Q&A for ptbsearch

[chris_c]

Ahh!!!! I think I made a mistake! A blank line is ignored (without incrementing y pos)by ptbsearch, right?.

I've never looked at that part of the code in detail, but yes, it does look like blank lines are completely ignored and do not translate into an empty row of cells in the input pattern. Instead I guess you need to put a single "." or something like that.

[Scorbie]

Glad to find out the problem. After the thing is stable I'll post my ptbsearch-symm config and settings.

[Saka]

This IS the ptbsearch Q&A, so...
How do I compile and use ptbsearch? Do I compile each .c? What are the .h files for? How do I use it? Thanks!

[Scorbie]

If you're on windows running make.bat (Or simply typing make on the command line of the ptbsearch directory) would suffice. (Makefiles don't work, dunno why and it was a long time ago.)
If you're on Linux/ OS X, just type make in the console.

Google something like "How to compile multiple files" "Makefile tutorial" etc. if you want to know what exactly you are doing.

And one tip: the rock branch is quite nice so you might want to try that.

EDIT: And how to use ptbsearch: viewtopic.php?p=23900#p23900

[Scorbie]

Okay, narrowed down a mysterious bug in my script. @chris when does the survive take snapshots in your symmetry hack? The snapshots are same for different patterns, so when I run uniq, some useful patterns are filtered out.

[chris_c]

Okay, narrowed down a mysterious bug in my script. @chris when does the survive take snapshots in your symmetry hack? The snapshots are same for different patterns, so when I run uniq, some useful patterns are filtered out.

The version of survive in the symmetry branch outputs the pattern if and only if it lasts for at least 400 generations without being permanently damaged or becoming P2.

EDIT: And to actually answer the question, the last snapshot will be the last moment that all the catalyst cells are alive. I think because of the line "if(fail>150) break;" the snapshot will always come from generation (400-150) = 250 or later.

[Scorbie]

Thanks I actually made a mistake with my code, not yours... But your explanation helped me remember what parts of the code do and fix the bug. I think it's fixed now

[Scorbie]

Is is just me, or does ptbsearch really misses some collisions?
presgr.in

Code: Select all

..........**............
.....**..*.*............
.....*...*..............
......***.****..........
........z.z..*..........
....***.z.z.............
...*..z.zz..............
...**...................
........................
...**...................
...*..*...bb............
....**z..b..b...........
.***......bb............
*...*........aa.........
.***.........a.a........
....**z......a..........
...*..*..........a......
...**..........a........
...................a....
.................a......
.....................a..
...................a....
.......................a
.....................a..

test.list

Code: Select all

.C$C.C$C.C$.C!

Running the following:

Code: Select all

./ptb2 presgr.in test.list 45 0 25 1

generates nothing, while I expected:

Code: Select all

x = 16, y = 26, rule = B3/S23
10b2o$5b2o2bobo$5bo3bo$6b3ob4o$8bobo2bo$4b3obobo$3bo2bob2o$3b2o2$3b2o$
3bo2bo3b2o$4b3o2bo2bo$b3o6b2o$o3bo8b2o$b3o9bobo$4b3o6bo$3bo2bo$3b2o5$
9bo$8bobo$8bobo$9bo!

I don't think there are other causes to this problem because when I try the same search with different (opaque, by the way) catalysts, they seem to output some results.

[chris_c]

Is is just me, or does ptbsearch really misses some collisions?

I don't think the beehive is restored for long enough to qualify as a valid result.

Call to "restored" is here. Definition of "restored" is here.

Meaning that after the placement of the final catalyst we need all of the catalysts to be cleanly restored for at least 10 generations out of the next 50 to output the result.

It looks like the beehive in the reaction you give is only cleanly restored for 2 generations.

[Scorbie]

Ahh, you're right... I remembered you saying that before... Sorry for the inconvenience.