triplets and quads statistics

[hexagonvex]

Hi Stan,

Thanks for the info.

Another thing. Since Pairs and Neighbours have been reorganized in the statistics section maybe it would be wise to rename the filters since it might to lead to some confusion with the naming conventions across the program.

Pairs filter & Triplets filter -> Neighbours filter with an option for 2 succ. numbers or 3 succ. numbers (2 succ. numbers being the default).

Just an idea....what do others think?

And another....in the statistics page....the successive numbers could be integrated into neighbours (maybe with an overview tab)?

Greets.

[stan]

good ideas, i'll think about it.
thanx

[Bobes]

Hi all,

In the previous version it was possible to identify the most frequent pairs only (Statistics - inversed matrix ). However the WN history filter logic is designed in such a way that usually the "either/or" rule applies for the adjacent pairs. Thus it was not so easy to find the most frequent triplet when the remainig set in the package was large after filtering (WN History Min, Max). Now you have possibility to identify the most frequent triplet in the package using the appropriate table of triplets. This feature is very useful in connection with the WN History filter (particularly when the condition "at least 8 to 11 must pass" applies resulting in very large package contents while allowing up to three mistakes in estimation). Naturally you can use the "Best tickets filter", however it can be very time consuming process particulary for large packages (50k of tickets per hour on my computer). Thus this filter (triplets) is very useful for preprocessing of the package contents also. :vogel: