Welcome Guest
Log In | Register )
You last visited January 22, 2017, 7:07 am
All times shown are
Eastern Time (GMT-5:00)

All The Brains

Topic closed. 130 replies. Last post 6 years ago by Stack47.

Page 4 of 9
PrintE-mailLink
RL-RANDOMLOGIC's avatar - usafce

United States
Member #59354
March 13, 2008
4091 Posts
Offline
Posted: June 1, 2011, 2:36 am - IP Logged

time*treat

I did not track this information so I don't know.  What I wanted to show was a problem with time based RNG's

used for simulations.   I generated what would be the main draw first and then added a 3 second delay before

running the 5 million simulated QP's.  The RNG used here uses the system ticks as a seed but a randomize timer

statement is also used to make the seed value even more random.  This produces a very nice random set but

when the time between sets is reduced to a very small value then the sets can repeat.  I would guess that some

sets appear over 100 times as I have seen this in the past.   

 

All state lotteries use RNG's but most think that because the final set is drawn by a ball dropper that it is better. 

Think about jades post on 1 for 10 or 10 for 1,  does it matter if 100,000 RNG QP's are drawn first and then ball

drop the last set or 100,000 ball drops followed by a RNG set.   How to properly analyze a lottery is beyond me

but if I had the brain power I would be looking at time for any RNG based game.  Even a system that uses a RNG

will do better or worse based on the time it is ran.  When I first started out I would set the system clock to match

the draw time but I no longer use a RNG in my software.  I now use a NSG or numerical set generator which was

not feasible back then. 

 

I could write the sets to a file but the time needed for each write statement would effect the outcome so I don't think

that the data would not mean much.  The more time I put between the sets the fewer sets that are needed to match a 

5 of 5.  Adding 100th of a second delay between each of the 5,000,000 sets would add 13.8 hours to the run time.

 

RL

    time*treat's avatar - radar

    United States
    Member #13130
    March 30, 2005
    2171 Posts
    Offline
    Posted: June 1, 2011, 3:32 am - IP Logged

    Ah, I knew simulations were useless for a reason. Wink

    In neo-conned Amerika, bank robs you.
    Alcohol, Tobacco, and Firearms should be the name of a convenience store, not a govnoment agency.


      United States
      Member #93947
      July 10, 2010
      2180 Posts
      Offline
      Posted: June 1, 2011, 9:26 am - IP Logged

      Those here who are having trouble producing random number sequences with acceptable cycle lengths and distributions by repeatedly calling the Randomize Function in Microsoft Basic might learn something if they look back at what the people at the Nuclear Energy Agency made available back in 1984.

      http://www.oecd-nea.org/tools/abstract/detail/nea-0939

      And for a discussion and the full source code for a RNG used under extreme scrutiny in the casino world of the 21st Century, click here:

      http://wizardofodds.com/askthewizard/randomnumbers.html

      When simulations and backtesting suggest that systems are capable of no more than controlling the distribution of winnings but not the long term amount, the tactics that some people will resort to in their desperate attempts to obfuscate this knowledge, is pathetic.

        RL-RANDOMLOGIC's avatar - usafce

        United States
        Member #59354
        March 13, 2008
        4091 Posts
        Offline
        Posted: June 1, 2011, 10:51 am - IP Logged

        Time*treat

        I tracked the match 0of5 to 4of5 and here is the result. 

         

        wins            QP's            0of5        1of5          20f5        3of5      4of5   Numbers drawn
        no winner    5000000    2406977  2011841  518219   62961         1    31  25  12  36  27
        no winner    5000000    2451204  1997521  507059   41295    2920   29  27  22  8  15
        no winner    5000000    2449719  2004747  483557   58115    3861   6  5  32  18  31
        no winner    5000000    2415969  1993471  537112   51981    1466   6  10  1  5  34
        no winner    5000000    2379156  2051243  511717   56005    1878   32  33  29  17  6
        no winner    5000000    2367475  2036339  545314   48440    2431   34  2  16  15  14
        no winner    5000000    2468170  1955529  522946   51372    1982   1  23  33  5  24
        no winner    5000000    2389704  2038142  519017   51210    1926   24  19  2  1  17
        no winner    5000000    2473843  1990527  499059   35104    1466   3  13  9  20  22
        no winner    5000000    2437658  1995089  519731   46566     955    25  37  14  17  39
        no winner    5000000    2392968  2038617  517042   49924    1448   39  5  23  28  22
        no winner    5000000    2422951  2030649  502706   41753    1940   17  12  7  9  10
        no winner    5000000    2449036  2001939  504303   42761    1960   19  27  16  6  24
        no winner    5000000    2337893  2067227  528065   64911    1903   5  39  32  37  4
        no winner    5000000    2439627  1973245  528616   54033    4478   21  28  26  34  35
        no winner    5000000    2436613  1972726  551092   37615    1953   18  3  10  26  9
        no winner    5000000    2355997  2044328  554953   41863    2858   5  20  13  23  29
        no winner    5000000    2431880  2021065  483913   60948    2193   21  27  32  34  33
        no winner    5000000    2409822  2008381  534731   47064          1   14  28  37  12  31
        no winner    5000000    2446711  1981450  526365   44510      963   25  11  4  9  28
        no winner    5000000    2420314  2030845  492733   56107          0   10  15  2  14  29
        no winner    5000000    2389731  2047808  520193   40810    1457   14  18  38  33  26
        no winner    5000000    2441285  1993508  521159   43562      485   9  21  17  26  31
        no winner    5000000    2384688  2067790  509372   38149          0   14  35  26  7  29
        no winner    5000000    2446095  1983715  513783   56405          1   35  33  3  23  11

        RL

          RL-RANDOMLOGIC's avatar - usafce

          United States
          Member #59354
          March 13, 2008
          4091 Posts
          Offline
          Posted: June 1, 2011, 11:13 am - IP Logged

          Those here who are having trouble producing random number sequences with acceptable cycle lengths and distributions by repeatedly calling the Randomize Function in Microsoft Basic might learn something if they look back at what the people at the Nuclear Energy Agency made available back in 1984.

          http://www.oecd-nea.org/tools/abstract/detail/nea-0939

          And for a discussion and the full source code for a RNG used under extreme scrutiny in the casino world of the 21st Century, click here:

          http://wizardofodds.com/askthewizard/randomnumbers.html

          When simulations and backtesting suggest that systems are capable of no more than controlling the distribution of winnings but not the long term amount, the tactics that some people will resort to in their desperate attempts to obfuscate this knowledge, is pathetic.

          Jimmy

          I think the code below belongs to you.

          "RANDOMIZE 11  ' Try different seeds and observe results"

          When he was making a point and teaching us fools it was ok then.  I have never thought MS's randomize

          was a good seeding tool.  I think I might use another RNG and see what happends just to be fair.  I wonder

          if the lottery would show the code they use so we could test it.

          RL

            RJOh's avatar - chipmunk
            mid-Ohio
            United States
            Member #9
            March 24, 2001
            19901 Posts
            Offline
            Posted: June 1, 2011, 11:14 am - IP Logged

            Hi,

              One of the reasons you are not seeing a group effort here is because of the members ending up treating each other when posting a system. It sometimes become a contest of compentacy and ego, rather than helpful suggestions. Then there are those who refuse to stay on topic. How many threads end up as some unrelated topic to the original post? Then there are those who expect other members to spend the rest of thier lifetime going through previous LP threads. I'm not sure that a lot of people here know how to complement, enhance, and aid in the building of any suggested method.

              As for me, I just work my own systems and come to the LP looking for any ideas that might complement the methods I use. I have no desire to post my methods here because of attitudes, egos, and lack of common courtesy. 

              A model of the way people should repond can be seen almost every day through RJOH. Even though our approaches are different, I totally respect the way he trys to help every member in a constructive, friendly, and courteous mannner.

              In a nutshell, until the overall attitude enviroment changes at the LP, the likelyhood of group built lottery system is slim at best.

              Oppotunity to learn a system or improve your existing systems is here for the taking by all if things could be just a little bit different.

              Best of luck to all.

            JKING,

            Thanks for your kind remarks. 

            I try to keep my criticism of others ideas to a minimum and give my best evaluation of those ideas when asked.  One doesn't need an idea to criticize those of others. 

            As far as I know other than buying a ticket, no one yet has come with a better idea that resulted in them winning a lottery.  The closest I ever came was in 2002 when I matched 5of6 in the Ohio Super Lotto (6/49+1) and I didn't have a clue of what I did right except buy a ticket with ten combinations I picked.

            There's been lots of ideas shared at LP and I believe members best efforts to use them are done privately, members can make their own conclusions about what has been posted publicly. 

            Members generally have no desire to be part of a team and when it comes to winning a lottery, the line "one for all and all for one" just doesn't apply.  Sharing of ideas is as close to a group effort that we will ever see at LP.

             * you don't need to buy more tickets, just buy a winning ticket * 
               
                         Evil Looking       

              RL-RANDOMLOGIC's avatar - usafce

              United States
              Member #59354
              March 13, 2008
              4091 Posts
              Offline
              Posted: June 1, 2011, 11:49 am - IP Logged

              JKING,

              Thanks for your kind remarks. 

              I try to keep my criticism of others ideas to a minimum and give my best evaluation of those ideas when asked.  One doesn't need an idea to criticize those of others. 

              As far as I know other than buying a ticket, no one yet has come with a better idea that resulted in them winning a lottery.  The closest I ever came was in 2002 when I matched 5of6 in the Ohio Super Lotto (6/49+1) and I didn't have a clue of what I did right except buy a ticket with ten combinations I picked.

              There's been lots of ideas shared at LP and I believe members best efforts to use them are done privately, members can make their own conclusions about what has been posted publicly. 

              Members generally have no desire to be part of a team and when it comes to winning a lottery, the line "one for all and all for one" just doesn't apply.  Sharing of ideas is as close to a group effort that we will ever see at LP.

              RJOh

              I use to share a similar opinion as JKING until you started sounding like jimmy.  When you said "Was it just an idea you

              thought up off the top of your head and never tried or tested before sharing at LP?" you lifted your skirt so to speak.

              I have found many post where I could have attacked your methods but chose not to say anything out of respect for

              a fellow system player.   I will still give you one last bit of advice and ask you to backtest your RNG to see how many

              sets it needs to generate to hit a 5 of 5 for you rolling cash 5.  Do this each day for a couple weeks and then see if it

              needs some tinkering.  Pre or post draw makes no difference but I would try the pre draw method.  Just run off a million

              sets or so and set them aside to check after the draw.   No filters are needed for this test.  In a 5-39 matrix every

              possible match 4 of 5 is within the first 73,815. 

               

              Hope you win soon.     

              RL

                time*treat's avatar - radar

                United States
                Member #13130
                March 30, 2005
                2171 Posts
                Offline
                Posted: June 1, 2011, 12:01 pm - IP Logged

                Time*treat

                I tracked the match 0of5 to 4of5 and here is the result. 

                 

                wins            QP's            0of5        1of5          20f5        3of5      4of5   Numbers drawn
                no winner    5000000    2406977  2011841  518219   62961         1    31  25  12  36  27
                no winner    5000000    2451204  1997521  507059   41295    2920   29  27  22  8  15
                no winner    5000000    2449719  2004747  483557   58115    3861   6  5  32  18  31
                no winner    5000000    2415969  1993471  537112   51981    1466   6  10  1  5  34
                no winner    5000000    2379156  2051243  511717   56005    1878   32  33  29  17  6
                no winner    5000000    2367475  2036339  545314   48440    2431   34  2  16  15  14
                no winner    5000000    2468170  1955529  522946   51372    1982   1  23  33  5  24
                no winner    5000000    2389704  2038142  519017   51210    1926   24  19  2  1  17
                no winner    5000000    2473843  1990527  499059   35104    1466   3  13  9  20  22
                no winner    5000000    2437658  1995089  519731   46566     955    25  37  14  17  39
                no winner    5000000    2392968  2038617  517042   49924    1448   39  5  23  28  22
                no winner    5000000    2422951  2030649  502706   41753    1940   17  12  7  9  10
                no winner    5000000    2449036  2001939  504303   42761    1960   19  27  16  6  24
                no winner    5000000    2337893  2067227  528065   64911    1903   5  39  32  37  4
                no winner    5000000    2439627  1973245  528616   54033    4478   21  28  26  34  35
                no winner    5000000    2436613  1972726  551092   37615    1953   18  3  10  26  9
                no winner    5000000    2355997  2044328  554953   41863    2858   5  20  13  23  29
                no winner    5000000    2431880  2021065  483913   60948    2193   21  27  32  34  33
                no winner    5000000    2409822  2008381  534731   47064          1   14  28  37  12  31
                no winner    5000000    2446711  1981450  526365   44510      963   25  11  4  9  28
                no winner    5000000    2420314  2030845  492733   56107          0   10  15  2  14  29
                no winner    5000000    2389731  2047808  520193   40810    1457   14  18  38  33  26
                no winner    5000000    2441285  1993508  521159   43562      485   9  21  17  26  31
                no winner    5000000    2384688  2067790  509372   38149          0   14  35  26  7  29
                no winner    5000000    2446095  1983715  513783   56405          1   35  33  3  23  11

                RL

                Thanks RL.

                We kick enough ideas around, something just might turn up. Cool

                In neo-conned Amerika, bank robs you.
                Alcohol, Tobacco, and Firearms should be the name of a convenience store, not a govnoment agency.

                  RJOh's avatar - chipmunk
                  mid-Ohio
                  United States
                  Member #9
                  March 24, 2001
                  19901 Posts
                  Offline
                  Posted: June 1, 2011, 2:45 pm - IP Logged

                  RJOh

                  I use to share a similar opinion as JKING until you started sounding like jimmy.  When you said "Was it just an idea you

                  thought up off the top of your head and never tried or tested before sharing at LP?" you lifted your skirt so to speak.

                  I have found many post where I could have attacked your methods but chose not to say anything out of respect for

                  a fellow system player.   I will still give you one last bit of advice and ask you to backtest your RNG to see how many

                  sets it needs to generate to hit a 5 of 5 for you rolling cash 5.  Do this each day for a couple weeks and then see if it

                  needs some tinkering.  Pre or post draw makes no difference but I would try the pre draw method.  Just run off a million

                  sets or so and set them aside to check after the draw.   No filters are needed for this test.  In a 5-39 matrix every

                  possible match 4 of 5 is within the first 73,815. 

                   

                  Hope you win soon.     

                  RL

                  After the post you made at the beginning of this thread about systems in general and yours in particular I had serious doubts about the time you spent researching your digit system since you seemed to be doubting some of your earlier posts about it and had to ask that question.  Feel free to criticize my posts when ever you think they deserve it, I won't be offended.

                  Time is limited and I hardly have enough to research my own ideas let alone those suggested by others.  Right now I'm concentrating on PowerBall and Classic Lotto and only updating my Rolling Cash5 and MegaMillions files. 

                  When ever I used my program to pick combinations I avoid repeating past combinations by comparing each combination generated with past drawings so I know occasionally all the numbers in a previous drawing are matched and rejected but that only effect my final picks as planned.

                  I hope you also win soon.

                  RJOh

                   * you don't need to buy more tickets, just buy a winning ticket * 
                     
                               Evil Looking       


                    United States
                    Member #93947
                    July 10, 2010
                    2180 Posts
                    Offline
                    Posted: June 1, 2011, 5:10 pm - IP Logged

                    time*treat

                    I did not track this information so I don't know.  What I wanted to show was a problem with time based RNG's

                    used for simulations.   I generated what would be the main draw first and then added a 3 second delay before

                    running the 5 million simulated QP's.  The RNG used here uses the system ticks as a seed but a randomize timer

                    statement is also used to make the seed value even more random.  This produces a very nice random set but

                    when the time between sets is reduced to a very small value then the sets can repeat.  I would guess that some

                    sets appear over 100 times as I have seen this in the past.   

                     

                    All state lotteries use RNG's but most think that because the final set is drawn by a ball dropper that it is better. 

                    Think about jades post on 1 for 10 or 10 for 1,  does it matter if 100,000 RNG QP's are drawn first and then ball

                    drop the last set or 100,000 ball drops followed by a RNG set.   How to properly analyze a lottery is beyond me

                    but if I had the brain power I would be looking at time for any RNG based game.  Even a system that uses a RNG

                    will do better or worse based on the time it is ran.  When I first started out I would set the system clock to match

                    the draw time but I no longer use a RNG in my software.  I now use a NSG or numerical set generator which was

                    not feasible back then. 

                     

                    I could write the sets to a file but the time needed for each write statement would effect the outcome so I don't think

                    that the data would not mean much.  The more time I put between the sets the fewer sets that are needed to match a 

                    5 of 5.  Adding 100th of a second delay between each of the 5,000,000 sets would add 13.8 hours to the run time.

                     

                    RL

                    Did you notice that my call to Randomize in my example simulation was made ONCE, at the top of the program, to allow for code testing?  Your references to Randomize and the System Clock are really not very transparent.

                    You said earlier that your sequences were repeating too often but you've ignored my links which provide you with random numbers that would solve all of your problems.

                    You force me to repeat myself:

                     

                    Those here who are having trouble producing random number sequences with acceptable cycle lengths and distributions by repeatedly calling the Randomize Function in Microsoft Basic might learn something if they look back at what the people at the Nuclear Energy Agency made available back in 1984.

                    http://www.oecd-nea.org/tools/abstract/detail/nea-0939

                    And for a discussion and the full source code for a RNG used under extreme scrutiny in the casino world of the 21st Century, click here:

                    http://wizardofodds.com/askthewizard/randomnumbers.html

                    When simulations and backtesting suggest that systems are capable of no more than controlling the DISTRIBUTION of winnings but not the long term amount, the tactics that some people will resort to in their desperate attempts to obfuscate this knowledge, is pathetic.

                    BTW, I thought you had officially "retired" from LP...

                      RL-RANDOMLOGIC's avatar - usafce

                      United States
                      Member #59354
                      March 13, 2008
                      4091 Posts
                      Offline
                      Posted: June 1, 2011, 5:40 pm - IP Logged

                      Did you notice that my call to Randomize in my example simulation was made ONCE, at the top of the program, to allow for code testing?  Your references to Randomize and the System Clock are really not very transparent.

                      You said earlier that your sequences were repeating too often but you've ignored my links which provide you with random numbers that would solve all of your problems.

                      You force me to repeat myself:

                       

                      Those here who are having trouble producing random number sequences with acceptable cycle lengths and distributions by repeatedly calling the Randomize Function in Microsoft Basic might learn something if they look back at what the people at the Nuclear Energy Agency made available back in 1984.

                      http://www.oecd-nea.org/tools/abstract/detail/nea-0939

                      And for a discussion and the full source code for a RNG used under extreme scrutiny in the casino world of the 21st Century, click here:

                      http://wizardofodds.com/askthewizard/randomnumbers.html

                      When simulations and backtesting suggest that systems are capable of no more than controlling the DISTRIBUTION of winnings but not the long term amount, the tactics that some people will resort to in their desperate attempts to obfuscate this knowledge, is pathetic.

                      BTW, I thought you had officially "retired" from LP...

                      Jimmy

                      When my membership ends so will my post.  It was not nessary to repeat yourself nice move though.

                      RL

                        RJOh's avatar - chipmunk
                        mid-Ohio
                        United States
                        Member #9
                        March 24, 2001
                        19901 Posts
                        Offline
                        Posted: June 1, 2011, 6:19 pm - IP Logged

                        Did you notice that my call to Randomize in my example simulation was made ONCE, at the top of the program, to allow for code testing?  Your references to Randomize and the System Clock are really not very transparent.

                        You said earlier that your sequences were repeating too often but you've ignored my links which provide you with random numbers that would solve all of your problems.

                        You force me to repeat myself:

                         

                        Those here who are having trouble producing random number sequences with acceptable cycle lengths and distributions by repeatedly calling the Randomize Function in Microsoft Basic might learn something if they look back at what the people at the Nuclear Energy Agency made available back in 1984.

                        http://www.oecd-nea.org/tools/abstract/detail/nea-0939

                        And for a discussion and the full source code for a RNG used under extreme scrutiny in the casino world of the 21st Century, click here:

                        http://wizardofodds.com/askthewizard/randomnumbers.html

                        When simulations and backtesting suggest that systems are capable of no more than controlling the DISTRIBUTION of winnings but not the long term amount, the tactics that some people will resort to in their desperate attempts to obfuscate this knowledge, is pathetic.

                        BTW, I thought you had officially "retired" from LP...

                        I've never noticed that problem with the RNG in my program.  I always enter the same seed which is my birthday and when I've rerun the program  I seldom get the same results farther out than 2 or 3 combinations.  It must be how I use it and some of the parameters I enter.  Actually since numbers can only be used in certain positions and add up to particular sums and can't be repeated until all numbers have been used once of a loop is create that can only be broken when the maximum times a number can be used is increase and are only acceptable if they have certain profiles I probably don't need a RNG, the method is chaotic enough on its own.

                         * you don't need to buy more tickets, just buy a winning ticket * 
                           
                                     Evil Looking       

                          RJOh's avatar - chipmunk
                          mid-Ohio
                          United States
                          Member #9
                          March 24, 2001
                          19901 Posts
                          Offline
                          Posted: June 1, 2011, 6:24 pm - IP Logged

                          Jimmy

                          When my membership ends so will my post.  It was not nessary to repeat yourself nice move though.

                          RL

                          Your membership never ends, just your premium status.  You don't think all those 100,000+ members are still posting or even still alive, do you?

                           * you don't need to buy more tickets, just buy a winning ticket * 
                             
                                       Evil Looking       


                            United States
                            Member #93947
                            July 10, 2010
                            2180 Posts
                            Offline
                            Posted: June 1, 2011, 6:53 pm - IP Logged

                            Did you notice that my call to Randomize in my example simulation was made ONCE, at the top of the program, to allow for code testing?  Your references to Randomize and the System Clock are really not very transparent.

                            You said earlier that your sequences were repeating too often but you've ignored my links which provide you with random numbers that would solve all of your problems.

                            You force me to repeat myself:

                             

                            Those here who are having trouble producing random number sequences with acceptable cycle lengths and distributions by repeatedly calling the Randomize Function in Microsoft Basic might learn something if they look back at what the people at the Nuclear Energy Agency made available back in 1984.

                            http://www.oecd-nea.org/tools/abstract/detail/nea-0939

                            And for a discussion and the full source code for a RNG used under extreme scrutiny in the casino world of the 21st Century, click here:

                            http://wizardofodds.com/askthewizard/randomnumbers.html

                            When simulations and backtesting suggest that systems are capable of no more than controlling the DISTRIBUTION of winnings but not the long term amount, the tactics that some people will resort to in their desperate attempts to obfuscate this knowledge, is pathetic.

                            BTW, I thought you had officially "retired" from LP...

                            I just took another look at the output from a computer simulation above of (5,39) Lotto.  In 25 runs of 5,000,000 draws each, I see NO Jackpots (5/5).  The distribution of (4/5) hits average close to expected but with some choppiness at the low end.  ANYTHING is possible, but since chance would cause me to expect to see around 8 or 9 Jackpots on average in 5,000,000 draws, if this were my program, I would turn on the DEBUGGER, disable all the timer stuff, and take a serious look at my algorithms!  Just sayin' !

                              Coin Toss's avatar - shape barbed.jpg
                              Zeta Reticuli Star System
                              United States
                              Member #30470
                              January 17, 2006
                              10392 Posts
                              Offline
                              Posted: June 1, 2011, 7:15 pm - IP Logged

                              Does anyone think the lotteries just might be running their own tests?

                              While people here are trying to "crack the code" with relatively limited resources do you think that the lotteries just might have people employed to make sure the code can't be cracked, with up to the minute technology, and a steady stram of millions of dollars pouring in every week to finance that technology?

                              Those who run the lotteries love it when players look for consistency in something that's designed not to have any.

                              Lep

                              There is one and only one 'proven' system, and that is to book the action. No matter the game, let the players pick their own losers.