Has anyone tried neural-lotto.net?

Thanks for the link. Here are 2 quotes that caught my untrained eye :

There are up to 999 hidden layers each with up to 990 neurons, coupled with more than 20,000 perceptrons which results in over 1 million active neurons. This complex network features dynamic multithreaded backpropagation, evolutionary algorithmic and gene expression programming, probabilistic metaheuristics and expectation-maximization, with a core non-parametric statistical model.

The NeuralReality AI Engine — the Neural-Lotto core, took nearly 150,000 man-hours to program, test and deploy, and more than US$3.2 million to implement. Over the course of 8 years, Neural-Lotto underwent many software and hardware changes and upgrades, aimed soley at improving overall performance, stability and efficiency.

That was a lot of time and effort.

HotPicks

2019-07-03 06:03:16 UTC

Post by HotPicks
Hi,
Since they do not provide statistics, I found an hidden page at https://www.neural-lotto.net/index.php/en/component/k2/item/26-free-promo-code
where you can get a promo code for two games, so if many people get a promo we can check how well it performs.

That is for the NeuralReality AI Engine, but they also have the KGA6, but I do not know if the free promo code is valid also for the KGA6, I tried the NeuralReality AI Engine, four free tickets in a row, but I only got one number. I also challenged them on facebook, but I did not get any reply.

So when you get the free promo code use them on the KGA6 which seems more advanced.

I personally do not believe that AI or Neural Networks will work with the lottery.

2019-07-03 09:03:40 UTC

I believe that there is a pattern to the numbers drawn.

I think AI and neural networks stand the best chance of identifying this
pattern. So I am disinclined to just dismiss the whole idea as out of hand.

On the other hand,

If you invented an algorithm that could predict numbers, would you share it
with the world?

And what's more how long before you were banned from playing.

2019-07-03 09:06:27 UTC

Correction : Scratch out the best, and leave it at a chance...

HotPicks

2019-07-03 09:34:03 UTC

Well, obviously their algorithm is not perfect to get the right draws every time otherwise they would not need to sell it to the public.

I do not think there is a pattern, and the reason is that I made an experiment:

I re-wrote Saliu software:power632 and made it using 2000 layers instead of 4.

I analyzed 86 draws and yes, there was a pattern, for instance the easiest filter called ALL repeated 7 times in 86 draws in a range from 37 to 40 layers on 2000. So I created the files with those positions and for one year I applied those filters without success and even when a got a result it was a wrong result.

So seems that mathematics is an opinion.

2019-07-04 01:05:46 UTC

Post by HotPicks
Well, obviously their algorithm is not perfect to get the right draws every time otherwise they would not need to sell it to the public.
I re-wrote Saliu software:power632 and made it using 2000 layers instead of 4.
I analyzed 86 draws and yes, there was a pattern, for instance the easiest filter called ALL repeated 7 times in 86 draws in a range from 37 to 40 layers on 2000. So I created the files with those positions and for one year I applied those filters without success and even when a got a result it was a wrong result.
So seems that mathematics is an opinion.

Can I ask :

1) You start of saying you don't think there is is pattern, but further down you say there is a pattern. Are you talking about two different things?

2) Why 2000 ? It seems an arbitary number.
How many nodes per layer ?
Was it feed forward ?
With back-prop, how long did a epoch take ?
Did you settle upon using unsupervised learning ?

3) Why 86 draws ? ( 50/52 or 100/104 is what I expected for a year )

4) Which lottery ( /country ). My interest lies in the UK & Euro Lotteries
rather than the multitude out there.

HotPicks

2019-07-04 06:32:33 UTC

1) There is a pattern but not with AI or Neural Network, I used Saliu software (power632) modified to use 2000 layers instead of 4 layers.
Power632 uses 4 layers of 10 filters. So I re-wrote the software to use 2000 layers and analyzed each filter, the filter that repeated more in 86 draws is the ALL filter, 7 times on 86 draws from 37 to 40 positions, but after a year those filter either did not re-appear o did not give the right result.
So Mathematics is an opinion.

2) 2000 because the limitation of history, no AI or Neural Network for me, they are a waste of time.

3) 86 draws, because I had to do combinations of up to 7 at the time out of 86 draws -> 5373200880 combinations, it took more then one week. So I could not analyze more than 86 draws without a supercomputer.

4) Old UK lottery 6/49

The only lottery worldwide worth playing are:

1) Euromillions hot pick (5 numbers £1000000)

2) Italy lotto (5 numbers 6000000 euros, but too many numbers to choose from (90))

3) MegaMillions and Powerball

4) Euromillions

2019-07-04 10:20:37 UTC

Thank you for sharing.

When you mentioned "layers" I jumped to the idea you were using a neural
net but this does not appear to be the case.

I am not familiar with power632.

I dont know if you have applied the software to the recent 6/59. I would be
interested to see if you found a pattern in there.

As a bit of background:

I got interested in the 6/49 because I noticed a pattern too. The problem
was that whilst I ( human ) can see a pattern, its not so easy to teach
a machine to see the same. I tried a few algorithmns but working on previous
draws is no guarantee those patterns will continue going forwards.

Motivation is also a factor. If something doesnt work over the next two or
three draws, I give up and onwards to the next idea.

When 6/59 came along, those same patterns fizzled and I lost interest in
that.

I have looked at one other lottery which was 6/45. I assumed I could take
ideas from 6/59 and apply it to 6/45 but it didnt seem to translate.

HotPicks

2019-07-08 09:32:38 UTC

Post by NN
Thank you for sharing.
When you mentioned "layers" I jumped to the idea you were using a neural
net but this does not appear to be the case.
I am not familiar with power632.
I dont know if you have applied the software to the recent 6/59. I would be
interested to see if you found a pattern in there.
I got interested in the 6/49 because I noticed a pattern too. The problem
was that whilst I ( human ) can see a pattern, its not so easy to teach
a machine to see the same. I tried a few algorithmns but working on previous
draws is no guarantee those patterns will continue going forwards.
Motivation is also a factor. If something doesnt work over the next two or
three draws, I give up and onwards to the next idea.
When 6/59 came along, those same patterns fizzled and I lost interest in
that.
I have looked at one other lottery which was 6/45. I assumed I could take
ideas from 6/59 and apply it to 6/45 but it didnt seem to translate.

I went back to check it, and in my research, I used both 6/49 & 6/59 that is because 6/59 was just out, so there were no enough draws to use 2000 layers.
So even the 6/49 draws were decreased 59 times in order to use both 6/49 and 6/59 together. Now because the UK lottery has 2456 draws so far, you can have more than 2000 layers. E.g. 2400 layers and 56 to check against those layers.

HotPicks

2019-07-04 10:38:52 UTC

No, after the filters either did not occur again or did not give the right result I was puzzled because why they reoccur seven times on 87 and then either not reappear for a year or did not give the right result, so I assumed that either statistics is bullshit or life is against me. As I said it took me a week to check all the combinations of 7/87 and after a year not getting those filters again a gave up and starting using another strategy.

I recently switched to Euromillions since lotto is not worth the effort 6/59 for a small jackpot, not even lotto hot picks is worth since you can get more money with Euromillions hot picks.

I believe the only way is to brute force, my software run on gpu and when I was still looking into uk lotto 6/59 I was able to check 50 filters x 45057474 combinations = 2252873700 in less than a millisecond.

This is a link to a strategy of power632:

https://saliu.com/bbs/messages/701.html

Ion Saliu

2019-07-05 10:44:34 UTC

Post by HotPicks
No, after the filters either did not occur again or did not give the right result I was puzzled because why they reoccur seven times on 87 and then either not reappear for a year or did not give the right result, so I assumed that either statistics is bullshit or life is against me. As I said it took me a week to check all the combinations of 7/87 and after a year not getting those filters again a gave up and starting using another strategy.
I recently switched to Euromillions since lotto is not worth the effort 6/59 for a small jackpot, not even lotto hot picks is worth since you can get more money with Euromillions hot picks.
I believe the only way is to brute force, my software run on gpu and when I was still looking into uk lotto 6/59 I was able to check 50 filters x 45057474 combinations = 2252873700 in less than a millisecond.
https://saliu.com/bbs/messages/701.html

Kulai?

Here is the latest in W6 and MD6 reports. The real lotto game is 6/49. The values for ‘median’ and ‘average’ are far more precisely calculated by the software.

You did have a bright idea in applying the ‘Least6’ file based on the ‘least pairings’ for each number. If you look at the PairGrid6.1 report, you notice that in a majority of cases at least one of the pairs has a rank above 30. So, you generate all combinations for all the numbers in the game (e.g. 4) with their top-30 pairings.

But you will be wrong: The winning combination will be missing because a pairing for a number in the draw was NOT ranked under 30. That’s when the potent ‘LIE elimination’ strategy steps in. You only need to be right once and hit that big jackpot!

• https://saliu.com/lie-lottery-strategies-pairs.html
• LIE Elimination: Lottery, Lotto Strategy in Reverse for Pairs, Number Frequency.

Best of luck, axiomaticule!

* LOTTO-6 Winning Pattern - Layer 1 W6.1 *
* Draws analyzed: 1000

Line Sum Pot One Two Three Four Del4 Del5 Del6 Bun6
no. 1 1 1 1 1 1 1 1 1 1
-----------------------------------------------------------------------------
Med.> 87 8 0 4 35 695 284 9781 794862 9
-----------------------------------------------------------------------------
Avg.> 248 9 0 5 52 1104 466 17483 1364351 14
-----------------------------------------------------------------------------
StDv> 1405 6 1 7 52 1234 604 25002 1691627 15
-----------------------------------------------------------------------------
1 399+ 8+ 0 1- 17- 1931+ 113- 374- 212099- 13+
2 22- 5- 0- 2+ 56- 442+ 406- 15598+ 1662789- 8-
3 258+ 7- 1+ 1 148+ 361- 463+ 4801- 1750926+ 14-
4 161- 11 0 1- 130+ 1364- 350+ 49574+ 1361406+ 39+
5 163+ 11+ 0 2- 53+ 5074- 1- 10821- 929243- 18+
6 88- 10+ 0 8+ 41- 5791+ 15- 19026+ 2766012+ 5+
7 398+ 4- 0 0- 117+ 161- 26- 14424+ 684156+ 0-
8 273+ 11- 0- 4- 11- 633- 343+ 4431- 449080+ 1-
9 18- 15+ 1+ 5+ 63- 1539+ 13- 12736- 39989- 19+
10 2623+ 5- 0 3+ 125+ 132- 132+ 17977+ 40617- 16-
11 130+ 13+ 0 1- 29- 1112+ 31- 416+ 476770+ 25+
12 100- 7+ 0 5+ 41+ 962- 271+ 271- 28357- 21-
13 121- 5- 0 0- 27- 3020+ 9- 17278+ 252896- 28+
14 499+ 7- 0- 8 28+ 66- 733+ 12424+ 1818203- 8+
15 70- 15- 2+ 8+ 13- 183- 700+ 1446- 1828786+ 1-
…

* MD6 Winning Reports - Layer 1 MD6.1 *
* Draws analyzed: 1000

Line Ion1 Ion2 Ion3 Ion4 Ion5 Any1 Any2 Any5 Any6 Ver1 Ver2 Ver5 Ver6
no. 1 1 1 1 1 1 1 1 1 1 1 1 1
---------------------------------------------------------------------------------
Med> 14 41 5 2 12 0 2 10 17 2 6 32 64
---------------------------------------------------------------------------------
Avg> 21 111 9 3 195 1 2 11 18 3 8 51 149
---------------------------------------------------------------------------------
StDv> 22 448 27 4 2082 1 2 5 10 4 7 93 565
---------------------------------------------------------------------------------
1 25+ 11- 5+ 1- 46- 1+ 1 12- 17- 5+ 12+ 146+ 236+
2 20- 94- 2 4 267+ 0- 1 14- 22- 0- 2- 30- 69-
3 35+ 313+ 2- 4+ 4- 1+ 1 15+ 26 22+ 37+ 48+ 75+
4 30+ 29- 5- 3+ 17+ 0 1 7+ 26+ 0 1- 38+ 40-
5 1- 116+ 8+ 1- 6 0- 1- 6- 21- 0- 7- 22- 41-
6 15+ 52+ 5- 3+ 6- 2+ 3+ 13+ 38+ 4+ 11+ 38+ 49+
7 1- 26- 39+ 0- 32+ 0 0- 4- 7- 3+ 4+ 24+ 31-
8 9+ 38+ 2- 1- 24+ 0- 2- 11- 12- 0- 2- 11- 55-
9 3- 2- 18+ 12+ 18- 1+ 5+ 29+ 32+ 1- 5+ 58+ 70-
10 6- 826+ 10- 2+ 646+ 0- 2+ 8+ 25+ 2- 3- 25- 171+
11 31+ 0- 23+ 0- 14+ 1- 1- 3- 10- 3- 6- 54+ 83+
12 3+ 15- 14+ 2- 1- 3+ 4+ 30+ 32+ 4+ 9+ 31- 41-
13 1- 168+ 2+ 5+ 67+ 0 0- 10+ 16- 0 2- 129+ 168+
14 2- 31- 0- 3 25+ 0- 3+ 8 21+ 0- 6- 75+ 79+
15 15- 46- 1- 3+ 7+ 1+ 2+ 8 13+ 5+ 13+ 39+ 54+
…

LOTTO-6 Pairing & Frequency Ranks - Layer 1 PairGrid6.1
Draws Analyzed: 2000
Pair ParpaluckP: 98 ~ Frequency ParpaluckF: 49

----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
P A I R I N G S R A N K S FREQUENCY RANKS - Any FREQUENCY RANKS - Pos
Line Num Num Num Num Num Num FIVE Num Num Num Num Num Num Num Num Num Num Num Num
no. 1 2 3 4 5 6 Filter 1 2 3 4 5 6 1 2 3 4 5 6
----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
1 2 14 21 44 47 2 10 13 35 41 2 11 16 18 44 11 17 27 42 43 12 17 24 36 39 13 22 27 30 42 1000 24 15 2 12 26 5 2 17 12 26 16 7
2 22 33 37 45 47 3 15 16 22 47 20 33 39 40 46 2 22 40 41 46 8 29 32 33 34 15 17 18 26 40 1000 13 15 39 24 44 22 12 7 11 11 13 2
3 7 17 31 40 44 19 23 24 29 30 2 17 23 29 48 16 17 26 41 43 8 10 14 26 32 17 26 33 36 40 1000 10 45 16 41 35 20 11 16 21 34 5 8
4 16 17 23 39 45 27 37 40 45 46 7 13 22 40 44 5 8 17 28 45 22 23 33 36 46 3 8 29 39 46 1000 15 28 22 23 26 34 5 32 23 7 14 3
5 1 3 5 20 42 11 14 34 42 48 2 13 14 19 41 10 19 21 44 48 1 5 8 16 41 1 5 17 30 34 1000 24 6 8 28 20 29 2 4 4 15 23 2
6 3 8 22 30 38 3 17 36 38 39 22 28 35 36 44 4 20 27 29 36 22 36 38 39 44 6 13 18 34 37 1000 37 28 14 2 49 47 3 2 28 12 17 12
7 12 18 19 24 48 10 18 23 24 48 6 11 19 26 27 23 25 30 33 42 17 25 27 33 35 3 16 21 29 48 1000 42 32 39 1 36 38 9 5 15 1 14 4
8 5 10 28 40 45 16 27 33 43 45 16 24 26 27 35 7 15 21 42 47 15 16 20 39 42 1 13 17 26 35 1000 22 43 6 13 45 47 1 12 30 29 15 29
9 3 16 19 34 36 12 21 23 31 42 1 3 21 22 35 11 21 26 31 41 1 8 15 36 39 4 8 21 27 35 1000 41 48 35 1 31 29 20 13 5 1 43 4
10 5 38 41 43 46 26 30 36 37 38 26 27 35 44 46 1 17 25 38 42 22 33 38 42 43 17 29 30 36 44 1000 20 48 35 46 30 27 6 18 21 13 4 1
11 5 8 9 17 29 6 9 17 26 32 3 17 20 21 35 11 12 18 24 36 2 5 11 36 41 8 12 16 22 23 1000 22 6 8 3 2 41 21 7 2 7 21 5
12 22 25 29 30 45 15 35 41 44 45 17 19 28 31 41 7 18 24 30 32 7 26 31 41 45 6 11 12 31 42 1000 32 16 22 6 25 27 4 7 7 2 5 6
13 4 5 13 28 35 9 15 18 41 43 15 17 29 39 48 7 23 26 41 43 3 26 35 41 43 2 4 10 31 36 533 17 38 11 37 23 41 3 4 6 4 5 13
14 7 12 28 42 47 3 8 22 31 39 1 2 8 31 48 2 10 19 28 31 11 28 34 39 42 8 17 25 28 39 1000 26 5 27 13 18 46 10 7 14 3 11 6
15 10 16 18 26 32 1 7 11 18 30 23 24 26 34 39 3 10 24 32 35 12 16 24 33 34 5 13 16 21 27 1000 19 33 40 8 34 11 4 9 8 10 11 1
…

Ion Saliu

2019-07-05 10:57:44 UTC

Post by Ion Saliu
LOTTO-6 Pairing & Frequency Ranks - Layer 1 PairGrid6.1
Draws Analyzed: 2000
Pair ParpaluckP: 98 ~ Frequency ParpaluckF: 49
----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
P A I R I N G S R A N K S FREQUENCY RANKS - Any FREQUENCY RANKS - Pos
Line Num Num Num Num Num Num FIVE Num Num Num Num Num Num Num Num Num Num Num Num
no. 1 2 3 4 5 6 Filter 1 2 3 4 5 6 1 2 3 4 5 6
----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
1 2 14 21 44 47 2 10 13 35 41 2 11 16 18 44 11 17 27 42 43 12 17 24 36 39 13 22 27 30 42 1000 24 15 2 12 26 5 2 17 12 26 16 7
2 22 33 37 45 47 3 15 16 22 47 20 33 39 40 46 2 22 40 41 46 8 29 32 33 34 15 17 18 26 40 1000 13 15 39 24 44 22 12 7 11 11 13 2
3 7 17 31 40 44 19 23 24 29 30 2 17 23 29 48 16 17 26 41 43 8 10 14 26 32 17 26 33 36 40 1000 10 45 16 41 35 20 11 16 21 34 5 8
4 16 17 23 39 45 27 37 40 45 46 7 13 22 40 44 5 8 17 28 45 22 23 33 36 46 3 8 29 39 46 1000 15 28 22 23 26 34 5 32 23 7 14 3
5 1 3 5 20 42 11 14 34 42 48 2 13 14 19 41 10 19 21 44 48 1 5 8 16 41 1 5 17 30 34 1000 24 6 8 28 20 29 2 4 4 15 23 2
6 3 8 22 30 38 3 17 36 38 39 22 28 35 36 44 4 20 27 29 36 22 36 38 39 44 6 13 18 34 37 1000 37 28 14 2 49 47 3 2 28 12 17 12
7 12 18 19 24 48 10 18 23 24 48 6 11 19 26 27 23 25 30 33 42 17 25 27 33 35 3 16 21 29 48 1000 42 32 39 1 36 38 9 5 15 1 14 4
8 5 10 28 40 45 16 27 33 43 45 16 24 26 27 35 7 15 21 42 47 15 16 20 39 42 1 13 17 26 35 1000 22 43 6 13 45 47 1 12 30 29 15 29
9 3 16 19 34 36 12 21 23 31 42 1 3 21 22 35 11 21 26 31 41 1 8 15 36 39 4 8 21 27 35 1000 41 48 35 1 31 29 20 13 5 1 43 4
10 5 38 41 43 46 26 30 36 37 38 26 27 35 44 46 1 17 25 38 42 22 33 38 42 43 17 29 30 36 44 1000 20 48 35 46 30 27 6 18 21 13 4 1
11 5 8 9 17 29 6 9 17 26 32 3 17 20 21 35 11 12 18 24 36 2 5 11 36 41 8 12 16 22 23 1000 22 6 8 3 2 41 21 7 2 7 21 5
12 22 25 29 30 45 15 35 41 44 45 17 19 28 31 41 7 18 24 30 32 7 26 31 41 45 6 11 12 31 42 1000 32 16 22 6 25 27 4 7 7 2 5 6
13 4 5 13 28 35 9 15 18 41 43 15 17 29 39 48 7 23 26 41 43 3 26 35 41 43 2 4 10 31 36 533 17 38 11 37 23 41 3 4 6 4 5 13
14 7 12 28 42 47 3 8 22 31 39 1 2 8 31 48 2 10 19 28 31 11 28 34 39 42 8 17 25 28 39 1000 26 5 27 13 18 46 10 7 14 3 11 6
15 10 16 18 26 32 1 7 11 18 30 23 24 26 34 39 3 10 24 32 35 12 16 24 33 34 5 13 16 21 27 1000 19 33 40 8 34 11 4 9 8 10 11 1
…

Look also at the ‘Frequency’ sections. Load the ‘PairGrid6.1’ report in Notepad (with word-wrap off).

• For ‘Frequency regardless of position’: at least one of the numbers in the drawing has a rank above 25 in the vast majority of drawings. Generate combinations for the top-25 lotto numbers, then ‘LIE eliminate’ them. You get rid of a very large amount of unnecessary combinations.

• For ‘Frequency by position’: at least one of the numbers in the drawing has a rank above 10 in the vast majority of drawings. Generate combinations for the top-10 lotto numbers in-strict-position, then ‘LIE eliminate’ them. You get rid of a very large amount of unnecessary combosnations.

HotPicks

2019-07-05 11:01:12 UTC

Saliu,

Your software has great potential, and I love it, but it needs a lot of work done on it.

Ion Saliu

2019-07-05 11:20:50 UTC

Post by HotPicks
Saliu,
Your software has great potential, and I love it, but it needs a lot of work done on it.

1) First things first: Are you Kulai, or not? I no longer hold any grudge. I might fight now and then with one guy, Norman Wattenberger (royalty-name Shkiottorban), of ‘Qfit’ infamy. I respond selectively when he tries to kick my arse. He is the only skumbullow who succeeded in taking my website off line:
• https://saliu.com/gambling-fights.html
• Norman Wattenberger, Gambling Formula, Law Abuse Narcissism.

2) Secondly, the programs in the 'Power632' are long obsolete. They are superseded by the ‘Bright’ and especially ‘Ultimate Software’ packages.

But they still require diligent effort and patience. But, again, just one hit is more than enuff in life…

HotPicks

2019-07-05 11:25:07 UTC

Post by HotPicks
Saliu,
Your software has great potential, and I love it, but it needs a lot of work done on it.

• https://saliu.com/gambling-fights.html
• Norman Wattenberger, Gambling Formula, Law Abuse Narcissism.
2) Secondly, the programs in the 'Power632' are long obsolete. They are superseded by the ‘Bright’ and especially ‘Ultimate Software’ packages.
But they still require diligent effort and patience. But, again, just one hit is more than enuff in life…

1) NO I'm not Kulai and I do not have any grudge against you.

2) I meant any software, not just power632 or wheel632.

But if you like to know how to make your software more powerful let me know

Ion Saliu

2019-07-05 12:25:42 UTC

Post by HotPicks
But if you like to know how to make your software more powerful let me know

Thanks for the offer. I believe my software is already very powerful. Maybe too powerful, hence extra difficulty/effort for some users.

Meanwhile, "2000 layers" sounds way out of bounds! Mission impossible! One lotto-6 layer requires 3000000 (3 million) combinations. Multiply that by 2000 and you are way out of the 2GB size-limit of files in Windows (including 64bit)!!!

HotPicks

2019-07-05 12:37:36 UTC

Post by HotPicks
But if you like to know how to make your software more powerful let me know

Thanks for the offer. I believe my software is already very powerful. Maybe too powerful, hence extra difficulty/effort for some users.
Meanwhile, "2000 layers" sounds way out of bounds! Mission impossible! One lotto-6 layer requires 3000000 (3 million) combinations. Multiply that by 2000 and you are way out of the 2GB size-limit of files in Windows (including 64bit)!!!

Well I disagree, You see one of the most useless thing in your software, is using random numbers, there is no numbers generator in the world that will not repeat numbers, so to use your software you have to wait hours and hours until you are satisfied that no more combinations will be generated. Would it be more simple and faster to generate all the combinations and than use one at the time?
That is because powerbasic can not do parallel.

2000 layers is not mission impossible, the only thing is that you have to modify the way your software works and with 2000 layers you can use your draws history instead of generating a random one.

Ion Saliu

2019-07-05 18:30:17 UTC

Post by HotPicks

Post by HotPicks
But if you like to know how to make your software more powerful let me know

Thanks for the offer. I believe my software is already very powerful. Maybe too powerful, hence extra difficulty/effort for some users.
Meanwhile, "2000 layers" sounds way out of bounds! Mission impossible! One lotto-6 layer requires 3000000 (3 million) combinations. Multiply that by 2000 and you are way out of the 2GB size-limit of files in Windows (including 64bit)!!!

Well I disagree, You see one of the most useless thing in your software, is using random numbers, there is no numbers generator in the world that will not repeat numbers, so to use your software you have to wait hours and hours until you are satisfied that no more combinations will be generated. Would it be more simple and faster to generate all the combinations and than use one at the time?
That is because powerbasic can not do parallel.
2000 layers is not mission impossible, the only thing is that you have to modify the way your software works and with 2000 layers you can use your draws history instead of generating a random one.

Skatolchah:

Looks like you are playing mockery with me. * Nobody can handle 2000 layers! * Additionally, you playing the “Fake-English” game.

Again and again, you are using software I discontinued long ago. Still, nobody can or may rewrite my software.

And you are wrong once more. My software generates:
• random combinations – unwheeled (Combine6.exe)
• random combinations – wheeled (Wheel6.exe)
• combinations in lexicographic order – all combos in the set (Lexico6.exe).

HotPicks

2019-07-06 07:12:31 UTC

Post by HotPicks
But if you like to know how to make your software more powerful let me know

Thanks for the offer. I believe my software is already very powerful. Maybe too powerful, hence extra difficulty/effort for some users.
Meanwhile, "2000 layers" sounds way out of bounds! Mission impossible! One lotto-6 layer requires 3000000 (3 million) combinations. Multiply that by 2000 and you are way out of the 2GB size-limit of files in Windows (including 64bit)!!!

Well I disagree, You see one of the most useless thing in your software, is using random numbers, there is no numbers generator in the world that will not repeat numbers, so to use your software you have to wait hours and hours until you are satisfied that no more combinations will be generated. Would it be more simple and faster to generate all the combinations and than use one at the time?
That is because powerbasic can not do parallel.
2000 layers is not mission impossible, the only thing is that you have to modify the way your software works and with 2000 layers you can use your draws history instead of generating a random one.

Looks like you are playing mockery with me. * Nobody can handle 2000 layers! * Additionally, you playing the “Fake-English” game.
Again and again, you are using software I discontinued long ago. Still, nobody can or may rewrite my software.
• random combinations – unwheeled (Combine6.exe)
• random combinations – wheeled (Wheel6.exe)
• combinations in lexicographic order – all combos in the set (Lexico6.exe).

HotPicks

2019-07-06 07:58:54 UTC

Post by HotPicks
But if you like to know how to make your software more powerful let me know

Thanks for the offer. I believe my software is already very powerful. Maybe too powerful, hence extra difficulty/effort for some users.
Meanwhile, "2000 layers" sounds way out of bounds! Mission impossible! One lotto-6 layer requires 3000000 (3 million) combinations. Multiply that by 2000 and you are way out of the 2GB size-limit of files in Windows (including 64bit)!!!

Well I disagree, You see one of the most useless thing in your software, is using random numbers, there is no numbers generator in the world that will not repeat numbers, so to use your software you have to wait hours and hours until you are satisfied that no more combinations will be generated. Would it be more simple and faster to generate all the combinations and than use one at the time?
That is because powerbasic can not do parallel.
2000 layers is not mission impossible, the only thing is that you have to modify the way your software works and with 2000 layers you can use your draws history instead of generating a random one.

First of all, yes you can do 2000 layers, here how:
Instead of doing 1-2000, 2001-4000, 4001-5000,5001-6000 you will do (for example,for the 59 numbers lottery):
1-59,60-118,119-177 and so on. You will not have to use random numbers in the D file, but the original history file modified to ..............

What is the Fake-English game? (Well, English is not my first language)

As I said, I did re-write your software.

And yes, I know that your software generates random combinations [well pseudo-random combinations], and that one thing that is wrong with your software, because you will have to wait hours and hours and hours until you are satisfied that no more combinations passed the filters.

Ion Saliu

2019-07-06 08:54:58 UTC

Post by HotPicks
1-59,60-118,119-177 and so on. You will not have to use random numbers in the D file, but the original history file modified to ..............
What is the Fake-English game? (Well, English is not my first language)
As I said, I did re-write your software.
And yes, I know that your software generates random combinations [well pseudo-random combinations], and that one thing that is wrong with your software, because you will have to wait hours and hours and hours until you are satisfied that no more combinations passed the filters.

“(Well, English is not my first language)”

I think it is. You created here a parody of Ion Saliu’s software. My educated guess is that you are Kulai (the real one) or Doollow. You must be an old-timer here since you knew about that royalty-name list (buried under gazillions of posts).

If you create 2000 “layers”, you’ll need 59*2000 = 118000 real drawings. At 100 lottery drawings a year, you’ll have to live 1180 years to get that many real draws without random combinations!!!

You can’t even count to 2000 in a reasonable time. Only crazy people would try that.

Which brings me to the conclusion: If you are not a parodist, you must be CRAZY-CRAZY!

BTW –
Combinations in lexicographical order are NOT random or pseudorandom. The generation starts at index #1 (e.g. 1,2,3,4,5,6) and ends at the last lexicographic rank (e.g. 54,55,56,57,58,59).

HotPicks

2019-07-06 09:28:34 UTC

“(Well, English is not my first language)”
I think it is. You created here a parody of Ion Saliu’s software. My educated guess is that you are Kulai (the real one) or Doollow. You must be an old-timer here since you knew about that royalty-name list (buried under gazillions of posts).
If you create 2000 “layers”, you’ll need 59*2000 = 118000 real drawings. At 100 lottery drawings a year, you’ll have to live 1180 years to get that many real draws without random combinations!!!
You can’t even count to 2000 in a reasonable time. Only crazy people would try that.
Which brings me to the conclusion: If you are not a parodist, you must be CRAZY-CRAZY!
BTW –
Combinations in lexicographical order are NOT random or pseudorandom. The generation starts at index #1 (e.g. 1,2,3,4,5,6) and ends at the last lexicographic rank (e.g. 54,55,56,57,58,59).

First of all I am new to the scene, I am not an old-time or parodist, if you want the proof let's have a skype session.

Well I am disappointed, I thought, judging by your software that you could think outside the box, however, although I did not want to say here how is it done, I will tell you: The trick is to using the history file, but every draws has to be decreased by one, e.g. 7 9 12 15 25 36 -> 6 8 11 14 24 35 -> 5 7 10 13 23 34 ...............8 10 13 16 26 37, so here we go you will have 2000 layers.
Now in the national lottery so far 2455, but when I did the test was over 2000, so 2000 x 59 (for lottery of 59 numbers) = 118000 draws.

As for the combinations, if you generate combinations in lexicographical order and I thought that you generated them in random fashion, is better, sorry about that.

HotPicks

2019-07-06 09:57:54 UTC

“(Well, English is not my first language)”
I think it is. You created here a parody of Ion Saliu’s software. My educated guess is that you are Kulai (the real one) or Doollow. You must be an old-timer here since you knew about that royalty-name list (buried under gazillions of posts).
If you create 2000 “layers”, you’ll need 59*2000 = 118000 real drawings. At 100 lottery drawings a year, you’ll have to live 1180 years to get that many real draws without random combinations!!!
You can’t even count to 2000 in a reasonable time. Only crazy people would try that.
Which brings me to the conclusion: If you are not a parodist, you must be CRAZY-CRAZY!
BTW –
Combinations in lexicographical order are NOT random or pseudorandom. The generation starts at index #1 (e.g. 1,2,3,4,5,6) and ends at the last lexicographic rank (e.g. 54,55,56,57,58,59).

I would love to know why would I hide from you, you are not a contract killer, are you? Do I need to fear you?

HotPicks

2019-07-06 08:24:05 UTC

Post by HotPicks
But if you like to know how to make your software more powerful let me know

Thanks for the offer. I believe my software is already very powerful. Maybe too powerful, hence extra difficulty/effort for some users.
Meanwhile, "2000 layers" sounds way out of bounds! Mission impossible! One lotto-6 layer requires 3000000 (3 million) combinations. Multiply that by 2000 and you are way out of the 2GB size-limit of files in Windows (including 64bit)!!!

Well I disagree, You see one of the most useless thing in your software, is using random numbers, there is no numbers generator in the world that will not repeat numbers, so to use your software you have to wait hours and hours until you are satisfied that no more combinations will be generated. Would it be more simple and faster to generate all the combinations and than use one at the time?
That is because powerbasic can not do parallel.
2000 layers is not mission impossible, the only thing is that you have to modify the way your software works and with 2000 layers you can use your draws history instead of generating a random one.

First of all, yes you can do 2000 layers, here how:
Instead of doing 1-2000, 2001-4000, 4001-5000,5001-6000 you will do (for example 59 numbers lottery):
1-59,60-118,119-177 and so on. You will not have to use random numbers as the D file, but the original history file modified to.....

What is the Fake-English game? (Well, English is not my first language)

As I said I did re-write your software, but just to test my hypothesis.

Yes, I know your software generates random combinations and that is one thing that is wrong, since you will have to wait hours and hours and hours until you are satisfied that no more combinations passed the filters

HotPicks

2019-07-05 12:40:00 UTC

Post by HotPicks
But if you like to know how to make your software more powerful let me know

Thanks for the offer. I believe my software is already very powerful. Maybe too powerful, hence extra difficulty/effort for some users.
Meanwhile, "2000 layers" sounds way out of bounds! Mission impossible! One lotto-6 layer requires 3000000 (3 million) combinations. Multiply that by 2000 and you are way out of the 2GB size-limit of files in Windows (including 64bit)!!!

Why 2000 layers, well the answer is simple, there will be a chance that in 2000 layers many of them will be reoccur again and that will never be with 4 layers.

HotPicks

2019-07-05 12:04:05 UTC

Ok, I found the list:

Kotkoduck (aka Rober Perkis, founder)
Kokostirk (aka Joe Roberts, aka CDEX, sworn brother of Kotkoduck)
Psychosama (aka Shobolan, aka John Griffin)
Krokodick (aka Dick Adams)
Kuppuck (aka Royce Penny)
Skitser (aka Scott Ruby; the man who first reacted to my software by
promising to help me “write bad software, but with nice graphics;
that’s what people really need use; nobody should ever offer good
lottery software!”)
Shkitser (aka Mike Sharkey, the openly acknowledged member of the
academia)
Karaklonchah (aka Paracelsus)
Kotskarr (aka Druncan Smith, aka John Dingley, aka Jameson Somethin’)
Doollow (aka Gel Nigel)
Mockofun (aka Gooskey, aka gARY)
Boozedoogahn (aka Gerry)
Pulicker (aka Paul McCoy)
Koochew (aka Nick UK)
Kokolosh (aka Harry Scott)
Barbalick (aka Icirc, aka Lotto Hat Ricci)
Kulai (aka Nik Barker)
Sticklete (aka Stig Holmquist)
Kloshker (aka F-coli, aka Colin Fairbrother; relatively a newcomer,
full of hatred towards me, but his entire activity consists only of
plagiarism and piracy of some of my stuff).

I always wondered what happened to him (Nik Barker) either is rich or dead.

Ion Saliu

2019-07-05 12:16:03 UTC

Post by HotPicks
Kotkoduck (aka Rober Perkis, founder)
Kokostirk (aka Joe Roberts, aka CDEX, sworn brother of Kotkoduck)
Psychosama (aka Shobolan, aka John Griffin)
Krokodick (aka Dick Adams)
Kuppuck (aka Royce Penny)
Skitser (aka Scott Ruby; the man who first reacted to my software by
promising to help me “write bad software, but with nice graphics;
that’s what people really need use; nobody should ever offer good
lottery software!”)
Shkitser (aka Mike Sharkey, the openly acknowledged member of the
academia)
Karaklonchah (aka Paracelsus)
Kotskarr (aka Druncan Smith, aka John Dingley, aka Jameson Somethin’)
Doollow (aka Gel Nigel)
Mockofun (aka Gooskey, aka gARY)
Boozedoogahn (aka Gerry)
Pulicker (aka Paul McCoy)
Koochew (aka Nick UK)
Kokolosh (aka Harry Scott)
Barbalick (aka Icirc, aka Lotto Hat Ricci)
Kulai (aka Nik Barker)
Sticklete (aka Stig Holmquist)
Kloshker (aka F-coli, aka Colin Fairbrother; relatively a newcomer,
full of hatred towards me, but his entire activity consists only of
plagiarism and piracy of some of my stuff).
I always wondered what happened to him (Nik Barker) either is rich or dead.

Great find, axiomaticule! Honest-honest, are you on the list?

Kulai (aka Nik Barker) must be alive and kicking. He was a top-notch user of my software, plus a keen strategy discoverer. Chances are excellent he won at least one jackpot.

HotPicks

2019-07-05 12:20:58 UTC

Great find, axiomaticule! Honest-honest, are you on the list?
Kulai (aka Nik Barker) must be alive and kicking. He was a top-notch user of my software, plus a keen strategy discoverer. Chances are excellent he won at least one jackpot.

Fortunately, I am not on the list

Ion Saliu

2019-07-05 12:44:03 UTC

Post by HotPicks
Fortunately, I am not on the list

Every human deserves a royalty name and should have one. Your royalty-name is Skatolchah; mine is Parpaluck.

HotPicks

2019-07-05 12:48:17 UTC

Post by HotPicks
Fortunately, I am not on the list

Every human deserves a royalty name and should have one. Your royalty-name is Skatolchah; mine is Parpaluck.

Yes, but I do not want to be on the list of undesirable people.

2019-07-05 12:17:30 UTC

Cant see myself on the list. Obviously not good enough to make the shortlist.

Looks like I am gonna have to form my own band.

Ion Saliu

2019-07-05 18:14:12 UTC

Post by NN
Cant see myself on the list. Obviously not good enough to make the shortlist.
Looks like I am gonna have to form my own band.

No offense, Strumelleahg (aka NN). I am totally an egalitarian: Every human being deserves a royalty-name and should have one. Here is the updated list of royalty names on July 5, 2019.

Kotkoduck (aka Rober Perkis, founder)
Kokostirk (aka Joe Roberts, aka CDEX, sworn brother of Kotkoduck)
Psychosama (aka Shobolan, aka John Griffin)
Krokodick (aka Dick Adams)
Kuppuck (aka Royce Penny)
Skitser (aka Scott Ruby; the man who first reacted to my software by
promising to help me “write bad software, but with nice graphics;
that’s what people really need use; nobody should ever offer good
lottery software!”)
Shkitser (aka Mike Sharkey, the openly acknowledged member of the
academia)
Karaklonchah (aka Paracelsus)
Kotskarr (aka Druncan Smith, aka John Dingley, aka Jameson Somethin’)
Doollow (aka Gel Nigel)
Mockofun (aka Gooskey, aka gARY)
Boozedoogahn (aka Gerry, aka John Griffin)
Pulicker (aka Paul McCoy)
Koochew (aka Nick UK)
Kokolosh (aka Harry Scott)
Barbalick (aka Icirc, aka Lotto Hat Ricci)
Kulai (aka Nik Barker)
Sticklete (aka Stig Holmquist)
Kloshker (aka F-coli, aka Colin Fairbrother; relatively a newcomer,
full of hatred towards me, but his entire activity consists only of
plagiarism and piracy of some of my stuff)
• Parpaluck (aka Ion Saliu)
• Strumelleahg (aka NN)
• Martsopolleahk (aka MJ)
• Skatolchah (aka HotPicks)
• Kobzorock (aka Rich)
• Goolleeye (aka FrankM).
• Bullbooyak (aka Driveling Yak 驅動犛牛 aka Nan)

nigel

2019-07-18 10:47:37 UTC

• Parpacluck (aka Ion Saliu)

A number of promising PARP inhibitors are coming to market (eg
Rucaparib, Zejula, Lynparza, Olaparib etc).

Soon only the cluck will be left :)

Evil Nigel

2019-07-05 12:33:25 UTC

Marmaduke Jinks

2019-07-05 12:56:30 UTC

Post by HotPicks
No, after the filters either did not occur again or did not give the right
result I was puzzled because why they reoccur seven times on 87 and then
either not reappear for a year or did not give the right result, so I
assumed that either statistics is bullshit or life is against me. As I
said it took me a week to check all the combinations of 7/87 and after a
year not getting those filters again a gave up and starting using another
strategy.
I recently switched to Euromillions since lotto is not worth the effort
6/59 for a small jackpot, not even lotto hot picks is worth since you can
get more money with Euromillions hot picks.
I believe the only way is to brute force, my software run on gpu and when
I was still looking into uk lotto 6/59 I was able to check 50 filters x
45057474 combinations = 2252873700 in less than a millisecond.
https://saliu.com/bbs/messages/701.html

I was trying to imagine what kind of filters I could employ and the
idea of using 2000 was fascinating. But it also occurred to me that :

I have been analysing my own work and I realise that the fault is mine.

I like to tweak the variables in the hope that this will lead to better
results. And my software shows me I am doing better. In reality I am not
improving but only seeing what I want to see.

Maybe it is a case that less is more.

Has anyone else here ( this forum) had a similar experience and reached the
same conclusion as me ?

~~~~~~~

I remember when Mao Tse Tung was asked what he thought of the impact of the
French Revolution. He said it was too early to tell.

So, my guess is that it's not possible to say yet.

In the other thread though the occurrence of pairs was mentioned. Do you
think there's any mileage in considering
the low occurrence of some pairs and choosing them? Can't see it myself.

MJ

2019-07-05 17:53:59 UTC

Post by NN

Post by HotPicks
No, after the filters either did not occur again or did not give the right
result I was puzzled because why they reoccur seven times on 87 and then
either not reappear for a year or did not give the right result, so I
assumed that either statistics is bullshit or life is against me. As I
said it took me a week to check all the combinations of 7/87 and after a
year not getting those filters again a gave up and starting using another
strategy.
I recently switched to Euromillions since lotto is not worth the effort
6/59 for a small jackpot, not even lotto hot picks is worth since you can
get more money with Euromillions hot picks.
I believe the only way is to brute force, my software run on gpu and when
I was still looking into uk lotto 6/59 I was able to check 50 filters x
45057474 combinations = 2252873700 in less than a millisecond.
https://saliu.com/bbs/messages/701.html

I was trying to imagine what kind of filters I could employ and the
I have been analysing my own work and I realise that the fault is mine.
I like to tweak the variables in the hope that this will lead to better
results. And my software shows me I am doing better. In reality I am not
improving but only seeing what I want to see.
Maybe it is a case that less is more.
Has anyone else here ( this forum) had a similar experience and reached the
same conclusion as me ?
~~~~~~~
I remember when Mao Tse Tung was asked what he thought of the impact of the
French Revolution. He said it was too early to tell.
So, my guess is that it's not possible to say yet.
In the other thread though the occurrence of pairs was mentioned. Do you
think there's any mileage in considering
the low occurrence of some pairs and choosing them? Can't see it myself.
MJ

I think there is value. See :

http://lotto.merseyworld.com/Analysis/Pairs.html &

http://lotto.merseyworld.com/Analysis/Triples.html

I just haven't found a way to incorporate it into my algorithms yet.
Although I should look more as this would be useful for hotpicks,

Marmaduke Jinks

2019-07-07 16:10:24 UTC

Post by NN

Post by HotPicks
No, after the filters either did not occur again or did not give the right
result I was puzzled because why they reoccur seven times on 87 and then
either not reappear for a year or did not give the right result, so I
assumed that either statistics is bullshit or life is against me. As I
said it took me a week to check all the combinations of 7/87 and after a
year not getting those filters again a gave up and starting using another
strategy.
I recently switched to Euromillions since lotto is not worth the effort
6/59 for a small jackpot, not even lotto hot picks is worth since you can
get more money with Euromillions hot picks.
I believe the only way is to brute force, my software run on gpu and when
I was still looking into uk lotto 6/59 I was able to check 50 filters x
45057474 combinations = 2252873700 in less than a millisecond.
https://saliu.com/bbs/messages/701.html

http://lotto.merseyworld.com/Analysis/Pairs.html &
http://lotto.merseyworld.com/Analysis/Triples.html
I just haven't found a way to incorporate it into my algorithms yet.
Although I should look more as this would be useful for hotpicks,

My thoughts turn to that principle of the Law of Large numbers, that every
number combination, over time,will come up. So I think the pairs that have
not yet been "called" will appear. Now in teh Euromillions a pair isn't
very much cash though.

MJ

Marmaduke Jinks

2019-07-07 23:06:35 UTC

Post by NN

Post by HotPicks
No, after the filters either did not occur again or did not give the right
result I was puzzled because why they reoccur seven times on 87 and then
either not reappear for a year or did not give the right result, so I
assumed that either statistics is bullshit or life is against me. As I
said it took me a week to check all the combinations of 7/87 and after a
year not getting those filters again a gave up and starting using another
strategy.
I recently switched to Euromillions since lotto is not worth the effort
6/59 for a small jackpot, not even lotto hot picks is worth since you can
get more money with Euromillions hot picks.
I believe the only way is to brute force, my software run on gpu and when
I was still looking into uk lotto 6/59 I was able to check 50 filters x
45057474 combinations = 2252873700 in less than a millisecond.
https://saliu.com/bbs/messages/701.html

My thoughts turn to that principle of the Law of Large numbers, that every
number combination, over time,will come up. So I think the pairs that
have not yet been "called" will appear. Now in teh Euromillions a pair
isn't very much cash though.
MJ

Looking at the pairs it seems that every combo has come up, but there are
two languishing at 1 time only - 16, 27 and 33, 43 and 4, 23 having come up
25 times.

16&27 and 33&43 have a lot of catching up to do.

MJ

nigel

2019-07-07 11:03:19 UTC

Post by NN
I was trying to imagine what kind of filters I could employ and the
I have been analysing my own work and I realise that the fault is mine.
I like to tweak the variables in the hope that this will lead to better
results. And my software shows me I am doing better. In reality I am not
improving but only seeing what I want to see.
Maybe it is a case that less is more.
Has anyone else here ( this forum) had a similar experience and reached the
same conclusion as me ?

I'm experimenting with something I'm calling Dimensionally Asymmetric
Filtering. It should have no overall effect - Garbage In, Garbage Out,
etc - therefore no better or worse than a random filter. But actually
it's bad - so bad that if I can reverse the effect, it could be good.

Evil Nigel

Post by NN
~~~~~~~
I remember when Mao Tse Tung was asked what he thought of the impact of the
French Revolution. He said it was too early to tell.
So, my guess is that it's not possible to say yet.
In the other thread though the occurrence of pairs was mentioned. Do you
think there's any mileage in considering
the low occurrence of some pairs and choosing them? Can't see it myself.
MJ

Marmaduke Jinks

2019-07-08 16:06:05 UTC

I'm experimenting with something I'm calling Dimensionally Asymmetric
Filtering. It should have no overall effect - Garbage In, Garbage Out,
etc - therefore no better or worse than a random filter. But actually it's
bad - so bad that if I can reverse the effect, it could be good.
Evil Nigel

ps I realise you are missing the word Treatment after Dimensionally
Asymmetric
Filtering.

;-)

MJ

nigel

2019-07-09 10:05:36 UTC

Post by nigel
I'm experimenting with something I'm calling Dimensionally Asymmetric
Filtering. It should have no overall effect - Garbage In, Garbage Out,
etc - therefore no better or worse than a random filter. But actually it's
bad - so bad that if I can reverse the effect, it could be good.
Evil Nigel

ps I realise you are missing the word Treatment after Dimensionally
Asymmetric
Filtering.
;-)
MJ

Or Theory ...

I think Dimensionally Asymmetric Filtering Theory sounds impressively
snake-oilish.

:-)

Evil Nigel

Marmaduke Jinks

2019-07-09 13:48:28 UTC

Post by nigel
I'm experimenting with something I'm calling Dimensionally Asymmetric
Filtering. It should have no overall effect - Garbage In, Garbage Out,
etc - therefore no better or worse than a random filter. But actually
it's bad - so bad that if I can reverse the effect, it could be good.
Evil Nigel

ps I realise you are missing the word Treatment after Dimensionally
Asymmetric
Filtering.
;-)
MJ

Or Theory ...
I think Dimensionally Asymmetric Filtering Theory sounds impressively
snake-oilish.
:-)
Evil Nigel

I like it!

I'd buy that for a dollar. ;-)

MJ

2019-07-09 21:28:40 UTC