« first day (1 day earlier)      last day (1448 days later) » 

12:24 AM
@El'endiaStarman shorter hashes for varlife half done :D
 
is it okay if they are acronyms?
e.g. rule -> r
 
It's your code, not mine.
 
but
idk if people are gonna actually read the hash
 
I'm saying, it doesn't really matter to me.
 
 
9 hours later…
10:01 AM
@zyabin101 updated github.com/somebody1234/cellulose btw
 
@somebody Does it contain my electron flowing in a wire example?
 
?
wait
what example?
 
23 hours ago, by zyabin101
x = 8, y = 7, rule = Tetris
6C2.$B6A.$6C2.$6.A.$5.CAC$5.CAC$5.CAC!
 
um
broken for me
for some reason yellow ones don't work
oh wait a second
need to fix
 
@somebody It looks like the Tetris rule is broken. Can you provide a GIF?
Ninja'd me!
 
10:09 AM
i'll fix it
kinda fixed
not pushed yet
fixed and pushed, can you add a folder with some of the components in gates.js? (not the clocks, and you don't have to do all)
nvm
@zyabin101 maybe you could help fixing my or gate? it's one cell too fast if it's only one input
 
10:50 AM
@somebody I'm in, but what's the format? Also, I can't do it right now. I'll ping you when I'm home.
 
golly format for now
i can do them though
 
@somebody I mean, the format of patterns in the *.js.
 
oh
space = black
0 = blue, | = green
 
11:12 AM
@zyabin101 done
 
11:22 AM
@somebody 1) Wat~~?! How you~~ 2) Why the patterns are all in Macrocell format? Golly uses the Run Length Encoded format.
 
wait
idk
kills Linux Golly
 
@somebody From the Golly help center:
Golly prefers to store patterns and pattern fragments in a simple concise textual format we call "Extended RLE" (it's a modified version of the RLE format created by Dave Buckingham). The data is run-length encoded which works well for sparse patterns while still being easy to interpret (either by a machine or by a person). The format permits retention of the most critical data:

The cell configuration; ie. which cells have what values.
The transition rule to be applied.
Any comments or description.
Here is an example:
x = 3, y = 3, rule = B3/S23
b2o$2o$bo!
 
fixed
 
Yes?
 
nothing
 
11:31 AM
We don't need the extended RLE part. Can you delete the #CXRLE comments at the start of each RLE file?
 
done
 
11:43 AM
You have in the repository the Cellulose editor (or is it?). It is in HTML and JavaScript. Maybe publish it into GitHub Pages?
 
oh yeah
been trying to delete master branch
or should master be used for buggy updates?
actually, done
 
@somebody But which branch is the default branch?
 
gh-pages is now default, plus it is displayed
 
12:09 PM
Recently GitHub added and removed a thing from the web interface: You cannot edit files! You have now to upload them, which is added recently.
Plus: You can finally commit multiple files.
 
i can still edit files
from web
 
That means that was my imagination. :?
 
yes
what do you mean by you can finally commit multiple files?
 
@somebody You can upload multiple files in one go, meaning the commit will contain multiple files.
 
well, that's only because of drag and drop
*multiple files from the web interface via drag and drop
 
12:16 PM
Welcome, @Adnan.
You have just joined the quest for Tetris. Click this link for more.
 
i swear @Adnan just joined chat and went afk
 
Pretty colors!
@somebody Yeah, that is true :p
 
@somebody Almost 90% of chat joins are meaningful. [Source: Yes, I made that up.]
 
Source: 'I made that up Yes'? never heard of them before
 
@somebody I made the (fake) quote up myself.
I propose another quest: build a self-reproducing machine in PhiNotPi's CA.
 
12:25 PM
anyone know jquery well?
@zyabin101 we need components first
 
But that would be before or after step seven.
@somebody Err..... no.
@somebody ^^
 
@zyabin101 it's not possible, can't change rule
unless you start with full blue or something
 
@somebody Then the unbounded universe will be full of electrons flowing in the infinite wire.
@somebody No one follows your proposal. Also, I can't.
Maybe you can spread the word in the Nineteenth Byte?
Or in...

 Area 51

General discussion for area51.meta.stackexchange.com
 
12:40 PM
idk, need five questions first
nave no idea what to write for them
 
@somebody You need fourteen questions. Also, they all need to have a score of 10 or more.
You only have the limitation of five questions per proposal per user.
 
yeah
@zyabin101 you can't follow it?
 
One idea: How can I discover new things in Conway's Game of Life?
 
too broad?
 
@somebody Two problems: 1) I don't want to synchronize my account with an email 2) The second part is a sekrit ಠ͜ ಠ
@somebody Second idea: What is Catagolue and how can I contribute to it?
(Catagolue is this thing)
 
12:52 PM
2. bad question, solved by a google search
also, disapproving happy face?
 
@somebody Then third idea: I don't know what to do.
@somebody Yes. ಠ͜ ಠ
 
i tried to search the GoL forums for something
 
@somebody Ask the example questions in your proposal.
 
 
1 hour later…
1:55 PM
@zyabin101 example questions?
 
@somebody A proposal in the definition phase (or pre-classic phase, it's more geeky) requires example questions to get to the commitment phase (or classic phase).
 
ik
what i meant was i was looking in the GoL forums for questions
 
 
4 hours later…
6:55 PM
@El'endiaStarman hi
Oh, refreshing the page made yous avatar turn grey.
 
I think I found better diagonal wire.
@El'endiaStarman Actually, it works even better than expected.
 
I'm trying to render to gif
 
Woah, nice!
That's fantastic.
 
7:17 PM
Also, I found that your design for a parallel wire crossing is flawed.
Electrons sometimes leave with "doubled tails."
Next on the list is better delay wire.
There's two main uses for delay wire:
1) Use in a logical circuits, like to slow down a shorter data bus as to take the same amount of time as a longer bus.
2) Use inside of a single "logic tile." If some of our logic tiles introduce an unavoidable delay, we can add some delay wire onto the outputs of all other logic tiles, so that they all have the same delay.
 
7:41 PM
@El'endiaStarman Also something that we should do eventually is to load up some actual metapixels in a GoL sim to make sure that, when we give then different rules, stuff works the way it should.
 
@PhiNotPi Is that when there are simultaneous on inputs? I didn't design it with that use-case in mind.
@PhiNotPi Hmm, yeah.
I have no experience with Golly, though (which would probably be the best choice).
 
Oh, huh, whoops.
A flaw covered up by the smaller width earlier.
 
 
1 hour later…
8:53 PM
@El'endiaStarman I tested it out in Golly and it seems to work.
 
Excellent.
 
More specifically, I tested that changing the rule in a single cell works as expected (It's somewhat more difficult to build an actual wire)
 
oh, heh
 
What's also nice is that you don't have to infinitely tile the plane to get the metapixel to work: it treats edges as dead
 
Oh good.
 
8:57 PM
As far as I can tell, the original OTCA Metapixel team never envisioned mixing rules, it just happens to work out nicely that way.
 
Huh, cool.
That's super convenient for us, then.
 
I think it's funny to was people randomly drop in and leave.
 
ha, yeah
 
They enter thinking we're talking about Tetris, but leave once they realize they have no clue what is going on.
Golly is like, super-powerful by the way.
 
@PhiNotPi Hahahaha. Hooray for clickbait titles! :P
 
 
1 hour later…
10:24 PM
@El'endiaStarman ^ my current best delay wire. It takes 9 ticks for the electron to flow 5 cells across.
The 9:5 ratio is kinda ugly, though. I would much prefer a 4:2, 3:2, or 6:3.
 
yeah
@PhiNotPi Another delay circuit; delays electrons by 8 ticks. I used your idea and combined it with the fact that a blue wire of length 7 with the leftmost cell as the seed becomes static after several generations.
 
10:47 PM
^ has a kinda erratic delay, though
Oh, yeah, the design I just posted is basically unusable.
 
Problem with backscattering?
 
Yeah, the first 1 or 2 subunits have a smaller delay than the rest.
 
why delay?
 
If two wires aren't the same length, we need a way to slow one of them down.
 
@PhiNotPi when are two wires not the same length?
^- golly
 
10:54 PM
@somebody Well, I strongly doubt that we will be able to have all the timings correct just by adjusting wire lengths.
 
but
everything has a period of 9 gens right now
asuming we're using the or gate with red
also, somebody1234.github.io/cellulose has shorter hashes (plus it shows link) :(
 
@somebody The logic gates won't necessarily be the same distance from each other, though.
 
then don't you only need a delay 9 wire though?
 
We also don't know what period the gates will end up being, if some other useful features take more time, we can add delay to anything. And we also don't know how big each logic unit will end up being, so we need a way to construct an arbitrary delay for however big the logic units are.
 
but
don't we do that afterwards?
not very easy to do every single delay from 1->9
 
11:04 PM
Technically, we simply need relatively-prime delays....
 
I don't really think it matters when we figure out how to make delays.
I wonder, when will we be able to pick a size for the "logic units"?
IDK what we should call "logic units"
 
components
gates?
 
Ehh, I'll still call it units for now, since it's more of a unit of distance than a description of what they do.
 
components
because they're all part of a circuit
 
Anyways, we need each component large enough that we can make self-contained NOT gates.
Since each NOT gate needs to contain a clock of the correct period.
Since each component will have an odd-numbered length, we also need to make a clock of an odd-numbered size.
 
11:18 PM
so a period 9 clock?
 
Probably more.
 
why?
well, right now we only need a period 9 one
we already have a period 16 one
 
I don't think we can fit the clock and an XOR gate together in a 9x9 area.
 
which can be extended to 16+4n
@PhiNotPi but it will probably fit in an 18x9 area
also, it's possible to build the clock into the actual gate
 
I was thinking we should have the NOT gate fit into a single square.
 
11:21 PM
not = xor with clock right?
 
yes
 
11:34 PM
goo.gl/8HEQdJ modified, still can't fit the clock though
 

« first day (1 day earlier)      last day (1448 days later) »