NotesMade by Auto Ray Pty Ltd, Melbourne, Australia, 1949
John (Jack) Thomas Cooper (1901-1988), a Melbourne-based bookmaker entered into a partnership about 1940 with Joseph Ferris Reilly, a photographer with the Melbourne Sun Herald and part-time inventor.
John Cooper's experience as a bookmaker and punter inspired something of a personal crusade for the eradication of the corrupt practice of some race club judges to falsely declare the winners of close races. This led to his involvement in photo-finish technology. His aim was that it be used by all race clubs.
Reilly had invented a type of fish eye lens camera, which captured panoramic images. When he met Cooper he was in the process of developing an improved camera for photographing moving objects. Reilly's design used a still camera with a series of lenses linked together, with a rotating column inside with a film negative wrapped around it.
Cooper approached the Physics faculty of Melbourne University to assist with its development. The resulting prototype was a fifteen lens camera which took a total of 360 pictures per second. Following its successful trial at the Sandown Park greyhound track in Melbourne, Cooper and Reilly established a photo finish service, Auto-Ray P/L. The first multi-lens cameras were made using war surplus material, including the lenses which were sourced from Wellington bombers, where they had been used as part of the bomb sight cameras. According to Cooper's daughter, Judith Alexandrovics, the venture was initially funded from proceeds generated by his champion greyhound 'Tinkerman'. A patent for the multiple lens camera was applied for in 1945 and granted in 1947.
Conventional high-speed still cameras and the motion picture camera were unsuitable for photo-finish photography. The time interval between successive photos (or frames) was not sufficiently short to achieve the precision required to accurately capture the instant at which the foremost portion of the horses or other moving objects reached the winning post.
A new type of camera - the circular flow camera - had been invented in the 1930s especially for the purpose of photographing moving objects by Lorenzo Del Riccio, a well-known motion picture engineer who headed Paramount Pictures technical laboratories The first racing club to make use of Del Riccio's 'Photo-Chart' camera for photo-finishes was the famous Del Mar Turf Club in California at its inaugural meeting in 1937.
Unlike conventional cameras the circular flow camera had no shutter, operating by a strip film moving horizontally across a fine vertical slit located in the focal plane. This limited the field of vision to no more than a few inches, the restricted field being aligned with the vertical line on the winning post on which the lens was focused. The strip film moved across the slit in the opposite direction to the race and at substantially the same speed as the rate of movement of the image of the horses as it passed the finishing line. This kept the image of the horses more or less stationary with respect to the film. As soon as the first horse started to pass over the line, the camera began to record its image on the moving film from the nose backwards along the length of the body in succession. This produced a strip photographic record of the horses as they passed the vertical plane (winning post).
In 1945 B T Giles purchased the Australasian sales and manufacturing rights to the Photo Chart photo-finish system and began promoting it to racing clubs. The first to install it was the Sydney Turf Club at Rosehill Racecourse in 1946.
A disadvantage of this technology was that a photo of the horses at the finishing line was not obtained. Further, as the film was in continuous motion during the finish of the race stationary objects did not appear in the photo. The absence of an image of the winning post in the photograph created the potential for doubt as to whether the camera was actually in alignment with the finishing line at the time a photograph was taken. (This problem was later addressed by a drum ('spinner') set into the winning post, revolving at the same speed as the average horse, reproducing on film as vertical black and white lines. Their absence in the background of the photograph indicated that the camera had been looking either to the left or right).
Around 1946 a Melbourne-based scientist and inventor, Bertram Alston Pearl, developed the 'Camera Graph', an improved version of the continuous strip photo-finish camera. It incorporated a special type of neon light which was set in the winning post. This produced a vertical line in the photo which equated to the position of the winning post. Apart from confirming the proper alignment of the camera with the winning line, this 'neon tube tracery' also scaled the print into time intervals of one hundredth (1/100) of a second, the speed of the camera, making it possible to read the winning margin in time by direct examination and therefore providing greater accuracy in the determination of results. The Camera Graph system was therefore an important innovation in photo-finish technology. A patent for the Camera Graph was applied for in January 1947 (granted in August 1950).
Pearl entered into a partnership with well-known Melbourne fashion photographer, Athol Shmith, who worked on the faster processing of negatives. The Camera Graph was first trialled at Flemington Race Course in 1946. Pearl and Shmith then formed Amalgamated Photo Finish Pty Limited, together with Giles and his partner. The service eventually included the production of a photo print within sixty-five seconds. The Camera Graph became the standard system used by major metropolitan race clubs in Australia.
Auto Ray's multi-lens camera, on the other hand, was cumbersome to handle and expensive to operate because of the size of the film it used. At about the same time that Pearl had invented his Camera Graph, Reilly was designing another improved method using continuous film technology, the Rotoflo Circular Film Camera. This addressed the other major shortcoming of the Camera Graph system, perspective distortion. The use of straight film meant that perspective (parallax errors due to lens distortion) was not allowed for which, causing every horse to appear a different size according to its distance from the camera, did not produce a perfect record of their true position relative to the winning post. Notably, a horse on the far side of the track appeared smaller in comparison with the horses nearest the camera. As decisions were often made on fractions of an inch the resulting photo-finish record could be inaccurate.
In order to mitigate perspective distortion, the Camera-Graph system mounted a mirror facing the camera on the far side of the track above the electronic tube, providing a mirror image in the opposite direction as the camera pointed at it from an angle. The advantage of obtaining a reflected and reverse view of that of the camera simultaneously was that it assisted in distinguishing the positions of the leaders (such as the situation where two or more horses were in line and one's head blocked the view of the other).
With Reilly's Rotoflo Camera, however, such distortion was avoided altogether. It used a rotating circular film, the rotation speed directly related in ratio speed to that of the horses being photographed with the film turning faster on the outside of the disc for the outside horse and slower on the inside. This ensured that the images of the horses travelled at the correct speed in relation to the exposure time, irrespective of their position on the track.
The centre of the camera was determined by taking a photograph with a still camera of objects of equal size placed across the track and lined up on the winning line, using the same lens as that of the Rotoflo and from the same position. When the negative was obtained, a line was drawn along the perspective frame, where this line and the winning line intersect, which was the centre point of the Rotoflo Camera. This line - made by placing the wet negative on a circular glass with accurately etched lines radial from a centre peg - appeared on Rotoflo prints. The film was rotated to bring the noses of the horses up to a radial line. As the horses had been photographed with the slit radial to the centre of the camera (lined up on the winning line), this radial line was an equivalent of the actual winning line.
A patent was granted in 1949.