New Wheel Generator version 2.0 released on 10/1/2022!!!

Any latest news will be posted here...
Post Reply
User avatar
lottoarchitect
Site Admin
Posts: 1568
Joined: Tue Jan 15, 2008 5:03 pm
Location: Greece
Contact:

New Wheel Generator version 2.0 released on 10/1/2022!!!

Post by lottoarchitect » Wed Nov 17, 2021 3:51 pm

This is a teaser of what will come in the new WG version. It has been a long awaited for Wheel Generator to be able to support bigger values for k,t,m and I am pleased to announce a working version of Wheel Generator which can tackle sizes for k up to 25 and m up to 20. Not every possible combination of v,k,t,m is possible however, k and m new limits are the maximum permitted but actual construction might not be possible if requirements go beyond what can be handled. Along this new long awaited improvement, there is a new optimization algorithm called "Linear scan", it is similar to what Cyclic does but it performs a quite different approach to wheel building and has assisted several times in making new records, many of which are included in the list below. Also fixed a rarely occurring bug in Hits analysis where total occurrences report incorrect results (typically missing one combination unreported). Removed the alternative bias checkbox as it deemed not very useful for now (may be reintroduced after being adjusted accordingly). Speed has been increased overall a bit more too and there are a couple more improvements here and there, like an improvement to the hits analysis where we can limit the lower division hits evaluation to a more manageable size. What I am astonished about this WG engine is its ability to break records. This is a list of new (yet) unpublished records which till now was not possible to create in WG due to k,m parameter limitation of 12 and these where made using WG during the development of this new version:

33,13,4,4=101 (ex 103)
34,13,4,4=108 (ex 112)
35,13,4,4=127 (ex 135)
36,13,4,4=144 (ex 148)
37,13,4,4=157 (ex 160)

25,13,6,6=233 (ex 251)
26,13,6,6=313 (ex 348)
27,13,6,6=408 (ex 443)
28,13,6,6=576 (ex 606)
29,13,6,6=720 (ex 738)

36,14,4,4=104 (ex 108)
37,14,4,4=110 (ex 118)
38,14,4,4=116 (ex 138)
39,14,4,4=140 (ex 145)
40,14,4,4=153 (ex 155)
41,14,4,4=161 (ex 187)
42,14,4,4=194 (ex 216)

25,15,6,6=97 (ex 99)
27,15,6,6=124 (ex 126)

45,16,4,4=134 (ex 136)
46,16,4,4=151 (ex 157)

43,17,4,4=106 (ex 108)
44,17,4,4=109 (ex 114)
45,17,4,4=112 (ex 124)
46,17,4,4=115 (ex 127)
47,17,4,4=126 (ex 130)
48,17,4,4=134 (ex 138)

47,18,4,4=110 (ex 117)
48,18,4,4=113 (ex 122)
49,18,4,4=121 (ex 127)
50,18,4,4=129 (ex 131)

48,19,4,4=106 (ex 109)
49,19,4,4=108 (ex 114)
50,19,4,4=112 (ex 120)
51,19,4,4=114 (ex 125)
52,19,4,4=119 (ex 129)

51,20,4,4=107 (ex 112)
52,20,4,4=110 (ex 117)
53,20,4,4=113 (ex 119)
54,20,4,4=117 (ex 122)
55,20,4,4=120 (ex 126)
56,20,4,4=123 (ex 127)
57,20,4,4=132 (ex 135)
58,20,4,4=138 (ex 141)
59,20,4,4=145 (ex 153)
60,20,4,4=166 (ex 197)
61,20,4,4=185 (ex 210)
62,20,4,4=207 (ex 219)

there even more records made but I think the list is representative of what WG can do. The ex values are current records to the best of my knowledge. So, stay tuned, this new version will hopefully come out early new year or sooner. This post is made to illustrate what is already developed and currently tested and will become available soon. Also starting from this new version, we change the version naming of WG to a single number. This new version will be Wheel Generator 2.

UPDATE 10/1/2022
New Wheel Generator version 2 is available at the website! Check this post if you encounter 'access denied' or "error 2127' messages.

bluemoon
Regular
Regular
Posts: 40
Joined: Sat May 05, 2018 3:47 pm

Re: New Wheel Generator version will come out in the following months

Post by bluemoon » Sun Dec 05, 2021 4:21 am

To be honest, we are not interested in your WG program but when do you update your GAT program?

User avatar
lottoarchitect
Site Admin
Posts: 1568
Joined: Tue Jan 15, 2008 5:03 pm
Location: Greece
Contact:

Re: New Wheel Generator version will come out in the following months

Post by lottoarchitect » Sun Dec 05, 2021 10:34 pm

What update are you looking for in GAT? As for WG, personal opinions do not reflect everybody's opinion; there are people who are extremely interested in this, so please refrain from commenting like that.

bluemoon
Regular
Regular
Posts: 40
Joined: Sat May 05, 2018 3:47 pm

Re: New Wheel Generator version will come out in the following months

Post by bluemoon » Mon Dec 06, 2021 10:57 pm

Do you think the current GAT work?
If GAT does not provide any LOTTO #s, what good is WG?

User avatar
lottoarchitect
Site Admin
Posts: 1568
Joined: Tue Jan 15, 2008 5:03 pm
Location: Greece
Contact:

Re: New Wheel Generator version will come out in the following months

Post by lottoarchitect » Tue Dec 07, 2021 1:24 pm

I'm not following you. Does the program have an issue and does not display predicted numbers? Elaborate on what you say, what is the problem?

bluemoon
Regular
Regular
Posts: 40
Joined: Sat May 05, 2018 3:47 pm

Re: New Wheel Generator version will come out in the following months

Post by bluemoon » Thu Dec 09, 2021 4:21 am

OK, I'll tell you the problems of GAT 3.0 based on my experiences in detail later.

bluemoon
Regular
Regular
Posts: 40
Joined: Sat May 05, 2018 3:47 pm

Re: New Wheel Generator version will come out in the following months

Post by bluemoon » Sat Dec 25, 2021 12:00 am

Because my boss has kept me very busy recently, I'm late to reply until now. Sorry about this!
I'm a senior and serious GAT player. Based on my years of experience using GAT, I think the main problems of GAT or GAT 3.0 are as follows:

1) It can't generate enough Lotto #s or the best GAT IDs to win (get a real profit):
First of all, I like GAT Engine, whether its design concept or the implementation of the program. Moreover, the five methods summarized by Mr. AT in "TUTORIAL" are also completely correct. But the reality is that no matter what methods or strategies including Run Factor + 100/X or the "Low-->Hill" model at Blue/Red lines. ... or no matter everywhere including the Top-most of a column /the middle columns, ... have been tried, we can't get a real profit still!

In other words, it is almost impossible to get a Jackpot or second prize. Although we can get some small prizes, the gambling money paid for this far exceeds the bonus, that is, the gain is not worth the loss!

Based on my years of experience, I found that the key problem with GAT or GAT 3.0 is not that it doesn't have a predictive function but it is too low, that is, it generates the Lotto #s or the best GAT IDs are far less than expected, so not enough to help players win or get a real profit.
How to make the GAT Engine generate enough optimal GAT IDs should be the next improvement goal of GAT!

2) Try to develop another way: design a program that produces the least Lotto #s, that is, the ablation approach:
I personally think that deleting those loser numbers should be easier than getting the lotto numbers. In addition, I have read and tested all the posts about the ablation approach there. Although their methods have merits, the effects are very limited and unstable because the current GAT program is designed for getting the lotto #s, which is in the positive (+) direction. However, these interesting attempts by players also proved that GAT has the potential for reverse deletion, which is in the negative (--) direction.

Mr. AT, can you design a new version in which the main purpose is to help users safely eliminate loser numbers with as many as possible? Of course, you can also combine the two versions into one. If so, this will be the first software to predict Lotto #s by deleting loser numbers in the world! Why not try it?

Finally, Mr. AT, do you think there is room or possibility for improvement of the current GAT Engine?
If so, we look forward to your masterpiece coming soon. At that time, I will work hard to promote the new GAT on related websites!
If not, please forget what I said above!
Thanks for your reading.

Merry Christmas and Happy New Year,
Bluemoon

User avatar
lottoarchitect
Site Admin
Posts: 1568
Joined: Tue Jan 15, 2008 5:03 pm
Location: Greece
Contact:

Re: New Wheel Generator version will come out in the following months

Post by lottoarchitect » Mon Jan 10, 2022 1:01 pm

As of today, the new version of Wheel Generator 2 is available at the website. Check the history in help file for a complete list of changes that took place. The big addition in this version is the ability to work with higher k,t,m parameters as explained at the initial post among other small changes/bug fixes.

NOTE: After the installation of the setup, in some systems the user may get the error message "access denied" or on closing down the program another message "error 217" or something similar. This may occur due to some interference from antivirus programs which inspect the loaded file in real-time in their sandbox mode. Although the antivirus will not find anything, still this inspection process results in forbidding Wheel Generator from getting proper permissions to use/create its own files. If this happen to your system, go to menu Help->Data folder and delete both files "client.id" and "options.ini" if they are there. Re-activate the application after doing so. This should address the issue. I have experienced that with Avast antivirus.

Enjoy this fine release! There are a lot of new things coming in WG.

User avatar
lottoarchitect
Site Admin
Posts: 1568
Joined: Tue Jan 15, 2008 5:03 pm
Location: Greece
Contact:

Re: New Wheel Generator version 2.0 released on 10/1/2022!!!

Post by lottoarchitect » Tue Jan 11, 2022 10:48 pm

I have noticed over a forum board a discussion about WG and those folks there compare older versions of WG (1.8.4 and 1.8.5) to the newest one. They believe the older versions are more powerful compared to the latest one. What they do is they open up the programs (old vs new) set some v,k,t,m using the default settings at the optimization panel and click ok. After a while, the older versions supposedly produce a better outcome and they assume because of that the newer versions are bad or poorly performing.
Well, the case is all newer versions are better to older ones. These users fail to understand that a change occurred in the newer 1.8.5.1 (which re-introduced the bias multiplier) and this new version 2, something that lacks in the older version. The need for the introduction of the bias multiplier is due to a bug fix that was hidden for years but eventually surfaced and it had to be fixed. The bug which was fixed in 1.8.5.1 and version 2 is the following: if you adjust the priority slider of the main coverage engine, this should not affect the bias applied by the engine. Unfortunately, this internally was connected (THIS IS A BUG). Just try to alter the priority slider and see what you'll get in those older versions. What happens here and it is a pity for what these folks assume about the new versions is, the priority at 50% happens to be a sweet spot to the coverings they attempt for the targeted coverage. But this same setting is quite bad for coverings which are more to being closed down to 100%. On the other hand, fixing this bug resulted in the need to re-introduce the bias multiplier. The coverings those users attempt actually need a bias multiplier of x10 or x100. Yes, this is it. You'll get better coverings quicker. Check the manual, it discuss the necessity of bias multiplier. In reality WG 2 is the strongest ever made WG.

Post Reply

Who is online

Users browsing this forum: No registered users and 1 guest