Jumper to Jumper

Previously I’ve looked at fractals created by a point moving half-way towards the random chosen vertex of a polygon. But you can also choose an initial vertex, then choose a new vertex by adding a random number to that initial vertex. Then repeat. For example, if the polygon is a square and the initial vertex is v = 1, then choose v + 3 = 4 or v – 1 = 3, and so on.

You can then ban or un-ban the choice of vertex-jump as you can ban or un-ban direct choices of vertex. These two methods of random choice are effectively the same, but one can be simpler to program than the other. That’s why I’ve come across some new fractals by using vertex-jumps. Here they are:

vertices = 4, vertex-jump = (1,2,3,4), ban on same choice twice in a row


vertices = 4, vertex-jump = (1,2,3,4), ban on 2 in row (black-and-white version)


v = 4, vj = (1,2,3,4), ban on choice c + 2 from previous choice c


v = 4, vj = (1,2,3,4), ban c + 2 (animated gif)


vj = (1,2,3,4), ban c + 2 (black-and-white)


vj = (1,2,3,4), ban c + 0 at time t+1 unless c + 0 at time t-1


vj = (1,2,3,4), ban c + 0 at t+1, unless c + 0 at t-1 (black-and-white)


vj = (1,2,3,4,5), ban c + 0


vj = (0,1,2,3,4), ban c + 0


vj = (0,1,2,3,4), ban c + 0 (black-and-white)


vj = (1,2,3,4), ban c + 2 at t+1 unless c + 2 at t-1 (animated gif)


vj = (1,2,3,4), ban c + various at t+1 unless c + various at t-1 (animated gif)


vj = (1,2,3,4,5), ban c + 0 at t+1 unless c + 0 at t-1


vj = (-2,-1,0,1,2), ban c + 0


vj = (-2,-1,0,1,2), ban c + 0 (black-and-white)


vj = (0,1,2,3,4), ban c + va unless c + va


v = 5, vj = (1,2,3,4), ban c + 0


v = 5, vj = (1,2,3,4), ban c + 2


v = 5, vj = (0,1,2,3), ban c + 3


v = 6, vj = (0,1,2,3), ban c + 2


v = 5, vj = va, ban c + va (animated gif)


Appointment with Distality

distal, adj. Anat. Situated away from the centre of the body, or from the point of origin (said of the extremity or distant part of a limb or organ); terminal. Opp. to proximal. [← stem of dist- (in distant adj.) + -al, after dorsal, ventral, etc.] — Oxford English Dictionary

When a point jumps inside a triangle, moving halfway towards a randomly chosen vertex each time, a fractal known as the Sierpiński triangle appears:
chaos_triangle

Point jumping halfway towards random vertex of a triangle


chaos_triangle_bw

Point jumping inside triangle (black-and-white version)


But when a point moves at random in the same way inside a square, no fractal appears. Instead, the interior of the square gradually fills with a haze of pixels:
random_fill

Point jumping halfway towards random vertex of a square


Now trying imposing restrictions on the point jumping inside a square. If it can’t jump towards a vertex twice in a row, this fractal appears:
select_1_0

Ban consecutive jumps towards same vertex


select_1_0_bw

Ban consecutive jumps towards same vertex (black-and-white version)


Suppose the vertices are numbered from 1 to 4 and the point can’t jump towards the vertex one lower than the previously chosen vertex. That is, if it jumps towards vertex 3, it can’t jump next towards vertex 2, but it can jump towards vertices 1, 3, or 4 (if the vertex is 1, it’s banned from moving towards vertex 4, i.e. 1-1 = 0 = 4). Now this fractal appears:
select_1_1

Ban jump towards vertex v-1


select_1_1_bw


This is the fractal when the point can’t jump towards the vertex two places lower than the one it has just jumped towards:
select_1_2

Ban jump towards vertex v-2


select_1_2_bw


But if you can ban, you can also un-ban. Suppose the point jumps towards vertex v at time t and is then banned from jumping towards vertex v-2 at time t+1 unless it had jumped towards vertex v-1 at time t-1. This interesting fractal appears:
select_2_1_1_2

Ban jump v-2 at t+1 unless jump v-1 at t-1


Here are some more fractals using the ban / un-ban technique:
select_2_1_various

Ban / un-ban various


select_2_1_0_1

Ban jump v+0 at t+1 unless jump v+1 at t-1


select_2_1_1_3

Ban jump v+1 at t+1 unless jump v+3 at t-1


select_2_1_2_0

Ban jump v+0 at t+1 unless jump v+2 at t-1


select_2_1_2_2

Ban jump v+2 at t+1 unless jump v+2 at t-1


select_1_2_various

Ban / un-ban various


You can also impose or lift bans based not on the vertex the point jumps towards, but on the distance the point jumps. For example, take the radius r of the circle circumscribing the square and divide it into four segments, 0 to ¼r, ¼r to ½r, ½r to ¾r, and ¾r to r. When the point is going to jump towards vertex v, test whether its jump will land in the same segment, measured from the center of the circle, as it currently occupies. If it does, ban the jump and choose another vertex. Or unban the vertex if the point occupied segment s + x at time t-1. Here are some of the fractals produced using this technique:
dist_2_1_various

Ban / un-ban based on distance jumped


dist_center_1_0

Ban jump into segment s+0 of 4


dist_center_1_1

Ban jump into segment s+1 from center


dist_center_1_2

Ban jump into segment s+2


dist_center_-2_1_2_2

Ban jump into s+2 at t+1 unless jump into s+2 at at t-1


dist_xy_1_0

Ban jump into s+0 from present point


dist_xy_1_2

Ban jump into s+2 from present point


dist_xy_1_3

Ban jump into s+3 from present point


dist_xy_2_1_1_0

Ban jump into s+0 at t+1 unless jump into s+1 at at t-1


It’s easy to think of variants on all these themes, but I’ll leave them as an exercise for the interested reader.

The Swing’s the Thing

Order emerges from chaos with a triangle or pentagon, but not with a square. That is, if you take a triangle or a pentagon, chose a point inside it, then move the point repeatedly halfway towards a vertex chosen at random, a fractal will appear:

triangle

Sierpiński triangle from point jumping halfway to randomly chosen vertex


pentagon

Sierpiński pentagon from point jumping halfway to randomly chosen vertex


But it doesn’t work with a square. Instead, the interior of the square slowly fills with random points:

square

Square filling with point jumping halfway to randomly chosen vertex


As I showed in Polymorphous Perverticity, you can create fractals from squares and randomly moving points if you ban the point from choosing the same vertex twice in a row, and so on. But there are other ways. You can take the point, move it towards a vertex at random, then swing it around the center of the square through some angle before you mark its position, like this:

square_sw90

Point moves at random, then swings by 90° around center


square_sw180

Point moves at random, then swings by 180° around center


You can also adjust the distance of the point from the center of the square using a formula like dist = r * rmdist, where dist is the distance, r is the radius of the circle in which the circle is drawn, and rm takes values like 0.1, 0.25, 0.5, 0.75 and so on:

square_dist_rm0_05

Point moves at random, dist = r * 0.05 – dist


square_dist_rm0_1

Point moves at random, dist = r * 0.1 – dist


square_dist_rm0_2

Point moves at random, dist = r * 0.2 – dist


But you can swing the point while applying a vertex-ban, like banning the previously chosen vertex, or the vertex 90° or 180° away. In fact, swinging the points converts one kind of vertex ban into the others.

square_ban0

Point moves at random towards vertex not chosen previously


square_ban0_sw405

Point moves at random, then swings by 45°


square_ban0_sw360

Point moves at random, then swings by 360°


square_ban0_sw697

Point moves at random, then swings by 697.5°


square_ban0_sw720

Point moves at random, then swings by 720°


square_ban0_sw652

Point moves at random, then swings by 652.5°


square_ban0_swing_va_animated

Animated angle swing


You can also reverse the swing at every second move, swing the point around a vertex instead of the center or around a point on the circle that encloses the square. Here are some of the fractals you get applying these techniques.
square_ban0_sw45_rock

Point moves at random, then swings alternately by 45°, -45°


square_ban0_sw90_rock

Point moves at random, then swings alternately by 90°, -90°


square_ban0_sw135_rock

Point moves at random, then swings alternately by 135°, -135°


square_ban0_sw180_rock

Point moves at random, then swings alternately by 180°, -180°


square_ban0_sw225

Point moves at random, then swings alternately by 225°, -225°


square_ban0_sw315

Point moves at random, then swings alternately by 315°, -315°


square_ban0_sw360_rock

Point moves at random, then swings alternately by 360°, -360°


square_swing_vx0_va_animated

Animated alternate swing


square_circle_sw45

Point moves at random, then swings around point on circle by 45°


square_circle_sw67

Point moves at random, then swings around point on circle by 67.5°


square_circle_sw90

Point moves at random, then swings around point on circle by 90°


square_circle_sw112

Point moves at random, then swings around point on circle by 112.5°


square_circle_sw135

Point moves at random, then swings around point on circle by 135°


square_circle_sw180

Point moves at random, then swings around point on circle by 180°


square_circle_sw_animated

Animated circle swing


Tri Again (Again)

I didn’t expect to find the hourglass fractal playing with squares. I even less expected it playing with triangles. Isosceles right triangles, to be precise. Then again, I found it first playing with the L-triomino, which is composed of three squares. And an isosceles triangle is half of a square. So it all fits. This is an isosceles right triangle:
isosceles_right_triangle

Isosceles right triangle


It’s mirror-symmetrical, so it looks the same in a mirror unless you label one of the acute-angled corners in some way, like this:

right_triangle_chiral_1

Right triangle with labelled corner


right_triangle_chiral_2

Right triangle reflected


Reflection is how you find the hourglass fractal. First, divide a right triangle into four smaller right triangles.

right_triangle_div4

Right triangle rep-tiled


Then discard one of the smaller triangles and repeat. If the acute corners of the smaller triangles have different orientations, one of the permutations creates the hourglass fractal, like this:

right_triangle_div4_1

Hourglass #1


right_triangle_div4_2

Hourglass #2


right_triangle_div4_3

Hourglass #3


right_triangle_div4_4

Hourglass #4


right_triangle_div4_5

Hourglass #5


right_triangle_div4_6

Hourglass #6


right_triangle_div4_7

Hourglass #7


right_triangle_div4_8

Hourglass #8


right_triangle_div4_9

Hourglass #9


right_triangle_div4_123_010

Hourglass animated


Another permutation of corners creates what I’ve decided to call the crane fractal, like this:
right_triangle_div4_123_001

Crane fractal animated


right_triangle_div4_123_001_static

Crane fractal (static)


The crane fractal is something else that I first found playing with the L-triomino:

l-triomino_234

Crane fractal from L-triomino


Previously pre-posted:

Square Routes
Tri Again

Square Routes

One of the pleasures of exploring an ancient city like York or Chester is that of learning new routes to the same destination. There are byways and alleys, short-cuts and diversions. You set off intending to go to one place and end up in another.

Maths is like that, even at its simplest. There are many routes to the same destination. I first found the fractal below by playing with the L-triomino, or the shape created by putting three squares in the shape of an L. You can divide it into four copies of the same shape and discard one copy, then do the same to each of the sub-copies, then repeat. I’ve decided to call it the hourglass fractal:

l-triomino_124

Hourglass fractal (animated)


l-triomino_124_upright_static1

Hourglass fractal (static)


Then I unexpectedly came across the fractal again when playing with what I call a proximity fractal:
v4_ban15_sw3_anim

Hourglass animated (proximity fractal)


v4_ban15_sw3_col

(Static image)


Now I’ve unexpectedly come across it for a third time, playing with a very simple fractal based on a 2×2 square. At first glance, the 2×2 square yields only one interesting fractal. If you divide the square into four smaller squares and discard one square, then do the same to each of the three sub-copies, then repeat, you get a form of the Sierpiński triangle, like this:

sq2x2_123_1

Sierpiński triangle stage 1


sq2x2_123_2

Sierpiński triangle #2


sq2x2_123_3

Sierpiński triangle #3


sq2x2_123_4

Sierpiński triangle #4


sq2x2_123

Sierpiński triangle animated


sq2x2_123_static

(Static image)


The 2×2 square seems too simple for anything more, but there’s a simple way to enrich it: label the corners of the sub-squares so that you can, as it were, individually rotate them 0°, 90°, 180°, or 270°. One set of rotations produces the hourglass fractal, like this:

sq2x2_123_013_1

Hourglass stage 1


sq2x2_123_013_2

Hourglass #2


sq2x2_123_013_3

Fractal #3


sq2x2_123_013_4

Hourglass #4


sq2x2_123_013_5

Hourglass #5


sq2x2_123_013_6

Hourglass #6


sq2x2_123_013

Hourglass animated


sq2x2_123_013_static

(Static image)


Here are some more fractals from the 2×2 square created using this technique (I’ve found some of them previously by other routes):

sq2x2_123_022


sq2x2_123_022_static

(Static image)


sq2x2_123_031


sq2x2_123_031_static

(Static image)


sq2x2_123_102


sq2x2_123_102_static

(Static image)


sq2x2_123_2011


sq2x2_123_201_static

(Static image)


sq2x2_123_211


sq2x2_123_211_static

(Static image)


sq2x2_123_213


sq2x2_123_213_static

(Static image)


sq2x2_123_033_-111


sq2x2_123_033_-111_static

(Static image)


sq2x2_123_201_1-11_static

(Static image)


sq2x2_200_1-11_static

(Static image)


sq2x2_123_132

(Static image)


Tri-Way to L

The name is more complicated than the shape: L-triomino. The shape is simply three squares forming an L. And it’s a rep-tile — it can be divided into four smaller copies of itself.

l-triomino

An L-triomino — three squares forming an L


l-triomino_anim

L-triomino as rep-tile


That means it can also be turned into a fractal, as I’ve shown in Rep-Tiles Revisited and Get Your Prox Off #2. First you divide an L-triomino into four sub-copies, then discard one sub-copy, then repeat. Here are the standard L-triomino fractals produced by this technique:

l-triomino_123_134

Fractal from L-triomino — divide and discard


l-triomino_234


l-triomino_124


l-triomino_124_upright


l-triomino_124_upright_static1

(Static image)


l-triomino_124_upright_static2

(Static image)


But those fractals don’t exhaust the possibilities of this very simple shape. The standard L-triomino doesn’t have true chirality. That is, it doesn’t come in left- and right-handed forms related by mirror-reflection. But if you number its corners for the purposes of sub-division, you can treat it as though it comes in two distinct orientations. And when the orientations are different in the different sub-copies, new fractals appear. You can also delay the stage at which you discard the first sub-copy. For example, you can divide the L-triomino into four sub-copies, then divide each sub-copy into four more sub-copies, and only then begin discarding.

Here are the new fractals that appear when you apply these techniques:

l-triomino_124_exp

Delay before discarding


l-triomino_124_exp_static

(Static image)


l-triomino_124_tst2_static1

(Static image)


l-triomino_124_tst2_static2

(Static image)


l-triomino_124_tst1


l-triomino_124_tst1_static1

(Static image)


l-triomino_124_tst1_static2

(Static image)


l-triomino_134_adj1

Adjust orientation


l-triomino_134_adj2


l-triomino_134_adj3


l-triomino_134_adj3_tst3

(Static image)


l-triomino_134_adj4


l-triomino_134_exp_static

(Static image)


l-triomino_234_exp

For Revver and Fevver

This shape reminds me of the feathers on an exotic bird:

feathers

(click or open in new window for full size)


feathers_anim

(animated version)


The shape is created by reversing the digits of a number, so you could say it involves revvers and fevvers. I discovered it when I was looking at the Halton sequence. It’s a sequence of fractions created according to a simple but interesting rule. The rule works like this: take n in base b, reverse it, and divide reverse(n) by the first power of b that is greater thann.

For example, suppose n = 6 and b = 2. In base 2, 6 = 110 and reverse(110) = 011 = 11 = 3. The first power of 2 that is greater than 6 is 2^3 or 8. Therefore, halton(6) in base 2 equals 3/8. Here is the same procedure applied to n = 1..20:

1: halton(1) = 1/10[2] → 1/2
2: halton(10) = 01/100[2] → 1/4
3: halton(11) = 11/100[2] → 3/4
4: halton(100) = 001/1000[2] → 1/8
5: halton(101) = 101/1000[2] → 5/8
6: halton(110) = 011/1000 → 3/8
7: halton(111) = 111/1000 → 7/8
8: halton(1000) = 0001/10000 → 1/16
9: halton(1001) = 1001/10000 → 9/16
10: halton(1010) = 0101/10000 → 5/16
11: halton(1011) = 1101/10000 → 13/16
12: halton(1100) = 0011/10000 → 3/16
13: halton(1101) = 1011/10000 → 11/16
14: halton(1110) = 0111/10000 → 7/16
15: halton(1111) = 1111/10000 → 15/16
16: halton(10000) = 00001/100000 → 1/32
17: halton(10001) = 10001/100000 → 17/32
18: halton(10010) = 01001/100000 → 9/32
19: halton(10011) = 11001/100000 → 25/32
20: halton(10100) = 00101/100000 → 5/32…

Note that the sequence always produces reduced fractions, i.e. fractions in their lowest possible terms. Once 1/2 has appeared, there is no 2/4, 4/8, 8/16…; once 3/4 has appeared, there is no 6/8, 12/16, 24/32…; and so on. If the fractions are represented as points in the interval [0,1], they look like this:

line1_1_2

point = 1/2


line2_1_4

point = 1/4


line3_3_4

point = 3/4


line4_1_8

point = 1/8


line5_5_8

point = 5/8


line6_3_8

point = 3/8


line7_7_8

point = 7/8


line_b2_anim

(animated line for base = 2, n = 1..63)


It’s apparent that Halton points in base 2 will evenly fill the interval [0,1]. Now compare a Halton sequence in base 3:

1: halton(1) = 1/10[3] → 1/3
2: halton(2) = 2/10[3] → 2/3
3: halton(10) = 01/100[3] → 1/9
4: halton(11) = 11/100[3] → 4/9
5: halton(12) = 21/100[3] → 7/9
6: halton(20) = 02/100 → 2/9
7: halton(21) = 12/100 → 5/9
8: halton(22) = 22/100 → 8/9
9: halton(100) = 001/1000 → 1/27
10: halton(101) = 101/1000 → 10/27
11: halton(102) = 201/1000 → 19/27
12: halton(110) = 011/1000 → 4/27
13: halton(111) = 111/1000 → 13/27
14: halton(112) = 211/1000 → 22/27
15: halton(120) = 021/1000 → 7/27
16: halton(121) = 121/1000 → 16/27
17: halton(122) = 221/1000 → 25/27
18: halton(200) = 002/1000 → 2/27
19: halton(201) = 102/1000 → 11/27
20: halton(202) = 202/1000 → 20/27
21: halton(210) = 012/1000 → 5/27
22: halton(211) = 112/1000 → 14/27
23: halton(212) = 212/1000 → 23/27
24: halton(220) = 022/1000 → 8/27
25: halton(221) = 122/1000 → 17/27
26: halton(222) = 222/1000 → 26/27
27: halton(1000) = 0001/10000 → 1/81
28: halton(1001) = 1001/10000 → 28/81
29: halton(1002) = 2001/10000 → 55/81
30: halton(1010) = 0101/10000 → 10/81

And here is an animated gif representing the Halton sequence in base 3 as points in the interval [0,1]:

line_b3_anim


Halton points in base 3 also evenly fill the interval [0,1]. What happens if you apply the Halton sequence to a two-dimensional square rather a one-dimensional line? Suppose the bottom left-hand corner of the square has the co-ordinates (0,0) and the top right-hand corner has the co-ordinates (1,1). Find points (x,y) inside the square, with x supplied by the Halton sequence in base 2 and y supplied by the Halton sequence in base 3. The square will gradually fill like this:

square1

x = 1/2, y = 1/3


square2

x = 1/4, y = 2/3


square3

x = 3/4, y = 1/9


square4

x = 1/8, y = 4/9


square5

x = 5/8, y = 7/9


square6

x = 3/8, y = 2/9


square7

x = 7/8, y = 5/9


square8

x = 1/16, y = 8/9


square9

x = 9/16, y = 1/27…


square_anim

animated square


Read full page: For Revver and Fevver

Rep-tilian Rites

A pentomino is one of the shapes created by laying five squares edge-to-edge. There are twelve of them (not counting reflections) and this is the P-pentomino:

p_pentomino

But it’s not just a pentomino, it’s also a rep-tile, or a shape that can divided into smaller copies of itself. There are two ways of doing this (I’ve rotated the pentomino 90° to make the images look better):

p_pentomino_a


p_pentomino_b


Once you’ve divided the shape into four copies, you can divide the copies, then the copies of the copies, and the copies of the copies of the copies, and so on for ever:

p_pentomino_a_anim


p_pentomino_a_anim


And if you’ve got a reptile, you can turn it into a fractal. Simply divide the shape, discard one or more copies, and continue:

p_pentomino_a_124_1

Pentomino-based fractal stage 1


p_pentomino_a_124_2

Pentomino-based fractal stage 2


p_pentomino_a_124_3

Pentomino-based fractal stage 3


p_pentomino_a_124_4

Stage 4


p_pentomino_a_124_5

Stage 5


p_pentomino_a_124_6

Stage 6


p_pentomino_a_124_7

Stage 7


p_pentomino_a_124_8

Stage 8


p_pentomino_a_124_9

Stage 9


p_pentomino_a_124_10

Stage 10


Here are some more fractals created using the same divide-and-discard process:

p_pentomino_b_234

p_pentomino_b_234anim

Animated version


p_pentomino_b_134

p_pentomino_b_134anim

Animated version


p_pentomino_b_124

p_pentomino_b_124anim


p_pentomino_b_123

p_pentomino_b_123anim


p_pentomino_a_134anim

p_pentomino_a_134


p_pentomino_a_234anim

p_pentomino_a_234


p_pentomino_a_124

p_pentomino_a_124anim


p_pentomino_a_123

p_pentomino_a_123anim


You can also use variants on a standard rep-tile dissection, like rotating the copies or trying different patterns of dissection at different levels to see what new shapes appear:

p_pentomino_adj_13

p_pentomino_adj_anim13


p_pentomino_adj_6

p_pentomino_adj_anim6


p_pentomino_adj_anim5

p_pentomino_adj_5


p_pentomino_adj_3

p_pentomino_adj_anim3


p_pentomino_adj_2

p_pentomino_adj_anim2


p_pentomino_adj_1

p_pentomino_adj_anim1


p_pentomino_adj_17

p_pentomino_adj_anim17


p_pentomino_adj_15

p_pentomino_adj_anim15


p_pentomino_adj_16

p_pentomino_adj_anim16


p_pentomino_adj_8

p_pentomino_adj_anim8


p_pentomino_adj_10

p_pentomino_adj_anim10


p_pentomino_adj_11

p_pentomino_adj_anim11


p_pentomino_adj_14

p_pentomino_adj_anim14


p_pentomino_adj_anim4


p_pentomino_adj_anim12


p_pentomino_adj_anim9


p_pentomino_adj_anim7

Fragic Carpet

Maths is like a jungle: rich, teeming and full of surprises. A waterfall here, a glade of butterflies there, a bank of orchids yonder. There is always something new to see and a different route to try. But sometimes a different route will take you to the same place. I’ve already found two ways to reach this fractal (see Fingering the Frigit and Performativizing the Polygonic):

carpet2x2

Fractal Carpet


Now I’ve found a third way. You could call it the rep-tile route. Divide a square into four smaller squares:

square2x2

Add an extra square over the centre:

square2x2_1

Then keep dividing the squares in the same way:

carpet2x2_anim_1

Animated carpet (with coloured blocks)


carpet2x2_anim_2

Animated carpet (with empty blocks)


The colours of the fractal appear when the same pixel is covered repeatedly: first it’s red, then green, yellow, blue, purple, and so on. Because the colours and their order are arbitrary, you can use different colour schemes:

carpet2x2_col1

Colour scheme #1


carpet2x2_col2

Colour scheme #2


carpet2x2_col3

Colour scheme #3


Here are more colour-schemes in an animated gif:

carpet2x2_col

Various colour-schemes


Now try dividing the square into nine and sixteen, with an extra square over the centre:

carpet3x3

3×3 square + central square


carpet3x3_anim

3×3 square + central square (animated)


carpet4x4

4×4 square + central square


carpet4x4_anim

4×4 square + central square (animated)


You can also adjust the size of the square added to the 2×2 subdivision:

carpet2x2_1_2

2×2 square + 1/2-sized central square


carpet2x2_3_4

2×2 square + 3/4-sized central square


Elsewhere Other-Posted:

Fingering the Frigit
Performativizing the Polygonic

Polymorphous Pursuit

Suppose four mice are standing on the corners of a large square. Each mouse begins running at the same speed towards the mouse one place away, reckoning clockwise. The mice will meet at the centre of the square and the path taken by each mouse will be what is known as a pursuit curve:

v4_mi1

vertices = 4, mouse-increment = 1


v4_mi1_animated

v = 4, mi = 1 (animated)


As I showed in “Persecution Complex”, it’s easy to find variants on the basic pursuit curve. If mi = 2, i.e. each mouse runs towards the mouse two places away, the mice will run in straight lines direct to the centre of the square:

v4_mi2

v = 4, mi = 2


v4_mi2_animated

v = 4, mi = 2 (animated)


That variant is trivial, but suppose there are eight mice, four starting on the corners of the square and four starting on the midpoints of the sides. Mice starting on the corners will run different pursuit curves to those starting on the midpoints, because the corners are further from the centre than the midpoints are:

v4_si1_mi1

v = 4, si = 1, mi = 1


v4_si1_mi1_extra


If mi = 3, the pursuit curves look like this:

v4_si1_mi3

v = 4, si = 1, mi = 3


v4_si1_mi3_animated

v = 4, si = 1, mi = 3 (animated)


Suppose there are twelve mice, four on each corner and two more on each side. If each mouse runs towards the mouse four places away, then the pursuit curves don’t all meet in the centre of the square. Instead, they meet in groups of three at four points equidistant from the centre, like this:

v4_si2_mi4

v4_si2_mi4_curves

v = 4, si = 2, mi = 4


v4_si2_mi4_animated

v = 4, si = 2, mi = 4 (animated)


v4_si4_mi4_animated

v = 4, si = 4, mi = 4 (animated)


v4_si4_mi4_large

v = 4, si = 4, mi = 4 (zoom)


Now suppose each mouse become sophisticated and runs toward the combined positions of two other mice, one two places away, the other three places away, like this:

v4_si1_mi2_3

v = 4, si = 1, mi = (2, 3)


v4_si1_mi2_3_animated

v = 4, si = 1, mi = (2, 3) (animated)


These polypursuits, as they could be called, can have complicated central regions:

v4_si2_mi1_4

v = 4, si = 2, mi = (1, 4)


v4_si2_mi1_4_animated

v = 4, si = 2, mi = (1, 4) (animated)


v4_si_va_mi_va

v = 4, si = various, mi = various


And what if you have two teams of mice, running towards one or more mice on the other team? For example, suppose two mice, one from each team, start on each corner of a square. Each mouse on team 1 runs towards the mouse on team 2 that is one place away, while each mouse on team 2 runs towards the mouse on team 1 that is two places away. If the pursuits curves of team 1 are represented in white and the pursuit curves of team 2 in green, the curves look like this:

2v4_mi1_mi2

v = 4 * 2, vmi = 1, vmi = 2


2v4_mi1_mi2_green

v = 4 * 2, vmi = 1, vmi = 2


2v4_mi1_mi2_animated

v = 4 * 2, vmi = 1, vmi = 2 (animated)


Now suppose the four mice of team 1 start on the corners while the mice of team 2 start at the centre of the square.

v4_c4_vmi1_cmi2_white

v = 4, centre = 4, vmi = 1, cmi = 2 (white team)


v4_c4_vmi1_cmi2_green

v = 4, centre = 4, vmi = 1, cmi = 2 (green team)


v4_c4_vmi1_cmi2_both

v = 4, centre = 4, vmi = 1, cmi = 2 (both teams)


v4_c4_vmi1_cmi2_animated

v = 4, centre = 4, vmi = 1, cmi = 2 (animated)


Here are more variants on pursuit curves formed by two teams of mice, one starting on the corners, one at the centre:

v4_c4_vmi0_1_cmi0

v = 4, centre = 4, vmi = (0, 1), cmi = 0


v4_c4_vmi0_2_cmi0

v = 4, centre = 4, vmi = (0, 2), cmi = 0


v4_c4_vmi0_3_cmi0

v = 4, centre = 4, vmi = (0, 3), cmi = 0


2v4_mi1_mi2_both