« first day (1754 days earlier)      last day (3072 days later) » 

8:00 AM
@Vɪʜᴀɴ I edited my question (Ducttape the Ducttape). Is everything clear now? :D
 
@phase where are you geographically, if the ping isn't awful i'll melee netplay you some time
 
/me wishes I had a Wii/Gamecube and Melee
 
@El'endiaStarman
Dolphin is always an option if you aren't using a Raspberry Pi!
 
well
then the next problem is controller
playing smash with anything except gamecube controller is unacceptable
 
@Stefnotch Tried it. Runs at like 75%. Frustrating. :(
 
8:08 AM
@El'endiaStarman buy 25% more computers
 
I lack disposable income.
 
XD (Bank?)
 
@El'endiaStarman get more incomes
 
@orlp I'm trying...
AT LAST!
@Calvin'sHobbies!!!
 
Eh...what's that?
 
8:10 AM
Nice!
 
seems to be some floodfill color coded maze
did I nail it?
 
4
A: Sandbox for Proposed Challenges

El'endia StarmanBlack and White Rainbows code-golf image-processing graphical-output Given an image that has only black and white pixels and an (x,y) location that's a white pixel, color the white pixels based on their minimal Manhattan distance from (x,y) in a path that only involves traversing other white pi...

 
Wow, I love how it looks.
 
yes, yes I did nail it
 
I also know that doing that for the mazes won't look particularly special because there's not as much room to wrap around corners.
But I can do it for the spiral one...
 
8:12 AM
:D
 
WOW, that is cool!
 
How does the algorithm work for Euclid dist?
 
@Calvin'sHobbies euclid dist is not defined on a pixel grid without further explanation
 
Right. That's why I'm asking.
 
Okay. It's a bit complicated, but let's see if I can summarize it well enough...
 
8:14 AM
@El'endiaStarman by the way, it seems you're not using a proper color space
that's waaaay too much green
allow me to introduce you to
A Lab color space is a color-opponent space with dimension L for lightness and a and b for the color-opponent dimensions, based on nonlinearly compressed (e.g. CIE XYZ color space) coordinates. The terminology originates from the three dimensions of the Hunter 1948 color space, which are L, a, and b. However, Lab is now more often used as an informal abbreviation for the L-a-b representation of the CIE 1976 color space (or CIELAB, described below). The difference between the original Hunter and CIE color coordinates is that the CIE coordinates are based on a cube root transformation of the color...
 
@orlp All the green is equidistant away. That's the point
 
@Calvin'sHobbies no it's not
@Calvin'sHobbies unless there are multiple points involved
 
Seems pretty accurate to me. The yellower greens are closer and the bluish greens are farther.
 
1. Have a starting pixel.
2. Flood fill from that pixel.
3. For each reached pixel, take half-unit steps between that pixel and the start, following a straight line.
4. For each half step, `int()` their coordinates.
5. If the pixel at this coordinate is black, stop. Otherwise, continue.
 
since it's manhattan distance
you don't need to compute anything
just keep a priority queue to floodfill with
and you'll reach every point only once, at the fastest possible distance
 
8:18 AM
In my code, I do a fair bit more than this, but mainly for book-keeping.
 
@orlp The image above is not Manhattan distance, it's Euclidean. Thats what El'endia is describing
 
@orlp 1) I've already done Manhattan distance, and 2) you don't know the language I'm using. There is no easy way to do a queue.
Actually, I am using a queue, now that I think about it, but it's for starting points.
It's also quite slow, primarily because I need to allow color overwriting if the distance is shorter in that way.
About halfway done with the spiral one.
 
oh
I thought it was manhatten
 
If you check the Sandbox post, those are Manhattan.
 
either way
it's very evidently not a linear color space @Calvin'sHobbies
consider these two areas I marked
they're pretty much the same distance increase from the origin
(both manhatten and euclidean, difference is minimal)
 
8:26 AM
@orlp no problem
 
you can't possibly argue that this is a linear color space
 
@orlp Actually, the one in the green is more "sideways".
@orlp No, but it is the easiest.
I'm sure you could fix it in some graphics processing program.
 
the trick is to not use the mayflash driver, it's kinda bad. keep it in wii u mode and use the unofficial driver
you can also use a first party gc->wii u converter, but they're hard to find and the mayflash is just as good
 
@El'endiaStarman you'll get prettier results with a linear colour space though
but my point was just to illustrate to calvin that it is decidedly not linear
 
8:28 AM
Sure, but it already takes several minutes...and I'd have to write a new color function from scratch...
 
@El'endiaStarman minutes?
 
The language I'm using, by the way, is Blitz 2D/3D.
 
never heard of it
 
@orlp Blitz is not a fast language.
 
the only problem with melee netplay is that unless someone lives improbably close to you, you'll rarely get less than ~2 frames of lag
 
8:29 AM
any particular reason you're doing it with blitz?
 
Easy graphics.
 
@undergroundmonorail 60fps or 30fps?
 
Also one of my first programming languages and still one of my best.
 
@undergroundmonorail 33 ms and your frames are eaten
 
8:30 AM
yeah
 
Crap, it leaked through one of the centers of the spirals. It won't change the end result too much though...
 
it's kind of weird in dolphin and i'm not 100% sure how it works but
you're supposed to take your ping and divide by 16, and set that as your buffer, but 1 buffer only equals half a frame of lag
 
@El'endiaStarman what's the start cooridnate?
 
Of which one?
They're all in the Sandbox post.
 
@El'endiaStarman You could jump quarter steps or less to avoid that I suppose?
 
8:32 AM
ah
 
@Calvin'sHobbies I doubt it. Well, maybe. I don't really know how to fix it and I don't want to wait another half hour for it to finish.
It's almost done. Like, a couple minutes I think.
 
one time i played someone on netplay who kept changing characters after every couple of games. when we were done they told me i was "boring af" to play against because i only play captain falcon. i'm bad at falcon but they were garbage at like 5 different characters :P
 
I think you can tell which spiral leaked. :P
 
how are spirals leaking?
 
The color got through the center of one of them.
 
8:38 AM
what are these spirals?
 
New example:
 
ah nice!
 
8:39 AM
The spiral actually looks barely different in manhattan
 
Manhattan:
 
you get more straight lines at 45 degree angles (which is expected obviously) but other than that yeah they're pretty similar
 
Euclidean is way prettier though. :P
Also, they're only at 45 degree angles because the lines are at 90 degree angles.
 
they're both pretty awkward right in the middle, between the two lines that don't touch the edge
 
I wish I could mouseover and change the point on the fly :P
 
8:42 AM
oh that would be sick
 
That would be amazing.
And very hard.
New example:
 
what's it like non-manhatten?
 
8:44 AM
@El'endiaStarman I could do it in JS I bet (like this), but speed is the tricky part I expect
 
Euclidean way prettier
 
@Calvin'sHobbies if you have the code, I can help speeding it up
 
@El'endiaStarman Can you post the Chebyshev too?
 
@Optimizer I'll pastebin it in a bit.
 
8:47 AM
@El'endiaStarman JS ?
 
19 mins ago, by El'endia Starman
The language I'm using, by the way, is Blitz 2D/3D.
 
well, I was referring to JS speed up for Calvin
 
Sorry it's not a well-known language...
Also, you would probably be better off reimplementing it yourself instead of trying to translate.
 
"The Chebyshev distance between two spaces on a chess board gives the minimum number of moves a king requires to move between them." i want to see a distance defined by the number of moves it takes a knight
O
    is shorter than  OO
 O
that would be sick
 
......nnnno. :P
Also, it wouldn't look particularly different.
 
8:49 AM
i just meant "i want to see if someone's named this and described it" not "implement this in your program for my amusement" :P
 
Oh, okay, good. :P
 
i'm just looking at the Metric Geometry wikipedia category and clicking on things that end with "distance" now
 
oh sweet
ty
 
and yeah, it's mostly just a dithered gradient
 
8:56 AM
shhh, nobody tell Martin about all those martin-shops
 
Now with more cycles!
 
this is a whole wiki dedicated to
not programming
not chess
but programming chess
 
1. how is there a whole wiki for this 2. i love that there's a whole wiki for this
2
 
@undergroundmonorail the amount of literature available just on code and algorithm approaches to simulating the game of life is... impressive, as well
 
8:58 AM
the game of life makes sense to me because it's a math thing that people study and it happens to be the one popular cellular automaton or w/e
this is like
how to make a computer play a board game
 
A board game that is far more popular than the math thing.
 
that's true ig
it's not that i'm surprised that there's overlap between people who care about programming and people who care about chess i'm just surprised that the overlap even wanted a wiki
i didn't realize there would be much to say
 
I guess there's a lot of people who want to figure out how to make a program that can beat humans in chess. ... what's that? That's already been done? Well damn.
 
think I'm messing something up with the linearity
 
If you were to write a direct, very short introduction for Microsoft Office's word processor, it might be a...
...forward four-word foreword for Word.
7
 
9:13 AM
@orlp Or rather, it's just not as bright and saturated (i.e., luscious).
 
for the record i just stole the last thing from reddit
 
@El'endiaStarman it is 90% saturated
60% brightness
 
Ahh, yep.
I know it has to be like that to stay within the Lab color space.
 
I mean
you have to compromise
 
@Calvin'sHobbies: Do you think we should include Euclidean distance as a bonus or, perhaps, maybe, possibly, leave it for a separate contest?
 
9:17 AM
since blue can never be so bright as yellow
either you desaturize some colors
 
or you reduce light of some colors
I was wrong btw
last picture sacrifices saturatization
 
or the other way around
not sure which is which
but this is the 'pastel' approach
THAT'S what a linear colouring should look like
 
Huh, interesting.
 
9:20 AM
that's manhattan distance btw
 
I can see that. :)
 
the patterns a lot less obvious when you're using uniform colors :)
 
oh yeah, definitely
What language are you using?
 
Python
 
ahh, cool
 
9:21 AM
@El'endiaStarman I think it would be a good true bonus, -20% or something, if it can be accurately specified.
 
That's actually my other best language, but Blitz is so much easier with graphics stuff...
 
@El'endiaStarman really
 
@Calvin'sHobbies Do you think the summary I gave is good enough, provided that I fill in the details?
 
how many lines of code for the manhattan?
 
211 lines. But I'm not using any libraries.
 
9:23 AM
I do it in 26 with linear coloring :)
(with libraries)
 
And a lot of it is data structure stuff because it's less powerful in that way.
And probably about 20% of the lines are blank for readability. :P
 
@El'endiaStarman Probably, I got the gist
 
Euclidean, incidentally, is 327, but a significant chunk of that is making it so I can watch the color spread update instead of waiting forever for it to finish.
@Calvin'sHobbies Okay, I'll do that tomorrow. What times are you usually awake and alert?
 
I haven't optimized anything
it runs in ~5 sec
 
So I know when to expect review from you. :P
 
9:26 AM
Just ping me and I'll get it :P
 
@orlp Again, Blitz is slow. But it's about 2 seconds for that example. ... Huh. Faster than I thought! :P
Alright, good enough. :P :)
 
ah
you were comparing to the maze
ye, it seems quite a bit slower now
 
The example that you used for your images.
 
@El'endiaStarman Python is really slow though :P
 
Blitz is, honestly, probably slower than Python. :P
It's an interpreted language as well.
 
9:29 AM
Ugh. (off-topic moment of the internet):
 
@El'endiaStarman I used this btw husl-colors.org
 
@orlp Cool!
Okay, time for me to go to bed! G'night!
 
there's too many sandboxed questions with major ambiguities that have a lot of upvotes
the problem is there's no way to tell an "I've checked this and it's ready to go live" upvote from an "I think this is cool" upvote
 
9:39 AM
(...I hope those don't include any of mine...)
 
Just let that image stay in your mind throughout the night. c:
Sweet dreams. ccc:
 
9:50 AM
who can guess why i used to have an autohotkey script that typed "strawberryvanillachocolate" when i clicked mouse4
google is cheating
 
You really like neapolitan?
It's your password?
I really have no context. :D
 
i'm not a fan of neapolitan because i don't like strawberry ice cream and chocolate and vanilla are good but there's always that strawberry mixed in
it's not my password
your hints are: you're thinking about ice cream which is good, and i almost typed "neopolitan" by accident and that typo is relevant
 
how would you guys feel about a programming language that allows - in the middle of identifiers as an alias for _?
so instead of num_daily_collisions you can write num-daily-collisions
 
that sounds neat but in practice i feel like you'd always use one or the other
 
well, I'd always use the latter
however, I want to be compatible with other tools
so aliasing it to _ would seem the best way to do it
so you can also bind to foreign code and still use these-pretty-names
 
9:55 AM
that makes sense
 
like
the only disadvantage is that 5*x-y would have to be written 5*x - y
 
@xnor Indeed - tbh I don't vote much on the sandbox because I'm not sure what votes are good for there
 
10:15 AM
0
Q: Can you win with two more moves at Three Men's Morris?

insertusernamehereThe Game You remember the classic game "Nine Men's Morris" or simply "Mill"? There's variation called Three Men's Morris which is a bit like a mutable tic-tac-toe. Rules This is the blank board of the game: a b c 1 [ ]–[ ]–[ ] | \ | / | 2 [ ]–[ ]–[ ] | / | \ | 3 [ ]–[ ]–[ ] [ ...

 
Grrr. My CS class has a time requirement for how long our code takes to execute and they don't even allow us to specify -O3 (in fact, the compile it without any optimization option). Rather annoying.
I found that g++ has #pragma gcc optimize(3) (something like that), although that only works for each individual cpp file. So I do get the improvement from 2m47s to 1m37s, although I don't get the improvement to 36s. So sad.
 
use -O2
-O3 has badly tuned unroll settings and often performs worse than O2
 
Okay
 
@Justin they probably want you to use a better algorithm
 
@orlp I know
I'm using good algorithms, but I'm also using lambda functions a lot
And other little things like that
 
10:24 AM
what's the problem?
 
They take an extremely long time without optimization.
2m47s is a lot longer than 1m37s
 
no, I mean
the problem your CS class wants you to solve
 
Oh, we are interpreting Datalog programs.
 
i saw someone who wrote C or C++ (i don't recall which) for a class they had to take but was way below their level. they decided to write code that worked but was as confusing as possible. their "hello world" didn't have any code in it, main was defined as an array of ints
 
Honestly, I'm not that worried about the time requirement, but it bugs me that I can't just make it run as fast as possible.
@orlp -O2 vs -O3 makes no difference for this program
 
10:37 AM
I honestly wonder how much slower/faster my program is than other peoples' programs.
 
not sure what datalog is btw
 
I'm not sure either. It's a lot like Prolog.
 
I really don't like prolog
entirely backwards
 
Same.
@orlp You really don't need to be reading into this, but if you want to, this is the course wiki
Sometime I'm going to build an actual interpreter/compiler, for a real language, rather than this weird logic programming language. That will be a lot more fun than this.
 
doing that as we speak
 
10:43 AM
Fun
What does it mean if, when I ran my program, linux tells me "Killed"?
I'm assuming my program just consumed too much memory
 
10:55 AM
I think g++ requires -pthread to allow use of c++11's <thread>, but otherwise, I'd be using that for even more speed :-).
 
0
A: Sandbox for Proposed Challenges

TobstaTriangularity This is a challenge where you have to work out the length of a given w and h. To qualify as a correct solution: The program uses STDIN to get the values of w and h. The program uses STDOUT to output the value of a in the above diagram. The shortest code in bytes wins. Tags:...

 
@El'endiaStarman I'm curious how the image would look like with knight distance
 
11:10 AM
I used LEMON
it's pretty awesome
so far I've mostly laid down the infrastructure though
no actual programming language yet
 
 
1 hour later…
12:20 PM
@orlp When life gives you Lemon, you ... code in it?
4
 
1:14 PM
0
A: Sandbox for Proposed Challenges

WallyWest​Standalone Array Pop Function Without using any native array functions, methods or properties create a standalone array pop function, that taking an array, providing there is at least one element, will pop the last element off that array, return that value, resulting in the original array being...

 
 
1 hour later…
2:34 PM
-1
A: Sandbox for Proposed Challenges

VoteToCloseWhat time is it again? popularity-contestdate-time The challenge is simple: output the system type in any way you see fit. This includes using any standard loopholes or joke answers - whatever you want. Rules (limited they may be): The only standard loophole you cannot break is the new langu...

Feedback, anyone?
(keep in mind that this is supposed to be simple. Ish.)
 
It seems to suggest that answers don't even have to follow the only spec (outputting the system time)...
It's really broad, even for a popcon.
 
Hmm. So, remove that implication/part of the question. Aside of that? I'm just asking for system time of any method.
 
I meant more this part:
> In the case of a tie, the shortest answer that actually finds system time wins.
 
Oh, yeah, I meant to say shortest code. :P Fixing.
 
2:49 PM
If the challenge is "output the system time", there might be a dupe (not sure), but would be better as a golf either way. If it's "do something creative with time output", I think it's waaaaay too broad.
 
Output the system time is the challenge.
 
So why popcon?
It's just going to end up as "prettiest clock" which is an art challenge, not a programming one.
 
...I honestly don't know. XD I'll fix that.
Okay. Fixed that. I have to go offline for now, but I'll be back in 2(ish) hours.
 
> This includes using any standard loopholes - whatever you want.
Mine uses hardcoding:
System.out.print("now");
It doesn't even have to loop or be re-ran to stay current. Just run it once and you can glance back at the output whenever you want to know what time it is.
 
@Geobits System.out.print("wow");
> output the system type
typo?
And.... wait a second... how is bicycles.stackexchange.com graduated, but we haven't??
 
3:03 PM
They don't hold cycling contests; it's an actual Q&A site.
3
That probably helped.
 
:(
 
November's not over yet... I think you'll be pleasantly surprised soon.
6
Also, they have 7k questions to our 4k.
But they should definitely hold cycling contests.
 
That might get them downgraduated.
 
I fail to see how that affects me, though :P
 
I thought downgraduation would be right up your alley.
 
3:09 PM
I meant some sort of community gathering, though. Like an annual meetup of cyclists to have a race or something. I don't see how that would be a bad thing for them.
Sure, not everyone would be able to attend, but if you throw in a picnic/barbecue, it would be a bunch of fun.
 
(It wouldn't.)
 
Damn internet. I meant it wouldn't be bad, not that it wouldn't be fun.
 
oh okay
i thought you were like
dissing the bicycles.se community
"a meetup wouldn't be fun because you'd be meeting up with other bicycles.se users! hey-o!"
but i get it now
 
Haha.
 
3:16 PM
2
Q: Catalogue Challenge: 99 Bottle of Beer

GamrCorpsCatalog Challenge: Write a program that outputs the lyrics to 99 Bottles of Beer, in the shortest amount of bytes. Rules: Your program must log to STDOUT or an acceptable alternative, if STDOUT is not available. Your program must be a full, runnable program or function. Unlike our usual rules,...

 
3:30 PM
Would this be rejected because it takes it out of the snippet, which is usually bad? I wouldn't want my code taken out of a snippet.
actually it makes it so it can't run, which is good because running it just prints the code
@Dennis help me out here
 
The only positive to the snippet I see is that it's collapsed/hidden by default instead of having a large code block there. Would probably leave as-is, but I don't feel strongly enough about it either way to approve/reject.
It'd be nice if there was some way to collapse content without snippetizing it though.
 
3:46 PM
@phase I'm not sure about this. It isn't a snippet, but un-snippeting conflicts with the author's wishes. You could just wait for Pietu1998 to make the call.
 
0
Q: String with direction from vector

Filipe Almeidain Javascript (any version) can you make this shorter? var out = '': if(y>0){ out='Y'; }else if(y<0){ out='^'; } if(x>0){ out='>'; }else if(x<0){ out='<'; } console.log(out) Explanation: x and y are given numbers representing a vector output is a console.log of string with one or two ch...

 
4:29 PM
0
Q: Write a code that lists every ASCII character not in the code itself

ghosts_in_the_codeThe question title says more or less everything. Write a program/function that outputs every 8-bit ASCII character not present in the code itself, onto a file (or the screen itself; I'm not sure if that is even possible for certain control characters). Obviously, your code must be written in AS...

 
@NewMainPosts oh ffs
I had dibs on that
grr
Did he even sandbox it?
 
Sorry @Mego for taking your 99BB Seriously builtin.
 
4:49 PM
user image
3
The deed is done.
The formatting is a little strange because I couldn't fit all the text on the card.
@Dennis all hail
 
Hahahaha. :D
 
:D
Now we need one for Alex and Doorknob
 

« first day (1754 days earlier)      last day (3072 days later) »