Tuesday, August 20, 2013

Deja vu

I feel that I have already posted this one...  And just right now I figured out why.  The previous pattern was called jungle from 2006  and was made with a simple zig-zag augmentation to my original pent program.

pent { sides 5 ifon none { paint blue 5 { create pent right} } }

I guess it was a while back and my older stuff isn't saved to picassa so it's harder for me to find.

This ones nicer anyway.

Monday, August 19, 2013

Seeing Double

This is a simple pattern yet it has some appeal. Although it seems like disconnected, the local patterns are connected.  The touching pentagons have the same orientation which means that a regular pentagon would fit between them.

Somehow the colouring I choose has an interesting effect of seeing double.  I originally choose the colours to visually explain that the touching pentagons are related by an invisible pentagon (not by an arbitrary angle).  Then I noticed this doubled effect was kind of cool.  Perhaps the two sizes of triangular gaps add a bit the distortion effect.  Whatever the biggest contributor to making this pattern interesting, I'm happy with the end product.

Sunday, August 18, 2013

Mesmerising Maze

This is a bit of a maze of local patterns. Probably the easiest way to break it down is to focus on the 3-way propellers because they are unique.  The Y gaps seem especially difficult to figure out since there are so many slightly different versions in the overall pattern.

I'm a big fan of the pattern.  It took me a while to incorporate the Y gaps with the triangular gaps. Sometimes it's a matter of walking away and coming back with fresh eyes.  I'll probably revisit these local patterns again. Maybe tomorrow I'll look for a perspective that will make the pattern of construction a bit more transparent. I'll start by looking for a way to augment a triangle grid....

Thursday, August 15, 2013


I'm stuck on pentagons and triangles.  Here, I'm looking at the gaps that don't look like Ys.  I call them herons.  They have 6 different orientations in this arrangement.  It always takes more time to write the code than I expect.
I may look into a faster to generate the code. Currently, I play with Tileland for ideas then as soon as I understand the repeating pattern, I transcribe it into PolygonR&D.  It might not be too bad to grab a selection of polygons in Tileland and then create the shortest polygon path which would give parts of the code that could be then augmented with programming structures to make code with loops and branches. It would save me lots of time with transcription....but it's debatable whether it's worth the effort since currently I'm the only PolygonR&D coder I know :)

Probably right now, I should just worry about porting Tileland to Android...but first another distracting pattern :)

Popular Posts