that ... If
this is what it was meant to do, even back in '81, I would have expected to
see a couple of CMOS memory chips, and a Z80 or 8080 processor, maybe.

What am I missing here ? Could it be that they are mistaken in thinking that
it can be used to properly call a game of bingo ? Is it in fact a raffle
ticket drawing machine, where you would set the serial numbers of the first
and last tickets sold out of a batch of perhaps 1000 and then draw maybe 10
prize ticket numbers with the chances of it duplicating any when it's only
generating ten random numbers out of a pool of 1000, being at worst, slim ?

Haven't managed to get hold of the owners again yet to check if they feel
that it has ever worked as a bingo calling machine, but at the moment, my
feeling is that it is a 'new acquisition' to some little club, and that it
has been given to them after sitting long forgotten in an attic somewhere
for 20 years.

Any thoughts anyone ?

Arfa

[William Sommerwerck]

My gut reaction is that you're correct -- it has no memory. I don't see
anything in it that could be used as a memory.

You're right -- the "correct" way to implement a bingo counter would require
a microprocessor and some memory.

[N_Cook]

Arfa Daily wrote in message
...
Booked in by shop as a 'Bingo calling unit'. It has a large 5 digit 7 seg
display made from discrete LEDs. Has two 5 digit BCD encoded switch units
to
set start and end number. Has jack input to which is connected a hand-held
push button unit. Chip date codes indicate that this thing is from 1981.
It
looks 'professionally' hand built, if you see what I mean. It has stickers
on with a company name, but that could be largely meaningless. Two
problems
reported. First that it had a problem with the LEDs, and secondly that it
had "lost its memory".

LED problem was straightforward enough. One segment of one digit was not
lighting due to one LED in the series chain of five which made up the
segment, being open circuit. A replacement LED restored that.

Now, the second problem. For playing bingo, the start number would be set
to
"1", and the end number to "90". All ok so far, and if this is done, it
does
indeed generate random numbers in the range 1 to 90, one per button push.

The next requirement is that the unit is going to need to know for any
game
in progress, which of the numbers between 1 and 90 that it has already
produced so that they are not duplicated, and this is where it all goes
tits
up.

After a while, numbers *do* start to duplicate. If you set the start and
end
numbers to say "1" and "10", it might take 15 button pushes to get all ten
numbers out, the other five being one or more times dups.

So I have a good look at what's in it. 2 x 4027 dual J-K, 3 x 4011 quad
NAND, 2 x 4081 quad AND,
1 x 4040 12 stage counter, 1 x 4020 14 stage counter, 1 x 555 timer
and
last but not least, 1 x MK50395 decoder / driver. Now I don't know if I'm
just being dumb-arsed here, or having a senior moment, but I'm buggered if
I
can see any chips in that line up that could form a memory and comparator
for potentially 99999 numbers. Or even for 90 numbers come to that ... If
this is what it was meant to do, even back in '81, I would have expected
to
see a couple of CMOS memory chips, and a Z80 or 8080 processor, maybe.

What am I missing here ? Could it be that they are mistaken in thinking
that
it can be used to properly call a game of bingo ? Is it in fact a raffle
ticket drawing machine, where you would set the serial numbers of the
first
and last tickets sold out of a batch of perhaps 1000 and then draw maybe
10
prize ticket numbers with the chances of it duplicating any when it's only
generating ten random numbers out of a pool of 1000, being at worst, slim
?

Haven't managed to get hold of the owners again yet to check if they feel
that it has ever worked as a bingo calling machine, but at the moment, my
feeling is that it is a 'new acquisition' to some little club, and that it
has been given to them after sitting long forgotten in an attic somewhere
for 20 years.

Any thoughts anyone ?

Arfa




I repaired something similar about 15 years ago, it had upper and lower
bounds set on thumbwheel sw on the top
I think you could set it for auction rostrum incrementing or a random number
generator but no repeat delete facility.


--
Diverse Devices, Southampton, England
electronic hints and repair briefs , schematics/manuals list on
http://home.graffiti.net/diverse:graffiti.net/

[jakdedert]

Arfa Daily wrote:
Booked in by shop as a 'Bingo calling unit'. It has a large 5 digit 7 seg
display made from discrete LEDs. Has two 5 digit BCD encoded switch units to
set start and end number. Has jack input to which is connected a hand-held
push button unit. Chip date codes indicate that this thing is from 1981. It
looks 'professionally' hand built, if you see what I mean. It has stickers
on with a company name, but that could be largely meaningless. Two problems
reported. First that it had a problem with the LEDs, and secondly that it
had "lost its memory".

LED problem was straightforward enough. One segment of one digit was not
lighting due to one LED in the series chain of five which made up the
segment, being open circuit. A replacement LED restored that.

Now, the second problem. For playing bingo, the start number would be set to
"1", and the end number to "90". All ok so far, and if this is done, it does
indeed generate random numbers in the range 1 to 90, one per button push.

The next requirement is that the unit is going to need to know for any game
in progress, which of the numbers between 1 and 90 that it has already
produced so that they are not duplicated, and this is where it all goes tits
up.

After a while, numbers *do* start to duplicate. If you set the start and end
numbers to say "1" and "10", it might take 15 button pushes to get all ten
numbers out, the other five being one or more times dups.

So I have a good look at what's in it. 2 x 4027 dual J-K, 3 x 4011 quad
NAND, 2 x 4081 quad AND,
1 x 4040 12 stage counter, 1 x 4020 14 stage counter, 1 x 555 timer and
last but not least, 1 x MK50395 decoder / driver. Now I don't know if I'm
just being dumb-arsed here, or having a senior moment, but I'm buggered if I
can see any chips in that line up that could form a memory and comparator
for potentially 99999 numbers. Or even for 90 numbers come to that ... If
this is what it was meant to do, even back in '81, I would have expected to
see a couple of CMOS memory chips, and a Z80 or 8080 processor, maybe.

What am I missing here ? Could it be that they are mistaken in thinking that
it can be used to properly call a game of bingo ? Is it in fact a raffle
ticket drawing machine, where you would set the serial numbers of the first
and last tickets sold out of a batch of perhaps 1000 and then draw maybe 10
prize ticket numbers with the chances of it duplicating any when it's only
generating ten random numbers out of a pool of 1000, being at worst, slim ?

Haven't managed to get hold of the owners again yet to check if they feel
that it has ever worked as a bingo calling machine, but at the moment, my
feeling is that it is a 'new acquisition' to some little club, and that it
has been given to them after sitting long forgotten in an attic somewhere
for 20 years.

Any thoughts anyone ?

Arfa


Well, for a proper game of Bingo, you need more than numbers, ie the
letters b, i, n, g, & o. They're combined with a random number to score
plays on a card, which might contain many duplicate numbers across the
numeric grid, but not in the same column.

So I'd say this could only be 'half' of a Bingo number generator.
Something has to generate the letters. Combined with that, the numbers
could indeed duplicate, just not any specified combination of number and
letter.

jak

jak

[Jerry]

On Jan 19, 7:17*am, "Arfa Daily" wrote:


So I have a good look at what's in it. *2 x 4027 dual J-K, *3 x 4011 quad
NAND, * 2 x 4081 quad AND,
*1 x 4040 12 stage counter, * 1 x 4020 14 stage counter, * 1 x 555 timer and
last but not least, 1 x MK50395 decoder / driver. *Now I don't know if I'm
just being dumb-arsed here, or having a senior moment, but I'm buggered if I
can see any chips in that line up that could form a memory and comparator
for potentially 99999 numbers. Or even for 90 numbers come to that ... If
this is what it was meant to do, even back in '81, I would have expected to
see a couple of CMOS memory chips, and a Z80 or *8080 processor, maybe.

First, a caveat. It's been nearly 20 years since I drifted away from
digital hardware design to the dark arts of software. That said....

The 555 and the counters must be part of the logic to _generate_ the
random number, based on randomly stopping the counters with the button
push, the other stuff looks like it would be used for "glue" logic for
the counters, and comparison for the start and stop numbers. You're
right, I don't see anything that would serve as a memory to be used to
check for duplicates.

I think what you have is just a random number generator, and the
"memory" was a paper-and-pencil operation.

Jerry

[Arfa Daily]

"jakdedert" wrote in message
...
Arfa Daily wrote:
Booked in by shop as a 'Bingo calling unit'. It has a large 5 digit 7 seg
display made from discrete LEDs. Has two 5 digit BCD encoded switch units
to set start and end number. Has jack input to which is connected a
hand-held push button unit. Chip date codes indicate that this thing is
from 1981. It looks 'professionally' hand built, if you see what I mean.
It has stickers on with a company name, but that could be largely
meaningless. Two problems reported. First that it had a problem with the
LEDs, and secondly that it had "lost its memory".

LED problem was straightforward enough. One segment of one digit was not
lighting due to one LED in the series chain of five which made up the
segment, being open circuit. A replacement LED restored that.

Now, the second problem. For playing bingo, the start number would be set
to "1", and the end number to "90". All ok so far, and if this is done,
it does indeed generate random numbers in the range 1 to 90, one per
button push.

The next requirement is that the unit is going to need to know for any
game in progress, which of the numbers between 1 and 90 that it has
already produced so that they are not duplicated, and this is where it
all goes tits up.

After a while, numbers *do* start to duplicate. If you set the start and
end numbers to say "1" and "10", it might take 15 button pushes to get
all ten numbers out, the other five being one or more times dups.

So I have a good look at what's in it. 2 x 4027 dual J-K, 3 x 4011 quad
NAND, 2 x 4081 quad AND,
1 x 4040 12 stage counter, 1 x 4020 14 stage counter, 1 x 555 timer
and last but not least, 1 x MK50395 decoder / driver. Now I don't know
if I'm just being dumb-arsed here, or having a senior moment, but I'm
buggered if I can see any chips in that line up that could form a memory
and comparator for potentially 99999 numbers. Or even for 90 numbers come
to that ... If this is what it was meant to do, even back in '81, I would
have expected to see a couple of CMOS memory chips, and a Z80 or 8080
processor, maybe.

What am I missing here ? Could it be that they are mistaken in thinking
that it can be used to properly call a game of bingo ? Is it in fact a
raffle ticket drawing machine, where you would set the serial numbers of
the first and last tickets sold out of a batch of perhaps 1000 and then
draw maybe 10 prize ticket numbers with the chances of it duplicating any
when it's only generating ten random numbers out of a pool of 1000, being
at worst, slim ?

Haven't managed to get hold of the owners again yet to check if they feel
that it has ever worked as a bingo calling machine, but at the moment, my
feeling is that it is a 'new acquisition' to some little club, and that
it has been given to them after sitting long forgotten in an attic
somewhere for 20 years.

Any thoughts anyone ?

Arfa
Well, for a proper game of Bingo, you need more than numbers, ie the
letters b, i, n, g, & o. They're combined with a random number to score
plays on a card, which might contain many duplicate numbers across the
numeric grid, but not in the same column.

So I'd say this could only be 'half' of a Bingo number generator.
Something has to generate the letters. Combined with that, the numbers
could indeed duplicate, just not any specified combination of number and
letter.

jak

jak

'Standard' UK bingo just requires the numbers 1 to 90. It is played on a
'set' of six tickets, each having 15 numbers arranged as 3 horizontal lines
of five numbers, placed within 9 vertical columns, one each for the 10's,
20's, 30's etc

http://www.coronetbingo.com/userimag...icketSmall.jpg

so all 90 numbers appear. Usually played as 'treble chance' games so prizes
for one line, two lines and a full house. Some games such as the high value
National game is played as a full house only, but you only find this sort of
game in proper commercial bingo clubs.

In those establishments, the numbers are computer generated, and supposedly
totally random, but they are actually far from it, with easily detectable
patterns, if you watch carefully ...

Arfa

[Arfa Daily]

"Jerry" wrote in message
...
On Jan 19, 7:17 am, "Arfa Daily" wrote:


So I have a good look at what's in it. 2 x 4027 dual J-K, 3 x 4011 quad
NAND, 2 x 4081 quad AND,
1 x 4040 12 stage counter, 1 x 4020 14 stage counter, 1 x 555 timer and
last but not least, 1 x MK50395 decoder / driver. Now I don't know if I'm
just being dumb-arsed here, or having a senior moment, but I'm buggered if
I
can see any chips in that line up that could form a memory and comparator
for potentially 99999 numbers. Or even for 90 numbers come to that ... If
this is what it was meant to do, even back in '81, I would have expected
to
see a couple of CMOS memory chips, and a Z80 or 8080 processor, maybe.

First, a caveat. It's been nearly 20 years since I drifted away from
digital hardware design to the dark arts of software. That said....

The 555 and the counters must be part of the logic to _generate_ the
random number, based on randomly stopping the counters with the button
push, the other stuff looks like it would be used for "glue" logic for
the counters, and comparison for the start and stop numbers. You're
right, I don't see anything that would serve as a memory to be used to
check for duplicates.

I think what you have is just a random number generator, and the
"memory" was a paper-and-pencil operation.

Jerry

Hi Jerry. Yes, that was exactly my take on what the hardware did. Just
wanted a few other opinions so that I can be sure of what I'm saying, when I
throw it back at the shop, who will then have to tell their customer ...
:-(

Arfa

[jakdedert]

Arfa Daily wrote:
"jakdedert" wrote in message
...
Arfa Daily wrote:
Booked in by shop as a 'Bingo calling unit'. It has a large 5 digit 7 seg
display made from discrete LEDs. Has two 5 digit BCD encoded switch units
to set start and end number. Has jack input to which is connected a
hand-held push button unit. Chip date codes indicate that this thing is
from 1981. It looks 'professionally' hand built, if you see what I mean.
It has stickers on with a company name, but that could be largely
meaningless. Two problems reported. First that it had a problem with the
LEDs, and secondly that it had "lost its memory".

LED problem was straightforward enough. One segment of one digit was not
lighting due to one LED in the series chain of five which made up the
segment, being open circuit. A replacement LED restored that.

Now, the second problem. For playing bingo, the start number would be set
to "1", and the end number to "90". All ok so far, and if this is done,
it does indeed generate random numbers in the range 1 to 90, one per
button push.

The next requirement is that the unit is going to need to know for any
game in progress, which of the numbers between 1 and 90 that it has
already produced so that they are not duplicated, and this is where it
all goes tits up.

After a while, numbers *do* start to duplicate. If you set the start and
end numbers to say "1" and "10", it might take 15 button pushes to get
all ten numbers out, the other five being one or more times dups.

So I have a good look at what's in it. 2 x 4027 dual J-K, 3 x 4011 quad
NAND, 2 x 4081 quad AND,
1 x 4040 12 stage counter, 1 x 4020 14 stage counter, 1 x 555 timer
and last but not least, 1 x MK50395 decoder / driver. Now I don't know
if I'm just being dumb-arsed here, or having a senior moment, but I'm
buggered if I can see any chips in that line up that could form a memory
and comparator for potentially 99999 numbers. Or even for 90 numbers come
to that ... If this is what it was meant to do, even back in '81, I would
have expected to see a couple of CMOS memory chips, and a Z80 or 8080
processor, maybe.

What am I missing here ? Could it be that they are mistaken in thinking
that it can be used to properly call a game of bingo ? Is it in fact a
raffle ticket drawing machine, where you would set the serial numbers of
the first and last tickets sold out of a batch of perhaps 1000 and then
draw maybe 10 prize ticket numbers with the chances of it duplicating any
when it's only generating ten random numbers out of a pool of 1000, being
at worst, slim ?

Haven't managed to get hold of the owners again yet to check if they feel
that it has ever worked as a bingo calling machine, but at the moment, my
feeling is that it is a 'new acquisition' to some little club, and that
it has been given to them after sitting long forgotten in an attic
somewhere for 20 years.

Any thoughts anyone ?

Arfa
Well, for a proper game of Bingo, you need more than numbers, ie the
letters b, i, n, g, & o. They're combined with a random number to score
plays on a card, which might contain many duplicate numbers across the
numeric grid, but not in the same column.

So I'd say this could only be 'half' of a Bingo number generator.
Something has to generate the letters. Combined with that, the numbers
could indeed duplicate, just not any specified combination of number and
letter.

jak

jak

'Standard' UK bingo just requires the numbers 1 to 90. It is played on a
'set' of six tickets, each having 15 numbers arranged as 3 horizontal lines
of five numbers, placed within 9 vertical columns, one each for the 10's,
20's, 30's etc

http://www.coronetbingo.com/userimag...icketSmall.jpg

so all 90 numbers appear. Usually played as 'treble chance' games so prizes
for one line, two lines and a full house. Some games such as the high value
National game is played as a full house only, but you only find this sort of
game in proper commercial bingo clubs.

In those establishments, the numbers are computer generated, and supposedly
totally random, but they are actually far from it, with easily detectable
patterns, if you watch carefully ...

Arfa


How interesting...an unsuspected illustration of the breadth of the gap
across the pond.

Here in the States, a BINGO card consists of a grid of five columns with
the top row being 'bingo', the letters. Every card has a number of
digits in the rows below the five letters (don't recall how many) which
are combined with the letter above to produce a winning call, for
example: B-29, G-50 etc.

Winning any combination of numbers each of the five BINGO rows (five
across, but not necessarily in a straight row) is a winning card...and
the winner calls, loudly enough for the caller to hear: "Bingo!"

There are probably other permutations with which I'm not familiar (There
are, see: http://en.wikipedia.org/wiki/Bingo_(U.S. ); but that's the
main gist of it.

jak

jak

[Arfa Daily]

Well, for a proper game of Bingo, you need more than numbers, ie the
letters b, i, n, g, & o. They're combined with a random number to score
plays on a card, which might contain many duplicate numbers across the
numeric grid, but not in the same column.

So I'd say this could only be 'half' of a Bingo number generator.
Something has to generate the letters. Combined with that, the numbers
could indeed duplicate, just not any specified combination of number and
letter.

jak

jak

'Standard' UK bingo just requires the numbers 1 to 90. It is played on a
'set' of six tickets, each having 15 numbers arranged as 3 horizontal
lines of five numbers, placed within 9 vertical columns, one each for the
10's, 20's, 30's etc

http://www.coronetbingo.com/userimag...icketSmall.jpg

so all 90 numbers appear. Usually played as 'treble chance' games so
prizes for one line, two lines and a full house. Some games such as the
high value National game is played as a full house only, but you only
find this sort of game in proper commercial bingo clubs.

In those establishments, the numbers are computer generated, and
supposedly totally random, but they are actually far from it, with easily
detectable patterns, if you watch carefully ...

Arfa
How interesting...an unsuspected illustration of the breadth of the gap
across the pond.

Here in the States, a BINGO card consists of a grid of five columns with
the top row being 'bingo', the letters. Every card has a number of digits
in the rows below the five letters (don't recall how many) which are
combined with the letter above to produce a winning call, for example:
B-29, G-50 etc.

Winning any combination of numbers each of the five BINGO rows (five
across, but not necessarily in a straight row) is a winning card...and the
winner calls, loudly enough for the caller to hear: "Bingo!"

There are probably other permutations with which I'm not familiar (There
are, see: http://en.wikipedia.org/wiki/Bingo_(U.S. ); but that's the
main gist of it.

jak

A similar game is played here during the intervals in the main bingo. Each
'main' session is divided into sub-sessions, where up to 7 games of the
six-ticket (aka "cards") bingo is played in silence, apart from the caller's
voice. There is some big money at stake in these games, hence the quiet and
concentration. Each number takes about 2 seconds to call with about a half
to one second gap between numbers. All 'in club' games have the numbers
'single-called' as in "Two and three - twenty three". Big 'network' games
(aka "link" games) played between many clubs, and the National Game (which I
once won a big share of !) have the numbers double called as in " Twenty
three. Two and three - twenty three". The overall call time remains the
same however, as the gap between calls now shrinks to almost nothing. It's a
pleasure to listen to and play with a good caller, who gets a real
sing-songy rhythm going. A poor caller is actually difficult to play along
with. The game is stopped by calling anything you like. "YES !!" is quite
common, but "BINGO" is also sometimes used. "LINE !" or "HOUSE !" sometimes,
also. One guy in the club that I used to go to regularly, used to stop the
game by calling "TOP BANANA !". Another always used to bellow "HERE you are
then !!"

Between these main sessions, a different type of bingo known as cash bingo
or party bingo or prize bingo is played. This makes use of a fully
electronic touch board with fixed numbers, set in each table top. The
numbers are again in columns, but this time, each column has a colour, the
equivalent I guess of each of your columns each having a letter assigned to
them. There are 16 numbers in total and most games are played to win by four
across, four down, four on either diagonal, or four corners. Some in-house
games are played as a full house where you need to light all sixteen
numbers, and all network or linked-up games between clubs are full house.
It's called very quickly with no gaps at all between the calls. Caller tends
to sound a bit like an auctioneer. I can't for the life of me remember the
colour allocations, but it is called like "Red five yellow forty red
fourteen blue thirty six" and so on like a machine gun firing. Total time
for a four winning number game is typically 20 seconds. Games are paid for
by inserting coins at the table.

With it being fully electronic, all wins are just automatically flagged at
the caller's station, but you still have to stop the game by waving your
hand over the 'win' lozenge on the board. Some callers will pause a short
time when they see on their board that there is a winner, but most just
carry right on, and if you miss it, that's just your fault. Prizes are paid
in cash and brought to your table by a runner. The electronic touch boards
are quite new. It used to be done with a board with mechanical shutters over
the numbers. The board that you were playing was identified to the system by
the coin mech that you had put your money in. You used to have to be careful
that the board number matched the coin mech number, and that no one had
moved the boards between tables ... ! The tabletop bingo is played with a
lot of noise in the club, being between sub sessions. People are up and on
the move getting food, drinks, playing slots etc (oh, and going outside to
the smoking area, since it has been banned inside any public place in the UK
now ).

All of which has absolutely nothing to do with electronics, but as you say,
it is interesting to highlight the differences between our cultures, which
many people perceive as being broadly the same.

I've never really managed to get my head around Keno when I've been in
Vegas, but that's another story ... :-)

Arfa

[William Sommerwerck]

The lack of subjective randomness is not necessarily a problem, as the
players cannot choose their own cards.

And the repetition of numbers probably doesn't matter, as long as they all
eventually come up.

[Arfa Daily]

"William Sommerwerck" wrote in message
...
The lack of subjective randomness is not necessarily a problem, as the
players cannot choose their own cards.

And the repetition of numbers probably doesn't matter, as long as they all
eventually come up.



First point agreed. See my reply to Baron. As to the other point, it does
matter as, if there is no comparison to check if a number has been generated
already and thus needs another to be generated if it has, there could be no
average game length, which is very important in a commercial club where they
have to be ready at very specific times to jump online for linked games.
Games are surprisingly consistent at 5 minutes apiece, including the win
checks.

Arfa

[Eeyore]

Arfa Daily wrote:

Booked in by shop as a 'Bingo calling unit'. It has a large 5 digit 7 seg
display made from discrete LEDs. Has two 5 digit BCD encoded switch units to
set start and end number. Has jack input to which is connected a hand-held
push button unit. Chip date codes indicate that this thing is from 1981. It
looks 'professionally' hand built, if you see what I mean. It has stickers
on with a company name, but that could be largely meaningless. Two problems
reported. First that it had a problem with the LEDs, and secondly that it
had "lost its memory".

LED problem was straightforward enough. One segment of one digit was not
lighting due to one LED in the series chain of five which made up the
segment, being open circuit. A replacement LED restored that.

Now, the second problem. For playing bingo, the start number would be set to
"1", and the end number to "90". All ok so far, and if this is done, it does
indeed generate random numbers in the range 1 to 90, one per button push.

The next requirement is that the unit is going to need to know for any game
in progress, which of the numbers between 1 and 90 that it has already
produced so that they are not duplicated, and this is where it all goes tits
up.

After a while, numbers *do* start to duplicate. If you set the start and end
numbers to say "1" and "10", it might take 15 button pushes to get all ten
numbers out, the other five being one or more times dups.

So I have a good look at what's in it. 2 x 4027 dual J-K, 3 x 4011 quad
NAND, 2 x 4081 quad AND,
1 x 4040 12 stage counter, 1 x 4020 14 stage counter, 1 x 555 timer and
last but not least, 1 x MK50395 decoder / driver. Now I don't know if I'm
just being dumb-arsed here, or having a senior moment, but I'm buggered if I
can see any chips in that line up that could form a memory and comparator
for potentially 99999 numbers. Or even for 90 numbers come to that ... If
this is what it was meant to do, even back in '81, I would have expected to
see a couple of CMOS memory chips, and a Z80 or 8080 processor, maybe.

What am I missing here ? Could it be that they are mistaken in thinking that
it can be used to properly call a game of bingo ? Is it in fact a raffle
ticket drawing machine, where you would set the serial numbers of the first
and last tickets sold out of a batch of perhaps 1000 and then draw maybe 10
prize ticket numbers with the chances of it duplicating any when it's only
generating ten random numbers out of a pool of 1000, being at worst, slim ?

Haven't managed to get hold of the owners again yet to check if they feel
that it has ever worked as a bingo calling machine, but at the moment, my
feeling is that it is a 'new acquisition' to some little club, and that it
has been given to them after sitting long forgotten in an attic somewhere
for 20 years.

Any thoughts anyone ?

I think you're spot on.

Graham