Enumerating spaceships?

[codeholic]

There has been a number of programs, that enumerated still-lifes and low period oscillators. If I'm not missing anything, the latest and the most advanced one was written by Mark Niemiec in late 90s -- more than 15 years ago!

I wonder, if it would be possible now, having more powerful computers and maybe introducing more specialized heuristics in the algorithm, to enumerate spaceships at least up to 20 cells (the minimum population of the loafer) and answer the question: what are the top 5 smallest spaceships in the Game of Life?

[codeholic]

I appears that such program was written by David Eppstein in 1998, gsearch. Does anyone know how deeply was the search performed back then?

[codeholic]

It seems to me as if gsearch output only unique velocities, e. g. if there has been a p4 c/2 spaceship, then it won't report other spaceships of that period and velocity. Otherwise I cannot explain why it reports LWSS and the Schick engine, but neither MWSS nor HWSS. Am it getting it right?

[Sokwe]

I don't know the answer to your question, but I have another question. The description given in the source code for gsearch (https://www.ics.uci.edu/~eppstein/ca/gsearch.c) heredoes not seem to match the description on David Eppstein's website (here). For example, the source code says it tests random patterns and the description on the website says it enumerates all patterns in a small box. The code also seems to only check boxes of size NxN, but the description on the website claims that it checks boxes of size NxM. Does anyone know the reason for these discrepancies?

[codeholic]

The more I think the more I'm inclined towards the thought that zdr's search program could be written using a brute force approach I described in the first post or a similar one. In the end it would need to enumerate patterns only up to 14 cells, if symmetries are taken into account.

EDIT: Another question is why gsearch didn't find the copperhead

EDIT 2: I was wrong. zdr posted their search program.