Golly 3.0b2

For general discussion about Conway's Game of Life.
User avatar
muzik
Posts: 5614
Joined: January 28th, 2016, 2:47 pm
Location: Scotland

Re: Golly 3.0b2

Post by muzik » August 9th, 2017, 9:43 pm

How about non-totalistic versions of LtL rules, at least with range 2?

User avatar
praosylen
Posts: 2443
Joined: September 13th, 2014, 5:36 pm
Location: Pembina University, Home of the Gliders
Contact:

Re: Golly 3.0b2

Post by praosylen » August 9th, 2017, 10:18 pm

muzik wrote:How about non-totalistic versions of LtL rules, at least with range 2?
Let's not even go there.
former username: A for Awesome
praosylen#5847 (Discord)

The only decision I made was made
of flowers, to jump universes to one of springtime in
a land of former winter, where no invisible walls stood,
or could stand for more than a few hours at most...

User avatar
muzik
Posts: 5614
Joined: January 28th, 2016, 2:47 pm
Location: Scotland

Re: Golly 3.0b2

Post by muzik » August 9th, 2017, 11:11 pm

A for awesome wrote:
muzik wrote:How about non-totalistic versions of LtL rules, at least with range 2?
Let's not even go there.
It's not that bad, I could have said non-isotropic rules instead.

User avatar
gameoflifeboy
Posts: 474
Joined: January 15th, 2015, 2:08 am

Re: Golly 3.0b2

Post by gameoflifeboy » August 13th, 2017, 11:38 pm

I propose that the B/S notation be extended to describe Generations rules, and that this extension be included in the next version of Golly. The format I suggest to follow is B<birth conditions>/S<survival conditions>/G<# of cell states>. My reasons for this proposal are many:
First, the lack of a consistency between the notations I am aware of for Generations and 2-state isotropic rules has always seemed like an incompleteness.
Second, it will allow Generations rules to be written without slashes.
Third, the B/S notation is already the accepted standard for Catagolue rulestrings. Allowing it in Golly will make it so that soup RLEs and rule URLs use the same notation, which will become important soon when apgsearch starts allowing generations rules.
Fourth, the letter G is not only the first letter of 'generations' but one of the few remaining letters not used in the notation for isotropic non-totalistic rules. Even though I independently thought of this letter, I feel it is so obvious it practically chooses itself. Another possibility is the letter D for decay, unless this letter is used for something I am unaware of. I'm not as partial to this idea though, because the number following is the total number of states, not the number of decay states.

When a 'G' occurs in the middle of a rulestring, the interpreter should read the digits following the G, terminating at a non-digit character or the string's end, and interpret these digits as a base-10 number for the number of states; it should not stop reading after 3 digits simply because rules with more than 256 states don't get support. If no digits follow the G, it could default to two states or alternatively throw an error; G0 and G1 would also throw errors. When multiple G's appear in one string, the interpreter could ignore all G's (and their corresponding numbers) except the last one, or it could throw an error; I personally prefer the second choice.

This is still all assuming that a solution hasn't been thought of yet; maybe Calcyman already has his own notation.

User avatar
gameoflifeboy
Posts: 474
Joined: January 15th, 2015, 2:08 am

Re: Golly 3.0b2

Post by gameoflifeboy » August 14th, 2017, 12:38 am

I noticed that the dimensions of a torus in LtL rules can't be set to a height or width less than half the neighborhood radius. However, this still happens when one of the dimensions is set to 0, meaning unbounded in that direction in effect. As a result, I can't create LtL patterns in cylinders.

Also, when specifying multiple dimensions on a sphere (which shouldn't be allowed at all), the result sometimes comes in what appears to be a one-by-infinity bounded grid. So far, I have only seen this happen when the rest of the string is 'B3/S23' or '23/3'; for example, try 'B3/S23:S15,30'. Similar things happen when specifying more than two arguments on the other topologies.

User avatar
rowett
Moderator
Posts: 3776
Joined: January 31st, 2013, 2:34 am
Location: UK
Contact:

Re: Golly 3.0b2

Post by rowett » August 14th, 2017, 6:17 am

gameoflifeboy wrote:I noticed that the dimensions of a torus in LtL rules can't be set to a height or width less than half the neighborhood radius. However, this still happens when one of the dimensions is set to 0, meaning unbounded in that direction in effect. As a result, I can't create LtL patterns in cylinders.
Correct, cylinders are not supported.
gameoflifeboy wrote:Also, when specifying multiple dimensions on a sphere (which shouldn't be allowed at all), the result sometimes comes in what appears to be a one-by-infinity bounded grid. So far, I have only seen this happen when the rest of the string is 'B3/S23' or '23/3'; for example, try 'B3/S23:S15,30'. Similar things happen when specifying more than two arguments on the other topologies.
This was a bug and is fixed in the next release. Thanks for reporting!

User avatar
Apple Bottom
Posts: 1034
Joined: July 27th, 2015, 2:06 pm
Contact:

Re: Golly 3.0b2

Post by Apple Bottom » August 14th, 2017, 6:24 am

gameoflifeboy wrote:This is still all assuming that a solution hasn't been thought of yet; maybe Calcyman already has his own notation.
Generations rules on Catagolue will apparently be named e.g. g3b2s for B2/S/G3, aka B2/S/C3, aka /2/3, aka Brian's Brain. (Whew, so many akas.)
If you speak, your speech must be better than your silence would have been. — Arabian proverb

Catagolue: Apple Bottom • Life Wiki: Apple Bottom • Twitter: @_AppleBottom_

Proud member of the Pattern Raiders!

User avatar
muzik
Posts: 5614
Joined: January 28th, 2016, 2:47 pm
Location: Scotland

Re: Golly 3.0b2

Post by muzik » August 14th, 2017, 7:23 am

Can it be made so that any number typed into the rule specification box without context is read as a rule integer and is converted into the respective outer-totalistic rule? Currently (at least on the iOS version) typing numbers in such a way results in them being read as a rule with no birth conditions and said numbers as survival conditions.

User avatar
PHPBB12345
Posts: 1096
Joined: August 5th, 2015, 11:55 pm
Contact:

Re: Golly 3.0b2

Post by PHPBB12345 » August 14th, 2017, 7:45 am

Maybe totalistic "Large than Life" algorithm uses greedy algorithm or "integer" FFT algorithm?

drc
Posts: 1664
Joined: December 3rd, 2015, 4:11 pm

Re: Golly 3.0b2

Post by drc » August 14th, 2017, 7:21 pm

Andrew wrote:@drc: Please try this version (3.0b3) and let us know if it solves all your rendering problems:
Late reply, but it seems to be fixed! *applause*

User avatar
Apple Bottom
Posts: 1034
Joined: July 27th, 2015, 2:06 pm
Contact:

Re: Golly 3.0b2

Post by Apple Bottom » August 15th, 2017, 5:50 am

Here's a suggestion, if I may--

It's not currently able to interact with Golly's menu bar using just the keyboard. Usually (in Windows programs anyway), hitting Alt will highlight the menu bar, and allow you to navigate the menus using the cursor keys; alternatively, menus (and their menu items) will also have hotkeys assigned to them, so you'll be to use key combinations like Alt-R I G to quickly access the right menu item.

I'd love to see this in Golly, too; it would be really useful to be able to hit, say, Alt-C R to get the "Set Rule" dialog, key in a new rule and hit Enter to set it, rather than having to reach back and forth between keyboard and mouse.

I realize that this might clash with people's custom keybindings. For that matter I also realize that it's already possible to set custom individual key combinations (though not sequences). So perhaps this isn't particularly high priority, but I still think it would be nice to have, and make Golly just that little bit more awesome.

Thanks. :)
If you speak, your speech must be better than your silence would have been. — Arabian proverb

Catagolue: Apple Bottom • Life Wiki: Apple Bottom • Twitter: @_AppleBottom_

Proud member of the Pattern Raiders!

User avatar
Andrew
Moderator
Posts: 919
Joined: June 2nd, 2009, 2:08 am
Location: Melbourne, Australia
Contact:

Re: Golly 3.0b2

Post by Andrew » August 15th, 2017, 9:34 am

Apple Bottom wrote:I realize that this might clash with people's custom keybindings. ...
Not "might", but definitely would! Scripts that use g.getevent to detect alt key presses would also be affected. Sorry, but you'll just have to use Prefs > Keyboard to create key combinations for the menu actions you want.
Use Glu to explore CA rules on non-periodic tilings: DominoLife and HatLife

User avatar
muzik
Posts: 5614
Joined: January 28th, 2016, 2:47 pm
Location: Scotland

Re: Golly 3.0b2

Post by muzik » August 15th, 2017, 2:04 pm

If i try to run this pattern, i get spammed with a bunch of errors and end up crashing the program.

Code: Select all

x = 1, y = 1, rule = R500,C0,M0,S0..0,B1..1,NM
o!

drc
Posts: 1664
Joined: December 3rd, 2015, 4:11 pm

Re: Golly 3.0b2

Post by drc » August 15th, 2017, 3:44 pm

muzik wrote:If i try to run this pattern, i get spammed with a bunch of errors and end up crashing the program.

Code: Select all

x = 1, y = 1, rule = R500,C0,M0,S0..0,B1..1,NM
o!
Can confirm. I can't upload the attachment here, so the best I can do imgbb since you asked for no imgur:
Image
Also I accidentally included all my background things oops

User avatar
Apple Bottom
Posts: 1034
Joined: July 27th, 2015, 2:06 pm
Contact:

Re: Golly 3.0b2

Post by Apple Bottom » August 15th, 2017, 4:33 pm

Andrew wrote:Not "might", but definitely would! Scripts that use g.getevent to detect alt key presses would also be affected. Sorry, but you'll just have to use Prefs > Keyboard to create key combinations for the menu actions you want.
Fair enough. I thought I could get away with being lazy. ;) Thanks for the heads-up!
muzik wrote:If i try to run this pattern, i get spammed with a bunch of errors and end up crashing the program.

Code: Select all

x = 1, y = 1, rule = R500,C0,M0,S0..0,B1..1,NM
o!
Works for me (in 3.0b3), but after 7 generations Golly pops up a message saying "Sorry, but the universe can't be expanded that far."
sorrybut.png
sorrybut.png (4.18 KiB) Viewed 9343 times
If you speak, your speech must be better than your silence would have been. — Arabian proverb

Catagolue: Apple Bottom • Life Wiki: Apple Bottom • Twitter: @_AppleBottom_

Proud member of the Pattern Raiders!

User avatar
muzik
Posts: 5614
Joined: January 28th, 2016, 2:47 pm
Location: Scotland

Re: Golly 3.0b2

Post by muzik » August 15th, 2017, 4:38 pm

Apple Bottom wrote:
muzik wrote:If i try to run this pattern, i get spammed with a bunch of errors and end up crashing the program.

Code: Select all

x = 1, y = 1, rule = R500,C0,M0,S0..0,B1..1,NM
o!
Works for me (in 3.0b3), but after 7 generations Golly pops up a message saying "Sorry, but the universe can't be expanded that far."
sorrybut.png
That's my point. The errors get made extremely rapidly as well, so much so that the faint shadow behind them transforms into a black mass.

User avatar
dvgrn
Moderator
Posts: 10612
Joined: May 17th, 2009, 11:00 pm
Location: Madison, WI
Contact:

Re: Golly 3.0b2

Post by dvgrn » August 15th, 2017, 11:34 pm

drc wrote:
Andrew wrote:@drc: Please try this version (3.0b3) and let us know if it solves all your rendering problems:
Late reply, but it seems to be fixed! *applause*
Golly 3.0b3 looks good on my old laptop also -- no more display glitches in multistate rules with lots of states.

User avatar
Andrew
Moderator
Posts: 919
Joined: June 2nd, 2009, 2:08 am
Location: Melbourne, Australia
Contact:

Re: Golly 3.0b2

Post by Andrew » August 16th, 2017, 8:20 pm

muzik wrote:If i try to run this pattern, i get spammed with a bunch of errors and end up crashing the program.
Much thanks for the report. It's a Windows-only bug which I've fixed for the next release. You are meant to get that message once (when the LtL universe can't be expanded any further), but the bug was causing it to appear in an infinite loop.
Use Glu to explore CA rules on non-periodic tilings: DominoLife and HatLife

User avatar
Saka
Posts: 3627
Joined: June 19th, 2015, 8:50 pm
Location: Indonesia
Contact:

Re: Golly 3.0b2

Post by Saka » August 16th, 2017, 8:22 pm

After all these things are done for Golly-PC, will you be able to work on Golly Mobile? Please make it natively support non-totalistic rules. I'm tired of making a new table for every pattern I want to run.

User avatar
Andrew
Moderator
Posts: 919
Joined: June 2nd, 2009, 2:08 am
Location: Melbourne, Australia
Contact:

Re: Golly 3.0b2

Post by Andrew » August 16th, 2017, 8:42 pm

Saka wrote:After all these things are done for Golly-PC, will you be able to work on Golly Mobile? Please make it natively support non-totalistic rules.
If I do any more work on the iOS/Android versions (and frankly that's a big if) then they will automatically support all the new rules that are in the desktop version (including LtL). You clearly have access to a desktop Golly so I'm a bit mystified why you'd bother creating rule tables for the mobile version.
Use Glu to explore CA rules on non-periodic tilings: DominoLife and HatLife

User avatar
Saka
Posts: 3627
Joined: June 19th, 2015, 8:50 pm
Location: Indonesia
Contact:

Re: Golly 3.0b2

Post by Saka » August 16th, 2017, 8:45 pm

Andrew wrote: You clearly have access to a desktop Golly so I'm a bit mystified why you'd bother creating rule tables for the mobile version.
Sometimes I'm just on my phone and no access to my computer and I want to see non-tot patterns, for example when going to the mall. I'm a bit impatient.

User avatar
Saka
Posts: 3627
Joined: June 19th, 2015, 8:50 pm
Location: Indonesia
Contact:

Re: Golly 3.0b2

Post by Saka » August 16th, 2017, 10:10 pm

Golly crashes when I run this rule: R7,C5,M0,S39..52,B19..54,NN
Try running a few state 1 and 2 soups or scribbling denesly with state 1.
Why??

User avatar
toroidalet
Posts: 1514
Joined: August 7th, 2016, 1:48 pm
Location: My computer
Contact:

Re: Golly 3.0b2

Post by toroidalet » August 16th, 2017, 10:30 pm

Saka wrote:Golly crashes when I run this rule: R7,C5,M0,S39..52,B19..54,NN
Try running a few state 1 and 2 soups or scribbling denesly with state 1.
Why??
I forgot to save the pattern that did this but here's a screenshot: (I selected everything with ctrl-a)
It also produced a weird &quot;agar&quot; of state-1 cells and then deleted a huge portion at other generations (Golly crashed so I couldn't save these).
It also produced a weird "agar" of state-1 cells and then deleted a huge portion at other generations (Golly crashed so I couldn't save these).
Screen Shot 2017-08-16 at 7.26.12 PM.png (43.99 KiB) Viewed 9260 times
Strangely, this only happened when manually running it by pressing the spacebar. Similar things go on in B0 generations rules, where upping the step size generates an explosion or can cause ships to suddenly die out:

Code: Select all

x = 47, y = 22, rule = 012/03/23
46.A$40.6A$37.3A$33.4A$29.4A$20.9A16$A!

Code: Select all

x = 6, y = 6, rule = 012/03/23
A4.A5$A!
And that's before the anisotropy.
The 2nd pattern produces a single glider from the top right cell. Running at step 8^1 generates a space-filling replicator that fills space with the density 1 agar. Changing to step 8^0 causes the space filled to start dying.
Any sufficiently advanced software is indistinguishable from malice.

drc
Posts: 1664
Joined: December 3rd, 2015, 4:11 pm

Re: Golly 3.0b2

Post by drc » August 16th, 2017, 11:25 pm

It works with this pattern:

Code: Select all

x = 7, y = 7, rule = R7,C5,M0,S39..52,B19..54,NN
7A$7A$7A$7A$7A$6A$7A!
Also worth noting is that it erases your clipboard if you have that pattern copied. (Or selected. I don't know which.)

User avatar
Saka
Posts: 3627
Joined: June 19th, 2015, 8:50 pm
Location: Indonesia
Contact:

Re: Golly 3.0b2

Post by Saka » August 16th, 2017, 11:27 pm

drc wrote: Also worth noting is that it erases your clipboard if you have that pattern copied. (Or selected. I don't know which.)
Yes, I think this is the most bizarre part.

Post Reply