Open Cover database and WG (Wheel generator)

Talk about anything Lotto Architect related which doesn't fit in the other forums
Post Reply
draughtsman
Site Admin
Posts: 112
Joined: Thu Jan 17, 2008 4:22 am
Location: Melbourne, Australia

Open Cover database and WG (Wheel generator)

Post by draughtsman » Sun Jan 27, 2008 7:38 am

TRANSFERED FROM OLD FORUM

Hi everyone! As you already know, there is the WG1.4 program that develops optimal wheels. I've put this tool to work to construct or improve existing wheels.
As I have already announced, there will be an open-cover database available at v2.3 release. However, v2.2 users can still access an open-cover database. I'll build a small open-cover database for the most common wheel sizes. Everyone who wishes to get this database, feel free to send an e-mail at the contact address listed on the website and I'll provide the download link. I'll make this database available by the end of the year.
Also, I'll mostly include Pick 6 open-cover wheels for now. I'll add later other categories too. The full open-cover database will be available on v2.3 release.

How to use the open-cover database in v2.2

The database will be accessed as a Custom/user designed database. So, if you have created your own wheels in the program, you have to make a backup of the 'wheel-user.dat' file located in the installation folder of the program. Then, extract the zip file in the installation directory.

The zip file will contain some files: the database (wheel-user.dat) and some txt files with descriptions of the wheels included (coverage offered). As v2.2 cannot display such information (v2.3 will), those files serve the purpose of wheels identification. More information of the open-cover wheels to be included, available at the wheels page of the website.

Hi LA,

This is maybe the most famous wheel of Robert Serotic, lotto 6, 20 numbers in 39 tickets, Nick Koutras get impressed with the wheel performance, Could WG1.4 improve this wheel?, Could you take the experiment LA?

1 2 3 4 9 12
1 2 5 6 10 13
1 2 7 8 11 14
1 3 5 7 9 12
1 3 5 7 10 13
1 3 5 7 11 14
1 3 6 8 9 12
1 4 5 8 10 13
1 4 6 7 11 14
1 5 15 16 17 18
1 5 15 17 19 20
1 5 16 18 19 20
2 3 5 8 11 14
2 3 6 7 10 13
2 4 5 7 9 12
2 4 6 8 9 12
2 4 6 8 10 13
2 4 6 8 11 14
2 6 15 16 19 20
2 6 15 17 18 19
2 6 16 17 18 20
3 4 5 6 11 14
3 4 7 8 10 13
3 7 15 16 17 19
3 7 15 16 18 20
3 7 17 18 19 20
4 8 15 16 17 18
4 8 15 16 19 20
4 8 17 18 19 20
5 6 7 8 9 12
9 10 11 12 15 16
9 10 11 13 17 18
9 10 12 14 17 18
9 10 13 14 19 20
9 11 12 13 19 20
9 11 13 14 15 16
10 11 12 14 19 20
10 12 13 14 15 16
11 12 13 14 17 18

Covermaster detailed report 3 hits

Code: Select all

6 5 4 3 % Total Acc % 
--------------------------------------------------------------------------------------
- - - - 600 52.63158 52.63158 
- - - 1 372 32.63158 85.26316 
- - - 2 96 8.42105 93.68421 
- - - 3 72 6.31579 100.00000 
Covermaster detailed report 4 hits

Code: Select all

6 5 4 3 % Total Acc % 
--------------------------------------------------------------------------------------
- - - 1 1932 39.87616 39.87616 
- - - 2 840 17.33746 57.21362 
- - - 3 580 11.97110 69.18473 
- - - 4 648 13.37461 82.55934 
- - - 5 240 4.95356 87.51290 
- - - 6 66 1.36223 88.87513 
- - - 8 6 0.12384 88.99897 
- - 1 0-6 495 10.21672 99.21569 
- - 2 0-4 24 0.49536 99.71104 
- - 3 0 14 0.28896 100.00000 
Covermaster detailed report 5 hits

Code: Select all

6 5 4 3 % Total Acc % 
--------------------------------------------------------------------------------------
- - - 3 2664 17.18266 17.18266 
- - - 4 2604 16.79567 33.97833 
- - - 5 1078 6.95304 40.93137 
- - - 6 1272 8.20433 49.13571 
- - - 7 822 5.30186 54.43756 
- - - 8 536 3.45717 57.89474 
- - - 9 32 0.20640 58.10114 
- - - 10 84 0.54180 58.64293 
- - - 12 24 0.15480 58.79773 
- - 1 0-9 4788 30.88235 89.68008 
- - 2 1-8 978 6.30805 95.98813 
- - 3 1-6 370 2.38648 98.37461 
- - 4 3-8 18 0.11610 98.49071 
- 1 0-2 1-8 234 1.50929 100.00000 
Covermaster detailed report 6 hits

Code: Select all

6 5 4 3 % Total Acc % 
--------------------------------------------------------------------------------------
- - - 6 1008 2.60062 2.60062 
- - - 7 2412 6.22291 8.82353 
- - - 8 2724 7.02786 15.85139 
- - - 9 636 1.64087 17.49226 
- - - 10 984 2.53870 20.03096 
- - - 11 504 1.30031 21.33127 
- - - 12 276 0.71207 22.04334 
- - 1 4-14 13104 33.80805 55.85139 
- - 2 3-14 9168 23.65325 79.50464 
- - 3 2-12 2916 7.52322 87.02786 
- - 4 2-12 1727 4.45562 91.48349 
- - 5 0-8 171 0.44118 91.92466 
- - 6 2-4 52 0.13416 92.05882 
- - 7 0 25 0.06450 92.12332 
- - 9 0 1 0.00258 92.12590 
- - 12 0 1 0.00258 92.12848 
- 1 0-5 2-10 2748 7.08978 99.21827 
- 2 0-3 2-8 264 0.68111 99.89938 
1 0 2-6 0-6 39 0.10062 100.00000 

Well, this is a C(6,20,3,4)=39 wheel. The world record needs 35 tickets. Obviously this wheel has a different plan in its design. I may try to put WG to generate such a wheel with 39 tickets and see what I'll get. However, this is not a competition of wheels or the best designer tool. WG is designed to minimize cost (fewer tickets) whilst improving coverage and multiple hits in mind, which is not the case here. I assume it will find a very good wheel even if the reduction tickets scheme is disabled. If I find some time, I may try this wheel.

Hi LA,

Of course there is no doubt for me that WG1.4 is the best tool for minimize cost and improving coverage. I have to say that many of the Robert Serotic wheels have more tickets than world record wheels, could the high overlap justify the addition of tickets?, Could WG1.4 with the same number of tickets get better results than the high overlap of Robert Serotic wheels?

The result of the experiment will be very interesting. WG1.4 I know you could beat this wheel, don't worry for the result, the best wheels I have are the wheels improved by WG1.4. But sometimes good competition is needed to test system performance against similar wheels.

Regards

Well, maybe yes maybe no. There is a major constraint in WG which is to try to minimize the required tickets and then trying to find the best overlap of all minimized sets detected. This constraint does not exist of the wheels of Robert Serotic (they can be of any ticket size, greater than the minimum possible). They directly go for the best overlap on doubles or triples, given a ticket size. This means WG has not to take into account the minimization parameter but as this is the search engine's primary function, therefore the system is not free of this parameter. In short, WG is not designed to produce bigger wheels than their minimum possible size.
Wheels of Robert Serotic and Nick Koutras are based on pairs and triples overlap. This is a totally different approach to wheel designs and currently not supported in WG. WG's main concern is to minimize first and then improve if possible. However, WG can scan a wheel of a given size (e.g. 39 tickets here) but the search engine will try to reduce the wheel size (although the reduction parameter will be disabled). This means there are good chances not to go through the best 39 ticket wheels (due to the reduction that tries to achieve) but of those detected, it will find many better ones. But the design at this given ticket size will not allow for great improvements as WG detects it can reduce the tickets size and will try to do this first. The conclusion is that Robert Serotic wheels (and Nick Koutras) cannot be compared with WG's wheels. WG is best when we need the lowest possible tickets. Other designs are best when we do not care of the minimum possible tickets (and definitely more than the lowest possible). I have some plans to integrate in WG a design approach similar to Serotic & Nick Koutras wheels, besides the current functionality.

Post Reply

Who is online

Users browsing this forum: No registered users and 9 guests