Page **1** of **1**

### B36/S23, B37/S23, B38/S23

Posted: **September 4th, 2016, 8:55 am**

by **Rich Holmes**

I'm sure there must be something on this subject written up somewhere but I have no idea how to find it if so.

A 32x32 50% soup in B36/S23 (HighLife) typically stabilizes in about 1000 generations.

A similar soup in B38/S23 also typically stabilizes in 1000 generations or so, though not uncommonly it persists a few thousand generations.

A similar soup in B37/S23 (DryLife) almost always explodes.

Question: Why? Why, when B36/S23 and B38/S23 are stable, is B37/S23 explosive?

### Re: B36/S23, B37/S23, B38/S23

Posted: **September 4th, 2016, 9:13 am**

by **Bullet51**

Along the lines:

Why is B3S23 and B35S23 stable, but B35cS23 explosive?

### Re: B36/S23, B37/S23, B38/S23

Posted: **September 4th, 2016, 4:44 pm**

by **Rich Holmes**

In fact B37c/S23 is explosive, while B37e/S23 isn't. I have no idea why.

### Re: B36/S23, B37/S23, B38/S23

Posted: **September 4th, 2016, 5:11 pm**

by **BlinkerSpawn**

Rich Holmes wrote:In fact B37c/S23 is explosive, while B37e/S23 isn't. I have no idea why.

Presumably because preloaves occur more often than pies.

(Right?)

### Re: B36/S23, B37/S23, B38/S23

Posted: **September 5th, 2016, 1:35 am**

by **Rich Holmes**

BlinkerSpawn wrote:Rich Holmes wrote:In fact B37c/S23 is explosive, while B37e/S23 isn't. I have no idea why.

Presumably because preloaves occur more often than pies.

(Right?)

Presumably it's not solely the frequency that matters, but the consequences when they do occur.

If you consider an isolated preloaf and pi in each of the two rules...

Code: Select all

```
x = 31, y = 12, rule = B37c/S23
5$6b2o16b3o$5bobo18bo$5b3o16b3o!
```

Code: Select all

```
x = 31, y = 12, rule = B37e/S23
5$6b2o16b3o$5bobo18bo$5b3o16b3o!
```

the pi takes more time and more space to stabilize in B37c/S23 than in B37e/S23 while the preloaf in B37e/S23 does not get that large nor take that long to settle.

But patterns containing non-isolated preloaves and pis need to be considered too.

### Re: B36/S23, B37/S23, B38/S23

Posted: **September 5th, 2016, 11:09 am**

by **Bullet51**

Rich Holmes wrote:But patterns containing non-isolated preloaves and pis need to be considered too.

It's not solely the frequency that matters, but the circumstances when they do occur.

An isolated B-heptomino in B37e/S23 is much more chaotic than that in B37c/S23, but after adding a still life, it turned out to be the opposite.

B+SL reactions are far less chaotic in B37e/S23, while the same reactions are often catastrophic in B37c/S23.

(Right?)

Rich Holmes wrote:
the pi takes more time and more space to stabilize in B37c/S23 than in B37e/S23 while the preloaf in B37e/S23 does not get that large nor take that long to settle.

Such self-sustaining behavior may be the key to explain why B37c/S23 allows chaotic dynamics to appear in the center of a soup, while B37e/S23 only allow chaotic dynamics to appear on the edge of a soup.