Development updates

Any latest news will be posted here...
User avatar
lottoarchitect
Site Admin
Posts: 1460
Joined: Tue Jan 15, 2008 5:03 pm
Location: Greece
Contact:

Re: Development updates

Post by lottoarchitect » Tue May 29, 2018 7:58 am

At this point such discussions are pointless, is it 0.325% better or -1.2% worse or 4.2% better here and 3% worse there or whatever? Can we please stop this never ending discussion? If I had something precise to show to you I would, I don't have any reason not to. I believe you have understood by now it is not easy to compute a rating like 2/5 or 3/5 as you think it is. So, please don't bring up this question again, I have no more data to add, even the data above are practically pointless at this point and discussiong this right now wastes time, both mine and yours.

bluemoon
Regular
Regular
Posts: 25
Joined: Sat May 05, 2018 3:47 pm

Re: Development updates

Post by bluemoon » Tue May 29, 2018 1:30 pm

So, overall the new changes improved slightly the overall outcome. However the core GAT ID picking logic hasn't changed for the expert mode. Now with the introduction of the automatic scan, I expect most users will use that mode instead which will try to approach the game in a most favorable way on its own and just present the result (final prediction) to the user it could come up with.
I’m still vague with above sentences. It seems there are two modes: the Expert and Automated in GAT Engine. What’s the Expert mode and Automated mode? There maybe is the Expert mode only in 2.3b but both modes of the Expert & Automated in 3.0. If so, users would select a mode from the both to scan in GAT 3.0, e.g. if selected the Automated mode, users would get some better performing GATs, otherwise would get the same results with 2.3b. Is it correct?

User avatar
lottoarchitect
Site Admin
Posts: 1460
Joined: Tue Jan 15, 2008 5:03 pm
Location: Greece
Contact:

Re: Development updates

Post by lottoarchitect » Tue May 29, 2018 4:11 pm

The automatic mode does everything by itself. During the process it proposes the best numbers it could come up with directly, there are no GAT tables or any panorama to inspect. This mode underneath utilizes various techniques and decide among them automatically to pick the one approach that thinks will yield the better outcome overall. Still automatic also rely on Run Factor utilization but the user has nothing to do really, just wait for the process to perform a scan - still open ended - and the user just picks the numbers that show up. The picked numbers show up as soon as it can make an initial prediction based on the estimations it has gathered up to that point but as the scan progresses, these picked numbers may and probably change to reach an equilibrium state which presumably will be the best it can come up with. Basically, this automatic mode consists of two different scans: the first scan which takes the most time, is to establish the preferred estimations and approaches to use - this scan also makes a proposal of numbers to use during the scan. For the following draws, the system has already the information needed to perform the scan from that initial run (this is where RF comes to play), thus these scans are much quicker. The outcome of automatic mode is directly the numbers to play. After 10 plays or so, we repeat the automatic process to establish new estimations and approaches, all this works behind the scenes.
The expert mode, called manual scan mode in GAT, is the same to 2.3b - if you think of 2.3 as the expert GAT version - the core engine has been improved slightly compared to 2.3b but how much better can't get quantified as I have explained but overall it produces some more good hits in the panorama but not in all situations. The core logic now has incorporated the information of total hits/delay/regularity in the core engine so the need to inspect the graphs for regularity is reduced, although still worthy for the case bizarre graphs may make it to the panorama (which is now greatly reduced). Checking the graphs is an optional step really, even at the current 2.3 engine. The original concept of Run Factor assumes you pick any of the proposed GATs and keep using it and the promise is, within the 100/X span the desired hit is expected to show up. In 3.0 engine, this is more true than ever before. In 2.3 this is true but since the engine didn't take into account concepts like regularity, this is the part needed to be checked by the user (thus the real expert mode). Another enhance is the automatic stat.data evaluation, applicable both to automatic and expert mode, this feature alone also is expected to give some slight boost to hits production as it tries to establish the most favorable setting for stat.data. It examines the settings that yield the most GATs that produce the most hits and picks the one that delivers the best overall performance. In reality the improvement is slight but it is there. Still the default 20 stat.data is a good setting for all situations but if the automatic stat.data evaluation picks another value, I would prefer that instead for the slight boost it will provide over the default setting. In my test runs it has even picked values like 120 or 175 and when compared to the standard 20 setting, indeed I see some better hits production at the various panorama displays (tested on Australia's Monday Lotto mostly) which ultimately means more chances to pick even the one GAT that will produce the good hit at the next draw (this regardless the RF utilization which is the primary usage).

I would post some pics of how this new version forms but the various panels are not finalized yet, especially the options part. I'll do that when automatic is fully operational along with a detailed review of what has changed.

bluemoon
Regular
Regular
Posts: 25
Joined: Sat May 05, 2018 3:47 pm

Re: Development updates

Post by bluemoon » Wed May 30, 2018 3:53 pm

Sounds good!
Two questions following:

1)How many best numbers would be generated at least or in default by the Auto mode? e.g. for 5/45 and 6/49 or your favorite games?

2)You used the word: “BEST” to describe the automated numbers. Could you tell me how about the “BEST” ? e.g. for 5/45 and 6/49 or your favorite games?

Sincerely,

User avatar
lottoarchitect
Site Admin
Posts: 1460
Joined: Tue Jan 15, 2008 5:03 pm
Location: Greece
Contact:

Re: Development updates

Post by lottoarchitect » Wed May 30, 2018 7:05 pm

For question 1, you can ask as many numbers as you want. What changes is the system does everything under the hood. For question 2, the word "best" refers to the process that tries to identify the best overall picking method to use, similar to the various approaches currently a user does to decide among the displayed GATs. The "overall" part means it targets to pick the method that does deliver the most hits production over a range of draws, not any particuar draw, which means better stability. If you refer to hit percentages, please refer to my answer on the previous posts.

bluemoon
Regular
Regular
Posts: 25
Joined: Sat May 05, 2018 3:47 pm

Re: Development updates

Post by bluemoon » Wed May 30, 2018 10:54 pm

For question 1, you can ask as many numbers as you want.
The more numbers to use the much money for bets. My question is what would be the minimum number generated by this mode? For example, for a 5/45 game, the mode would generate 12 best numbers at least and for a 6/49 game, would be 15 #s at least.

OR, no limit in the setting. e.g. users can select 5-45 for a 5/45 and 6-49 for a 6/49. If so, does the mode give out some recommendation values for the setting of best numbers?
What changes is the system does everything under the hood.
What’s the hood? I don’t understand what’s the relationship between my Question 1 and your answer above.
For question 2, the word "best" refers to the process that tries to identify the best overall picking method to use, ……The "overall" part means it targets to pick the method that does deliver the most hits production over a range of draws which means better stability.
I understood your explanations above: The auto mode would generate out the numbers which are based on the best overall picking method that delivers the most hits over a range of draw with the better stability from the displayed GATs. So, those numbers would be the best e.g. 5/5, 6/6 or the worse e.g. 0/5, 0/6,....etc. Is it right?

Could you tell me what’s the current results according to your back tests for the living lotto games by using the mode?

Sincerely,

User avatar
lottoarchitect
Site Admin
Posts: 1460
Joined: Tue Jan 15, 2008 5:03 pm
Location: Greece
Contact:

Re: Development updates

Post by lottoarchitect » Thu May 31, 2018 4:20 am

The more numbers to use the much money for bets. My question is what would be the minimum number generated by this mode? For example, for a 5/45 game, the mode would generate 12 best numbers at least and for a 6/49 game, would be 15 #s at least.

OR, no limit in the setting. e.g. users can select 5-45 for a 5/45 and 6-49 for a 6/49. If so, does the mode give out some recommendation values for the setting of best numbers?
If you mean which amount of proposed numbers is the best to use, no it can't provide that.
What’s the hood? I don’t understand what’s the relationship between my Question 1 and your answer above.
"Under the hood" is an expression, meaning everything happens without user intervention.
So, those numbers would be the best e.g. 5/5, 6/6 or the worse e.g. 0/5, 0/6,....etc. Is it right?
I'm not sure I understand this. The best set of numbers it can find, nothing more or less.
Could you tell me what’s the current results according to your back tests for the living lotto games by using the mode?
We are not yet at this stage. Automatic is still under development so any such questions are very premature.

bluemoon
Regular
Regular
Posts: 25
Joined: Sat May 05, 2018 3:47 pm

Re: Development updates

Post by bluemoon » Fri Jun 01, 2018 1:01 am

I'm not sure I understand this. The best set of numbers it can find, nothing more or less.
Mr. Lottoarchitect:

I’m afraid that the Auto mode would be a Quick Picker similarly.
Because the system does everything under the hood and no one knows which result would be gotten from the Auto mode. Not sure when and where, users would perhaps get a best number e.g. 5/5 ,6/6 or a worst number as 0/5, 0/6, or a better number 3/5,4/5, 4/6, 5/6, depending on their luck by using the Auto mode, wouldn’t they?

Sincerely,

User avatar
lottoarchitect
Site Admin
Posts: 1460
Joined: Tue Jan 15, 2008 5:03 pm
Location: Greece
Contact:

Re: Development updates

Post by lottoarchitect » Fri Jun 01, 2018 1:46 am

The whole point of automatic is to not burden the user with decision making at any point - which is the exact opposite to manual mode. The manual mode will be always available however if automatic doesn't suit your needs. Most people however prefer simple to use operations, not matter how complex they may seem to be. This is the primary reason I develop automatic and it depends on the original GAT table data, so it is definitely not a quick picker.

bluemoon
Regular
Regular
Posts: 25
Joined: Sat May 05, 2018 3:47 pm

Re: Development updates

Post by bluemoon » Fri Jun 01, 2018 3:52 am

Mr. Lottoarchitect:
I appreciate your reply with confidence greatly!
Do hope the smart Auto mode would have a brilliant performance in the future.
…..tries to identify the best overall picking method to use, similar to the various approaches currently a user does to decide among the displayed GATs. The "overall" part means it targets to pick the method that does deliver the most hits production over a range of draws, not any particular draw, which means better stability.
If my understand to above is correct, the Auto mode can select out some better GATs based on the multiple factors e.g. Hits Produced, Hits and Hits Delay,…etc. through “observing & analyzing” the all GATs on the Panorama first, then it can decide out a best GAT from the better GATs, finally it can generate out the lotto numbers as good as possible directly from the best GAT. Is it right?

Sincerely,

Post Reply

Who is online

Users browsing this forum: No registered users and 1 guest