« first day (3031 days earlier)      last day (1809 days later) » 

1:00 AM
Ok, I definitely think it would be easier to program an algorithm for the Finite Promise Games than the Greedy Clique Sequences
 
 
2 hours later…
2:31 AM
3
Q: Expand a hexagon

xnorGiven an ASCII art hexagon as input, output one whose sides are all one unit longer. _____ ____ / \ / \ / \ / \ / \ \ \ => \ \ \ / \ / \____/ \ / ...

 
2:58 AM
Today I learned why lock(thing) { await do_stuff(); } is a compiler error in C# by spending 4 hours fixing errors where threads would exit before releasing the locks they acquired.
 
 
7 hours later…
10:00 AM
I was reading a children's math book.. it says... On planet binary kids have only 1 finger on each hand. They count 1,10,... what are the next five numbers?
If you only have 1 finger on each hand, can you have more than 4 numbers in a sequence?!
 
 
2 hours later…
12:06 PM
added a new question to the sandbox.. scared :)
 
0
A: Sandbox for Proposed Challenges

AnushOutput digits of \$\pi^{1/\pi}\$ forever This challenge is to produce the shortest code for the constant \$\pi^{1/\pi}\$. Your code must output consecutive digits of \$\pi^{1/\pi}\$ forever. This is code golf, so the shortest submission (in bytes) wins except that it must output the first 10,0...

 
please let me know what you think
 
12:21 PM
@NewSandboxedPosts imo that'd be more interesting as a "how many digits can you print in 10 seconds" challenge without code-golf
 
@dzaima OK. How would those 10 seconds be measured objectively?
 
@Anush the same way you'd measure them in the current challenge, whatever that is
there, ideally, should be no distinction between the two
the usual is to have some individual score every answer on their machine
but idk, it might do well as-is currently
 
I think it would be interesting as code-golf, but that 10,000 digits is a lot. Maybe 1000? I also think I would personally be more interested to solve it if I had to only output a fixed number of digits, kolmogorov style
 
also i have no idea how hard is it to meet that speed requirement, which matters a lot - I don't think anyone would enjoy golfing a 400 byte answer in Jelly
 
@H.PWiz thanks
@dzaima thanks. I hope to get more views
 
12:31 PM
@dzaima I don't know either, but to get 1000 digits per second sounds hard without good optimisation
 
@H.PWiz pretty sure the time requirement is there for requiring a lot of optimization
 
@H.PWiz if you do it kolmogorov style how do you ensure the code actually outputs the digits before the heat death of the universe?
 
(keep some time limit)
Maybe like this
Or say within 10 seconds on a reasonable machine like before
 
imo "runnable enough for you to verify it works", if applicable, is better than "on a reasonable PC", but both achieve different things
 
Do they?
(I also like the first better) It gives feels more freeing
 
12:37 PM
@H.PWiz one makes constants not matter much as 10x worse speed just means posting the answer a bit later, the other makes them pretty important, especially if the limit is as low as 10s
e.g. Kotlins Hello World takes 8s on TIO for some reason, giving only 2s freedom (i have no doubt that's because of compilation/restarting everything from scratch for sandboxing, but still)
 
good point
 
maple can evalulate pi^{1/pi} to 10,000 digits in less than a second
 
@Anush I'd assume the source code of maple is a bit bigger than a couple thousand characters
 
I should probably exclude a built in for Pi or trigonometric functions
@dzaima good assumption :)
I meant just in case if you were wondering if it was possible
 
@Anush here I don't doubt it's possible as pi's been calculated to way too many digits way too many times, generating it is a known problem
 
12:41 PM
@dzaima this is about pi^{1/pi} which is not the same as computing pi
 
@Anush the cost of that extra calculation shouldn't be much, is it?
 
@dzaima it's not 100% clear how to do it naively
I mean it's not enough to compute pi to 10,000 places and do the arithmetic
and how would you do that one digit at a time anyway?
you will have to do something different I feel
I added the answer to the sandbox
 
@Anush start with a big enough precision to get exactly 1000 decimal places, then after that, multiply precision by 9, start again, output the 1001th digit, multiply precision by 9, output 1002th, etc. Should work pretty fast for the requirement and work at all after
 
I reduced it to 1000 too
mostly so I could post the answer in a reasonable amount of space
 
it'll be horrendously inefficient after 1000 digits, but that doesn't matter
if that's not enough, I'd assume precision = precision^precision definitely should be
 
 
3 hours later…
3:36 PM
@dzaima I look forward to seeing it!
 
 
2 hours later…
5:24 PM
0
A: Sandbox for Proposed Challenges

jimmy23013Output the 3x3x3 cube as a graph Your task is to generate a graph with 54 vertices, each corresponds to a facet on a Rubik's cube. There is an edge between two vertices iff the corresponding facets share a side. Rules You may choose to output an adjacency list, adjacency matrix, edge list, or...

 
 
2 hours later…
6:58 PM
the sandbox link has gone from codegolf.stackexchange.com
 
@Anush huh, the tag is still there. damn you SE
 
Anonymous
7:13 PM
@Anush There's a limit to how many things can be on the sidebar. The update to the incident post knocked the Sandbox off. I removed the featured tag from Best of 2018, so the Sandbox will be back on the sidebar in some indeterminate amount of time.
 
Cool
 
 
2 hours later…
9:21 PM
1
A: Sandbox for Proposed Challenges

MilkyWay90code-golfnumber To exponential digit growth and beyond! Challenge Given a base (which is less than 10) and a term, your mission (should you choose to accept it) is to: Make a number which will be the base added by 1. We will call this "the current number" Do the following the term number of ...

Repost
Oh I already reposted
 
10:19 PM
I was wondering about this passage in the finite promise game specification

Z is the set of all integers. N is the set of all nonnegative
integers. All letters represent integers.

We say that T:N^k into N is PL if and only if it is piecewise linear
with integer coefficients. I.e., T can be defined by various affine
functions with integer coefficients on each of finitely many pieces,
where each piece is defined by a finite set of linear inequalities
with integer coefficients.
It says "finite set of linear inequalities with integer coefficients"
Which since there are `k` inputs to each PL function, I'm guessing it's a set of inequalities like

x1 < 3, 0 < x2 < 16, x3 > 1
 
 
1 hour later…
11:43 PM
@ThePlasmaRailgun Confusion 100
 

« first day (3031 days earlier)      last day (1809 days later) »