This technology history page contains a photograph which is one of several belonging to the photo gallery pages which are part of several pages relating to the invention of the world's first automatic totalizator in 1913 and Automatic Totalisators Limited, the Australian company founded to develop, manufacture and export these systems.
Old Dividends and Odds Display Generator
This is an image of the equipment used to generate odds and dividends displays long ago, prior to the advent of Video Character Generators. It is an interesting glimpse at the march of technology and the way it used to be. This image was taken at the Gabba Greyhounds in Brisbane and was part of the equipment in the old main tote house in use when the old electromechanical Julius Tote was in operation. The PDP11 minicomputer based totalisator system which I moved to Queensland with replaced this equipment and the Julius Tote. The new system had an electronic CCTV display generating system manufactured by Automatic Totalisators Limited. This allowed the computer systems to directly update the screen displays via the video character generators. This image shows how video character displays were generated prior to this development.
More after the image...
Click on the image to go back to the photo gallery
This is an Automatic Totalisators photograph from Neville Mitchell's collection
For the Dividends Display, a video camera was placed in front of a static display board where moulded characters were manually placed to provide the required information. This can be seen bottom right in the image above. Half of the display board is obscured by the carry case and the first line on the board reads RACE 10. The complete contents of this board can be read on the TV set in the image above. The camera video was then switched to the closed circuit television system for distribution to the public display televisions around the track. Neville Mitchell, a long serving Automatic Totalisators Engineer and Manager, wrote the following about this. Neville mentions Charlie Barton who was the Chief Engineer of the Julius Totes in Queensland and The Gabba installation was one of the totalisator systems in his jurisdiction:
The results and dividend display used two MOVITEX boards that were quickly set up when the dividends were calculated, these boards were common in picture theatre foyers noting upcoming films etc.. I was never at the Gabba when this system was in use. Charlie often remarked on how well it all worked.
For the Odds Display, we have the prominent panel with all the rotary switch knobs on it, which for the want of a better name I will call the Odds Input Panel. This panel has the television monitor on top of it and is to the left of the dividends display camera. It was used to input odds which were displayed on an Odds Display Panel which is not visible in the image above. This Odds Display Panel had an associated video camera like the dividends display camera so the odds could be transmitted on the CCTV system. The dividends display board was updated whilst the television system was still showing the odds display.
On the left hand edge of the above image, the right hand end of a Counter Tote can be seen sitting on the bench. Between the Counter Tote and the Odds Input Panel is a black oblong Counter Tote Display Panel. The Counter Tote is described in another image file in the Photo Gallery of this website, accessible by clicking on the image above and selecting the second following image thumbnail with corresponding descriptive text starting with An ATL product information brochure on the Counter Totes.
Lying on top of the Odds Input Panel, to the right of the television monitor, is a VDA (Video Distribution Amplifier.) This amplifies the signal from the camera and makes it suitable for transmission on coaxial cable and also provides the ability to support multiple transmission cables, which gives rise to the word Distribution in the name Video Distribution Amplifier. One cable will drive the neighbouring TV and presumably multiple cables will go to different locations around the track. At these destinations around the track other VDAs can be cascaded providing support for numerous television sets in each area. Call me picky, but this is a sloppy piece of installation work and looks very much like finding a home for this VDA was an afterthought. This VDA looks quite compact and modern in comparison to the IRT VDAs, which were plug-in modules with a substantial metal case and were prevalent around the Queensland tracks when I started work there. There were many IRT VDAs that were still working when I retired over three decades later, although there had been additional frequency division multiplexing RF video distribution systems installed during my time as a result of expansion of facilities. Consequently, as the VDA in the image above seems to be out of place, I suspect it was not part of an original installation.
Following is an image of a couple of Counter Totes which are also at the Gabba Greyhounds. They provide a better view of what is sitting on the bench at the left hand edge of the image above. Having just mentioned The Gabba again I will take this opportunity to explain where this name originated. The Gabba Greyhounds track used to surround the Brisbane Cricket Ground which is in the suburb Woolloongabba. Gabba in The Gabba Greyhounds is a contraction of this suburb name and The Gabba is the name of the Brisbane Cricket Ground.
It can be seen by the pull out handle on the near side of the nearest Counter Tote in the image below, that this was portable equipment. There are handles at both ends of these Counter Tote cases and I suspect it took two people to carry them, although I have seen technicians performing herculean tasks moving equipment. Similarly, in the image above, a handle can be seen on the right hand side of the Counter Tote at the left hand edge of the image, as well as one on the right hand side of the carry case for the Dividends Display and Odds Input Panel.
A Counter Tote at The Gabba Greyhounds Brisbane
Following are more recollections from Neville Mitchell. He mentions George Klemmer, who I remember very well as an iconic Automatic Totalisators design engineer. He also mentions Charlie Barton who I worked with and knew well. Charlie was the last Chief Engineer of the Julius Totes in Brisbane. As the Julius Totes were replaced by the PDP11 computer based tote systems in 1979, Charlie continued to work for the company on the new computer totes. Neville also mentions Charlie doing the upgrade, of which the equipment in the image at the top of this page is a part, which took place during a period when the Gabba acquired new facilities. This would have taken place some time before 1979 when this equipment was rendered redundant. The Odds Input Panel I mentioned, Neville refers to as the input console:
I remember this set of CCTV gear designed especially for the Gabba Greyhound track.
The input console was designed by a young cadet engineer working for George Klemmer. Each odds display, two digits and a decimal point, were controlled by rotary switches and a toggle switch for the decimal point, as shown on the photograph of the display console.
The rotary switches were custom made by AWA's OAK Switch division at Ashfield. The switch had seven switch banks one for each digit segment. The knobs were off the shelf standard instrument 12 numbered position, made by Agis a local company.
The small monochrome CCTV camera and its distribution system was supplied by others most likely AWA Rediffusion.
The system included a W & P counter tote, meant to be transportable. The doubles /Quinella was on punch tape machines.
Charlie Barton did the installation work it was about the time when the Gabba got new facilities, Grandstand upgrade, new tote selling line under the control room.
The Odds Display Panel, which is controlled by what Neville calls the input console, is not visible in the images above. There were multiple devices like the one in the image at the top of this page at the Gabba, when I first started working on the replacement of these systems in Brisbane. Neville added the following relating to the miniature Odds Display Panel used for the CCTV system shown in the image at the top of this page minus the panel. Neville describes the display elements that display a single character that are used to build the complete Odds Display Panel:
The small Gabba 7 segment digital display was a glass encapsulated device with 7 filaments fixed to 8 connecting pins at the back, each device was about 12 mm wide 20 mm high and 10 mm deep. They were from Asia possibly Taiwan the Stanley Lamp Company which was also making our small lamps automotive style we used in 6" and 3 -1/4" lamp displays.
It is now January 2018 and I am adding the following explanation that Neville has just sent of the display panel which is visible in the image of the top of this page. It is part of the Counter Tote and can be seen between the Odds Input Panel on the right and the section of the counter tote on the left hand edge of the image. It occurs to me that the last full year the equipment in the above images operated was 1978, which happens to be forty years ago. The computer totalisators that my team and I introduced superseded this equipment in January 1979. Neville wrote:
Re Panel on the left:: It's a commonly used counter tote with mechanical Sedco counters, the design allowed expansion up to 24 starters. If you look you can see the white digits on some of the lower rows of counters.
I installed a Quinella version of these counter totes in Sabah circa 1970.
The panel with all the knobs is for 9 starters win & place which set up the odds on the seven segment displays videoed for distribution.
Of course during a meeting, all the knobs Neville mentions, on the Odds Input Panel, would have to be regularly manipulated during a race, to reflect the changing trends of betting. It is interesting to note, on the subject of odds displays prior to the advent of video character generators, that when Automatic Totalisators Limited started to introduce the world's first computer totalisators, a method of providing odds display was to have the computer systems print out the odds at regular intervals and these printed sheets were then placed in front of a video camera which broadcast the odds printouts around the track on the CCTV system. This is a logical step forward, with very little change required to the video side of the pre-existing installation, yet providing a significant improvement in the update rate as printing a page of odds and placing it in front of a camera is a lot quicker than mechanically twiddling all the knobs. No doubt this printout method would have had a significant improvement in accuracy as well.
When looking at the equipment in the image at the top of this page, I wonder what the delivery of odds information from this system might have been like. Regarding my experience with odds displays, which was during the era when computer totes were already well established, any perceived delay of an expected odds update by the savvy punters, even a few seconds, would generate complaints. This was particularly so when the shrewd high value punters had a means of comparing odds information, for instance when Teletext became available and odds could be retrieved from TV sets with this feature. Naturally they believed that the on-course displays should be ahead of those coming from a Television station.
Two things immediately come to mind when thinking about the odds displays generated by the equipment shown in the image at the top of this page. What is the update repetition rate and what is the time taken to perform the update. All the odds display systems I have worked on, have divided the lead-up to a race into update rate periods. For example longer than two hours till the race starts, between two and one hour to the race, between one and half an hour, between half an hour and 15 minutes, between 15 minutes and 5 minutes and within five minutes to the race start. The fastest update rate was usually every 30 seconds, which in this example would be in the last five minutes. At the other extreme, outside of two hours to the race, 30 minute updates were generally acceptable.
Even if someone was permanently assigned to reading the investment totals on the Counter Tote Display Panel and calculating the odds, then updating the knob and switch driven Odds Input Panel, I wonder how often a new set of odds would be displayed. It is probable that the person doing this would have had other duties to perform as well. One thing is fairly evident, new information would not have been quick to be displayed. As I have indicated however, it is only as the race time draws near that more frequent updates are required.
Regarding the time taken to update screens, in the computer systems I worked on, the update took place in a split second. Obviously that is radically different to what is achievable using the equipment in the image at the top of this page, as reading through a list of odds and entering every odd using the knobs and switches would have been time consuming. With 90 knobs and their associated switches representing 45 two digit odds with a decimal point to control, a gun operator may achieve an odd in 5 seconds giving an update time nearing 4 minutes. In this manual operation, accuracy will be inversely proportional to speed and a lengthening of the update time would probably be necessary to improve accuracy with most operators. Then consider that this equipment was operating a race at a time for a single venue on what would now be considered limited pools. Supporting multiple pools or venues required duplication of equipment in the pre computer era. During the high value punter era, probably in its heyday during the 1980s and 1990s, there were instant complaints if there was any delays whatsoever in the 30 second repetition rate period closest to the race. Odds displays were of paramount importance during my time. From an engineering perspective, I recall we ensured that the write operations to the video frame buffer of the CCTV generating equipment was done during the television vertical retrace interval to eliminate the flashing effect associated with updating it during the active line period. In conclusion, the odds displays are very important to the tote, as Roger Penwarden, an ex ATL Queensland Branch Manager once commented, the odds displays are the shop window of the tote, without them punters do not spend.
Neville made the following comment about my musings in the last paragraph:
Regarding Gabba odds displays, your remarks re how long it took to update would be firstly how many TAB inputs were received and the Counter Tote totals added and using a odds calculator adjust the 90 knobs. In those days the TAB very much overshadowed the oncourse wagering total so there was not a lot of odds movement.
Neville appropriately describes above how the equipment shown in the images above, was used during the TAB era, however I will point out that these systems originally operated without the TAB as they pre-dated the TABs.
In conclusion, although computerisation was a milestone in the evolution of totalisator systems, the prior technologies including mechanical and electromechanical equipment served the industry extremely well for several decades. It is only when you become accustomed to something better that you realise the limitations of what you had. Similarly, when totalisator systems first became mechanised, the previous manual systems seemed to many, to be intolerable.
- Thanks to Neville Mitchell for the images above and his recollections of these display systems.
- Thanks to Charlie Barton for all the things he taught me about these historic systems.