Pattern viewer for forum threads

[rowett]

The cogwheel button in this pattern occasionally fails.

If by fails you mean gets greyed out then this will happen if the height of the viewer is small. This is by design.

[confocaloid]

The pattern in the following quote doesn't display correctly (due to CXRLE it settles into a still life), and there seems to be no error message to inform the user that something went wrong:

Yup, it's easy to construct agars with arbitrarily large periods.

What about naturally occurring ones, though?

The other day I found this one, which appears to be p9440 on the face of it, though it's actually only p590 if you take the symmetry into account:

Code: Select all

#CXRLE Pos=0,0 Gen=0
x = 31, y = 32, rule = B3/S23:T31,32
5b2o12b2ob2o$19b2obo3b2o$21bo4b2o7$16b3o$15bo2bo$14bo4bo$12b2o2bob3o5b
o$15bo2bo5bo2bo$11bobobo8bo2b2o$13bo2bo10b2o$11b2o12b2ob2o$21b2o3bob2o
$21b2o4bo7$2o28bo$2bo27bo$3bo25bo$bo2b2o16bo5b3o$2bo18bo2bo5bo$2bobobo
13b2o2bo$bo2bo15b2o!

It works correctly without CXRLE:

Code: Select all

x = 31, y = 32, rule = B3/S23:T31,32
5b2o12b2ob2o$19b2obo3b2o$21bo4b2o7$16b3o$15bo2bo$14bo4bo$12b2o2bob3o5b
o$15bo2bo5bo2bo$11bobobo8bo2b2o$13bo2bo10b2o$11b2o12b2ob2o$21b2o3bob2o
$21b2o4bo7$2o28bo$2bo27bo$3bo25bo$bo2b2o16bo5b3o$2bo18bo2bo5bo$2bobobo
13b2o2bo$bo2bo15b2o!

[rowett]

The pattern in the following quote doesn't display correctly (due to CXRLE it settles into a still life), and there seems to be no error message to inform the user that something went wrong

There's no error here. The pattern is positioned correctly with the CXRLE specification.

[unname4798]

The pattern in the following quote doesn't display correctly (due to CXRLE it settles into a still life), and there seems to be no error message to inform the user that something went wrong

There's no error here. The pattern is positioned correctly with the CXRLE specification.

The pattern is positioned incorrectly with the CXRLE specification.

[rowett]

The pattern is positioned incorrectly with the CXRLE specification.

Try the pattern in Golly.

[unname4798]

The pattern is positioned incorrectly with the CXRLE specification.

Try the pattern in Golly.

I can't have an app, and I can't run Golly on web

[rowett]

I can't have an app, and I can't run Golly on web

OK. It's correct and it functions the same way in Golly.

[confocaloid]

When the LtL notation (with either "M0" or "M1" specified) is combined with weighted neighbourhood ("NW..."), there are some inconsistencies between Golly and LifeViewer:

• in Golly 4.3, it seems like the "M..." part is ignored, and the meaning of each numeric survival condition is determined solely by the "NW..." part.
  In other words, the weight of the middle cell specified in the "NW..." part overrides the choice of either "M0" or "M1".
• In LifeViewer build 1201, the "M0,NW" cases seem to be interpreted differently (see examples below).

Examples / test cases:

"M0": all three examples define Conway's Life in Golly 4.3; however, the "M0,NW" examples work differently in LifeViewer.

Code: Select all

x = 3, y = 3, rule = R1,C0,M0,S2..3,B3..3,NM
b2o$2o$bo!

Code: Select all

x = 3, y = 3, rule = R1,C0,M0,S2..3,B3..3,NW111101111
b2o$2o$bo!

Code: Select all

x = 3, y = 3, rule = R1,C0,M0,S3..4,B3..3,NW111111111
b2o$2o$bo!

"M1": all three examples define Conway's Life in Golly 4.3

Code: Select all

x = 3, y = 3, rule = R1,C0,M1,S3..4,B3..3,NM
b2o$2o$bo!

Code: Select all

x = 3, y = 3, rule = R1,C0,M1,S2..3,B3..3,NW111101111
b2o$2o$bo!

Code: Select all

x = 3, y = 3, rule = R1,C0,M1,S3..4,B3..3,NW111111111
b2o$2o$bo!

Hexagonal cells: all examples define B2/S34H in Golly 4.3; however, the "M0,NW" examples work differently in LifeViewer.

Code: Select all

x = 3, y = 4, rule = R1,C0,M0,S3..4,B2..2,NH
obo$obo$2bo$bo!

Code: Select all

x = 3, y = 4, rule = R1,C0,M0,S3..4,B2..2,NW110101011H
obo$obo$2bo$bo!

Code: Select all

x = 3, y = 4, rule = R1,C0,M0,S4..5,B2..2,NW110111011H
obo$obo$2bo$bo!

Code: Select all

x = 3, y = 4, rule = R1,C0,M1,S4..5,B2..2,NH
obo$obo$2bo$bo!

Code: Select all

x = 3, y = 4, rule = R1,C0,M1,S3..4,B2..2,NW110101011H
obo$obo$2bo$bo!

Code: Select all

x = 3, y = 4, rule = R1,C0,M1,S4..5,B2..2,NW110111011H
obo$obo$2bo$bo!

I think a reasonable possibility here would be to completely disallow rulestrings where both "M..." and "NW..." appear simultaneously.

(When the user specifies a weighted neighbourhood via "NW...", they are explicitly specifying the weight for the middle cell, too; and that weight should determine whether or not the middle cell is "counted as a neighbour".

When the user specifies either "M0" or "M1", that part should determine whether or not the middle cell is "counted as a neighbour".

When both "M..." and "NW..." are given, it is unclear which of two should be followed, so it would make sense to consider such rulestrings invalid.)

[rowett]

When the LtL notation (with either "M0" or "M1" specified) is combined with weighted neighbourhood ("NW..."), there are some inconsistencies between Golly and LifeViewer:

Thanks. It's on the backlog.

[get_Snacked]

could this be added sometime soon?
(like, being able to use the notation without having to create a ruletable)

[wirehead]

When a pattern is such that LifeViewer can run it at 64x without any throttling, I find it actually restrictive that the maximum step is 64x. Can there be in option to "run as fast as possible without freezing the browser"?

[rowett]

When a pattern is such that LifeViewer can run it at 64x without any throttling, I find it actually restrictive that the maximum step is 64x. Can there be in option to "run as fast as possible without freezing the browser"?

You can use Settings>Playback>Go To Gen (hotkey Shift N) and then type +10000 in the dialog and click OK. That will tell LifeViewer to advance 10000 generations as fast as it can. You can use any number instead of 10000 as you wish.

[rowett]

could this be added sometime soon?
(like, being able to use the notation without having to create a ruletable)

Not by me. Perhaps someone else can take it.

[get_Snacked]

Not by me. Perhaps someone else can take it.

hmm... i think using the script to convert from rulestring to ruletable should be easy enough to make in LV.

[confocaloid]

I think it would be helpful to do some more brainstorming on what are different existing possibilities for such a notation.

Designing a useful notation for multistate CA is a highly nontrivial problem. So far, the proposals seem to be unnecessarily complicated, and/or insufficiently well-understood at this point. It is better to avoid "setting in stone" something that is likely to need revisions or even a complete redesign/rethinking/restarting from scratch.

Among other things, the scope would be necessarily limited (there would be always many "interesting" cellular automata that cannot reasonably be defined by a single-line string, and would still require a .rule file). So a proposal for a single-line notation would need to explain what's the intended scope.

Further, the notation needs to be simple, straightforward, easy to explain, easy to understand and easy to use.
(That implies limitations on what the notation can do in a reasonable way; however, see above, the scope would be necessarily limited anyway and many things would still require a .rule file.)

could this be added sometime soon?
(like, being able to use the notation without having to create a ruletable)

Not by me. Perhaps someone else can take it.

hmm... i think using the script to convert from rulestring to ruletable should be easy enough to make in LV.

[Citation needed]

I'm re-quoting this post so it can get more attention.

Can you allow changing individual colors in waypoints at specific points?
Can you also allow overwriting colors?
Example cases:

Code: Select all

#R B3/S23
bo$3o$o!
[[ COLOR DEADRAMP 255 255 255 ]]
[[ ZOOM 8 AUTOSTART T 100 ]]
[[ PAUSE 1 COLOR DEADRAMP 0 0 0 ]]
[[ T 200 ]]

[unname4798]

Bug:

Screenshot_2024-11-12-17-54-17-171.jpeg (600.36 KiB) Viewed 6933 times

[rowett]

Bug:

Please provide steps to reproduce the issue.

[unname4798]

Bug:

Please provide steps to reproduce the issue.

Click the cogwheel button. Click "Themes" button, then click any theme twice or more times.

[rowett]

Click the cogwheel button. Click "Themes" button, then click any theme twice or more times.

Which device and browser are you using?

[unname4798]

Click the cogwheel button. Click "Themes" button, then click any theme twice or more times.

Which device and browser are you using?

Lenovo Tab M10 HD (2nd generation), Android 10, RS Browser

[rowett]

Lenovo Tab M10 HD (2nd generation), Android 10, RS Browser

I can't reproduce it on my Android tablet. Try using Chrome. Also which LifeViewer build are you using?

[unname4798]

Lenovo Tab M10 HD (2nd generation), Android 10, RS Browser

I can't reproduce it on my Android tablet. Try using Chrome. Also which LifeViewer build are you using?

I can't have Chrome. I use LifeViewer build 1208.

[rowett]

I can't have Chrome. I use LifeViewer build 1208.

Try build 1209. I can't reproduce the issue but it may help.

[confocaloid]

An idea for a new feature: allow running two or three two-state patterns in different "layers" within a single viewer instance, "on top of each other".
With three "layers", one could use the RGB scheme (red = alive in layer 1, green = alive in layer 2, blue = alive in layer 3).

Possible scenarios include:

• two or three (slightly?) different patterns, all in the same two-state cellular automaton. (E.g. showing propagating effects of initial differences; showing different ways to use a component.)
• a single initial pattern, in two or three (slightly?) different two-state cellular automata, to show how the evolutions diverge.

For range-1 Moore CA, this is partially covered by RuleLoader tables like Rule:TripleLife. However, RuleLoader doesn't work for higher ranges.

A single pattern in two different 2-state CA
For example, it would be nice to be able to run the following pattern in the specified range-2 CA and in Conway's Life at the same time:

Code: Select all

#C Source: https://conwaylife.com/forums/viewtopic.php?p=175692#p175692
x = 30, y = 28, rule = R2,C0,S18-24,26-35,60,B27-35,NW00010001000109090901000900090001090909010001000100
20b2o$20b2o5$17bo$18bo3bo$13b2o3bo4b2obo$13bo3bo2bo2b2obo$13bobo3b4o3b
o$14b2o3b2o$2o$2o7$8b2o$8b2o$20b2o2bo$20bo2bo$22bobo3b2o$20b3o5bo$20bo
7bo$28bo!
#C [[ THEME MCell ZOOM 8 TRACKLOOP 96 8/96 8/96 ]]

Code: Select all

x = 30, y = 28, rule = B3/S23
20b2o$20b2o5$17bo$18bo3bo$13b2o3bo4b2obo$13bo3bo2bo2b2obo$13bobo3b4o3b
o$14b2o3b2o$2o$2o7$8b2o$8b2o$20b2o2bo$20bo2bo$22bobo3b2o$20b3o5bo$20bo
7bo$28bo!
#C [[ THEME MCell ZOOM 8 TRACKLOOP 96 8/96 8/96 ]]

Two different patterns in a single 2-state CA
For another example, it would be nice to be able to run these two patterns "on top of each other" (here they are shown side-by-side):

Code: Select all

#C Source for the merge circuit: https://conwaylife.com/forums/viewtopic.php?p=170752#p170752
#C [[ T 347 PAUSE 1 STOP 533 HARDRESET ]]
x = 343, y = 87, rule = B3/S23
40b2o3b2o193b2o3b2o$40bo5bo23b2o168bo5bo23b2o$37b2obo5bob2o20b2o5b2o
158b2obo5bob2o20b2o5b2o$36bobob2o3b2obobo26b2o157bobob2o3b2obobo26b2o$
35bo6bobo6bo183bo6bobo6bo$35b7o3b7o4bo178b7o3b7o4bo$56b3o16b2o179b3o
16b2o$37b3o7b3o9bo15b2o160b3o7b3o9bo15b2o$36bo2bo3bo3bo2bo7b2o21b2o
153bo2bo3bo3bo2bo7b2o21b2o$30b2o4b2o4bobo4b2o30b2o147b2o4b2o4bobo4b2o
30b2o$30b2o10bobo185b2o10bobo$43bo199bo2$61bo199bo$61bo199bo$16b2o32b
2o9b3o152b2o32b2o9b3o$17bo32b2o11bo153bo32b2o11bo$17bobo197bobo$18b2o
198b2o3$54b2o198b2o$48b2o5bo192b2o5bo$48bobob3o18b3o10b2o160bobob3o18b
3o10b2o$50bobo20bo11bo2bo161bobo20bo11bo2bo$18bo23b2o6b2o20b2o11b2obo
129bo23b2o6b2o20b2o11b2obo$18bobo21b2o44bob2o126bo23b2o44bob2o$18b3o
64b2o2bobo126b3o64b2o2bobo$20bo63bo2b2o131bo63bo2b2o$85b2o198b2o$30b2o
54bo143b2o54bo$30b2o2b2o50bobo141b2o2b2o50bobo$34bobo48b2ob3o143bobo
48b2ob3o$36bo54bo144bo54bo$36b2o25b2o20b2ob3o145b2o25b2o20b2ob3o$64bo
20b2obo175bo20b2obo$64bobo14b2o181bobo14b2o$65b2o14bobo181b2o14bobo$
82b2o198b2o2$32b2o45b2o151b2o45b2o$32b2o45bobo150b2o45bobo$81bo28b2o
169bo28b2o$48bo10b2o20b2o28bo136bo10b2o20b2o28bo$46b3o10b2o49bo135b3o
10b2o49bo$45bo64b2o133bo64b2o$45b2o198b2o$141bo199bo$81bo58bobo138bo
58bobo$81bo58bobo138bo58bobo$70b2o9b3ob2o54bo128b2o9b3ob2o54bo$16b2o
52b2o11bob2o26b2o15bo85b2o52b2o11bob2o26b2o15bo$15bo2bo2b2o90b2o15bo
84bo2bo2b2o90b2o15bo$16b2o2bobo107b3o83b2o2bobo107b3o$18b2o112bo85b2o
112bo$18bo199bo$15b2obo196b2obo$15bo2bob2o44b2o147bo2bob2o44b2o$17b2ob
2o44b2o149b2ob2o44b2o3$108b2o198b2o$108b2o198b2o$119b2o198b2o$120bo
199bo$14b2o3b2o73b2o21b3o94b2o3b2o73b2o21b3o$14b2o3b2o74bo21bo96b2o3b
2o74bo21bo$92b3o197b3o$92bo199bo$104b2o198b2o$42b2o60bo137b2o60bo$42bo
bo60b3o134bobo60b3o$2o42bo62bo92b2o42bo62bo$2o42b2o154b2o42b2o6$33b2o
198b2o$33b2o198b2o2$10b2o198b3o$11bo8b2o189bo8b2o$11b3o6bo190b3o6bo$
21b3o197b3o$23bo199bo!

This case is currently "solvable via RuleLoader" as shown below. However, higher-range CA would require another approach.

Code: Select all

#C Source for the merge circuit: https://conwaylife.com/forums/viewtopic.php?p=170752#p170752
#C [[ T 347 PAUSE 1 STOP 533 HARDRESET ]]
x = 143, y = 87, rule = TripleLife
40.2G3.2G$40.G5.G23.2G$37.2G.G5.G.2G20.2G5.2G$36.G.G.2G3.2G.G.G26.2G$
35.G6.G.G6.G$35.7G3.7G4.G$56.3G16.2G$37.3G7.3G9.G15.2G$36.G2.G3.G3.G
2.G7.2G21.2G$30.2G4.2G4.G.G4.2G30.2G$30.2G10.G.G$43.G2$61.G$61.G$16.
2G32.2G9.3G$17.G32.2G11.G$17.G.G$18.2G3$54.2G$48.2G5.G$48.G.G.3G18.3G
10.2G$50.G.G20.G11.G2.G$18.G23.2G6.2G20.2G11.2G.G$18.G.D21.2G44.G.2G$
18.3G64.2G2.G.G$20.G63.G2.2G$85.2G$30.2G54.G$30.2G2.2G50.G.G$34.G.G
48.2G.3G$36.G54.G$36.2G25.2G20.2G.3G$64.G20.2G.G$64.G.G14.2G$65.2G14.
G.G$82.2G2$32.2G45.2G$32.2G45.G.G$81.G28.2G$48.G10.2G20.2G28.G$46.3G
10.2G49.G$45.G64.2G$45.2G$141.G$81.G58.G.G$81.G58.G.G$70.2G9.3G.2G54.
G$16.2G52.2G11.G.2G26.2G15.G$15.G2.G2.2G90.2G15.G$16.2G2.G.G107.3G$
18.2G112.G$18.G$15.2G.G$15.G2.G.2G44.2G$17.2G.2G44.2G3$108.2G$108.2G$
119.2G$120.G$14.2G3.2G73.2G21.3G$14.2G3.2G74.G21.G$92.3G$92.G$104.2G$
42.2G60.G$42.G.G60.3G$2G42.G62.G$2G42.2G6$33.2G$33.2G2$10.2GC$11.G8.
2G$11.3G6.G$21.3G$23.G!