Escher on the cheap.

This was an experiment in creating a winged imp tesselation.

I’ve always liked the prints of M C Escher, for their beauty, and complexity, but have always found them a little daunting as a model to emulate.  Escher always stated that he knew nothing about science or maths, but I think that this was modesty or at best, a half-truth.

Now I really am somebody who knows nothing about maths (I’m not bragging, just being honest) and looking at some of the concepts that apparently appear in his prints are enough to make a numerical numbskull like me flee for the hills.

Escher is all about measurement, precision and detail, and although a conventional artist usually has more than enough of these things in their work, the amount of it in Escher's, pushes it into the extraordinary.

For my Zazzle site, I wanted to design something that gave me the 'feel' of Escher, without any of that fiddly measurement, precision and detail stuff - in fact I wanted Escher 'on the cheap', or maybe that might be better expressed as 'speedy Escher.'

Another imp tesselation experiment, combining three different representations.

I did a lot of drawing on paper, trying to get my shapes to add up into tesselations that worked and found it to be a pretty thankless task.  In the above experiment I think it fails because they aren't full figures as Escher would have done, but partial forms.

A lot of his work revolves around interlocking elements.  These are quite hard to achieve, each element of the design has to interlock with all others without any change.  The whole can have a number of different shapes, but they have to be designed so that they all interlock perfectly with each other in a repeating pattern.  Tricky.

Another reason I think it fails is that the above experimental tesselation can be seen as one element, and I don't remember ever seeing Escher combine many figures onto one element.  His elements are always a single human or animal figure.  I suppose there are no hard and fast rules, but it still seems like cheating.

The above element, combined into a full scale tesselation.

Another approach I tried was a tesselation of a number of rider figures, and I chose as the theme a man riding a dragon.  Here you see a black line drawing of one element.

A single element of the tesselation of my Dragon Rider design.

If you're familier with Escher's work you'll recognise that this is based on one of his prints called 'horsemen'.  His work is not just a tesselation, but also approximates a moebius strip, a mathematical body constituting a strip in the form of a loop with a half turn.  The resulting loop has only one side and one edge.  Clever, and not something I'm going to be attempting any time soon.  Here's the finished tesselation.

The full tessalation of the Dragon Riders.

Apart from the fact that the dragons all look as if they're humping one another, this design suffers from the same problems.  Two figures on the same element, and the dragon design is a little awkward. 

As I said earlier in the post I wanted a feel of Escher's work, but quicker; for me tesselations are too time consuming and I don't think my results are as good as I want, so I began working towards a look -  something that suggested Escher and had something (I hope) of their charm.

First up we have a roundel that originated as a one figure design for a t-shirt, a strong colourful image that could be seen across a room.  I combined it into this roundel, and liked the effect so much, I began to look at other designs I could use in the same way.

An Imp roundel.

I especially designed this next image as a roundel, and found even this was quite tricky to pull off, as each figure had to be positioned carefully to get the swords in the right place.

Another six politicians in the making.

Their feet had to be on the red ring only, I didn't want them overlapping too much into the central part, and of course their tails had to go under and over each others swords.  Not as complex as a tesselation but it had its moments.


Here's my website 

My Flickr page.