Welcome Guest
Log In | Register )
You last visited December 4, 2016, 3:15 pm
All times shown are
Eastern Time (GMT-5:00)

2 million dollar Supercomputer?

Topic closed. 41 replies. Last post 10 years ago by JAP69.

Page 1 of 3
PrintE-mailLink
Avatar

Honduras
Member #20982
August 29, 2005
4715 Posts
Offline
Posted: February 21, 2007, 9:13 pm - IP Logged

I was reading this article:

            And if you read the last paragraph you will know that IBM built (or is building) a supercomputer that can make 7 trillion calculations per second (or teraflop) for 24 million dollars dubbed "Blue Sky"..Some of you may have already heard me saying that with a supercomputer you could perhaps analyzed in depth and crack the lotto...Some of you may disagree...
        What got me thinking is if IBM charged 24 million dollars for a supercomputer that can do 7 trillion calculations per second how much will they charge to build you a supercomputer that can do 20 Billion calculations per second (modest enough to browsed to all of megaMillion's combinations over and over and do all kinds of experiments, e.g take the square root of all the possible combinations, 90% percentile of all combinations total sums, analyzing en depth all keno combinations, etc, etc, etc all in less than one second)...
        Guys doing the calculations, i say they will charge you about 2 millions perhaps 1 million...So you know if you ever win the lottery and will like to buy yourself a modest (i don't see nothing modest about it) supercomputer to do lotto research/experiments at least you know that perhaps with less than 2 million dollars you could get yourself one...Also if you buy it from IBM, the supercomputer comes with a new thechnology named "eLiza" that heal itself saving you money in repairing cost.........
          NOw my question to you is how much do you think IBM will charge to build you supercomputer that can do 20 Billion calculations per second...Will you be happy with 20 Billion calculations per second or with less?
        A supercomputer like that is not that bad, you can always action it and perhaps get 75% of your money back...After all, it could be a good investment..If you cracked the lotto do you know how much money that will be? Imagine winning every state's pick6 or international lottos...
    Note: When i say trying to crack the lotto i mean pick6 and above..
    Raven62's avatar - binary
    New Jersey
    United States
    Member #17843
    June 28, 2005
    49674 Posts
    Offline
    Posted: February 21, 2007, 9:34 pm - IP Logged

    No News is better than Old News (5+ years old).

    A mind once stretched by a new idea never returns to its original dimensions!

      Avatar

      Honduras
      Member #20982
      August 29, 2005
      4715 Posts
      Offline
      Posted: February 21, 2007, 9:48 pm - IP Logged

      I say it could be a good investment because if you crack or do not crack the lotto you will be at the forefront of lottery research and you would have done more than anybody has ever done for lottery research and to your credit/credential you can claim you have a supercomputer to prove what you said...So if you don't get to crack the lottery after years of research & hard work, you can state that: "You have done everything possible to crack the lottery and that everything has been done to crack the lottery even with the use of a supercomputer" which you will claim you have...After that you write a book and whether to prove that the lottery can be cracked or to prove that it can't be cracked and that book should stay in the # 1 bestseller for a while, generating you lots of money...YOu probably come out on TV, cnn...

       

       

      "Keno is "the demon who makes trophies of men".......        "You can't see it......          "And it skins them........."                          from movie "Predator 1"......

        Raven62's avatar - binary
        New Jersey
        United States
        Member #17843
        June 28, 2005
        49674 Posts
        Offline
        Posted: February 21, 2007, 10:30 pm - IP Logged

        If you believe a Supercomputer is required to beat the Games of Chance then:

        You should find this article interesting!

        A mind once stretched by a new idea never returns to its original dimensions!

          Avatar

          Honduras
          Member #20982
          August 29, 2005
          4715 Posts
          Offline
          Posted: February 21, 2007, 11:16 pm - IP Logged

          Some of you think that supercomputers are the size of a door or cabinets...Some of them are, but some are the size of a 39 inch T.V.They got some that are the size of regular home computer..Here are pictures/photos of some of them...

          http://www.cs.cmu.edu/~scandal/gifs.html

          Look at the "Convex Examplar", the "J90", the "T3E", the "SP1", Look at NEC's SX-4/32 is not that big (fits in a room), look at the "SGI Origin2000", look at the "Gigabooster", and the "SGI Power Challenge XL"...They are all small...

          My favorite is the "Gigabooster" & "SGI" or you want want to call it "SG-1" (T.v show)...

           

           

          "Keno is "the demon who makes trophies of men"........."          "You can't see it..........."        "And it skins them......"                                        from movie "Predator 1"

            Avatar

            Honduras
            Member #20982
            August 29, 2005
            4715 Posts
            Offline
            Posted: February 22, 2007, 12:01 am - IP Logged

            I just realized something..I said earlier that you could study in depth MegaMillion/PB with a supercomputer, but if you wanted to crack the lotto, the best thing to do is (for the starting payout that megaMillion or Powerball give out, of 12 or 10 million) is to study and try to crack the state's Pick6 6/49...Wait 'till it grow and hit it a couple of times or different states, and you will get what mega million or powerball starting jackpot gives out..It is not advisable to go after MegaMillion or Powerball...Instead you should go after every state's pick6 and if you can go after California's Superlotto...CA's Superlotto sometimes climbs....and it starts at 7 million dollars...But first you got after each state's Pick 6/49 only...After you have accumulated 20 million dollars, you then go abroad and wait until you have accumulated another 20 million dollars (the U.K, Germany, France, Swizerland, Japan (if it has a 6/49 game), Spain, Denmark, Sweeden, etc) and then you should have 40 million dollars..After those 40 million dollars you then go after California's Superlotto...After you finish with california you invest (you will have between 47 & 60), (is then than you can decide to write a book) and you then continue with California Superlotto (you will have your money on the stock market and continue to hit CA superlotto) until you get tire and decide to retire...Is then when you can give yourself the luxury of buying a replica of the lotto machine that the CA lottery uses...You can then also start a prediction website and charge customers whatever you want... And i will advise you to keep Maximum security wherever you keep that Supercomputer...If you have to keep it with family/relatives so be it...

             

             

             

            "Keno is "el diablo cazador the hombres"....."            "YOu can't see it......."          "And it skins them......."                              from movie "Predator 1"

              Raven62's avatar - binary
              New Jersey
              United States
              Member #17843
              June 28, 2005
              49674 Posts
              Offline
              Posted: February 22, 2007, 1:36 am - IP Logged

              Maybe you could use this Supercomputer to break the code!

              A mind once stretched by a new idea never returns to its original dimensions!

                MillionsWanted's avatar - 24Qa6LT

                Norway
                Member #9517
                December 10, 2004
                1272 Posts
                Online
                Posted: February 22, 2007, 1:40 am - IP Logged

                If you can afford something like that, save the money instead. No need to play the lottery.

                  four4me's avatar - gate1
                  MD
                  United States
                  Member #1701
                  June 18, 2003
                  8360 Posts
                  Offline
                  Posted: February 22, 2007, 1:58 am - IP Logged

                  Pumpi. if you had a hunded supercomputers running at the same time there is no way they could just calculate what the winning numbers are going to be before the drawing takes place with just one ticket.

                   I'm not talking about a bunch of tickets one would have to play.

                  that pc in front of you can calculate all the possible combinations.

                    LANTERN's avatar - kilroy 28_173_reasonably_small.jpg
                    Tx
                    United States
                    Member #4570
                    May 4, 2004
                    5180 Posts
                    Offline
                    Posted: February 22, 2007, 5:38 am - IP Logged

                    My friend Pumpi.

                    A supercomputer or just any computer is just a very expensive paper weight and won't do anything for anybody by itself.

                    Without one or more computer programs it is not worth anything to anybody.

                    But you, of course, already knew that. 

                    Speed of calculations is of course very important, but only if the right calculations are being made.

                    Look for a program called MaxHitPro using google, download it and then test it, there is a demo or trial version, I think that now in the newest version it does not ever time out in the demo mode so perhaps you should be able to test it forever if you want to.

                    I am telling you about that program just to give you an example of what you are talking about, that is a program that will test thousands of "SYSTEMS" (but not so many of them in the demo version) on the past draws, that is using the past draws.

                    As can more often than not be seen on that program, such a thing can be useless, because what worked for the past draws or on the past draws, more often than not, does not work on the future draws to come.

                    Why? It has to do with the prediction logic used on those thousands of "SYSTEMS".

                    Simple systems that use "Wrong Prediction(s) Logic(s), can't cope with a process or processes that are much more complicated than the simple thousands or billions of prediction systems.

                    -----------------

                    Prediction systems that work well and or often enough can't be created by those whose prediction logic(s) is or are wrong.

                    --------------

                    One must have a prediction logic or predictions logics that do work O.K. or well enough, before they can be put into a program or programs and used.

                    ---------------------------------

                    In other words: If you don't know how to predict right or well often enough, then any prediction program that you make or can make won't predict right often enough.

                    ------------------

                    What you say is that possible prediction logics must be put into a program, perhaps thousands, millions or billions of them and then tested to see if one or more of them will work well enough on future draws.

                    -----------

                    But Who will know What prediction systems, methods or logics to code into the program so they can be tested by the computer or supercomputer?

                    Wrong Prediction Logic in, Equals Wrong Prediction Out..

                    At least this is how it was with a prior version of "MaxHitPro that I tested for the pick 3 game, thousands of useless systems, that is useless for the next future draw, which is the draw that counts.

                    --------------

                    And a person who might already know how to predict right, if there is any such a person, surely does not need a supercomputer, he needs his theories put into programs and tested and almost any "Regular" computer should do for that.

                    Once he refines one or more working systems or methods then they can be put into a new program or programs and used to predict with.

                    Later newer ideas can again be put into newer programs and tested also, until that person finds his best possible prediction system or method for that particular game.

                    -------------

                    A person can't just test any possible prediction methods (Look at MaxHitPro) there could trillions or more of such useless systems.

                    A person most learn what prediction is all about, how to develop and or find and test possible winning systems, not trash systems.

                    ----------------------------

                    First of all: Past game's statistics must be used as a base for any future predictions.

                    Scientific prediction based on past draws is the only right way to go, unless you are a good seer such as NostraDamus and or the Oracle of Delphi or at Delphi.

                    I call it "Statistical Prediction"

                    But as with "True Visions" or any other forms of prediction, proper interpretation(s) are a must or needed or the Statistics, Visions, Cards, Astrological or other Reading(s)are useless.

                    -------------------------

                    Statistics there are many, but perhaps more often than not we fail in their right interpretation.

                    -----------

                    Just to show you the power of their "Right" or of their "Near Right" interpretation, here is a past example of "Statistical Scientific Prediction".

                    That past prediction was posted by me at or on The Pick 3 Forum nearly a year ago. 

                    -------------------

                     Texas Evening 18 may 06

                    Pick 3  3-4-7


                    Or reduced a little more.
                    023, 093, 129, 138, 143, 149, 183, 189, 203, 219, 235, 237, 239, 243, 246, 249, 253, 264, 269, 273, 279, 280, 289, 291, 293, 297, 302, 314, 318, 324, 325, 327, 345, 347, 354, 374, 381, 384, 387, 390, 392, 408, 413, 419, 423, 426, 429, 435, 437, 453, 473, 480, 483, 487, 491, 495, 523, 534, 624, 723, 729, 734, 789, 798, 813, 819, 837, 847, 879, 897, 908, 912, 914, 918, 924, 927, 945, 946, 978
                    76 straight combos.

                    --------------

                    I got the 347 Straight 1 time and 4 times also boxed on that one same prediction, in other words I got it 5 times out of the 6 possible times, as pick 3 singles are 6 way numbers.

                    ---------------------------------- 

                    Now Next that past prediction Deflated or using Todd's Deflate Pick 3 System:

                    Output

                    • Results: 27

                    Tier 5
                    138(183,318,381,813) 345(354,435,453,534) 347(374,437,473,734) 789(798,879,897,978)

                    Tier 4
                    129(219,291,912) 149(419,491,914) 235(253,325,523) 237(273,327,723) 246(264,426,624) 279(297,729,927)

                    Tier 3
                    023(203,302) 143(314,413) 189(819,918) 239(293,392) 243(324,423) 249(429,924)

                    Tier 2
                    093(390) 384(483) 387(837) 408(480) 487(847) 495(945)

                    Tier 1
                    269 280 289 908 946

                    Input:
                    023 093 129 138 143 149 183 189 203 219 235 237 239 243 246 249 253 264 269 273 279 280 289 291 293 297 302 314 318 324 325 327 345 347 354 374 381 384 387 390 392 408 413 419 423 426 429 435 437 453 473 480 483 487 491 495 523 534 624 723 729 734 789 798 813 819 837 847 879 897 908 912 914 918 924 927 945 946 978

                    Filters: No filtering selected

                    ---------------

                    As you can see, it can be Deflated to only 27 Boxed picks. or to:

                    20 Straight pick 3 numbers if only Tier 5 is used, which was the most likely right predicted tier for that particular prediction method used at that time.

                    ---------------

                    So 720 straight singles could be reduced or predicted down to only 20 straight singles.

                    ---------------------------- 

                    But four4me is right, it is very unlikely for numbers to be predicted right down to only 1 of them, not so much with the pick 3 but for other lottery games.

                    ---------------

                    By the way, Don't ask me about that particular prediction of mine or about any other one, they are all more or less unique and after I make them, I hardly ever remember how they are mad, as I don't keep records of how particular predictions are made, some of them more or less similar, but never exactly in the very same way, so far anyway.

                    -------------

                    Your friend Fernando. 

                    --------------- 

                    Well! I have already wasted several of my free uses of the Deflate 3 and also of the Inspector 3 trying to prove points to people, but is O.K. I still have enough free tries.

                     

                      LANTERN's avatar - kilroy 28_173_reasonably_small.jpg
                      Tx
                      United States
                      Member #4570
                      May 4, 2004
                      5180 Posts
                      Offline
                      Posted: February 22, 2007, 6:15 am - IP Logged

                      Perhaps it might be best if you all see the original post as it was posted first:

                      http://www.lotterypost.com/thread/134410

                      Try page # 3 or so. 

                      And or:

                      http://www.lotterypost.com/thread/134410/606727?q=lantern 

                      You might have to copy and paste it into your browser's address window and then click enter.

                        guesser's avatar - Lottery-017.jpg

                        United States
                        Member #41383
                        June 16, 2006
                        1969 Posts
                        Offline
                        Posted: February 22, 2007, 2:21 pm - IP Logged

                        This is hilarious.

                        I saw the name of this thread, then I bet myself a million dollars as to who started it, and I was right. Now I have to pay up, LOL.

                        Anyway, 20 billion MIPS - great - but in what language do you write in order to execute the machine code in the most efficient manner: how clean is the code ? 

                          Avatar
                          NY
                          United States
                          Member #23835
                          October 16, 2005
                          3474 Posts
                          Offline
                          Posted: February 22, 2007, 4:16 pm - IP Logged

                          It isn't computing power that's the roadblock. If you had the right information far enough ahead of time you could do it on a home pc that's several years old. If you were patient you could probably do it with a slide rule. The problem you're up against is the GIGO rule. Putting garbage into a faster computer gets you an answer in less time, but the answer that comes out will still be garbage.

                          The problem is that you don't have values for the 57,386 variables that affect each of the balls, and the 48,916 variables that affect the machine(s) every time there's a drawing, and you will never have that information. Even if the equipment was set up ahead of time and you had a week to investigate the best you could hope for would be to come close, because some of those variables change constantly.

                            LANTERN's avatar - kilroy 28_173_reasonably_small.jpg
                            Tx
                            United States
                            Member #4570
                            May 4, 2004
                            5180 Posts
                            Offline
                            Posted: February 22, 2007, 5:00 pm - IP Logged

                            It isn't computing power that's the roadblock. If you had the right information far enough ahead of time you could do it on a home pc that's several years old. If you were patient you could probably do it with a slide rule. The problem you're up against is the GIGO rule. Putting garbage into a faster computer gets you an answer in less time, but the answer that comes out will still be garbage.

                            The problem is that you don't have values for the 57,386 variables that affect each of the balls, and the 48,916 variables that affect the machine(s) every time there's a drawing, and you will never have that information. Even if the equipment was set up ahead of time and you had a week to investigate the best you could hope for would be to come close, because some of those variables change constantly.

                            That is exactly why "Statistical Prediction" is the only form of lottery prediction that is both: Practical and that does have a much better chance of working well enough.

                            Right interpretation and use of the statistical data is the KEY to proper prediction. 

                            Fernando has known that for a very long time. 

                            Statistical prediction does not care any about the physics of lottery drawings.

                            It only cares about statistical data of past draws, it works with random and physical laws only tru the statistics of past draws in other words IMPERSONALLY, not directly.

                            There is no prediction formula based on physics and never will be, not one that will work well enough.

                            There can only be statistical formulas that work well enough.

                            Fernando. 

                              guesser's avatar - Lottery-017.jpg

                              United States
                              Member #41383
                              June 16, 2006
                              1969 Posts
                              Offline
                              Posted: February 22, 2007, 5:00 pm - IP Logged

                              It isn't computing power that's the roadblock. If you had the right information far enough ahead of time you could do it on a home pc that's several years old. If you were patient you could probably do it with a slide rule. The problem you're up against is the GIGO rule. Putting garbage into a faster computer gets you an answer in less time, but the answer that comes out will still be garbage.

                              The problem is that you don't have values for the 57,386 variables that affect each of the balls, and the 48,916 variables that affect the machine(s) every time there's a drawing, and you will never have that information. Even if the equipment was set up ahead of time and you had a week to investigate the best you could hope for would be to come close, because some of those variables change constantly.

                              Yep.

                              Not only do the variables change, but the dynamics of those variables change also.

                              Sure, maybe you can program in the trajectory of how balls bounce, but you cannot predict that one speck of dust, the temperature or humidity on the inside of the ball machine, or the actual current going to the blowers that move the balls - just to name a few.....

                              I have theories, I see trends that hit with 80% accuracy - but they change.  Last week's 88% is next week's 10%, the problem is by the time you notice a trend, it's usually too late to play it, so you just have to track it and see if you can predict when it will come around next.

                              And even at that, 88% is probably not good enough.

                              And I am not even counting the red ball, that one kills me, I suppose I should pay more attention to it.....