22-bit still life syntheses

AGreason · Post by **AGreason** » November 14th, 2022, 9:08 pm

DON'T PANIC - this thread is at this time essentially a placeholder, created to acknowledge that the xs21 synthesis project has been completed. There has been minimal automated effort put into the xs22, and no manual synthesis effort is requested. For the next while the primary objective of this thread is to provide a place to track and discuss automated xs22 synthesis efforts.

I've kicked off a depth-1 all-targets transfer search, dunno how long it'll take with the newly-expanded target set.

I acknowledge that people (including me) have largely stopped caring about these projects at this point.

11/14/22: 672172 unsynthed strict xs22
11/27/22: 2804
02/21/23: 1926
09/13/23: 1913
09/21/23: 1519
03/12/24: 1146
03/20/24: 687

May13 · Post by **May13** » November 19th, 2022, 10:57 am

Have any efforts been made on this project? It looks like the full list of 22-bit strict still lifes is still not loaded to Catagolue.
There are 64325 22-bit strict still lifes with syntheses (still life #64325 is the Cis-mirrored worm with 4G synthesis), so there are 672172-64325=607847 still lifes remaining.
The full list of apgcodes of strict xs22's is here. Use bash command xz -d 22_bits_strict_apgcodes.txt.xz to decompress this file.

carsoncheng · Post by **carsoncheng** » November 19th, 2022, 11:43 am

May13 wrote: ↑
November 19th, 2022, 10:57 am
Have any efforts been made on this project? It looks like the full list of 22-bit strict still lifes is still not loaded to Catagolue.
There are 64325 22-bit strict still lifes with syntheses (still life #64325 is the Cis-mirrored worm with 4G synthesis), so there are 672172-64325=607847 still lifes remaining.
The full list of apgcodes of strict xs22's is here. Use bash command xz -d 22_bits_strict_apgcodes.txt.xz to decompress this file.

It appears that the depth-1 transfer.py run is not complete yet, given that I can't even see a merge request on the Shinjuku repository. For example, in the 21-bit synthesis thread, it took 2 days to process most of the 216425 unsynthesized xs21s at the start of the project. Given that the list of unsynthesized 22-bit still lifes is more than three times greater, at 672172, it appears that more time is needed to process these still lifes. After that, most of the 22-bit still lifes will be synthesized.

hkoenig · Post by **hkoenig** » November 21st, 2022, 9:56 am

There are 14 17-bit oscillators for which Catagolue doesn't have a construction:

xp2_35g41h54gzx3x3
xp2_j92g55hz11x3
xp2_m24g441lpz03
xp2_31a0nzwc850e
xp2_31a0nzw4490ac
xp2_k51h4gljzw3
xp2_35g411l4z22403
xp2_0m24gl3z2a882
xp2_3hk2060szw2251
xp2_03l0i4kz642x23
xp2_318czw315b8o
xp2_0b1205jgz1140c02
xp2_62k030288zx3048d
xp2_62k030288zx1148d

AGreason · Post by **AGreason** » November 21st, 2022, 4:26 pm

carsoncheng wrote: ↑
November 19th, 2022, 11:43 am
Given that the list of unsynthesized 22-bit still lifes is more than three times greater, at 672172, it appears that more time is needed to process these still lifes. After that, most of the 22-bit still lifes will be synthesized.

Yes. I'm running an all-targets depth-1 transfer run, but I also included already synthesized sls and sls with soups but no synths in the target set in addition to the xs22, resulting in a target set of roughly 3.5 million objects. It's roughly half done, if nothing interrupts the job it should finish in about a week.

AGreason · Post by **AGreason** » November 27th, 2022, 9:22 pm

Transfer pass (and some recursive searches, which knocked down the after-depth-1 remaining count of about 10k down significantly) complete, 2804 unsynthed
commit: https://gitlab.com/parclytaxel/Shinjuku ... 2de1b48bd6
It will take an unknown amount of time for this to be fully reflected on catagolue - it exceeds the size of the largest previous component dump by more than a factor of four.

May13 · Post by **May13** » November 28th, 2022, 11:59 am

1. While I was digging among new synthesized 22-bit still lifes, I found this (xs22_0695m88cz4a9611):