The Nineteenth Byte

For those who aren't familiar, it's basically a generalised Fibonacci operator, so f¡@ for a dyadic function f calculates the Fibonacci sequence but with + subbed with f

StackMeter

did @RedwolfPrograms ragequit or is he flagging my account

Redwolf Programs

I don't really mind that, kOTh is just painful :p

hyper-neutrino

4:16 PM

says the person who apparently title-cases SQL

Redwolf Programs

The worst part is the documentation actually does that

pxeger

lowercase SQL is the only valid way to go

Redwolf Programs

Agreed

Wezl

How do you pronounce sql

hyper-neutrino

i use lowercase sql cuz i find it easier to read than shouting

@Wezl squill

Redwolf Programs

4:17 PM

@Wezl Ess Que Ell

Wezl

I say Esquel

caird coinheringaahing

@RedwolfPrograms Thanks for the Reviewer badge :D

Redwolf Programs

Nice!

caird coinheringaahing

@Wezl Structured Query Language

hyper-neutrino

i usually say "sequel" but the correct answer is "however your employer/boss pronounces it"

Redwolf Programs

4:19 PM

No, the correct answer is "sorry I don't know SQL"

Wezl

imagine having an employer/boss

@RedwolfPrograms how do you pronounce that?

Redwolf Programs

Feb 28 at 0:59, by Redwolf Programs

Imagine having a job

caird coinheringaahing

@RedwolfPrograms I know the only relevant bit of SQL: DROP TABLE :P

Same way that the only bash I know is rm -rf / --no-preserve-root

Redwolf Programs

CMC: Take a positive integer as input, then return a program in the same language. Its length should be the inputted number. No exact scoring criterion because this is a CMC and I'm lazy, but the programs outputted should product the largest output possible.

caird coinheringaahing

@RedwolfPrograms Jelly, 3 bytes

Not the largest possible output, but a big output

hyper-neutrino

4:25 PM

@RedwolfPrograms Jelly, 5 bytes

New Sandboxed Posts

1

A: Sandbox for Proposed Challenges

Make it prime with the smallest suffix primes code-golf Given a positive integer as input, output the smallest positive integer such that appending its digits (in base 10) to the end of the input number will form a prime number. Examples 1 --> 1 2 --> 3 8 --> 3 9 --> 7 11...

0

A: Sandbox for Proposed Challenges

Given a table(2D array of strings), arrange width of each column to shorten total width, such that containment in each cell can be fully shown with a space after it, being allowed to occupy next several cells if they are empty. In another word, for each column provide a width \$W_j\$, making \$\s...

caird coinheringaahing

@RedwolfPrograms Jelly, 6 bytes

Redwolf Programs

Imagine having the ability to use TIO

Wasif

@RedwolfPrograms Python 2, 22 bytes

caird coinheringaahing

I think repeated application of ! produces a larger number than repeated application of * (power), right? 9!!!! is larger than ((9*9)*9)*9?

hyper-neutrino

4:27 PM

I had pretty much the same idea lol, ”!ẋṖ9;

Adám

Not sure this is optimal.

hyper-neutrino

@cairdcoinheringaahing oh it's not even close lol

((9^9)^9)^9 can compute

9!! doesn't even compute cuz

Adám

You might be better off using some bytes for a quick to repeat application of a few factorials.

hyper-neutrino

362880 factorial like

have fun

(9!)! is on the order of 10^1.86mil

Adám

1.6097144004100126211034436107333177265055203976228861583 × 10^1859933

pxeger

4:30 PM

(10^1.8) million, or 10^(1.8 million)?

caird coinheringaahing

@Adám Hang on, yeah, that'll be huge, gimme a second

Adám

@pxeger The latter. The former is tiny in comparison.

rak1507

APL: ,∘'9'⍴∘'!'⍤-∘≢⍨

Wasif

Python 2, 25 bytes Try it online!

Adám

@rak1507 Simples: {1↓⍵1/'!9'}

rak1507

4:31 PM

lol that is slightly easier

Adám

But still not optimal. You want ⍣

Wasif

Can't we just append comments at the end of answer? It increases our code length (score) too

rak1507

yeah

Adám

!⍣9⍣9⍣9…!9 or something.

rak1507

@Wasif the code length isn't the score , the goal is to output a valid program with that length, that produces a big number

caird coinheringaahing

4:33 PM

@RedwolfPrograms Jelly, 11 bytes

Essentially, we apply ! 9! times, then do that again \$2^{n-4}\$ times

Adám

@cairdcoinheringaahing Can you not make it repeat 9! times or something?

@rak1507 Maybe !⍣r⍣r⍣r…!r←!9 ?

rak1507

@Adám yeah, something like that

there will be sufficiently many bytes after a while that you could implement something that grows faster but that's probably a waste of time trying

caird coinheringaahing

@Adám Yeah, that's better, for only 1 extra byte: 12 bytes

For an input of 10, it calculates:

a = 9

9! times do:
  9! times do:
    9! times do:
      9! times do:
        a = a!

print(a)

hyper-neutrino

that's gotta be like at least 3 or maybe even 4

caird coinheringaahing

Might be 5, still working it out

Adám

4:38 PM

@cairdcoinheringaahing Can you not do that !9 times?

caird coinheringaahing

@Adám I don't think so, but I think my explanation might be wrong, and it's actually bigger

Adám

I want 9! times do:9! times do:…(!9 times)…9! times do:a=a!

caird coinheringaahing

Correction, for an input of 10, it does:

a = 9

a! times do:
  a! times do:
    a! times do:
      a! times do:
        a = a!

print(a)

Where a! is recalculated each time it begins a new loop

hyper-neutrino

reminder to everyone: nominations for TNB room owner close at UTC midnight today (in 7 hours) (though everyone who intends to nominate has probably done so already; just giving a last reminder)

caird coinheringaahing

As an example of just how big ¡ can go, this is a version that starts with a = 2 and replaces ! with increment

(Yeah, it times out)

hyper-neutrino

4:44 PM

o.O wow. yeah i think that's gotta be close to optimal lol

Adám

@cairdcoinheringaahing Ouch.

ais523

@cairdcoinheringaahing you can get way higher than that, look at this

I have a program that's one byte shorter than yours, and outputs a program of the same form as yours but much, much longer

the extra byte could be used to stick a V in there and eval the generated program

caird coinheringaahing

Forgot about ȷ, that makes it insanely larger

ais523

with a few more bytes you could apply this technique recursively to itself

Adám

We need a bigger universe.

5

caird coinheringaahing

4:52 PM

@ais523 You need a ḣ at the end to comply with the original CMC (output length = input)

ais523

ah, I was ignoring the original question, just showing off a technique

caird coinheringaahing

Ignoring the CMC, we can make it even larger by using Ø% (\$2^{32}\$)

ais523

how new is that one?

caird coinheringaahing

I think it was added with the latest bunch of atoms

ais523

also, that's two bytes long, and using larger constants is normally one of the worst ways to increase the size of a big-number-calculating program, the extra byte may well be usable elsewhere

caird coinheringaahing

4:56 PM

Yeah, using ȷ! as the first 2 bytes would be much larger, I was sticking with the "constant followed by a bunch of !¡s" pattern

ais523

there's actually an interesting question about PPCG rules: a golfing language would benefit from a builtin that expands to some specific, incredibly large number (for algorithms where you want to do something a large but finite number of times); larger numbers would be usable with more algorithms, but if it's large enough the programs will never run to completion so there'd be no way to test them

pxeger

@ais523 How about a "run for exactly 200 quadrillion years" builtin?

ais523

@pxeger that would be a similar idea and have the same problem, I think

although you'd want an exact number of instructions executed, rather than realtime, so that you got deterministic results despite things like CPU load and processor frequency scaling

(as a nondeterministic result is less generally usable)

caird coinheringaahing

@cairdcoinheringaahing Starting with a = 1 and it still times out:

1‘ followed by n ‘¡

n  out
0  2
1  3
2  7
3  2047
4  times out

ais523

that function appears to be named w_n(2)

according to a quick OEIS search

(although the results are off by 1)

I guess the actual function is therefore \n.w_n(2)-1

where \ is a lambda function definition

Newly Featured Questions

5:18 PM

38

Q: Division and remainder

This challenge, while probably trivial in most "standard" languages, is addressed to those languages which are so esoteric, low-level, and/or difficult to use that are very rarely seen on this site. It should provide an interesting problem to solve, so this is your occasion to try that weird lang...

math

5:33 PM

@cairdcoinheringaahing I was just thinking if the name Jay in your profile des is your real name or it comes from Jelly. :P

caird coinheringaahing

@math It's the name I go by IRL, nothing to do with Jelly :P

math

oh, ok.

user

Ooh, Low Blow just dropped a position!

caird coinheringaahing

Yeah, Simple Killer's so much better against Tit for tat, Low Blow can't keep up on points

Still number 1 in the overall rankings tho \o/

user

Can't you detect Tit for Tat in some way?

And then just return 1 each time?

caird coinheringaahing

5:38 PM

I can and do, but Simple Killer plays 100 against Tit for tat, which plays it back, so Simple Killer gets ~100000 points against Tit for tat, while Low Blow gets a few hundred

user

Oh

Then can you return 100 each time?

caird coinheringaahing

As soon as you play anything other than 100 to Tit for tat, the strategy is to drive it down to 1, otherwise it'll just beat you

@user No, cause I don't know which bot I'm up against in the first round, and playing 100 first means I get fewer points overall than just driving Tit for tat down

user

Hmm, then you'd have to open with 100, thus also gaining 100 from Dedicated Counter. Edit: nvm

Redwolf Programs

I think I'm going to make a computer in Minecraft this afternoon

user

@RedwolfPrograms See you in a few years :P

Redwolf Programs

5:40 PM

Might do it on a creative server so y'all can pop in and laugh at me from time to time

caird coinheringaahing

The wins for Low Blow is annoying, it wins against Hard-coded 90% of the time, must have just gotten unlucky this time

user

I hate non-deterministic things :(

Probabilities are evil.

~~Probably~~ Definitely why I failed stats and am failing quantum physics

Redwolf Programs

CMQ: What review queue do you have the fewest reviews in?

caird coinheringaahing

Reopen Votes

Redwolf Programs

Same here, 31 :/

caird coinheringaahing

5:46 PM

It's likely to be LA or RV for most users, seeing as they have <10k total reviews

user

Can you see all your stats for review queues in the same place?

caird coinheringaahing

No, but you can just click the "Stats" button for all 6

user

Oh ok

caird coinheringaahing

There's only 7 relevant numbers on each :P

user

First Posts/Late Answers (2 for each)

Redwolf Programs

5:47 PM

I have the most of FP

(142)

user

@cairdcoinheringaahing ?

caird coinheringaahing

@RedwolfPrograms Same here, 442

user

@RedwolfPrograms CMQ: Fun misinterpretations of FP (don't actually answer this)

caird coinheringaahing

@user Need Review, Reviews Today, All time reviews, the 3 badges progress bars and your all time reviews

user

Oh, I see

caird coinheringaahing

5:51 PM

CMP from the SO blog: what's your favourite badge on CGCC?

user

Nice Answer/Question. It makes me feel appreciated, more so than Famous Question.

Wezl

What's a good high-performance fungeoid I should use in my next project?

caird coinheringaahing

6:07 PM

@Lyxal I still only have 68 unique badges, but I just hit 69 silver badges ;P

@Wezl IIRC Befunge-98 has stuff like shell commands available to it, so maybe that?

user

6:24 PM

Grim Trigger looks like a fun bot

It's not going to win a lot, but it could decrease other bots' points

caird coinheringaahing

It's giving Simple Killer more points (and so is High to Low) :/

Plus it means I have to do more bot analysis to stay competitive. I definitely should've created a bot that learns, rather than gets taught

user

6:41 PM

Time to train a neural network :P

Redwolf Programs

Do 10k users get notified of any flag on chat, or only ones for sites they're on?

user

Any room's flags

I've been notified about flags in a room on Portuguese SO (or some other language) before :P

caird coinheringaahing

@user Grim Trigger and High to Low have secured Naiive's points win over Low Blow :/ Naiive now gets ~60k more than Low Blow

user

⍨

caird coinheringaahing

Naiive was definitely far too OP to be an example bot :P

user

6:54 PM

Just Random would have been way better, honestly

OP's too smart for their own good :P

caird coinheringaahing

1: Crab with 15/16 wins
1: Low Blow with 15/16 wins

Isn't Crab supposed to win every match?

user

...

I don't even know what happened there

I guess there was a tie? ¯\_(ツ)_/¯

caird coinheringaahing

Me neither, I'm running a 10 match competition so I turned off my score reports :/

I'll see if I can repo it again

Redwolf Programs

Probably a tie yeah

caird coinheringaahing

I think it must have been that Meandian returned 1 on it's first round, so no-one got any points

user

7:04 PM

I just wrote my first answer in Stack Exchange! Let me know if it's invalid.

Redwolf Programs

avoids urge to comment "Nice first answer! Welcome to Stack Exchange!"

user

As an aside, I actually modified the parser to just print a warning instead of erroring on unmatched closing braces and parentheses so I could treat them as comment characters, but 1) that approach didn't pan out 2) it would've been longer than 2 bytes and 3) it was cheating to modify my parser for this challenge.

@RedwolfPrograms :P

Redwolf Programs

The physics differences between SMB1 and SMB3 are so annoying

Most of the time when I die in SMB1 it's because I jump straight off a cliff expecting to be able to do some specific thing

user

Does that make your head hurt?

Redwolf Programs

:p

Considering switching to :P

user

7:08 PM

laughs in q:

caird coinheringaahing

@user Not as much as multithreaded Node :P

user

I really want to answer this in my language :P

Redwolf Programs

I don't really know why this wasn't very well received. Any ideas?

caird coinheringaahing

@RedwolfPrograms It got distilled down into an "easy" formula that every answer used

Redwolf Programs

Hmm, true.

user

7:28 PM

Is a language not Turing complete if it can't produce a quine?

Unrelated String

a language can be Turing complete without having I/O

user

It has IO, just not strings

Redwolf Programs

A string can be defined as a list of ASCII code points

user

And I have no idea if it can produce a quine even if it had strings

John Dvorak

or, a language can be written in a weird code page yet have no non-ASCII support

Redwolf Programs

7:30 PM

But I'd think any otherwise TC language (capable of IO which can represent itself in some way) has a quine, right?

user

But how do you choose which representation is fair to use?

@RedwolfPrograms Idk

Redwolf Programs

Well, how does your interpreter take input?

If it's ASCII/UTF-8, then that

user

It's pushed onto the stack

Redwolf Programs

Wait no

user

@RedwolfPrograms No strings, although I can represent strings as stacks of stacks of empty stacks

Redwolf Programs

7:31 PM

I meant to ask how the programs are formatted. It's ASCII, right?

user

Yeah

Redwolf Programs

I guess you could do stacks of stacks, where the length of the stack is a code point

John Dvorak

Magic the Gathering is TC

4

user

@JohnDvorak ???

There's no way you can make a quine with that

John Dvorak

@user arxiv.org/abs/1904.09828

user

7:34 PM

Wow

@RedwolfPrograms You were right

The only problem is finding the quine now :(

Wezl

@JohnDvorak of course it's a Trading Card game :P

caird coinheringaahing

I was very confused about the pinned message, before remembering that mods can edit any message at any time :/

hyper-neutrino

7:50 PM

@cairdcoinheringaahing :P yeah i decided not to post another message because i've already posted a reminder earlier but this makes the pin more visible and maybe draws attention to the fact that it's closing in 4 hours?

again, not expecting new nominations, it's gone about how i'd expect it to

rak1507

poor wezl :(

hyper-neutrino

yeah :c

Wezl

heh

caird coinheringaahing

Wasif and Wezl were the only users who got downvoted, and Wezl's on +4/-6 :/

rak1507

4/6 vs 4/1 is very surprising to me...

caird coinheringaahing

7:57 PM

@Wezl FYI, Hawk returns floats (.0 floats but still). I'd suggest changing sum(map(lambda t: t[0], last_results)) // (len(last_results)*1.5) to int(sum(map(lambda t: t[0], last_results)) / (len(last_results)*1.5))

Wezl

* Wezl is very very slightly offended at the python version being shown first

Transcript for