The Nineteenth Byte

PhiNotPi

12:12 AM

@TheBestOne What is the exact value of that constant?

TheBestOne

Uh... I used a graphing calculator to approximate it.

I have no idea.

PhiNotPi

okay

The first value of .735... is (Log[2] + 1)/Log[10]

TheBestOne

A + ((5 + integrate(max(0, log10(x) + A) from 0 to 5)) / 5) -> A

Converges to about .7187493558

PhiNotPi

What about changing 5 to 50?

TheBestOne

log10(10*x) + A = log10(x) + A + 1 for all A.

New Sandboxed Posts

12:22 AM

0

A: Sandbox for Proposed Challenges

Compressing the Atomic Ionization Energies This is a new type of compression challenge. In a normal compression challenge, you are required to recreate a list exactly. Here, you are allowed to round the values in any way you wish. What's the catch? Your errors are penalized based on how wron...

PhiNotPi

Does A stand for the same thing in both equations?

TheBestOne

Yes.

A + ((5 - integrate(max(0, log10(x) + A) from 0 to 5)) / 5) -> A

I typed the wrong sign the first time.

Geobits

Do you happen to know the full rounding penalty for either "output 1 for all inputs" or "truncated at decimal"? Those could be very short on bytes.

TheBestOne

Is 3d space path-finding a viable challenge ?

Martin Büttner

@TheBestOne on a lattice? euclidean space?

Geobits

12:29 AM

outer space, like through a dense field of asteroids ;)

TheBestOne

No lattice, with gravity

PhiNotPi

@Geobits The penalty function does need a little bit of work, such as taking into account the decimal point.

13.598434005136 -> 13 is a penalty of about 12.51, but saves 13 chars.

because of the decimal point

saving a char without causing rounding

Geobits

Yea, that's what I was wondering about. Yay for balancing acts :P

TheBestOne

Change the values to have the same number of digits :P

7.6 -> 7.60000000

PhiNotPi

I could make everything the same length, although I don't think that would solve the problem by itself.

randomra

12:39 AM

anyone run the grid KOTH with the new bots?

PhiNotPi

I could make everything the same length and remove the decimal point from output.

Or simply remove the decimal point from output.

Geobits

I was more concerned with loop(print(1)) type answers. It only adds a couple points to each penalty (over truncating), but it saves a huge amount of bytes total, since that's the whole thing.

(or whatever number gives the least penalty overall)

PhiNotPi

There may not be a good way to prevent those answers.

Rounding everything to 1 would save about 900 chars, which means it has to occur a massive penalty to balance.

This might be hard to fix.

Geobits

1:05 AM

Hmm. Maybe you could add a rule that each number has to be +-1 from true to be considered a valid submission. Doesn't help truncating, but it should prevent any serious loop abuse, right?

New Sandboxed Posts

0

A: Sandbox for Proposed Challenges

Zeckendorf Number Representation Zeckendorf's Theorem states the every positive integer can be uniquely represented as a sum of one or more distinct Fibonacci numbers in such a way that does not include any two consecutive Fibonacci numbers. For example a valid Zeckendorf representation is: ...

PhiNotPi

1:36 AM

How do I find the size of a CJam array?

Or the dimension of it?

Sp3000

1:56 AM

@PhiNotPi If you mean length, then comma

PhiNotPi

Well, I found a slightly different CJam Sierpinski solution ],X+:X8mL

It's too similar to post as-is, and doesn't save any bytes.

Is there a way to... un-array something? [1 2 3] -> 1 2 3 on the stack?

Sp3000

Tilde

I have to say, that X solution you have is pretty trippy

PhiNotPi

2:13 AM

X~]:X,8mL also works

New Meta Posts

3:15 AM

0

Q: Would we benefit from a tag that indicates a text-block scoring criterion?

A number of times now I have posted challenges where a rectangular block of text is the submission requirement. These are usually source-layout challenges. Examples: (other users surely have some too) Sierpinskified Code Reading Code in 8 Ways to Output 8 Numbers Code that runs the Game of Life ...

luser droog

5:53 AM

Woohoo! Tips are no longer CW!

13

Q: Re-evaluating [tips] and tip-like questions

Currently, all tips questions on the main site are Community Wiki. Furthermore, there are not many other non-challenge questions on the main site. These types of questions are generally supressed on PPCG. This should change. The first point I'm going to attack is the fact that tips questions ar...

discussion new-genre

I just flagged one yesterday on the old policy. Not having seen this.

In other news, I bagged-up some tricks for golfing in postscript.

github.com/luser-dr00g/G

So, now I can start looking for new challenges where I can use it!

Optimizer

7:24 AM

nice! ** coming to Firefox and Chrome

Optimizer

8:13 AM

Piet is getting up votes just because its image based :D

Sp3000

Magic votes formula: Cleverness * IsPythonBonus * IsPietBonus * Humor * HNQ^2

Optimizer

8:48 AM

/isCJamBonus

feersum

8:58 AM

also writing out a number which is at least several hundred digits long is useful

Optimizer

as in ?

feersum

31

A: Smallest python script to print even numbers 0 to 100

Python 2 - 12 characters print 8**999 The decimal representation of all even numbers from 0 to 100 can be found in the output: 1537789902701396471164448516595940643300892369671042144707647536450073500768341185969200084798241824478037061567564756135641105226122796029481353102581685414043699187...

17

A: Finding Programs in the Primes

Lenguage, M = ∞ All of the programs start at the beginning of the string. The following poorly written Python program calculates how many characters are needed for a given M. def program_length(n): PLUS, MINUS, DOT = '000', '001', '100' i = 1 s = '' while n > 0: i += 1 ...

Martin Büttner

9:52 AM

@Sp3000 can the ><> quine be adapted for CH's challenge?

Martin Büttner

10:14 AM

does ><> error on unknown characters or ignore them?

Optimizer

iMMA TRY cjAM

Martin Büttner

hf

feersum

i'm starting with writing a regular quine for TECO

Martin Büttner

I was thinking Prelude (the score would be only be twice as large as a plain quine score, probably less), but then I couldn't be bothered

Sp3000

@MartinBüttner It errors, so no it can't

Optimizer

10:30 AM

hmm, having just 1 byte operator space makes it really tough

Sp3000

@MartinBüttner Also I'm not sure how the ><> quine would help because 1) You'll be opening/closing strings as you go and 2) I'd assume you'd need to take a log somewhere?

Optimizer

Math.log ?

feersum

well I've got my regular teco quine, It was actually easier than I thought.

Optimizer

are those : keystrokes too ?

feersum

no

Optimizer

10:37 AM

I mean, what is the final text in the editor window which gets evaluated ?

feersum