Page 1 of 4

Golly 2.9b1 and the overlay

PostPosted: November 14th, 2016, 6:55 am
by Andrew
The 1st beta version of Golly 2.9 is now available:

EDIT: Download links have been removed now that version 3.0b1 has been released.

This version has a new feature called the overlay -- a rectangular area of pixels that can be displayed above the current layer. Lua scripts have total control over what the overlay looks like and how the user interacts with it. Here's a simple example:

local g = golly()
local ov = g.overlay
-- create a 400x300 overlay in the middle of the layer
ov("create 400 300")
ov("position middle")
-- fill it with a translucent blue background
ov("rgba 0 0 255 128")
ov("fill")
-- draw some white text in the top left corner
ov("rgba 255 255 255 255")
ov("text msg Click to exit.")
ov("blend 1")
ov("paste 10 10 msg")
g.update()
-- wait for user to click in the overlay
while true do
    local event = g.getevent()
    if event:find("^oclick") then break end
end
ov("delete")   -- remove the overlay

All the overlay commands are documented in Help > Overlay.

The Scripts/Lua folder has a number of scripts that illustrate how the overlay can be used to extend Golly in all sorts of ways. A good place to start would be overlay-demo.lua. It contains examples of most of the overlay commands. More complex scripts are also supplied:

breakout.lua -- A script by Chris Rowett that shows how the overlay can be used to create an interactive game.

hexgrid.lua -- Creates a true hexagonal grid for rules that use a hexagonal neighborhood.

lifeviewer.lua -- Another script by Chris that displays patterns like LifeViewer, with history, smooth non-integer scaling, and rotation. Still a work in progress.

pop-plot.lua -- This script has been updated to use the overlay to display nicer looking plots that can be saved in a PNG file.

Apart from the overlay there are a number of other important changes in this version, so make sure you read Help > Changes.

Please note that we're still working on the overlay API. There's likely to be a number of beta releases before the final version is ready for sourceforge (probably well into next year).

Feedback, suggestions, bug reports are all welcome!

Re: Golly 2.9b1 and the overlay

PostPosted: November 14th, 2016, 11:48 am
by Scorbie
Oh my god, the overlay is really crazy!!! Thanks a lot for the work!!!
Eye candy for the hex grid was one of the reasons I really wanted to write my own CA simulator, but not it's here actually, I'd rethink about that....

Anyway this is really awesome; people should try this! Oh my god!!

Re: Golly 2.9b1 and the overlay

PostPosted: November 24th, 2016, 10:22 pm
by Saka
this new golly cant find the python library. Where should the dll be? it's not in the Python27 -> DLLs folder

Re: Golly 2.9b1 and the overlay

PostPosted: November 25th, 2016, 1:05 am
by Andrew
Saka wrote:this new golly cant find the python library. Where should the dll be? it's not in the Python27 -> DLLs folder

2.9b1 looks for the Python library in exactly the same way as 2.8 (and 2.7). Did you download the 32-bit or 64-bit build? Is that the same build as your previous Golly version? Does it match your Python bit size? What is the python_lib setting in your GollyPrefs file? That file should be in the same folder as Golly.exe or in one of these locations:

On Windows XP: C:\Documents and Settings\username\Application Data\Golly\
On Windows 7+: C:\Users\username\AppData\Roaming\Golly\

Re: Golly 2.9b1 and the overlay

PostPosted: November 25th, 2016, 1:34 am
by Saka
Before this I was using an ancient version of golly, Golly 2.6

Re: Golly 2.9b1 and the overlay

PostPosted: January 9th, 2017, 12:00 pm
by BlinkerSpawn
Do later betas exist yet or have I been seeing things?

Re: Golly 2.9b1 and the overlay

PostPosted: January 10th, 2017, 12:45 am
by Andrew
BlinkerSpawn wrote:Do later betas exist yet ... ?

No.

Re: Golly 2.9b1 and the overlay

PostPosted: January 14th, 2017, 10:18 pm
by GUYTU6J
Using lua technology,is it posible to make SeedOfDestruction game into Golly ?

Re: Golly 2.9b1 and the overlay

PostPosted: February 25th, 2017, 11:00 pm
by Andrew
GUYTU6J wrote:Using lua technology,is it posible to make SeedOfDestruction game into Golly ?

I've never been able to get SoD working on my Mac so not 100% sure, but looking at the docs and the screen shots it should be possible to create a nice-looking overlay script that does the same job (but possibly not as fast). Looks like a lot of work though.

Re: Golly 2.9b1 and the overlay

PostPosted: March 16th, 2017, 8:45 pm
by Andrew
I've edited the link for the Mac build in the 1st post to point to a signed dmg file. This should avoid various problems caused by an annoying security "feature" Apple recently added in Mac OS 10.12. The dmg file can also be unpacked on older Mac systems (right back to 10.6).

Re: Golly 2.9b1 and the overlay

PostPosted: March 16th, 2017, 9:17 pm
by drc
Can MAP be added in 2.9?

Re: Golly 2.9b1 and the overlay

PostPosted: March 17th, 2017, 12:25 am
by rowett
drc wrote:Can MAP be added in 2.9?

Yes it's already implemented. It will be available when 2.9b2 is released (or earlier if you wish to build from the source).

Re: Golly 2.9b1 and the overlay

PostPosted: March 17th, 2017, 11:12 am
by blah
Since we're asking for features, what about extended neighbourhoods in RuleLoader?

Re: Golly 2.9b1 and the overlay

PostPosted: March 21st, 2017, 10:34 pm
by wildmyron
Been very happily using Golly 2.9b1 on Windows x64 - thank you all for your great work.

I've noticed a minor issue with determining the canonical rulestring for isotropic rules:

1) Change the rule to B3-acei/S
*) Golly displays the rule as B3jknqry/S

2) Open the Set Rule dialogue and close it again
*) Golly displays the rule as B3-acei/S

A number of other actions also cause Golly to flip between different representations. In particular there is a strange interaction with the Undo/Redo mechanism

The same occurs for any similar string with 3 or 5 as the totalistic number and also for rulestrings such as B2-ac/S , B6-ac/S , and B4-aceij/S .

Re: Golly 2.9b1 and the overlay

PostPosted: March 21st, 2017, 11:27 pm
by rowett
wildmyron wrote:I've noticed a minor issue with determining the canonical rulestring for isotropic rules

This has been fixed. Thanks for reporting!

Re: Golly 2.9b1 and the overlay

PostPosted: March 22nd, 2017, 12:02 am
by wildmyron
rowett wrote:
wildmyron wrote:I've noticed a minor issue with determining the canonical rulestring for isotropic rules

This has been fixed. Thanks for reporting!

That was quick. Thank you!

Re: Golly 2.9b1 and the overlay

PostPosted: March 24th, 2017, 6:37 am
by Saka
Golly doesnt simulate b0 generations properly, I think.
Example case:
1. Change rulestring to 012/03/23
2. Soup-search until a "rocket" appears
3. Copy the rocket (in an even rule)
4. Open a new layer
5. Paste the copied rocket in
6. Run
As you can see, even the same pattern can behave differently in different layers

Re: Golly 2.9b1 and the overlay

PostPosted: March 24th, 2017, 7:28 am
by dvgrn
Saka wrote:As you can see, even the same pattern can behave differently in different layers

Or even in the same layer, actually:

x = 2, y = 2, rule = 012/03/23
2A$A!

Running this in Golly for 24 ticks produces a diagonal lightspeed rocket. But try

  • running a few dozen ticks with the spacebar;
  • hitting "z" once or twice at any point;
  • holding down the spacebar again.
The rocket mysteriously dies instead of continuing to travel. You have to undo all the way to T=0 before the original behavior is restored.

Copying and pasting to a new layer also seems to cause the ON cells to die off for some reason, which I think is the behavior Saka is describing.

The behavior in LifeViewer Build 222 (or Build 226) seems to be similar, leading to an early death for patterns in this rule.

Re: Golly 2.9b1 and the overlay

PostPosted: March 24th, 2017, 9:49 am
by blah
Here's the pre-block actually simulated for 500 generations in that rule in a bounded grid, using my own software. The more states the rule has, the slower patterns have to go, given that they can only expand 2 (?) cells per (amount of states).
x = 130, y = 130, rule = 012/03/23
130Q$130Q$130Q$130Q$130Q$130Q$130Q$130Q$130Q$130Q$130Q$130Q$130Q$130Q$130Q$130Q$130Q$130Q$130Q$130Q$130Q$130Q$130Q$130Q$130Q$130Q$130Q$130Q$130Q$130Q$130Q$130Q$130Q$130Q$130Q$130Q$130Q$130Q$130Q$130Q$130Q$130Q$130Q$62QO3A.10QA3.2AOA45Q$53QACA2Q3A3O3J3.QA2.A.2JAMIL2O2PA43Q$53Q2HCA3DPEOJMJ2AQ.APA3.2JMIMA2N.2P43Q$53Q2HAE2D.2EAGJAIK2Q.P2.2J2LA2MA2NA.PA42Q$51QA2PJK.DA.AF3G.IKAQA3.2J2LK3I2N2.A.42Q$50QA2P2AJD2.2DFA4IA3LK2AJAJAKIJ2IN.N2.42Q$50Q2PA2OM2DACEA3I.2H.LA2K3JA2.AGE2IAKA.42Q$48QA2P2NAOFDBCEC4IA2HJ2.KLKJA3.G3E.K3.42Q$47QA2P2N.N2FDBAUAU3IA.HA3.AVFGHA2DA.3AKA.42Q$47Q2PAN.N.ADA2V3U3V2.2V3.V2DFGF2D3.2JM2.42Q$44QA2HPAOAN.2A2C2VAUV.2V2.V.A.A3DB3FA2.2KJ2M43Q$44QC2HJA2OF2A2BVA2VSAV.A2.A2.UTDA3BF.2UA2KAM44Q$44QACAKJM2FDC2BU2V2SUV.A5.2TAI3B.ADUJ.AM45Q$45QAE.4DACVU.U3SUAVA4.A.TA2ICB.ADA2JNM45Q$45Q3D.D2B2VAVUA2.2S5.2I.UAIG2A3.BF2N2A45Q$44QA2DA.ACA4VS3.A5.AJ3AU2G2E.3FAGKN2L44Q$44QAD2.DCE2UAV2S2.3A.2AHI.J2D.EFE.A2FHI2GAML44Q$44QAPEADECA2U4S2A2BTV.HAIA2D3EAB2JAHAIG2MA44Q$44QO2E2FAI2UVA2US.A2BV3A3.JDAQVBAKJ2HIGM2NA44Q$43Q3OAGA3IV.2VA3.TVMP4Q.AJQTAV2DAJHAKAPNA44Q$43QAOJ2G4I2V2.V2.AVAPA4QVLQART.DEBA3KA.O45Q$43QAJMJG4I2V3A2.A.A2Q2SQALQ2LKA2.3ALK2I2OA44Q$43QA2JA.I.2A7.2HA2QSAT2LNA.K2S2.JLIA2I2O.44Q$43Q.JA3I2H7.AIA.3Q3TA2P.K2SA2.A4I2FA44Q$44Q.A2KA3H2VA2.IJ.I.2QAL2T2PA2K2U.D.L2.IA2FA44Q$44Q.2QAL.JAV4.IAJA2.V2LAPQP2K2A4D3A2HO2M44Q$44Q2.2Q2L3.A2.A.A2DJALQN3PAOA.DED.B.M.2HA2M44Q$45QA.ALA4.UT.UA3DJQLAPAKO.ABAD2F2A.2FNAN45Q$44QA2P.3K2.A3TAU.EAQAL2.2K5A3FA2D2F3N45Q$44Q.A2.AKLAV2D2AIG2EQTR4KA.BA2.2ACA3DA.3NA43Q$44Q4.AJKV2DA2I2GFEVATA2SUAD2A.A2C.C3.DCA2N.43Q$44QA.5JF2D2BIA2EABV2.2SUDEDFAC2G2A2.A.2CLA.43Q$44Q2.2JAJAGF3BCAE.BA2D2.A.2D2FAC2G.GAL.NALJA44Q$44Q2J2LJA.HGF3B2.AJKDEA2.2D.2FC.A.G.A2L.NLA2O.42Q$43QA2J2LA2.A3F3.2F2JABAJ2.DB3ACAG4.4LA.OA42Q$43Q.AMA2K2.2DF.2A.2FAHJ2ALALA.A2D2.2A4.LA.L2N.42Q$43Q.MIM2IAG2DAU2DBF4HKL2I.AM.2D2.2L2.A.2A2LNJA42Q$43Q.I2MIJGEA2.2UAFAIAIA2KAI.A.2FD.A.2L3.2BALAJA42Q$43QAL2A2I2E3.A2JN2GIG2K4I2H2FAD.N.2LA3BA2O2M42Q$43QAO3N2IEA.2K.JNK2GM2A3IA2H2N.2CANL2A2BNMPO2M42Q$43Q2O4NI.AJ2KANANAMNP.2O2FO2A2NAC3L.L2AMAJA2M42Q$43QAP.A2.AKA2JA2MAL2M2N3O2F2M4NLJ2A3LOPJHJPA42Q$44Q2P2.NK.K3M3Q2L3AQA.2A2M2Q2N2AO.2NA2OAJ2P43Q$44QA2PA.A.A.M18QA2.Q2ON2J3M2PA43Q$46QA6.23Q.A.2A3MA45Q$130Q$130Q$130Q$130Q$130Q$130Q$130Q$130Q$130Q$130Q$130Q$130Q$130Q$130Q$130Q$130Q$130Q$130Q$130Q$130Q$130Q$130Q$130Q$130Q$130Q$130Q$130Q$130Q$130Q$130Q$130Q$130Q$130Q$130Q$130Q$130Q$130Q$130Q$130Q$130Q$130Q$130Q!

This can probably be simulated with alternating rules, given how the state of the outside is always (generations%number of states), although I'm not sure about the details. Needless to say, Golly should at least display an error message.

Re: Golly 2.9b1 and the overlay

PostPosted: March 24th, 2017, 10:19 am
by rowett
dvgrn wrote:The behavior in LifeViewer Build 222 (or Build 226) seems to be similar, leading to an early death for patterns in this rule.

I think LifeViewer is correct. The pattern should not create rockets given the three cell starting configuration. I'll see if I can mend Golly.

Re: Golly 2.9b1 and the overlay

PostPosted: March 24th, 2017, 10:31 am
by blah
rowett wrote:I think LifeViewer is correct.

It isn't. Golly and LifeViewer both simulate the rule innacurately. As I posted earlier, the pattern slowly explodes. It does not move, nor does it die. How do LifeViewer and Golly even simulate this?

Edit: And on the subject of simulating it for infinite grids, maybe the different states could represent a difference from the outside state (t%states). For example, at t=20 in a 17-state rule, state 20%17=3 is treated as the live state, and state 2 would be the dead cells, and they would count up modularly. A 'still life' would constantly oscillate. At some point it would become state 0, but that wouldn't be a contradiction; it would have to have neighbours, othewise it would be s8 anyway.

Did I explain that well enough? I think that could work.

Re: Golly 2.9b1 and the overlay

PostPosted: March 24th, 2017, 11:19 am
by dvgrn
blah wrote:
rowett wrote:I think LifeViewer is correct.

It isn't. Golly and LifeViewer both simulate the rule [inaccurately]. As I posted earlier, the pattern slowly explodes. It does not move, nor does it die.

I can confirm this in Golly. Here's the equivalent 24-state rule table:

@RULE Generations-with-B03-S012-23states-plus-NEWOFF

Rule table simulation of B0 Generations rule, for test purposes
Start by building a block of state-23 cells to simulate OFF cells
(because using actual OFF cells has bad effects in Golly,
   rapidly increasing population even with a bounded grid)

@TABLE

n_states:24
neighborhood:Moore
symmetries:permute

var a={23,0,2,3,4,5,6,7,8,9,10,11,12,13,14,15,16,17,18,19,20,21,22}
var b={23,0,2,3,4,5,6,7,8,9,10,11,12,13,14,15,16,17,18,19,20,21,22}
var c={23,0,2,3,4,5,6,7,8,9,10,11,12,13,14,15,16,17,18,19,20,21,22}
var d={23,0,2,3,4,5,6,7,8,9,10,11,12,13,14,15,16,17,18,19,20,21,22}
var e={23,0,2,3,4,5,6,7,8,9,10,11,12,13,14,15,16,17,18,19,20,21,22}
var f={23,0,2,3,4,5,6,7,8,9,10,11,12,13,14,15,16,17,18,19,20,21,22}
var g={23,0,2,3,4,5,6,7,8,9,10,11,12,13,14,15,16,17,18,19,20,21,22}
var h={23,0,2,3,4,5,6,7,8,9,10,11,12,13,14,15,16,17,18,19,20,21,22}

var m={23,0,1,2,3,4,5,6,7,8,9,10,11,12,13,14,15,16,17,18,19,20,21,22}
var n={23,0,1,2,3,4,5,6,7,8,9,10,11,12,13,14,15,16,17,18,19,20,21,22}
var o={23,0,1,2,3,4,5,6,7,8,9,10,11,12,13,14,15,16,17,18,19,20,21,22}
var p={23,0,1,2,3,4,5,6,7,8,9,10,11,12,13,14,15,16,17,18,19,20,21,22}
var q={23,0,1,2,3,4,5,6,7,8,9,10,11,12,13,14,15,16,17,18,19,20,21,22}
var r={23,0,1,2,3,4,5,6,7,8,9,10,11,12,13,14,15,16,17,18,19,20,21,22}
var s={23,0,1,2,3,4,5,6,7,8,9,10,11,12,13,14,15,16,17,18,19,20,21,22}
var t={23,0,1,2,3,4,5,6,7,8,9,10,11,12,13,14,15,16,17,18,19,20,21,22}

# NEWOFF cell turns ON if 0 or 3 ON neighbors
23,a,b,c,d,e,f,g,h,1
23,1,1,1,a,b,c,d,e,1

# ON cells remain ON if 0, 1, or 2 neighbors
1,a,b,c,d,e,f,g,h,1
1,1,a,b,c,d,e,f,g,1
1,1,1,a,b,c,d,e,f,1

# otherwise ON cells become slowly-dying NEWOFF cells
1,m,n,o,p,q,r,s,t,2
2,m,n,o,p,q,r,s,t,3
3,m,n,o,p,q,r,s,t,4
4,m,n,o,p,q,r,s,t,5
5,m,n,o,p,q,r,s,t,6
6,m,n,o,p,q,r,s,t,7
7,m,n,o,p,q,r,s,t,8
8,m,n,o,p,q,r,s,t,9
9,m,n,o,p,q,r,s,t,10
10,m,n,o,p,q,r,s,t,11
11,m,n,o,p,q,r,s,t,12
12,m,n,o,p,q,r,s,t,13
13,m,n,o,p,q,r,s,t,14
14,m,n,o,p,q,r,s,t,15
15,m,n,o,p,q,r,s,t,16
16,m,n,o,p,q,r,s,t,17
17,m,n,o,p,q,r,s,t,18
18,m,n,o,p,q,r,s,t,19
19,m,n,o,p,q,r,s,t,20
20,m,n,o,p,q,r,s,t,21
21,m,n,o,p,q,r,s,t,22
22,m,n,o,p,q,r,s,t,23

@COLORS

23   96   96   96

And here's the pattern to run in that rule, with a simulated 100x100 bounded grid:

x = 100, y = 100, rule = Generations-with-B03-S012-23states-plus-NEWOFF
100W$100W$100W$100W$100W$100W$100W$100W$100W$100W$100W$100W$100W$100W
$100W$100W$100W$100W$100W$100W$100W$100W$100W$100W$100W$100W$100W$
100W$100W$100W$100W$100W$100W$100W$100W$100W$100W$100W$100W$100W$100W
$100W$100W$100W$100W$100W$100W$100W$100W$49W2A49W$49WA50W$100W$100W$
100W$100W$100W$100W$100W$100W$100W$100W$100W$100W$100W$100W$100W$100W
$100W$100W$100W$100W$100W$100W$100W$100W$100W$100W$100W$100W$100W$
100W$100W$100W$100W$100W$100W$100W$100W$100W$100W$100W$100W$100W$100W
$100W$100W$100W$100W$100W$100W!

After 500 ticks this produces the exact same pattern that blah posted, though it's hard to tell in LifeViewer due to the different colors. In Golly the colors are almost identical:

Code: Select all
x = 50, y = 49, rule = Generations-with-B03-S012-23states-plus-NEWOFF
50Q$50Q$21QO3AW10QA3W2AOA6Q$12QACA2Q3A3O3J3WQA2WAW2JAMIL2O2PA4Q$12Q2H
CA3DPEOJMJ2AQWAPA3W2JMIMA2NW2P4Q$12Q2HAE2DW2EAGJAIK2QWP2W2J2LA2MA2NAW
PA3Q$10QA2PJKWDAWAF3GWIKAQA3W2J2LK3I2N2WAW3Q$9QA2P2AJD2W2DFA4IA3LK2AJ
AJAKIJ2INWN2W3Q$9Q2PA2OM2DACEA3IW2HWLA2K3JA2WAGE2IAKAW3Q$7QA2P2NAOFDB
CEC4IA2HJ2WKLKJA3WG3EWK3W3Q$6QA2P2NWN2FDBAUAU3IAWHA3WAVFGHA2DAW3AKAW
3Q$6Q2PANWNWADA2V3U3V2W2V3WV2DFGF2D3W2JM2W3Q$3QA2HPAOANW2A2C2VAUVW2V
2WVWAWA3DB3FA2W2KJ2M4Q$3QC2HJA2OF2A2BVA2VSAVWA2WA2WUTDA3BFW2UA2KAM5Q$
3QACAKJM2FDC2BU2V2SUVWA5W2TAI3BWADUJWAM6Q$4QAEW4DACVUWU3SUAVA4WAWTA2I
CBWADA2JNM6Q$4Q3DWD2B2VAVUA2W2S5W2IWUAIG2A3WBF2N2A6Q$3QA2DAWACA4VS3WA
5WAJ3AU2G2EW3FAGKN2L5Q$3QAD2WDCE2UAV2S2W3AW2AHIWJ2DWEFEWA2FHI2GAML5Q$
3QAPEADECA2U4S2A2BTVWHAIA2D3EAB2JAHAIG2MA5Q$3QO2E2FAI2UVA2USWA2BV3A3W
JDAQVBAKJ2HIGM2NA5Q$2Q3OAGA3IVW2VA3WTVMP4QWAJQTAV2DAJHAKAPNA5Q$2QAOJ
2G4I2V2WV2WAVAPA4QVLQARTWDEBA3KAWO6Q$2QAJMJG4I2V3A2WAWA2Q2SQALQ2LKA2W
3ALK2I2OA5Q$2QA2JAWIW2A7W2HA2QSAT2LNAWK2S2WJLIA2I2OW5Q$2QWJA3I2H7WAIA
W3Q3TA2PWK2SA2WA4I2FA5Q$3QWA2KA3H2VA2WIJWIW2QAL2T2PA2K2UWDWL2WIA2FA5Q
$3QW2QALWJAV4WIAJA2WV2LAPQP2K2A4D3A2HO2M5Q$3Q2W2Q2L3WA2WAWA2DJALQN3PA
OAWDEDWBWMW2HA2M5Q$4QAWALA4WUTWUA3DJQLAPAKOWABAD2F2AW2FNAN6Q$3QA2PW3K
2WA3TAUWEAQAL2W2K5A3FA2D2F3N6Q$3QWA2WAKLAV2D2AIG2EQTR4KAWBA2W2ACA3DAW
3NA4Q$3Q4WAJKV2DA2I2GFEVATA2SUAD2AWA2CWC3WDCA2NW4Q$3QAW5JF2D2BIA2EABV
2W2SUDEDFAC2G2A2WAW2CLAW4Q$3Q2W2JAJAGF3BCAEWBA2D2WAW2D2FAC2GWGALWNALJ
A5Q$3Q2J2LJAWHGF3B2WAJKDEA2W2DW2FCWAWGWA2LWNLA2OW3Q$2QA2J2LA2WA3F3W2F
2JABAJ2WDB3ACAG4W4LAWOA3Q$2QWAMA2K2W2DFW2AW2FAHJ2ALALAWA2D2W2A4WLAWL
2NW3Q$2QWMIM2IAG2DAU2DBF4HKL2IWAMW2D2W2L2WAW2A2LNJA3Q$2QWI2MIJGEA2W2U
AFAIAIA2KAIWAW2FDWAW2L3W2BALAJA3Q$2QAL2A2I2E3WA2JN2GIG2K4I2H2FADWNW2L
A3BA2O2M3Q$2QAO3N2IEAW2KWJNK2GM2A3IA2H2NW2CANL2A2BNMPO2M3Q$2Q2O4NIWAJ
2KANANAMNPW2O2FO2A2NAC3LWL2AMAJA2M3Q$2QAPWA2WAKA2JA2MAL2M2N3O2F2M4NLJ
2A3LOPJHJPA3Q$3Q2P2WNKWK3M3Q2L3AQAW2A2M2Q2N2AOW2NA2OAJ2P4Q$3QA2PAWAWA
WM18QA2WQ2ON2J3M2PA4Q$5QA6W23QWAW2A3MA6Q$50Q$50Q!

Re: Golly 2.9b1 and the overlay

PostPosted: March 24th, 2017, 11:30 am
by rowett
blah wrote:How do LifeViewer and Golly even simulate this?

They use alternate algorithms for even and odd generations the same as for non-Generations rules.

Re: Golly 2.9b1 and the overlay

PostPosted: March 24th, 2017, 11:35 am
by rowett
So for clarity I'm working on two separate issues:
  1. LifeViewer and Golly behave differently on the same pattern for 012/03/23. They shouldn't since they use exactly the same algorithm
  2. I'm investigating the algorithm they both use to see why it doesn't produce the output blah and Dave generated

Re: Golly 2.9b1 and the overlay

PostPosted: March 24th, 2017, 11:42 am
by blah
Well, if they use two different rules for even and odd generations, that's the problem. These rules don't alternate every 2 generations, they oscillate through n generations (n being the number of states). The algorithm I came up with before should work.