Pair-golfing Twin primes and the Collatz sequence


This is a new kind of challenge inspired by the Recover the mutated source code problem.

You should write two programs or functions both in the same language. The first one should solve Task #1 and the second one should solve Task #2.

Your score will be the sum of the longer program and the Levenshtein distance between the two programs source code. Lower score is better so you should try to make the two solutions similar while keeping the lengths of the programs short.

Task #1

You are given a positive integer N and you should output the Collatz sequence of N separated by spaces or newline. Trailing separator is allowed.

The first element of the Collatz sequence is N. The rest of the elements are generated based on their successor \$a_{i-1}\$:

$$ a_i = \begin{cases} \frac{a_{i-1}}{2} & \text{ if } a_{i-1} \text{ is even} \\ 3a_{i-1} + 1 & \text{ if } a_{i-1} \text{ is odd} \end{cases} $$

As soon as the sequence reaches 1 no new elements are generated.

Input => Output examples:

6 => 6 3 10 5 16 8 4 2 1
8 => 8 4 2 1
1 => 1

Task #2

A pair of twin primes is a pair of positive integer whose difference is 2 and they are both primes.

You are given a positive integer N and you should output the smallest pair of twin primes where both primes are bigger than N The first number should be the smaller one and the two primes should be separated by spaces or newline. Trailing separator is allowed.

Input => Output examples:

6 => 11 13
42 => 59 61
1 => 3 5

Snippet for calculating the score

(Modification of the one in the Recover the mutated source code problem.)

var f=document.getElementById("f"),g=document.getElementById("s");function h(){var a=f.value,e=g.value,m=Math.max(a.length,e.length);if(10000<a.length)a="<span style='color:red'>First program is too long!</span>";else if(10000<e.length)a="<span style='color:red'>Second program is too long!</span>";else{if(0===a.length)a=e.length;else if(0===e.length)a=a.length;else{var d=[],b;for(b=0;b<=e.length;b++)d[b]=[b];var c;for(c=0;c<=a.length;c++)d[0][c]=c;for(b=1;b<=e.length;b++)for(c=1;c<=a.length;c++)d[b][c]=e.charAt(b-1)===a.charAt(c-1)?d[b-1][c-1]:Math.min(d[b-1][c-1]+1,Math.min(d[b][c-1]+1,d[b-1][c]+ 1));a=d[e.length][a.length]}a="Distance = "+a+"   Longer length = "+m+"   Score = <strong>"+(a+m)+"</strong>"}document.getElementById("d").innerHTML=a}f.onkeyup=h;g.onkeyup=h;
First program<textarea id=f rows=2 cols=80 style="width:100%"></textarea>Second program<textarea id=s rows=2 cols=80 style="width:100%"></textarea><p id=d></p>


In the answers' header let's use the format

[Language], [longer length] + [distance] = [final score].


Python 2, 60 + 32 = 92


Posted 2015-02-01T09:12:31.587

Reputation: 19 909



Pyth, 18 + 13 = 31

Collatz Sequence:


Twin Primes:


Try it here.

Some ideas thanks to FryAmTheEggman.


Posted 2015-02-01T09:12:31.587

Reputation: 39 268


CJam, 24 + 17 = 41 42

Collatz program:


Twin primes program:


Both take input from STDIN and print space/newline separated numbers to STDOUT

Try them online here


Posted 2015-02-01T09:12:31.587

Reputation: 25 836


CJam, 25 + 16 = 41

Collatz program:


Twin primes program:


Test it here.

I just golfed both of them for now. I'll see if I can reduce the score somehow.

Martin Ender

Posted 2015-02-01T09:12:31.587

Reputation: 184 808


05AB1E, 14 + 13 10 9 = 27 24 23

-4 score thanks to ASCII-only

Collatz sequence (14 bytes):


Try it online!

Twin primes (14 bytes):


Try it online!

Golfed Twin primes (11 bytes):


Try it online!


Posted 2015-02-01T09:12:31.587

Reputation: 554

Collatz link is incorrect – ASCII-only – 2019-01-19T01:22:02.990

Thanks, fixed it! – Wisław – 2019-01-19T01:32:26.950

primes, -3 score. probably worth keeping golfed version in the answer as a side note in case there's a different, better golf (what exactly does pair do, btw? like why do i have to do ,,, and why does only Ð, work there) – ASCII-only – 2019-01-19T01:42:18.983

-1 more (edit: never mind, first number should be smaller. score 24 for now – ASCII-only – 2019-01-19T01:57:08.350

22? – ASCII-only – 2019-01-19T02:02:38.600

Thanks! Guess I was more focused on trying to golf each individual challenge rather than minimize the score. I am not quite familiar with the Levenshtein distance :)

Pair pops a, b and pushes [a,b]. Note that the pair instruction is not a comma :). The comma is actually the print instruction ,. Compare these programs: two prints and [pair]( "05AB1E – Try It Online). Note that the two different characters look identical in the font. :)

– Wisław – 2019-01-19T02:08:28.297

24 is the shortest correct one i think – ASCII-only – 2019-01-19T02:19:08.477

score 23 if it is allowed to have output being a list of a list – Wisław – 2019-01-19T02:22:33.430

Let us continue this discussion in chat.

– ASCII-only – 2019-01-19T02:24:22.750


Pyth, 20 + 14 = 40 34



Prime Pairs:


Both are programs that take input from STDIN and output the numbers on newlines. Just golfed both answers while using the same looping primitive for now. This could probably be improved quite a bit.

Edit: Added better condition checking stolen from @isaacg. Seems like using filter is still shorter than using a while loop for the prime pairs.

Try it online here.


Posted 2015-02-01T09:12:31.587

Reputation: 16 206

This outputs 3 5 for the input 3 on prime pairs. It should output 5 7. – isaacg – 2015-02-01T23:40:30.807

@isaacg Fixed it, costed 3 :( – FryAmTheEggman – 2015-02-02T00:54:14.887


Java 8, 118 + 84 = 202



Twin primes:

n->{t:for(int p,z;;n++){for(z=n;z<n+3;z=z+2)for(p=2;p<z;p++)if(z%p<1)continue t;System.out.print(n+" "+(n+2));break;}}


Posted 2015-02-01T09:12:31.587

Reputation: 10 485

mention Java 8 .. – Optimizer – 2015-02-01T21:05:00.490


><>, 116 + 86 = 202

Collatz program (46):


Twin primes program (116):

0v?*r$0~&< .561~&<.1

Ouch. Both programs start with the same atoi function, but after that twin primes goes downhill. The same piece of code is repeated twice for primality check — it might be slightly shorter to somehow reuse the piece, but the setup for it wouldn't save many bytes.

Could do a lot better by throwing the back half of twin primes into the unused spots of the Collatz program, but I'm not sure if that's allowed.


Posted 2015-02-01T09:12:31.587

Reputation: 58 729

3"Golf it, Golf it NAO!" - Noanold Golfzenegger. That said, the question doesn't seem to say that adding in garbage to reduce the Levenshtein distance is taboo, so I would go nuts ;) – FryAmTheEggman – 2015-02-02T04:20:21.423


Mathematica, 53 + 29 = 82

Collatz Sequence:


Twin primes program:



Posted 2015-02-01T09:12:31.587

Reputation: 23 988


C++ Distance = 114 Longer length = 155 Score = 269

Task 1

void c(int N){while(N>1){cout<<N<<' ';N=(N%2==0)?N/2:N*3+1;}cout<<N;}

Task 2

int p(int x){int z=0;for(int i=2;i<x;i++){if(x%i==0){z=1;break;}}return z;}
void c(int N){N=(N%2==0)?N+1:N+2;int M=N+2;while(p(N)+p(M)>0){N=M;M+=2;}cout<<N<<' '<<M;}

Task 2 improved

int p(int N){int z=0;for(int i=2;i<N;i++){if(N%i==0){z=1;break;}}return z;}
void c(int N){N=(N%2==0)?N+1:N+2;while(p(N)+p(N+2)>0){N+=2;}cout<<N<<' '<<N+2;}


Posted 2015-02-01T09:12:31.587

Reputation: 1 235

I think you did not even try to reduce the score. For instance, you have different function names, different argument name different return type and what not which are increasing your distance. – Optimizer – 2015-02-01T11:15:23.190

@Optimizer I don't know of another way to test for primes? Other languages have this built in. – bacchusbeale – 2015-02-01T18:24:29.123

1I was not even talking about changing the algo. For instance, this code has only 102 distance: int p(int x){int z=0;for(int i=2;1<x;i++){cout<<x<<' ';x=(x%2==0)?x/2:x*3+1;}cout<<x;return z;} – Optimizer – 2015-02-01T18:32:04.550