~@Com~   On the 'Net since 1994	Black Jack Computers         Email
Links:  Photo Gallery  | Blackjack Project  |  Roulette Project  |  Roulette Timing  |  KIM-1 (6502)  |  Correspondence

Here are some images from our roulette project ca. 1984 - 1985.
We predicted the outcome of roulette in actual casino play by timing the ball and rotor of the roulette wheel.

Before we did anything, we gathered data.  At first we used our home-made "KIN" computer to time the ball and rotor (wheel) surreptitiously -- in a casino. The KIN computer was interfaced with a speech synthesizer; this hardware was left over from a previous project, the Black Jack machine.  With the help of my friends Rita and Richard, we three timed the ball and wheel in a casino. (At the time we did not own a roulette wheel.)  I wore all computer hardware concealed on my person; I think Rita had the our cassette recorder in her purse. One of the timing switches (Rita's switch) was attached to the rear of my belt. Rita would stand beside me and embrace me with one arm form behind, inside my overcoat, so that she could enter the revolution times of the rotor with the belt switch.  I timed the ball revolutions by using a switch in my trouser pocket.  The speech synthesizer spoke the revolution times onto a tape recorder.  We also spoke the landing pocket onto the recording. The speech output from the computer was temporally compressed so that it had a high pitch when played back at normal speed.  This compression was made so that everything would be recorded quickly in real time. A rheostat inserted into the recorder's drive motor allowed for slowing its motor speed, enabling us to play the tape so that it could be understood. Listen to a sample.  This data retrieval was time consuming, so we abandoned it, after having purchased the Radio Shack "PC2" (and the clone: Sharp PC-1500 Pic) computer pictured here.
Next, we did our analysis and prediction in two phases. In the first phase we used a concealed computer (Radio-Shack PC2) to gather data on ball and rotor speed, using a computer program that could time the ball and wheel speeds (revolution times). No attempt at prediction was made in this phase. The computer ran BASIC (WikiPedia)-- the timing routines were loaded as machine language, using "pokes". For each trial, data representing revolution times and the resulting winning number were then uploaded to an APPLE-II computer for analysis, Via a modem. The object of the analysis was to smooth the data points thus obtained and to determine the parameters of the decay curves for the ball and for the rotor. The PC2 computer was modified to interface with an APPLE II computer, (for uploading the timing data for analysis) as well as with with a speech synthesizer to announce the predicted winning number (in the second phase). Having determined these parameters from the uploaded data on the APPLE II, we typed them into the PC2 computer.  This information is unique to each wheel and was required top load the prediction program. (Shown here is the PC2 computer attached to its docking station. The docking station had a four-color plotter (left) and a modem (black box at rear).	Radio-Shack PC2 computer  Wiki
If you are seriously considering  this, you will find this book of IMMENSE interest. (I promise :) It's the story of techno-nerds who did  a Black Jack project using similar computer equipment as THIS. The Eudaemonic Pie by Thomas A. Bass   The single-zero wheel is used  in Asia and in Europe. This is picture is freely given at roulette tables at the Lisboa Casino. Our wheel
We required two experimenters for this operation during play; one to clock the ball and wheel, to enter the winning number and position on the rim of the wheel (between 12 o'clock and and 12 o'clock) Another experimenter placed the bets at the table. Communication was via infrared. During the second phase, at the roulette table, the actual predictions were made by the PC2 computer . The speech synthesizer transmitted the wining number to the second player via an infra-red link built into each of the two experimenter's shoes.

 

The PC2 Computer. Interface for timing switches, landing position switches, and speech synthesizer input is via an added connector at right.   The wires from the connector lead to various points  within the "works"  of the computer.   Keypad for entering switch-presses for timing - The data collected were: time for ball revolution time for rotor revolution winning number the position on the rim of the wheel where the ball touched down the number that would have won if no ball skip had occurred. We had a four buttons arranged in a square, strapped to the top of the leg with a Velcro strap. We cut a slit in the pants' pocket so we could press the switches directly, and epoxied a washer onto the top of some of the switches so that they could easily be identified by touch.   Beyond Belief Pick one up for a friend!       The speech synthesizer              Speech synthesizer and Infra-red  transmitter
Just for fun, here is a link to a web page which lets you write programs in AppleSoft BASIC. Basic is what we used to analyze our Roulette data on the Apple II.
Travel kit. A large suitcase contains APPLE II and floppy disk drive; a suitcase contains PC2 and  other hardware.	Everything which was worn by the     player who timed the ball and rotor.
One experimenter's shoe. The infrared receiver is embedded behind the wire grid in the sole.	This work was done by my then-friend Richard and me.    Here is the laboratory in which we did much of it.  (Photo ca. 1986)    Previously this lab was the site for attacking Black Jack.
Jan. 1984
Richard (right) about the time of inception of the Roulette project,     ca 1980. Mike Eggett (left) looks on.      Photo:  MINOX 8mm film	A picture of me ca 1984. John Huxley Wheel.     This wheel is FOR SALE.
Experimental setup for accurate measurement in the lab.      Roulette experiments ca. 1984.       John Huxley Wheel.	Black Jack predicting computer.     The homemade KIN computer is inserted into socket     on a test-kludge box (right); the KIM computer is in its     housing (center), ca. 1981.      Photo:  MINOX 8mm film.
( This casino quality double-zero John Huxley roulette wheel is for sale, by the way :)
ca. 1977
Links:    Photo Gallery    Back to Roulette Prediction Project Main page    Our Roulette Timing Data    Clone: KIN to KIM  --  Development    Casino Lisboa, Macao Pictures of our wheel    Pictures from our Black Jack project
We used a board with four buttons (arranged in a square) attached to the top of the leg with a velcro strap.  We cut a slit in the pocket so we could press the switches directly, and epoxied a washer to the top of some of the switches so they would stand out and give us a tactile sensation. As for timing, we used an arbitrary imagined "12 O'clock" position as a reference point on the rim of the wheel.  We used the same point for both the ball and the wheel.  When the ball or wheel passed the point, one of us would press the appropriate ball or wheel switch. Pressing for both ball and wheel at the same time was too confusing.  So we timed the wheel first for a few presses.  The wheel is quite predictable for several seconds in advance.  It slows at a very steady rate.  Once we had a few wheel presses as it passed 12 o'clock, we started pressing the ball switch as it passed the same 12 o'clock position.  The first 'wheel' press reset a timer and started counting in the wearable computer.  The timer counted in thousandths of a second and recorded the passing of the ball or wheel at 12 o'clock as the switch was pressed.  As the ball slowed to a certain point, as measured by how long it took to make a revolution of the ball, the computer projected how long it would be before the ball would fall from the rim and contact the wheel -- and also where the wheel would be at the time when the ball hit.  This was used to predict a winning number.  The information was then passed to the betting person via an infrared link (analog) fron the  a speech synthesizer that was integrated with the KIN computer.  The other two switches (four total) were used for data collection.  At the end of each play we recorded the place where the ball landed (clock position) and what number actually won.  We used that information later to assess our accuracy and adjust our predictions. "Prediction is very difficult, especially of the future"  --    Niels Bohr

Copyright @Com  2012.    Last updated  09.04.2015