Pattern viewer for forum threads

[rowett]

Although the anti-aliasing is gone, the annoying slowness that occurs at zoom levels with more than 1 pixel/cell is still there.

That should be gone for you also since it's due to high quality rendering which is now off by default.

I made a video showing rendering bugs

Thank you that's helpful. Which device are you using?

It also includes a centering problem

Does the centering problem only occur with high quality rendering? I can't reproduce it.

[Gustone]

Thanks, the blur is gone.

[muzik]

Since lifeviewer can support arbitrary history rules can it also be made to support arbitrary Extended Life rules?

Also is it possible to change rules without deleting the entire pattern?

[rowett]

Since lifeviewer can support arbitrary history rules can it also be made to support arbitrary Extended Life rules?

Please tell me more about Extended Life.

Also is it possible to change rules without deleting the entire pattern?

Yes: Settings->Pattern->Change Rule

[muzik]

Also is it possible to change rules without deleting the entire pattern?

Yes: Settings->Pattern->Change Rule

This doesn't seem to work if changing from a 2-state rule table to a 2-state normal rule.

[muzik]

Since lifeviewer can support arbitrary history rules can it also be made to support arbitrary Extended Life rules?

Please tell me more about Extended Life.

State 2: All 8 neighbouring cells will be born in the next generation (cells can still die within this region)
State 3: All 8 neighbouring cells will die in the next generation (cells can still be born within this region), same as gray cell from LifeHistory
State 4: Combination of the above, essentially inverts the states of all surrounding cells
State 5: Stops cells from being born on its position, but does not otherwise alter cell states
State 6: Essentially acts like a Life cell that cannot die, and can give birth to cells by themselves or in conjunction with other proper Life cells
State 7: Same as above but three of these cells cannot give birth to a normal Life cell on their own

[muzik]

Potential bug: the T menu does not show the alias for this rule.

Code: Select all

 x = 3, y = 1, rule = socrates
ob96o!

There's also this non-isotropic alias that I asked people to figure out the rulestring for years ago, but nobody cared: viewtopic.php?f=11&t=3097&p=51854#p51817

[muzik]

Here's the (not yet accepted in) merge request with the calculated rule aliases, which I'm praying are correct:

https://gitlab.com/apgoucher/catagolue/ ... equests/31

[rowett]

Potential bug: the T menu does not show the alias for this rule.

Code: Select all

 x = 3, y = 1, rule = socrates
ob96o!

It's because the alias is not canonical. MAP rules encode 512 bits of information (for Moore neighbourhood) into 86 base64 characters. Each base64 character is 6 bits so we get 86x6 = 516 bits. That means there are 4 "spare" bits at the end of the string that are not used.

The canonical form for MAP rules sets these "spare" bits to zero.

The final base64 character in the socrates alias is "s" which is the 44th base64 character and 101100 in binary. If we zero the last 4 "spare" bits this becomes the 32nd base64 character "g" or 100000 in binary.

LifeViewer is showing the canonical form of the rule in the T menu which does not match the non-canonical alias definition and so does not show the alias.

[muzik]

I've tagged on three Weighed Life rules to the merge request. Want to point out that Parity can be better expressed as B13/S13V, so it'd be better if LifeViewer were to display it as such.

edit: also B13/S024V is Fredkin2

[muzik]

I i've noticed that the grid no longer seems to oscillate for margolus patterns with the respective theme. Bug, or is there a new setting regarding it that I haven't noticed?

Also I've put up yet another merge request with aliases up (which still hasn't been merged in despite me having made it last decade ), so those too can be added to the LifeViewer list. I've considered maybe having some sort of central location where both LifeViewer and Catagolue could pull aliases from and that can be manually edited (maybe somewhere on the wiki?) but haven't thought about it that much.

Will PCA have custom colours per theme just like how generations rules do?

[rowett]

I i've noticed that the grid no longer seems to oscillate for margolus patterns with the respective theme. Bug, or is there a new setting regarding it that I haven't noticed?

It was a new setting introduced in Build 413.

Will PCA have custom colours per theme just like how generations rules do?

Probably not.

[rowett]

Build 479 is now live on the Forums and LifeWiki

Please note: you will need to refresh your browser to use the new build (Ctrl-F5 on Chrome)

Enhancements since the last released build:

• much faster high quality rendering
• don't use high quality rendering at integer zooms
• high quality rendering now works with [R]History rules

Press Ctrl+Q to toggle rendering quality:

Code: Select all

x = 111, y = 106, rule = LifeHistory
28$78.A$77.ABA$76.BABAB$75.3BA2B$74.7B$68.12B$68.10B$66.12B$66.11B$66.
8B$59.4B3.7B$58.2A5BA6BAB$58.2A4BAB2A4BAB$59.4B2AB2A4BAB$58.5B2A2B2A5B
$56.8B2ABA6B$54.12B2ABA3B$54.19B$53.14B3A3B$54.19B$53.21B$52.22B$51.4B
7.B2AB2A7B$50.4B7.BA3BAB2A5B$49.4B8.2BABA6B2A2B$48.4B9.3BA2BA4B2A2B$47.
4B9.4B6A6B$46.4B11.4BAB2A7B2A$45.4B12.4B3A8B2A2B$44.4B13.20B$43.4B15.
18B$42.4B16.2B3A10BA2B$41.4B16.B2AB2A9BA3B$40.4B16.2B2ABA10BABAB$39.4B
17.3B2AB3A8BA3B$38.4B18.7B2A13B$37.4B19.6B3A5B2A7B$36.4B21.11BABA8B$35.
4B24.8BA2BA8B$34.4B24.4B.3B6A6B$33.4B24.4B2.4BA2BA6B$32.4B24.4B2.6BAB
A7B$31.4B24.4B3.6B3A6B$30.4B24.4B4.15B$29.4B24.4B6.8B4.B$28.4B24.4B7.
8B$27.4B24.4B10.3B$26.4B24.A3B11.3B$25.4B24.A3B13.B$24.4B25.3A$23.4B$
22.4B$21.4B$20.4B$19.4B$18.4B$17.4B$16.4B$15.4B$14.4B$13.4B$12.4B$11.
4B$10.4B$9.A3B$9.ABA$9.2A!
[[ ZOOM 1.9 HEIGHT 380 QUALITY ]]

Fixes since the last released build:

• if pattern load failure is because of illegal cell state then don't assume rule is invalid

The updated hotkey map detailing the LifeViewer keyboard controls is here.

Comments, feedback, suggestions and bug reports welcome!

[GUYTU6J]

Just as an aside in the recent build of LifeViewer support was added for a smoother way of animating spaceships using the [[ TRACK X Y ]] command.

Code: Select all

#C [[ THEME 6 GRID GRIDMAJOR 0 THUMBLAUNCH AUTOSTART TRACK -1/8 0 GPS 4 THUMBNAIL ]]
#C A small c/8 orthogonal ship in the rule B3-ky8_S2-i3-y4qy, found by apgsearch
x = 8, y = 4, rule = B3-ky8_S2-i3-y4qy
b2o3bo$obo3b2o$obo3b2o$b2o3bo!

In the thumbnail, the black ON cells are stained by white dots.

[rowett]

In the thumbnail, the black ON cells are stained by white dots.

This should be fixed. Thanks for reporting!

[muzik]

Most of the merge requests for new rule aliases have been accepted in, so these should be fine for adding to LifeViewer:

https://gitlab.com/apgoucher/catagolue/ ... 93991b3e56 (Parity should instead be added as B13/S13V)

https://gitlab.com/apgoucher/catagolue/ ... 7ae5ad4e1e (Fredkin2 should instead be added as B13/S024V)

https://gitlab.com/apgoucher/catagolue/ ... 0a1f740139



Currently all that needs merged in is a correction to a mistake I made when identifying rules from mirek's site: https://gitlab.com/apgoucher/catagolue/ ... 2d7b2408fd

[rowett]

Most of the merge requests for new rule aliases have been accepted in, so these should be fine for adding to LifeViewer

Done, thanks!

[muzik]

Here's some more aliases that can be added, as well as some that should be removed (turns out they weren't actually Generations, maybe I read the notation wrong):

https://gitlab.com/apgoucher/catagolue/ ... d=69555748

[muzik]

Also, for oblique spaceships with named displacements (see https://www.conwaylife.com/wiki/Oblique_spaceship), can these be displayed in the results of Identify instead of just Oblique?

Code: Select all

x = 79, y = 31, rule = B3/S23
6bo3bob2o$6b3obo3bo18b3o$5b2o4b3o3b2o15b2o2b2o$5bo3bo6b2o14bo2bobo$o3b
o4b2obo3bo2bobo11bo4bobo$bob3o5b3o2bo4bo13bo4b2o$bo3b4obob2o3bob2obo2b
3o2bo3bobo4bobob2o$6b3o2bo2b2o5b2o2b2obo3b2o2b2o3bobob2o$7b2o2bo2b2o5b
2o2b2obobob2o4bo2bo11b2o$12bobo6b2o2bo2bob2obo5bob2o4bo5bobo$21bobob2o
b2o3b2o3b2o2b2o3bo5bo3bo$20bo5b3o5bob2obo4bo2bo5bo3b4o$20bo3b3o8bo9b2o
3bo2bo3bo5b3o$20bo24bob3obo6bo2bo$22bo22b3o3bo10bobob2o$45b2o2b2o3bo3b
2ob4obobo2b2o$46bob3obo5bo5bo7bo$45b2o2b2obo6bobo4bo5b3o$52b2o6bobobo
3bo4b2o$47bobo4b2obob2o2b3ob2ob2ob2o$48bo11b4obo4b2ob2o2$68bo6b2o$68bo
6bobo$68bo3bo2bo$69bobobo4bo$71bobo2b3o$67bo3bo2bo$67bo3bob2o$69bo2bo$
70bo!

In addition to the aliases in the above post, "Pulsar Life" should be added as an alternate alias for B3/S238, "Replicator 2" for B1357/S02468, "Anneal" for B4678/S35678, "Flakes" for B3/S012345678, and these aliases could be added:

Parity R2

Code: Select all

x = 65, y = 55, rule = R2,C0,S1,3,5,7,9,11,B1,3,5,7,9,11,NN
o!

Fredkin2 R2

Code: Select all

x = 65, y = 55, rule = R2,C0,S0,2,4,6,8,10,12,B1,3,5,7,9,11,NN
o!

Also, shouldn't the title "Non-Isotropic Aliases" be changed to "Non-Isotropic Non-Totalistic Aliases", since it's specified in both the isotropic non-totalistic titles and the button that makes you jump to it? On a similar note, shouldn't all the buttons starting with NT be changed to I NT to better fit in with the non isotropic button?

Since alternating aliases can't be added to catagolue here's two ones from the alternating rules thread:

B3/S23|B/S12345678 - Dotlesslife
B3/S23|B/S1c2345678 - SparklessLife

The s at the end of the alias for Socrates could be changed to a g to make it canonical, if I understand it right.

[rowett]

Also...

Done, thanks!

[muzik]

For the two range-2 vN rules, the C0 should be changed to a C2 so they get correctly recognised in the T menu.

There also seems to be something odd going on with the T menu here, as from what I can see these are the exact same rule but only the latter seems to have its alias actually recognised:

Code: Select all

x = 100, y = 1, rule = B134578/S02345678|B6i7/S268
bobo2bob5obobo3bob2obo11bobo2b5o4b2ob3ob2ob2obo2b3ob2o2bobo2bobobo4b4o
4bo!

Code: Select all

x = 100, y = 1, rule = Unidim1
bobo2bob5obobo3bob2obo11bobo2b5o4b2ob3ob2ob2obo2b3ob2o2bobo2bobobo4b4o
4bo!

Speaking of aliases, the Help > Aliases screen has the jump to section buttons going down off the screen and intersecting other buttons, so something should be done about this.

Is it possible to customise PCA cell colours? The scripts > colours section doesn't appear to list any of them.

Identify only seems to list the unique displacement name if the displacement cannot be simplified down, and does not list it for non-coprime pairs of that same displacement such as (66,22):

Code: Select all

x = 4, y = 2, rule = B2cek3aj4r5i/S2aen3jkqy4c5akn6e
bo$o2bo!

Code: Select all

x = 6, y = 5, rule = B2kn3-ekq4ceit5c/S23ijkqr4eikry6c8
bobo$2o$bobo$3bo$5bo!

And finally, another two new rules:
https://gitlab.com/apgoucher/catagolue/ ... 498dd909be

[rowett]

For the two range-2 vN rules, the C0 should be changed to a C2

Fixed.

... only the latter seems to have its alias actually recognised

It was an issue with reverse lookup and is now fixed.

the Help > Aliases screen has the jump to section buttons going down off the screen and intersecting other buttons, so something should be done about this.

Something will be. It's on the backlog.

Is it possible to customise PCA cell colours?

Yes use [[ COLOR 0..15 R G B|name ]].

...if the displacement cannot be simplified down

Fixed.

And finally, another two new rules

Added, thanks!

[muzik]

Fantastic - thank you for your continued efforts.

I feel as though I've reported this bug before, but it's either still here or has returned:

Code: Select all

x = 16, y = 1, rule = LifeHistory
A2.B2.C2.D2.E2.F!
[[ LAYERS 2 ]]

Also can PCA cells with more subcells inside of them be taller in Layers if they aren't already?

Is it possible to customise PCA cell colours?

Yes use [[ COLOR 0..15 R G B|name ]].

Will support for changing their colour via specifying their cell type e.g. NSW instead of their state number be supported at all?


EDIT: As we're on the topic of multistate rules, for the Select menu, could we have some sort of way to select which states can be present in a random fill and which ones cannot?

[rowett]

I feel as though I've reported this bug before

You did and my response was here.

[muzik]

Some suggestions relating to bounded grids:

- Could we be able to specify patterns to run on möbius strips or finite cylinders? Two opposing faces would act like plane and completely block birth, whereas the other two (the axis marked with a *) would be connected together. For example, a grid with cylinder 32*,32 would act like this on the y axis ends:

Code: Select all

x = 32, y = 32, rule = B3/S23:P32,32
11$14b3o3b3o$13bo2bo3bo2bo$16bo3bo$16bo3bo$13bobo5bobo!

but like this for the x axis ends:

Code: Select all

 x = 32, y = 32, rule = B3/S23:T32,32
13$16bo2bo$20bo$16bo3bo$17b4o4$17b4o$16bo3bo$20bo$16bo2bo!

Möbius strips would be similar - just flip the attached end.

- Sphere bounded grids act kind of weird and asymmetric compared to other bounded grids - as can be seen here, this barberpole rotor stretches around the top left to bottom right equator but dies when spanning the poles:

Code: Select all

x = 8, y = 8, rule = B3/S23:S8
7bo$obo2$2bobo2$4bobo2$6bo!

Code: Select all

x = 8, y = 8, rule = B3/S23:S8
o$5bobo2$3bobo2$bobo2$bo!

If a * is appended to the sphere grid, can this behaviour be flipped horizontally so that the lower RLE oscillates and the top dies?

- Can B0 Generations be permitted on bounded grids, since LtL/HROT-with-decay can be? These should be the same:

Code: Select all

x = 19, y = 15, rule = R1,C3,M1,S6..6,B0..8,NM:T300,300
8.6A$6.4A4.2A$5.A4.2A4.A$4.A7.A4.A$3.A8.A4.A$2.A8.2A4.A$.2A6.2A.A5.A$
A.A5.A3.A5.A$2A5.A4.A5.A$2A5.A3.2A5.A$2A4.5A7.A$2A15.A$2A14.A$A.A11.
2A$3.11A!
[[ STEP 3 THEME Blues ]]

Code: Select all

x = 19, y = 15, rule = B012345678/S6/G3:T300,300
8.6A$6.4A4.2A$5.A4.2A4.A$4.A7.A4.A$3.A8.A4.A$2.A8.2A4.A$.2A6.2A.A5.A$
A.A5.A3.A5.A$2A5.A4.A5.A$2A5.A3.2A5.A$2A4.5A7.A$2A15.A$2A14.A$A.A11.
2A$3.11A!
[[ STEP 3 THEME Blues ]]