Welcome Guest
Log In | Register )
You last visited January 18, 2017, 9:04 am
All times shown are
Eastern Time (GMT-5:00)

Info On RNG'S

Topic closed. 37 replies. Last post 9 years ago by psykomo.

Page 1 of 3
4.65
PrintE-mailLink
CARBOB's avatar - FL LOTTERY_LOGO.png
ORLANDO, FLORIDA
United States
Member #4924
June 3, 2004
5962 Posts
Offline
Posted: September 7, 2007, 6:30 pm - IP Logged

This is from a post on the tennessean.com web page about RNG'S with some links to other sites that pertain to the article.  I didn't write this, that person is listed at the bottm. Found it very informing. I would like some feedback.

 

 

 



Google Junky


Joined: 24 Aug 2007
Posts: 2
PostPosted: Fri Sep 07, 2007 3:04 am    Post subject: To the editor 

I think you ran a great story here. I have more information for you to help you make your next editorial. I did post this once, but not in its full content. So here you go.

It is kind of irritating that state lotteries are using or trying to use computers to randomize numbers rather than simple physics of plastic balls falling at random.
There are only 2 methods that can be used to generate random numbers through a computer and neither is true random.

1. The computer itself(which isn't random at all)
2. A hardware version through physical means

I'll cover them both

1. The Computer
If you know anyone that has ever written programming for a computer you could easily ask them if a computer can randomize. They will give a resounding NO.
A computers version of randomizing is called Pseudo-Random. I have a link to learn about Pseudo-Random farther down.

A little history lesson

Learn more here http://www.picotech.com/applications/colossus.htm

During WW2, the Germans relied mostly on the Enigma machine to encode messages. It attempted to randomize characters that could only be decoded by the receiving machine. The first computer ever built was called the Colossus and built in London at Bletchley Park in 1944. It decoded the code that the Germans had thought to be random. Unknown to the Germans there code had a sequence pattern known as Pseudo-Random.

To continue…
Even today computers are still not able to randomize numbers other than the Pseudo-Random way of doing it.
The technically explained version for Pseudo-Random you can find here
http://acm.uva.es/problemset/v3/350.html


2. A hardware version through physical means
There are only two ways this can be done.
a. decay times from a radio-active material
b. electrical noise from a resistor or semiconductor
An excerpt from a webpage by Robert Davies(Statistics Research Associates Limited)
You can read the full article here http://www.robertnz.net/hwrng.htm


"A hardware random number generator uses a physical phenomenon such as electrical noise from a resistor or semiconductor diode or the decay of a radioactive material for the initial source of randomness. The electronic circuitry of the generator converts this noise to bits and then assembles these into bytes or words for use by the computer.

There may be philosophical question as to whether processes used to generate the random noise are really random – as quantum theory suggests they are. Maybe we could calculate the movements of the electrons in the resistor and so model the noise as a deterministic process. This, of course, would impossible, because of huge number of electrons that would be involved even if the processes were really deterministic."

So knowing that these are the only two options to randomize, which do you choose? Do you choose little plastic balls bouncing at random in a box or a computer?Knowledge is power.

People should know why they shoudn't trust this new system. It isn't just that a configuration wasn't set correctly. Even when the setting is correct you will still have an error in randomizing.

Good luck in all you do,
Google Junky
    Guru101's avatar - rw6jhh
    Indiana
    United States
    Member #48725
    January 7, 2007
    1958 Posts
    Offline
    Posted: September 7, 2007, 7:47 pm - IP Logged

    This is from a post on the tennessean.com web page about RNG'S with some links to other sites that pertain to the article.  I didn't write this, that person is listed at the bottm. Found it very informing. I would like some feedback.

     

     

     



    Google Junky


    Joined: 24 Aug 2007
    Posts: 2
    PostPosted: Fri Sep 07, 2007 3:04 am    Post subject: To the editor 

    I think you ran a great story here. I have more information for you to help you make your next editorial. I did post this once, but not in its full content. So here you go.

    It is kind of irritating that state lotteries are using or trying to use computers to randomize numbers rather than simple physics of plastic balls falling at random.
    There are only 2 methods that can be used to generate random numbers through a computer and neither is true random.

    1. The computer itself(which isn't random at all)
    2. A hardware version through physical means

    I'll cover them both

    1. The Computer
    If you know anyone that has ever written programming for a computer you could easily ask them if a computer can randomize. They will give a resounding NO.
    A computers version of randomizing is called Pseudo-Random. I have a link to learn about Pseudo-Random farther down.

    A little history lesson

    Learn more here http://www.picotech.com/applications/colossus.htm

    During WW2, the Germans relied mostly on the Enigma machine to encode messages. It attempted to randomize characters that could only be decoded by the receiving machine. The first computer ever built was called the Colossus and built in London at Bletchley Park in 1944. It decoded the code that the Germans had thought to be random. Unknown to the Germans there code had a sequence pattern known as Pseudo-Random.

    To continue…
    Even today computers are still not able to randomize numbers other than the Pseudo-Random way of doing it.
    The technically explained version for Pseudo-Random you can find here
    http://acm.uva.es/problemset/v3/350.html


    2. A hardware version through physical means
    There are only two ways this can be done.
    a. decay times from a radio-active material
    b. electrical noise from a resistor or semiconductor
    An excerpt from a webpage by Robert Davies(Statistics Research Associates Limited)
    You can read the full article here http://www.robertnz.net/hwrng.htm


    "A hardware random number generator uses a physical phenomenon such as electrical noise from a resistor or semiconductor diode or the decay of a radioactive material for the initial source of randomness. The electronic circuitry of the generator converts this noise to bits and then assembles these into bytes or words for use by the computer.

    There may be philosophical question as to whether processes used to generate the random noise are really random – as quantum theory suggests they are. Maybe we could calculate the movements of the electrons in the resistor and so model the noise as a deterministic process. This, of course, would impossible, because of huge number of electrons that would be involved even if the processes were really deterministic."

    So knowing that these are the only two options to randomize, which do you choose? Do you choose little plastic balls bouncing at random in a box or a computer?Knowledge is power.

    People should know why they shoudn't trust this new system. It isn't just that a configuration wasn't set correctly. Even when the setting is correct you will still have an error in randomizing.

    Good luck in all you do,
    Google Junky

    My preference for plastic ball drawings is merely for security reasons. Do I think computers, when programmed appropriately, can generate numbers that are "random enough" to humans? Yes I do. Plastic ball drawings take full advantage of science, primarily physics. So basically plastic ball drawings work on a scientific level of randomization.  The human brain however can only view results generated by a randomization process up to a certain point. There are a lot of things that can be used in a combination for initializing an RNG. The most commonly used is time, which is used in a lot of programs that use randomization, but in the case of a lottery program, more variables should probably be taken into consideration. Instead of time alone, which narrows the current time down to a single second, to be taken into consideration for a seed, there are also other variables to look at. There's the current millisecond, current process utililization(includes multiple variables), current memory utilization(includes multiple variables), a psuedo-random number previously generated, as well as other variables. So scientifically speaking, it's not strictly that one process is random and one is not, but that one is more random than the other. Now, when security is taken into consideration, I think plastic ball drawings is ultimately the best process of randomization for picking lottery numbers.

    Gonna win.Big Smile

      LANTERN's avatar - kilroy 28_173_reasonably_small.jpg
      Tx
      United States
      Member #4570
      May 4, 2004
      5180 Posts
      Offline
      Posted: September 8, 2007, 4:09 am - IP Logged

      Do you want Random? Take a look at this:

      http://en.wikipedia.org/wiki/Electron_cloud

      Of course, when we don't know how and or why then we say "Random", that does not mean that there really is any such thing, the real meaning of "Random" might be: "Produced by unknown causes", what is unknown to us now, might not be unknown to people or to some people 20 thousand years from now and or long before that.

      While the changes of weather patterns might appear to be "random" there are definite reasons for it, it has been said that there is nothing as "Random" as the weather and yet now some weather patterns are predicted way in advance.

      That is no longer the unknown that it was. 

      BibleOnline  ParishesOnline  ChristianRadioOnline   MassOnline   Mass

      "Ten measures of beauty descended to the world, nine were taken by Jerusalem."

        LANTERN's avatar - kilroy 28_173_reasonably_small.jpg
        Tx
        United States
        Member #4570
        May 4, 2004
        5180 Posts
        Offline
        Posted: September 8, 2007, 4:50 am - IP Logged

        My preference for plastic ball drawings is merely for security reasons. Do I think computers, when programmed appropriately, can generate numbers that are "random enough" to humans? Yes I do. Plastic ball drawings take full advantage of science, primarily physics. So basically plastic ball drawings work on a scientific level of randomization.  The human brain however can only view results generated by a randomization process up to a certain point. There are a lot of things that can be used in a combination for initializing an RNG. The most commonly used is time, which is used in a lot of programs that use randomization, but in the case of a lottery program, more variables should probably be taken into consideration. Instead of time alone, which narrows the current time down to a single second, to be taken into consideration for a seed, there are also other variables to look at. There's the current millisecond, current process utililization(includes multiple variables), current memory utilization(includes multiple variables), a psuedo-random number previously generated, as well as other variables. So scientifically speaking, it's not strictly that one process is random and one is not, but that one is more random than the other. Now, when security is taken into consideration, I think plastic ball drawings is ultimately the best process of randomization for picking lottery numbers.

        Did you ever hear somebody said?: "Good enough for Government work"

        Yes, computers can produce patterns that to humans appear to be "Random" enough.

        All or most lottery players probably can't tell the difference between lottery patterns produced by an RNG and those of mechanical draw machines, at least this is what I think.

        As to security, they both have their own factors, a programmer might add code that is not well seen and or understood by those checking it, also there could be"Extra Hardware and or software" inside the Pc, including input and or output units, even or some sort of micro wireless receivers and or transmitters., a transmitter and or receiver can appear to be an electrolytic condenser and or some kind of transformer and or other electronic part(s), or even be inside the power supply and or the CD Rom Drive or elsewhere, the power supply would be a good place as it is connected to the wall's electric outlet and it also has many 'Power" wires going into the Pc.

        There could also be Rom Memory ICs somewhere inside that should not be there or something like that, including software and or hardware switches.

        BibleOnline  ParishesOnline  ChristianRadioOnline   MassOnline   Mass

        "Ten measures of beauty descended to the world, nine were taken by Jerusalem."

          Avatar
          Urbandale, IA
          United States
          Member #8624
          November 11, 2004
          115 Posts
          Offline
          Posted: September 14, 2007, 9:03 pm - IP Logged

          Did you ever hear somebody said?: "Good enough for Government work"

          Yes, computers can produce patterns that to humans appear to be "Random" enough.

          All or most lottery players probably can't tell the difference between lottery patterns produced by an RNG and those of mechanical draw machines, at least this is what I think.

          As to security, they both have their own factors, a programmer might add code that is not well seen and or understood by those checking it, also there could be"Extra Hardware and or software" inside the Pc, including input and or output units, even or some sort of micro wireless receivers and or transmitters., a transmitter and or receiver can appear to be an electrolytic condenser and or some kind of transformer and or other electronic part(s), or even be inside the power supply and or the CD Rom Drive or elsewhere, the power supply would be a good place as it is connected to the wall's electric outlet and it also has many 'Power" wires going into the Pc.

          There could also be Rom Memory ICs somewhere inside that should not be there or something like that, including software and or hardware switches.

          Computers are not normally TRUE-random, but they can be set up to be truly random.  There are a dozen RNG's operating at lotteries now that are TRUE-random machines.  They contain a radioactive element and the seed is taken from random decay.  That said, the Pseudo-RNG machines out there are still far good enough for a lottery drawing.  Lotteries are a double-random process anyway since the player number is "random" (okay Pseudo-RNG since picked by humans or the RNG in the terminal).

          As for the security measures – just about everyone can figure out how to cheat a ball machine (and has).  Cheating is tested and secured computer is much harder.  You don’t think these machines are hooked up to the Internet or are equipped with a wireless card?  They are treated just like the ball machines and are locked up and secured (or should be).  The machine is torn apart by independent testers.  They are aware of all of these "easy" could-be's and more.  The programs are examined line by line and recompiled.  Then the system goes through millions of test draws for analysis.

          RNG’s do bring a whole different set of problems with them, but they are in no way less secure than a ball machine.  I would expect players to want balls machines because they can NOT be perfectly random.  There are going to be inherent differences in weights, size, density that can be analyzed and used.  RNG’s give you little hope (unless you are aware of the different faults of RNG’s and play for those – like a cache not getting flushed out so that the same numbers come up, etc.).      

            LOTTOMIKE's avatar - cash money.jpg
            Tennessee
            United States
            Member #7853
            October 15, 2004
            11338 Posts
            Offline
            Posted: September 14, 2007, 9:15 pm - IP Logged

            how are you chuck?  i was wondering if powerball had any plans on going RNG since hot lotto made the switch recently?

                                                              thanks.

              Avatar
              Urbandale, IA
              United States
              Member #8624
              November 11, 2004
              115 Posts
              Offline
              Posted: September 14, 2007, 9:20 pm - IP Logged

              how are you chuck?  i was wondering if powerball had any plans on going RNG since hot lotto made the switch recently?

                                                                thanks.

              I'm Good (but incognito here - shhhh).  Powerball will not switch to RNG.  Just too much protest to make the change.  Kicking over the goofy Power Play machine (which can take the whole draw show to pick a number) is a good bet though - just for getting a result in a specific time. 

                LOTTOMIKE's avatar - cash money.jpg
                Tennessee
                United States
                Member #7853
                October 15, 2004
                11338 Posts
                Offline
                Posted: September 14, 2007, 9:29 pm - IP Logged

                i'm glad to hear that.especially after the dreaded tennessee debacle.

                  Todd's avatar - Cylon 2.gif
                  Chief Bottle Washer
                  New Jersey
                  United States
                  Member #1
                  May 31, 2000
                  23348 Posts
                  Offline
                  Posted: September 14, 2007, 9:30 pm - IP Logged

                  Computers are not normally TRUE-random, but they can be set up to be truly random.  There are a dozen RNG's operating at lotteries now that are TRUE-random machines.  They contain a radioactive element and the seed is taken from random decay.  That said, the Pseudo-RNG machines out there are still far good enough for a lottery drawing.  Lotteries are a double-random process anyway since the player number is "random" (okay Pseudo-RNG since picked by humans or the RNG in the terminal).

                  As for the security measures – just about everyone can figure out how to cheat a ball machine (and has).  Cheating is tested and secured computer is much harder.  You don’t think these machines are hooked up to the Internet or are equipped with a wireless card?  They are treated just like the ball machines and are locked up and secured (or should be).  The machine is torn apart by independent testers.  They are aware of all of these "easy" could-be's and more.  The programs are examined line by line and recompiled.  Then the system goes through millions of test draws for analysis.

                  RNG’s do bring a whole different set of problems with them, but they are in no way less secure than a ball machine.  I would expect players to want balls machines because they can NOT be perfectly random.  There are going to be inherent differences in weights, size, density that can be analyzed and used.  RNG’s give you little hope (unless you are aware of the different faults of RNG’s and play for those – like a cache not getting flushed out so that the same numbers come up, etc.).      

                  Not at all.  I want ball drawings because any layman can see or measure or detect problems or fraud.

                  That's why computerized drawings in California, Kansas, and Tennessee had problems that festered for weeks and in some cases months -- before PLAYERS in each case detected problems by looking at the poor probability of the bad drawings.

                  That could never have happened with ball drawings -- and HAS never happened.

                  The vast majority of lottery players who discuss this issue here and on other sites (like the Tennessean newspaper site) have been talking about this specific issue -- the possibility of committing fraud.  Only a very small percentage mentions that they think they can "beat" one or the other.

                  (In fact, I have seen far more who think they can "beat" computerized drawings, yet look at the polls, which each time show an overwhelming percentage of players against computerized drawings.)

                  Players don't want cartoons showing them the winning numbers.  It looks fake, and it IS fake.  The only excitement of a lottery drawing comes from watching the balls mix in the drum and get drawn one at a time.

                  Computerized drawings do have a purpose -- for games that would be unpractical to be drawn with lottery balls.  For example, 5-minute keno games and raffle drawings. 

                  For regular daily and periodic drawings, there is no excuse to use computers.  Out of the billions of dollars brought in by each state, the financial argument is ridiculous, and in fact the players who give up playing because of the fake drawings lower the revenue.  The Tennessee legislature is very worried about this, and they are not worried for no reason.

                   

                  Check the State Lottery Report Card
                  What grade did your lottery earn?

                   

                  Sign the Petition for True Lottery Drawings
                  Help eliminate computerized drawings!

                    Avatar
                    Urbandale, IA
                    United States
                    Member #8624
                    November 11, 2004
                    115 Posts
                    Offline
                    Posted: September 14, 2007, 9:32 pm - IP Logged

                    i'm glad to hear that.especially after the dreaded tennessee debacle.

                    You have to wonder if some disgruntled employee was at work there.  The RNG's were set wrong; the lottery phone number on brochures were replaced by an adult hot line number; the wrong jackpot amount is put on tickets?  What are the odds of writing down a phone number and having it be an adult hot line?

                      Avatar
                      Urbandale, IA
                      United States
                      Member #8624
                      November 11, 2004
                      115 Posts
                      Offline
                      Posted: September 14, 2007, 9:45 pm - IP Logged

                      Not at all.  I want ball drawings because any layman can see or measure or detect problems or fraud.

                      That's why computerized drawings in California, Kansas, and Tennessee had problems that festered for weeks and in some cases months -- before PLAYERS in each case detected problems by looking at the poor probability of the bad drawings.

                      That could never have happened with ball drawings -- and HAS never happened.

                      The vast majority of lottery players who discuss this issue here and on other sites (like the Tennessean newspaper site) have been talking about this specific issue -- the possibility of committing fraud.  Only a very small percentage mentions that they think they can "beat" one or the other.

                      (In fact, I have seen far more who think they can "beat" computerized drawings, yet look at the polls, which each time show an overwhelming percentage of players against computerized drawings.)

                      Players don't want cartoons showing them the winning numbers.  It looks fake, and it IS fake.  The only excitement of a lottery drawing comes from watching the balls mix in the drum and get drawn one at a time.

                      Computerized drawings do have a purpose -- for games that would be unpractical to be drawn with lottery balls.  For example, 5-minute keno games and raffle drawings. 

                      For regular daily and periodic drawings, there is no excuse to use computers.  Out of the billions of dollars brought in by each state, the financial argument is ridiculous, and in fact the players who give up playing because of the fake drawings lower the revenue.  The Tennessee legislature is very worried about this, and they are not worried for no reason.

                      Hey Todd, 

                      I'm not sure that any layman can see or measure or detect fraud on a ball machine.  At least not a person who is not at the studio with some testing equipment.  Magnets in balls; paint in Ping-Pong balls, etc. may not be detectable on a TV screen at all.  Players don't catch these problems because they are "fixed" once the fraud is committed.  I've never heard a person suggest a meaninful way to defaud an RNG (except for things like "well, I would write the code to draw my numbers".  Usually, they are concerned about radio or Internet connections or secret ROM chips and other things that just don't exist.  Lottery RNG's are not your average computer systems sitting on someone's desk.

                      I'm not sure that the majority of players care at all, but there are certainly a number of people who care deeply.  Certainly, the majority who bother to write about the subject (if not all of them) dispise RNG's.  I disagree with them that RNG's are easier to defraud, but they are the players and so you just gotta consider their feellings.

                      Frankly, it is not something that I'm all that passionate about (unlike taking the annuity).  Even though a tiny percentage of players really care about RNG's, they are customers and deserve to be heard.  I would not expect any major game to go RNG any time soon.

                        Todd's avatar - Cylon 2.gif
                        Chief Bottle Washer
                        New Jersey
                        United States
                        Member #1
                        May 31, 2000
                        23348 Posts
                        Offline
                        Posted: September 14, 2007, 10:44 pm - IP Logged

                        Hey Todd, 

                        I'm not sure that any layman can see or measure or detect fraud on a ball machine.  At least not a person who is not at the studio with some testing equipment.  Magnets in balls; paint in Ping-Pong balls, etc. may not be detectable on a TV screen at all.  Players don't catch these problems because they are "fixed" once the fraud is committed.  I've never heard a person suggest a meaninful way to defaud an RNG (except for things like "well, I would write the code to draw my numbers".  Usually, they are concerned about radio or Internet connections or secret ROM chips and other things that just don't exist.  Lottery RNG's are not your average computer systems sitting on someone's desk.

                        I'm not sure that the majority of players care at all, but there are certainly a number of people who care deeply.  Certainly, the majority who bother to write about the subject (if not all of them) dispise RNG's.  I disagree with them that RNG's are easier to defraud, but they are the players and so you just gotta consider their feellings.

                        Frankly, it is not something that I'm all that passionate about (unlike taking the annuity).  Even though a tiny percentage of players really care about RNG's, they are customers and deserve to be heard.  I would not expect any major game to go RNG any time soon.

                        Respectfully, I disagree.  I think the players who "don't care" are the ones who don't know about it, or are only casual players who play when the JP is big.

                        However, when even casual players find out about computerized drawings they are at first surprised (because the lottery tries to trick people into believing balls are still being used), and then the next reaction is "why?"  When they learn even more their reaction becomes, "we should switch back to balls."

                        These are not comments made by someone with a hidden agenda, or some anti-technology nut.

                        I am a "computer expert", and I have programmed computers for the entire 20 years of my career.  In fact, I have been programming computers for 8 years longer than that -- before i even started a career.  You are discussing this with someone who thinks the computer is the best invention ever made by mankind.  (No joke.)

                        Part of becoming a computer expert -- someone relied upon to deliver the best possible technology advice at both a high level and a macro level -- is knowing the proper application of computers  Not just "yes" and "no" to use computers or not, but also the degree of automation, the automation of business processes, the type of technology models in use, types of technology, etc., etc.  I have designed strategic systems for literally the largest companies in the world, and the smallest one-man shops.

                        I am telling you, with absolute certainty, both logically and physically, that:

                        1. There is no "unhackable" computer system,
                        2. A good programmer can cover their tracks so well that nobody in the world could know about the compromise,
                        3. Lottery balls, or anything else in the real, physical world is easier and more fool-proof to audit, and,
                        4. Compared to a mechanical drawing machine, computers have a vastly greater number of points of failure -- seen and unseen, known and unknown -- and the only advantage in using them is that it is easier for the lottery to conduct. 

                        There is no benefit to lottery players to having computerized drawings, and in fact they lose a lot by going to them -- namely,

                        1. Confidence in the results (as illustrated by the recent and past problems),
                        2. Lottery personalities,
                        3. Televised drawings,
                        4. Sense of anticipation and excitement.

                        Again, ALL of the benefit to computerized drawings falls in the lap of the lottery director, where it should not be.

                        I welcome the lively debate, and I would be interested in hearing how I am wrong about these specific points.

                         

                        Check the State Lottery Report Card
                        What grade did your lottery earn?

                         

                        Sign the Petition for True Lottery Drawings
                        Help eliminate computerized drawings!

                          Avatar
                          Urbandale, IA
                          United States
                          Member #8624
                          November 11, 2004
                          115 Posts
                          Offline
                          Posted: September 14, 2007, 11:19 pm - IP Logged

                          Respectfully, I disagree.  I think the players who "don't care" are the ones who don't know about it, or are only casual players who play when the JP is big.

                          However, when even casual players find out about computerized drawings they are at first surprised (because the lottery tries to trick people into believing balls are still being used), and then the next reaction is "why?"  When they learn even more their reaction becomes, "we should switch back to balls."

                          These are not comments made by someone with a hidden agenda, or some anti-technology nut.

                          I am a "computer expert", and I have programmed computers for the entire 20 years of my career.  In fact, I have been programming computers for 8 years longer than that -- before i even started a career.  You are discussing this with someone who thinks the computer is the best invention ever made by mankind.  (No joke.)

                          Part of becoming a computer expert -- someone relied upon to deliver the best possible technology advice at both a high level and a macro level -- is knowing the proper application of computers  Not just "yes" and "no" to use computers or not, but also the degree of automation, the automation of business processes, the type of technology models in use, types of technology, etc., etc.  I have designed strategic systems for literally the largest companies in the world, and the smallest one-man shops.

                          I am telling you, with absolute certainty, both logically and physically, that:

                          1. There is no "unhackable" computer system,
                          2. A good programmer can cover their tracks so well that nobody in the world could know about the compromise,
                          3. Lottery balls, or anything else in the real, physical world is easier and more fool-proof to audit, and,
                          4. Compared to a mechanical drawing machine, computers have a vastly greater number of points of failure -- seen and unseen, known and unknown -- and the only advantage in using them is that it is easier for the lottery to conduct. 

                          There is no benefit to lottery players to having computerized drawings, and in fact they lose a lot by going to them -- namely,

                          1. Confidence in the results (as illustrated by the recent and past problems),
                          2. Lottery personalities,
                          3. Televised drawings,
                          4. Sense of anticipation and excitement.

                          Again, ALL of the benefit to computerized drawings falls in the lap of the lottery director, where it should not be.

                          I welcome the lively debate, and I would be interested in hearing how I am wrong about these specific points.

                          Well, I don't know if I want to get into a big debate.  As I say, I'm not all that passionate about this subject. 

                          1.  There are indeed no "unhackable" computer systems - just as there are no "unhackable" physical draw systems.  Anything is possible, but we deal with probabilities.  The RNG needs to be built in guarded conditions, tested by an independent lab (fully with lines of code reviewed), systems built in later to self-test and create hashes, and then secured in the same way as all draw equipment.

                          2. Great programmers can do a lot but a great tester knows a lot of tricks too.  I would not recommend having just any John Doe testing RNG's.  It does need to be built by someone who can pass the background checks and it does need to be tested by the very best.  There is some great talent out there to check these machines.

                          3. Lottery balls are easier to audit for incompetent fraud, but not for someone who knows what they are doing.  I think the difference really is that there are no "incompetent" RNG frauds commited because you have to start from a much higher level to even attempt a thing.  Any Joe can try to stick some magnets into rubber balls or can try to inject white paint into Ping-Pong balls, but any Joe is not going to try to somehow take apart a machine (which is always under guard of course) and slip in a doctored ROM (such a thing would be caught by the systems hash checks anyway).  Any decent RNG has a specialized hash check with an algorithm that is held by yet another independent party that can check any draw report issued by the RNG.  Part of the problem is that everyone has a computer on their desk and they think it is the same as the lottery's RNG. 

                          4. I haven't really counted up the points of failure.  They exist with both systems, but there are so very many things that have gone wrong with mechanicals that I would even hate to start counting.  The RNG's are easier on the lottery to run and that is likely the reason that they are used.  Cost is there too, but I don't disagree that the bigger dollar picture makes all such costs pretty small.

                          The other points are more opinion items.  Draw machine problems also cause a loss of confidence.  You don't see a lot of problems with mechanical draws any more, but RNG's are still pretty new.  RNG problems will fade too.  TV stations are moving away from giving free air-time to lotteries and this if often the larger cost factor.  Viewership is way down for lottery draws - though I hope that can be changed (we'll see - watch this space, as they say).  I certainly can't argue with anticipation and excitement since those are personal, but very few watch the drawing any more.  The anticipation these days comes from the web site (most likely) or just stopping by to have the ticket checked at the retailer.

                          Just some quick comments here.  I'm not even going to take a stab at convincing you.  I wouldn't try to conivince the masses from 500 years ago that the world was not flat - just have to wait around a bit  <GRIN>

                          I love the computer stuff too and even had a hand in developing RNG's but I do still tend to like the mechanical draws (just not the Ping-Pong balls).  But then I also like to watch vinyl records go round and round (and those big reel tape machines - oooooohh).  If we could just get folks to watch the draws!

                          Again, I don't disagree that there some strong feelings out there against RNG's.  Even though the numbers don't seem to be there (sales don't change after a lottery moves to RNG's), there is really no rush to upset a player if you don't have to. 

                            four4me's avatar - gate1
                            MD
                            United States
                            Member #1701
                            June 18, 2003
                            8392 Posts
                            Offline
                            Posted: September 15, 2007, 12:01 am - IP Logged

                            Well, I don't know if I want to get into a big debate.  As I say, I'm not all that passionate about this subject. 

                            1.  There are indeed no "unhackable" computer systems - just as there are no "unhackable" physical draw systems.  Anything is possible, but we deal with probabilities.  The RNG needs to be built in guarded conditions, tested by an independent lab (fully with lines of code reviewed), systems built in later to self-test and create hashes, and then secured in the same way as all draw equipment.

                            2. Great programmers can do a lot but a great tester knows a lot of tricks too.  I would not recommend having just any John Doe testing RNG's.  It does need to be built by someone who can pass the background checks and it does need to be tested by the very best.  There is some great talent out there to check these machines.

                            3. Lottery balls are easier to audit for incompetent fraud, but not for someone who knows what they are doing.  I think the difference really is that there are no "incompetent" RNG frauds commited because you have to start from a much higher level to even attempt a thing.  Any Joe can try to stick some magnets into rubber balls or can try to inject white paint into Ping-Pong balls, but any Joe is not going to try to somehow take apart a machine (which is always under guard of course) and slip in a doctored ROM (such a thing would be caught by the systems hash checks anyway).  Any decent RNG has a specialized hash check with an algorithm that is held by yet another independent party that can check any draw report issued by the RNG.  Part of the problem is that everyone has a computer on their desk and they think it is the same as the lottery's RNG. 

                            4. I haven't really counted up the points of failure.  They exist with both systems, but there are so very many things that have gone wrong with mechanicals that I would even hate to start counting.  The RNG's are easier on the lottery to run and that is likely the reason that they are used.  Cost is there too, but I don't disagree that the bigger dollar picture makes all such costs pretty small.

                            The other points are more opinion items.  Draw machine problems also cause a loss of confidence.  You don't see a lot of problems with mechanical draws any more, but RNG's are still pretty new.  RNG problems will fade too.  TV stations are moving away from giving free air-time to lotteries and this if often the larger cost factor.  Viewership is way down for lottery draws - though I hope that can be changed (we'll see - watch this space, as they say).  I certainly can't argue with anticipation and excitement since those are personal, but very few watch the drawing any more.  The anticipation these days comes from the web site (most likely) or just stopping by to have the ticket checked at the retailer.

                            Just some quick comments here.  I'm not even going to take a stab at convincing you.  I wouldn't try to conivince the masses from 500 years ago that the world was not flat - just have to wait around a bit  <GRIN>

                            I love the computer stuff too and even had a hand in developing RNG's but I do still tend to like the mechanical draws (just not the Ping-Pong balls).  But then I also like to watch vinyl records go round and round (and those big reel tape machines - oooooohh).  If we could just get folks to watch the draws!

                            Again, I don't disagree that there some strong feelings out there against RNG's.  Even though the numbers don't seem to be there (sales don't change after a lottery moves to RNG's), there is really no rush to upset a player if you don't have to. 

                            The numbered ping pong balls they use today have seven coats of hardened lacquerer on them they are not easily punctured.

                            In the old days they were plain ping pong balls with silk screened numerals on them and before that they were hand painted numerals.

                            Want to know how i know all this it's because they make the ping pong balls used in most lotteries for the US and abroad here in Maryland. And i know some of the people who handled the job.

                              LANTERN's avatar - kilroy 28_173_reasonably_small.jpg
                              Tx
                              United States
                              Member #4570
                              May 4, 2004
                              5180 Posts
                              Offline
                              Posted: September 15, 2007, 4:34 am - IP Logged

                              http://www.random.org/integers/

                              Random Integer Generator

                              Here are your random numbers:

                              4   8   1
                              8   2   1
                              4   3   7
                              5   2   9
                              7   0   1
                              7   6   0
                              4   0   9
                              7   8   4
                              5   1   4
                              8   9   7
                              6   0   8
                              0   9   6
                              2   3   4
                              5   9   0
                              5   0   0
                              4   6   4
                              3   0   1
                              0   9   5
                              8   5   5
                              8   1   5
                              1   3   5
                              6   1   6
                              4   4   0
                              5   4   2
                              3   3   1
                              5   7   6
                              5   7   5
                              4   8   4
                              2   8   9
                              5   8   3

                              Timestamp: 2007-09-15 08:10:42 IST

                              Note: The numbers are generated left to right, i.e., across columns.

                              --------

                              As can be seen I got and or made those "Pick 3" numbers from  www.random.org so those "pick 3 numbers should be as "Random" as possible.

                              4   3   7      14 5 LLH EOO
                              5   2   9      16 7 HLH OEO
                              7   0   1      08 8 HLL OEO
                              7   6   0      13 4 HHL OEE
                              4   0   9      13 4 LLH EEO
                              7   8   4      19 1 HHL OEE
                              5   1   4      10 1 HLL OOE
                              8   9   7      24 6 HHH EOO
                              6   0   8      14 5 HLH EEE
                              0   9   6      15 6 LHH EOE
                              2   3   4      09 9 LLL EOE
                              5   9   0      14 5 HHL OOE
                              5   0   0      05 5 HLL OEE
                              4   6   4      14 5 LHL EEE
                              3   0   1      04 4 LLL OEO
                              0   9   5      14 4
                              8   5   5      18 9
                              8   1   5      14 5
                              1   3   5      09 9
                              6   1   6      13 4
                              4   4   0      08 8
                              5   4   2      11 2
                              3   3   1      07 7
                              5   7   6      18 9
                              5   7   5      17 8
                              4   8   4      16 7
                              2   8   9      19 1
                              5   8   3      16 7


                              Well, since Tx and the other lotteries do pre-tests, I can't say which appear to be more or less random, these or those.

                              But if the pre-tests are not taken into account, then I would say that the numbers produced by the Tx lottery do appear to be more random, to me, but what is random to me, it might not be to others.

                              To me, patterns that appear to me to be more random might be easier for me to predict, but one most do with what we have, if we have to, then we most find tricks to use, to some limits there most be 'LoopHoles" that can be exploited.

                              I myself don't care if the draws and their patterns appear to be random or not, so long as that is how they are produced by the machines, but if people thru the use of pre-tests or by other means get their own numbers and not those numbers that were meant to be then I do care and it is easier for people to get their numbers thru the use of RNG pre-tests than thru the use of machine drawn pre-tests, as it is much easier and faster to do RNG pre-tests.

                              To me the main problem is with the pre-tests and not so much with the RNGs themselves, the pre-tests and the operators of the machines.

                              Predicting winning lottery numbers is going to be hard no matter what, but it is much worse due to the lottery operators and their pre-tests, after all they quit selling numbers at least 30 minutes before the draws, that gives the lottery people 30 minutes, they should quit selling numbers 10 minutes before the draws. 

                              BibleOnline  ParishesOnline  ChristianRadioOnline   MassOnline   Mass

                              "Ten measures of beauty descended to the world, nine were taken by Jerusalem."