It’s working! … oh… wait…



The system has turned into a joy to work with.

One of my populations, precodedshort.pop, isn’t acting how it used to. I have a vague memory of doing some experiment, and accidentally saving a population in that file, overwriting what was there. Yeah, I even recall how I could get it back, and thinking how it didn’t matter anyway, I’m done with it.

Right, the other possibility is that something got messed up somewhere in my system, and it isn’t running right, or the same, somehow. However, every other population is acting how it used to. Before I remembered, just now, that I’d changed precodedshort.pop, I wanted to do another test of the system, to be certain all was well.

If I run the system with the usual seed for the random number generator, I’ll get a sequence of numbers in the report that I recognize. One of the things the report tells me, is what the largest name field value has been so far. The name field gets set to zero, whenever the entire population dies. The name field tells me what the longest line of replicators has been. That’s what I call it in the output: “longest.”

Over about a minute, several short-lived populations increase that value, so I’ll see, 77and a short while later, 884, then 923… then the stable population arrives, and the name field will just keep increasing. When I did that test, I saw the usual numbers.

I was thinking how that proves nothing has changed, because if the system had changed somehow, I’d get different numbers. For the hell of it, I simulated a system change by using the same seed but calling the random number generator in a place where it didn’t need to be called. Sure enough, I got a different set of numbers.

It was soooo easy. I didn’t have to dig through half a dozen files, and I didn’t have to keep track of the changes I made for the test to be able to take them out later. When I was done with the silly little test, I could just delete it without worrying that I could break something with something left over.

I spent months building these tools. I’m glad they’re working.

You know, I think I want the old precodedshort.pop back. I know how to generate that exact population again. I just use a different birth object and use the usual seed, wait until they’re stable, (roughly two seconds,) and save them.

What the hell, it will only take a moment.

The mystery is solved. It’s hours later. Not because I took that long to figure this out; it’s because I needed some sleep and to deal with the rest of the day.

Whatever I did to overwrite precodedshort.pop doesn’t matter. The trouble was that when that population was generated, the maximum size of a given figure was set to 75. Once I set the maximum size to 75 again, the population in precodedshort was stable.

When I generated precodedshort.pop, it was before the tools were done. At the time, any experiments I wanted to do required adding code to several different files. I’d simply forgotten about that modification. With the tools I have now, figuring out what was wrong and fixing it was simple. As I said, I’m glad they’re working.

Enough of this, I’ve else to do and get done.


Leave a Reply