pcallahan wrote: ↑
June 22nd, 2021, 10:26 am
The question gave me a flashback to 49P88
, which I have never thought of as a shuttle, despite taking a 180° turn. I still would not call it a shuttle. The classic shuttles are p30 and p46, which are both symmetric (ignoring catalysts) and can be thought of as flips as well as 180° turns. Maybe that is too strict as a definition, but I think it's my intuition when looking at something and deciding if it's a shuttle.
MathAndCode wrote: ↑
June 22nd, 2021, 10:34 am
I would arguable that 49P88 is more of a shuttle than the p29 pre-pulsar hassler, which is commonly called a shuttle, because in 49P88, the hassled region goes both directions in the same way, which is impossible for any odd-period oscillator.
The LifeWiki definition is close, but I don't think it's exactly correct, at least from my memory of how the term was initially used back in the Scientific American and Lifeline days. They key feature of shuttles (as distinguished from other pulsators) is that their active regions move a substantial distance. While it's true that all of Life's common natural shuttles (e.g. queen bee and twin bees) share two additional properties (i.e. bilateral glide symmetry, and the fact that their mechanisms must be assisted by additional non-moving components), these are not necessary properties, and there are common shuttles in other rules that don't share them, e.g. ones that operate without any outside assistance, or ones with C4 rotational symmetry, or ones with no glide symmetry, that move forwards and backwards using different mechanisms (so are capable of having odd periods).
E.g. in B3/S2ae3aeijr4-ckqy there is a natural P280 suttle with C4 glide symmetry; it moves a substantial distance every 70 generations, leaving huge clean plumes, and rotating 90°. It moves sufficiently far that two copies of it can co-exist in the same area. Placing two exactly 140 generations out of phase creates a true P140 oscillator, as the plumes interact trivially, but advancing one of the shuttles a bit (e.g. 8 generations here) allows the two copies to rotate totally independent of each other, resulting in two totally separate and independent shuttle oscillators that occupy the same space.
Code: Select all
x = 81, y = 68, rule = B3/S2ae3aeijr4-ckqy
49P88 would qualify as a shuttle, even under the narrower Lifewiki definition, because if you look at a 3D map of its active cells, at areas occupied at opposite ends of its travels (e.g. the pi heptomino evolutions) do not overlap.