ConwayLife.com

Forums for Conway's Game of Life
https://conwaylife.com/forums/

lifeviewer bug

https://conwaylife.com/forums/viewtopic.php?t=4742

Page 29 of 31

Re: lifeviewer bug

Posted: September 9th, 2025, 7:29 am
by rowett
muzik wrote: September 9th, 2025, 4:21 am It kept going until I woke up this morning, where I noticed that it had finally stopped identifying at generation 217232803617, without giving any actual results.

No idea if I've hit some sort of upper time limit in Identify, or something else ended up happening.
Analysis:
  • There's no time limit.
  • The generation number 217232803617 is well within range (Javascript integers can go up to 2^53 and that's around 2^38).
  • The buffer was 80% full at the last snapshot.
  • Identify stops without results if a) the buffer gets full or b) the pattern dies
From that I suspect the buffer got full. In which case it will have displayed a notification which likely was gone by the time you checked.

Side note: LifeViewer doesn't currently load the Identify snapshot. That's on the backlog.

Re: lifeviewer bug

Posted: September 11th, 2025, 12:45 pm
by R2INT
For some reason, when I reset a pattern on my laptop, it does not let me interact with LifeViewer until after the message saying 'Reset' closes.

Re: lifeviewer bug

Posted: September 11th, 2025, 6:16 pm
by b-engine
This may be a graphical issue.
Screenshot_20250912-061358.png
Screenshot_20250912-061358.png (42.12 KiB) Viewed 5840 times
Maybe draw the grid lines behind cells?

Re: LifeViewer bugs

Posted: September 13th, 2025, 11:27 am
by qqd
The versions of LifeHistory that are implemented in LifeViewer and Golly are different, and according to the article on the rule, only Golly implements it correctly. Run the following pattern (turn off any blue light filters beforehand):

Code: Select all

x = 15, y = 54, rule = LifeHistory
C2.C6.D2.D$4.C9.D$C3.C5.D3.D$.4C6.4D7$E2.E6.D2.D$4.E9.D$E3.E5.D3.D$.
4E6.4D7$C2.C6.C2.C$4.C9.C$C3.C5.C3.C$.4C6.4C7$E2.E6.C2.C$4.E9.C$E3.E
5.C3.C$.4E6.4C7$C2.C6.E2.E$4.C9.E$C3.C5.E3.E$.4C6.4E7$E2.E6.E2.E$4.E
9.E$E3.E5.E3.E$.4E6.4E!
[[ STOP 20 ]]
According to the article, state 5 (yellow) is treated as a live cell and will convert to state 4 (red) in the next generation. More importantly, it has no birth conditions and can therefore never appear in the evolution of a pattern whiteout any state 5 cells.

However, in the demo above, the last two rows evolve an LWSS made completely of state 5 cells, even though they were all in state 4 in generation 10. Apparently state 4 cells which were state 5 cells at a previous point in time evolve into state 5 cells rather than state 3 (white) cells as they should as demonstrated in the other rows. Fortunately, Golly does not experience this issue, but it makes me wonder how this bug could possibly happen with only 1 red state.

Re: LifeViewer bugs

Posted: September 13th, 2025, 12:03 pm
by muzik
qqd wrote: September 13th, 2025, 11:27 amHowever, in the demo above, the last two rows evolve an LWSS made completely of state 5 cells, even though they were all in state 4 in generation 10. Apparently state 4 cells which were state 5 cells at a previous point in time evolve into state 5 cells rather than state 3 (white) cells as they should as demonstrated in the other rows. Fortunately, Golly does not experience this issue, but it makes me wonder how this bug could possibly happen with only 1 red state.
This has been pointed out a handful of times es before, but the current behaviour is intentional: viewtopic.php?p=195406#p195406

Re: lifeviewer bug

Posted: September 14th, 2025, 10:15 pm
by Citation needed
Newly uploaded rules don't seem to work on "lazyslug.com".

Code: Select all

x = 0, y = 0, rule = ColourisedLife47s
GpDoQpNpJ Nb!
[[ ICONS ]]

Re: lifeviewer bug

Posted: September 15th, 2025, 1:18 am
by rowett
Citation needed wrote: September 14th, 2025, 10:15 pm Newly uploaded rules don't seem to work on "lazyslug.com".
Fixed, thanks!

Re: lifeviewer bug

Posted: September 16th, 2025, 4:22 am
by Citation needed
rowett wrote: September 15th, 2025, 1:18 am
Citation needed wrote: September 14th, 2025, 10:15 pm Newly uploaded rules don't seem to work on "lazyslug.com".
Fixed, thanks!
Unfortunately, OCA permalinks stopped working.

Re: lifeviewer bug

Posted: September 16th, 2025, 5:58 am
by rowett
Citation needed wrote: September 16th, 2025, 4:22 am Unfortunately, OCA permalinks stopped working.
They work fine:
http://lazyslug.com/?rle=3o$o$bo!&rule=b2-as12


See Edit 2 here.
Basically rule, name and viewerconfig are ignored if rle is not present.

Re: lifeviewer bug

Posted: September 26th, 2025, 3:35 am
by b-engine
The pop-up viewer flashes when scrolling.


On Firefox 143.0 (Build #2016113383), 62b30a28e7a4110ced2411d4b81bcb1256839616

GV: 143.0-20250908174027
AS: 143.0
OS: Android 9

Re: lifeviewer bug

Posted: October 4th, 2025, 6:02 pm
by jayden678910
Rule table just doesn’t work.

Code: Select all

x = 1, y = 1, rule = thingy
!
@RULE thingy
@TABLE
n_states:4
neighborhood:Moore
symmetries:rotate4reflect
var a = {0,1}
var b = a
var c = a
var d = a
var e = a
var f = a
var g = a
var h = a
var a1 = {0,3}
var b1 = a1
var c1 = a1
var d1 = a1
var e1 = a1
var f1 = a1
var g1 = a1
var h1 = a1
var a0 = {0,1,2,3)
var b0 = a0
var c0 = a0
var d0 = a0
var e0 = a0
var f0 = a0
var g0 = a0
var h0 = a0
0, 2,b,c,d,e,f,g,h, 1
# Death
1,a0,b0,c0,d0,e0,f0,g0,h0,0
@COLORS
1 255 0 0
2 255 128 0
3 180 0 255

Re: lifeviewer bug

Posted: October 4th, 2025, 10:08 pm
by rowett
jayden678910 wrote: October 4th, 2025, 6:02 pm Rule table just doesn’t work.

Code: Select all

var a0 = {0,1,2,3)
The quoted line has ) rather than }.

Re: lifeviewer bug

Posted: October 5th, 2025, 8:14 am
by jayden678910
rowett wrote: October 4th, 2025, 10:08 pm
jayden678910 wrote: October 4th, 2025, 6:02 pm Rule table just doesn’t work.

Code: Select all

var a0 = {0,1,2,3)
The quoted line has ) rather than }.
thanks

Re: lifeviewer bug

Posted: October 15th, 2025, 6:04 pm
by b-engine
Maximizing the viewer causes buttons to overlap:
Screenshot_20251016-060121.png
Screenshot_20251016-060121.png (22.37 KiB) Viewed 5487 times

Re: lifeviewer bug

Posted: October 22nd, 2025, 6:34 am
by b-engine
Now maximizing the viewer causes it to expand beyond bounds and render the maximize button as well as half of the UI inaccessible:
Screenshot_20251022-183307.png
Screenshot_20251022-183307.png (11.25 KiB) Viewed 5441 times

Re: lifeviewer bug

Posted: October 28th, 2025, 12:13 am
by WRAPboi
I pressed the reset button, and my patterns disappeared.
this happened with any ships that had been copied on to lifeviewer, i deleted it, reconstructed it myself, pressed play, reset, and it all disappeared. :( :shock: :!: :?: :|

Re: lifeviewer bug

Posted: October 28th, 2025, 8:17 am
by b-engine
WRAPboi wrote: October 28th, 2025, 12:13 am I pressed the reset button, and my patterns disappeared.
If you had drawn patterns after the 0th tick, resetting would remove the patterns. That's intentional [Citation needed].

Re: lifeviewer bug

Posted: October 30th, 2025, 7:20 am
by muzik
rowett wrote: March 7th, 2024, 4:56 amThe whole point of Theme Golly is to duplicate what Golly does.
This differs significantly:

Code: Select all

x = 19, y = 34, rule = LifeHistory
F22$9.D$7.D.D6.2E$7.3D5.2E$7.D8.2E$17.E7$18.F!
[[ THEME Golly
ZOOM 11 GPS 20 AUTOSTART
T 0 PAUSE 1
T 20 PAUSE 1
GPS 40 LOOP 200 ]]
Color definitions needed to make this accurate:

Code: Select all

x = 19, y = 34, rule = LifeHistory
F22$9.D$7.D.D6.2E$7.3D5.2E$7.D8.2E$17.E7$18.F!
[[ THEME Golly COLOR MARK1 192 255 192 COLOR ALIVE 0 255 0 COLOR ALIVERAMP 0 255 0 COLOR DEAD 0 0 128 COLOR DEADRAMP 0 0 128
ZOOM 11 GPS 20 AUTOSTART
T 0 PAUSE 1
T 20 PAUSE 1
GPS 40 LOOP 200 ]]

Re: lifeviewer bug

Posted: November 1st, 2025, 12:44 am
by rowett
muzik wrote: October 30th, 2025, 7:20 am This differs significantly:
Fixed in build 1343.

Re: lifeviewer bug

Posted: November 3rd, 2025, 11:41 am
by LWSSONHWSSONLWSS
This is from the trolls thread. Sometimes when you rewind the pattern, it follows a completely different sequence.
That is the problem with PASTE commands.

Code: Select all

 x = 10, y = 10, rule = B3/S23
5bobo$b2o2bobo$3o3bobo2$2bo6bo2$6b2obo$6b2o$6bo2bo$7b2o!




























































































































































































































































[[ PASTEMODE XOR PASTET EVERY 7 1 PASTEDELTA -1 0 PASTE oobb$bboo$oboo$boob! 0 1 ]]
[[ STARTFROM 1 ]]
[[ SHADER BASIC ]]

Re: lifeviewer bug

Posted: November 3rd, 2025, 4:09 pm
by Chris857
LWSSONHWSSONLWSS wrote: November 3rd, 2025, 11:41 am This is from the trolls thread. Sometimes when you rewind the pattern, it follows a completely different sequence.
That is the problem with PASTE commands.
I see issues when running it out somewhere 50-100 gen and rewinding.

Re: lifeviewer bug

Posted: November 3rd, 2025, 6:13 pm
by rowett
LWSSONHWSSONLWSS wrote: November 3rd, 2025, 11:41 am This is from the trolls thread. Sometimes when you rewind the pattern, it follows a completely different sequence.
That is the problem with PASTE commands.
Fixed in build 1345.

Re: lifeviewer bug

Posted: November 4th, 2025, 10:23 am
by b-engine
b-engine wrote: October 22nd, 2025, 6:34 am Now maximizing the viewer causes it to expand beyond bounds and render the maximize button as well as half of the UI inaccessible: Screenshot_20251022-183307.png
Bringing this back up because it's annoying me especially since I could easily mistap the maximize button.

Re: lifeviewer bug

Posted: November 8th, 2025, 2:40 pm
by LuveelVoom
New versions of lifeviewer seem to have a bug where the selection box's first corner is sticky: it will always be at the same spot even when you try to recreate the selection box somewhere else using the mouse or a touchscreen, unless you remove the entire selection box using the "remove selection" button. This is extremely annoying.

Re: lifeviewer bug

Posted: November 8th, 2025, 5:35 pm
by rowett
LuveelVoom wrote: November 8th, 2025, 2:40 pm New versions of lifeviewer seem to have a bug where the selection box's first corner is sticky: it will always be at the same spot even when you try to recreate the selection box somewhere else using the mouse or a touchscreen
I can't reproduce this. It works fine for me on Windows desktop (Chrome) and iPhone (Safari). Please let me know which device and browser you are using?
All times are UTC-04:00
Page 29 of 31

Powered by phpBB® Forum Software © phpBB Limited