simsim314 wrote:I'ts still interesting to beat the minimal designed ship record though.
At the moment I'd be tempted to do that by reworking the Demonoid, actually, since the current design is nowhere near optimal. I suspect it can be brought down to around 20,000 ON cells, which might be a hard target for a Caterloopillar to beat.
I haven't come up with a single-channel design that seems likely to improve on the 0hd model -- though it might happen, if single-channel elbow operation recipes turn out to be within a factor of two of the efficiency of 0hd glider-pair recipes.
However, it's definitely possible to find a slow-salvo recipe for the current construction arm constellation that's shorter than 540 gliders. That recipe was generated completely by hand, with no time spent on optimizing it for the 0hd elbow-op library... In fact, the 0hd library didn't even exist when that slow salvo was compiled.
A revised slow salvo using one of the recent eater2 seeds, and a few other optimizations like using the cheapest possible lane to delete extra debris, could easily bring the population below 24,000, and just possibly below 20,000 with enough work.
I can see that the Caterloopillar's initial stage of creating a target block and then pushing it way out, can be replaced by something much lower population, just by building the block in the right place to begin with (right?) But that only reduces a small fraction at the front of the ship.
simsim314 wrote:There are many alternatives which could improve specific speeds. In particular one can send gliders to destroy the salvos, thus the only limitation on the period would be the period of the reading head.
I don't have a clear picture of where those gliders would come from -- is this just for the negative helix, or can you bounce gliders up to the front somehow, too? (Without increasing the total population to something huge, that is, like the shield bug vs. the Centipede.)
What are other likely reduction points? Might there be a way to use only even-phase *WSSes, for example, instead of a mix of even and odd? That would cut down the minimum population by five percent or so, if it didn't increase the *WSS population (which it probably would).
Of course, the Caterloopillar is an order of magnitude ahead on the bounding box already -- there's no way the Demonoid is going to catch up on that metric.