Talk:Pattern Recognition

i'm not sure what all of this means. could you detail what the article is about any? 72.150.62.175 21:30, 17 February 2007 (EST)

This is a work in progress that is not even close to being finished yet. Basically, I'm detailing construction patterns involving 3-celled wells, soon to include 4(this in no way seeks to exclude 5+ celled techniques, but rather complement them). This serves as a visual representation of what a player sees in their head while they are stacking. A fast player is often unaware of these images because they have acquired this skill through experience. It's purpose is to enable a player to see patterns in advance and the end result the moment they look at a playfield position and the upcoming next pieces.

The two figures in each diagram are mirrors of each other.

The only rules so far: --75.26.5.235
 * 1) No I pieces are to be used in any pattern.
 * 2) The patterns only take duplicate next pieces into mind. (Triplets are implausible in modern randomizers)
 * 3) No figures will lead to a 020 end pattern because that will force the usage of an I piece.
 * 4) No soft dropping.

Can I make a suggestion please? Is there any way that this page can be done without needing 12000 images to be displayed? Even if that means using an alternate version of pfrow. And btw, I tidied up the list that you made in your previous comment here. Please remember to sign all your comments with  ~  and remember that registering has its benefits. --Lardarse 11:39, 18 February 2007 (EST)

Doesn't Playing forever adequately describe the 4-celled-well patterns, at least for 7-bag? --Tepples 13:39, 18 February 2007 (EST)

That article only describes a method of playing forever with the 7 given pieces. What Pattern Recognition aims for is optimal construction. Of course, that would mean 9-celled-well pattern recognition is the true goal, but that would be near impossible to document. (and what does that command do Lardarse? I'm a bit of a noob with code.)--DIGITAL

pfrow is the wiki template that generates the diagrams. It displays a 10-wide grid of 10x10 images. What I'm suggesting is that someone makes a narrower version of it, to decrease server load and improve loading times. --Lardarse 02:30, 19 February 2007 (EST)


 * You don't have to be an admin to create and edit templates on this wiki. To see how to make a template, compare pfrow to tnetrow. --Tepples 12:04, 19 February 2007 (EST)

There needs to be a way to properly caption the figures. In the page, the caption describes the picture below; this isn't obvious if you pay attention from the top. A divider of some sort would help128.163.239.233 13:35, 4 July 2007 (EDT)