There could be another compication ... we still do not know how the p8 pattern will be (rather optimaly) created with only one of 4 odd phases and 1 of 4 even phases. We can use pslmake modification to do so or create sleeping patterns of p8 reflectors to be activated by first incoming glider they have to reflect. Do we have such sleeping patterns for both bumper and bouncer?
Code: Select all
x = 105, y = 97, rule = B3/S23
bo$2bo$3o62$65bo$66bo$64b3o6$82b2o$82b2o3$87b2o$87b2o5$76b2o20b2o$75bo
bo20b2o3b2o$75bo27b2o$74b2o$84b2o$83bo2bo$83bobo15b2o$84bo16b2o2$79b2o
$80bo$77b3o$77bo22b2o$91b2o7b2o$91b2o!
I do have these for bouncer ... I do not think I can do it better.
Code: Select all
x = 313, y = 147, rule = LifeHistory
$128.B$128.B$128.B$126.3B$125.2BE6B$125.BEBE6B157.B$125.B2E7B157.B9.B
$125.10B20.B136.B9.B$127.8B20.B134.3B9.B$128.6B10.2B.2B6.B133.2BEB9.B
$122.2E5.4B9.3BE4B5.B133.BEBE$122.2E4.6B7.3BEBE4B4.B133.B2E7B$122.4B.
8B5.5BE5B138.10B$124.B.4B2.2B2D5.10B116.A23.8B$118.2EB2.B.4B4.2D2B5.
9B115.2AB14.2E7.6B$118.2EB.2B.3B6.BD2B2.11B115.ABAB12.E2.E2.2B3.4B$
114.E5.3BE3B8.13B.B117.4B11.E.E5B2.6B$114.3E3.2BEBE2.2B7.12B120.4B11.
E5B2.6B2D$117.E3.BEBE.3B8.12B120.4B11.5B.4B2.2DB5.9B$116.2E3.2BE2.4B
8.14B118.4B10.3B2.4B4.BD2B4.2B2E4B$116.4B2.2B.5BE6.11B2E121.4B8.3B2.
6B4.4B3.2BEBE3B$118.BEB4.4BE.E4.12B2EB121.4B7.2B2.2E2B.2B5.4B.4BE4B$
117.BEBE.B.2B2.E2.E3.4B2.9B123.4B9.E2BE2B.B2E4.7B.5B$114.2E.B2E2B7.2E
3.4B136.4B3.B.B2.B2E2B2.B2E6.4B3.4B$114.EC7B10.4B138.8B4.3B12.3B$115.
7B10.4B140.7BEB3.4B10.2B$115.7B9.4B142.5BEBEB4.2E9.2B$113.9B8.4B143.
5B2E2B4.2E8.3B$110.4BE8B6.4B143.8B.B2E11.4B$110.3BEBE2B2.4B4.4B137.5B
E7B3.BE.E9.4B$110.4B2E2B3.4B2.4B138.4BEBE2B.2BCE5.E8.4B$110.7B5.8B
139.5B2E2B3.2E5.2E6.4B$110.5B8.6B140.8B18.4B$110.4B10.4B141.6B19.4B$
110.3B10.6B141.4B19.4B$101.2B7.4B8.8B132.4B4.3B19.4B$100.3B7.5B6.4B2.
4B130.4B26.4B$100.3B9.4B4.4B4.2BAB129.4B25.4B$100.3B10.4B2.4B6.2B2A
128.4B24.4B$100.3B11.8B8.2A129.4B23.4B$101.2B12.6B141.3B12.D9.4B$102.
B13.4B143.2B12.BD7.4B$101.2B12.6B141.3B12.2DB5.4B$100.3B11.8B139.4B
12.4B3.4B$99.4B11.10B136.4B14.4B.4B$98.3CB12.7B2E2B134.2C2B16.7B$97.
3BC13.6BEBE2B133.CBCB17.7B2.B$96.3BC15.6BE3B132.3BC17.8B.3B$95.4B157.
4B17.14B$94.4B157.4B17.15B$93.4B157.4B18.13BE2B$93.3B157.4B19.12BEBEB
$93.2B157.4B20.13BEB$93.B157.4B21.16B$250.4B22.3B2E11B$113.3B133.4B
23.3B2EB4.5B$248.4B24.6B4.4B$247.4B25.6B4.4B$246.4B26.6B$245.4B$244.
4B29.6B$243.4B30.B2.7B$242.4B33.7B$241.4B35.5B$240.4B36.5B$239.4B$
238.4B$237.4B$236.4B$235.4B$234.4B$233.4B$232.4B$231.4B$230.4B$229.4B
$228.4B$227.4B$226.4B$225.4B$224.4B$223.4B60$2.3C158.2C$4.C157.C.C$3.
C160.C!
If we choose the sleeping reflectors version, the seed of destruction should not interract with the awakening reaction.
calcyman wrote: ↑November 19th, 2023, 6:35 pm
But if we are trying to minimize the number of bits, it would be better to restrict ourself to 2 given phases mod 8 in pslmake and extend the bespoke set to be able to construct all 8 phases directly. Hmm maybe better change to pslmake would be allowing gliders as an intermediate target forcing a turner be inserted immediately to get glliders propper orientation and timing. So including one time turners in bespoke...
Time to git clone
https://gitlab.com/apgoucher/pslmake.git ... . This salvo restriction and immediate gliders is one thing to do, the delaunay stuff, I was talking before is another one ...
Nice the pslmake uses 16 glider signatures (2 colors 8 phases in them) to allow filtering recipes not doable by DBCA/ECCA, unfortunately (seems to me) the database is not processed to support recipes with such restricted phases. (A lot of p2 recipes could be discarded by a pessimistic reasoning.) ... the line/phase agnosticised versions would serve much better (or recipes derived from the agnosticised ones). (Or the recipes should be marked as p1,p2,p8 and filtering by signatures of 1,2,4 bits ... and when not filtered out the appropriate version of gliders chosen.) It would be nice to cash recipes transated to alllowed signatures and filter out non matching recipes during their read rather then during attempt of their use. This actually leads to having separate recipes depending on target color (actually extended color (4 options) if we should distinguish phase 4 changes).
Even doing extended agnosticise while reading the recipe to either cache it or discard could be fine (for big patterns ... when it is done just once per a recipe (4 times considering extended colors)). This will be almost trivial for text (p2) recipes (if we consier recipe defning configuration after each glider reaction) ... doing agnosticise for indepenent p2 parts trying to change relative phases would lead to exponential compexity, but the indiviual recipes are short enough to think about it in later versions.
Eagnosticise would be important mainly on bespoke recipes. (Actually we neednot eagnosticise, we need filetring and translation to 4 target colors, "based on eagnsticise").
qd is used for dimers ordering ... it uses area of the enclosing rectangle, it could work well, but it is incompatible with the function description ...