@Runer112 there's a difference between not being allowed to win (and accepted) and not being allowed to answer

that certainly makes sense

People don't like my idea :( meta.stackexchange.com/a/250987/254466

But I don't understand the "Yes and No" versus the "Yes"

@Runer112 Optimizer is saying you're not allowed to answer if your language is specifically tailored to the task at hand.

@xnor because people like to be told, "yes, you're the best"?

so it's basically "Yes"

except with an added caveat

All the answers to the Language Design challenge are invalid :P

yes... and that caveat is "and no"

the bolded words being totally different from the "Yes" answer confused me

made me think it was somehow wildly different

I just realized I read APL without an internal voice in my head.

@PhiNotPi I don't think I'll post my Snake language. Most of the other languages are super-sets of it.

@Optimizer Base 95 doesn't have any benefit over base 85 unless you use big integers in the conversion.

@MartinBüttner but people who don't code in cjam or pyth or the like also want to be told "yes, you're the best" sometimes

I grew up with parents who told me that it was alright to come in second. I intend to tell my children that winning is awesome.

I'm curious to know @Geobits tactics with his child in regards to being a winner.

I don't need to be told that I'm the best.

He already knows through experience. While not everybody can come in first at everything, winning is a damn good feeling.

I didn't get that experience. I sucked at Basketball, Baseball, Soccer, and finally Football. Then I got into tennis and I was number one but 1) nobody cares about tennis and 2) I was already 17 by that point

@PhiNotPi See, this is what I meant.

Hmmm. That sentence needs revision

Two minutes to think before I become a whore

There we go.

Is it just me, or does the new UI for self-duping have the big "Yep, I'm a dupe" button a bit too close to the dupe target link? It's an irreversible action on the OP's part, right?

@Geobits Yeah, I saw that.

@Geobits It doesn't exactly say "Yep, I'm a dupe". I suggested that it should, though.

I saw. I'm undecided about the wording issue, though.

@TheBestOne How did you evolve it?

I'm about to run a tournament.

I tested it against the posted bots to get the fitness score. I used tournament selection with no crossover.

Pairwise Results:
CopyPasta Dwarf
CopyPasta Imp
Janitor CopyPasta
Evolved CopyPasta
Dwarf Imp
Janitor Dwarf
Evolved Dwarf
Janitor = Imp
Evolved Imp
Evolved Janitor
Leaderboard:
    8 - Evolved
    5 - Janitor
    4 - CopyPasta
    2 - Dwarf
    1 - Imp
The tounrnament took 2254.435503281 seconds, or 37.573925054683336 minutes.

Was that with the official parameters?

yes

Also, if you want to test multiple versions of the bot, you can put the old leaderboard (without your bot) into the leaderboards.txt file.

Did you run it on the "newer" version of the controller? (the latest "important" update occurred 8 hours ago)

The program I used to evolve directly called Game.run().

Okay

I downloaded it a couple hours ago.

Then this leaderboard can be official.

I'm going to try to dissect the evolved bot.

Results confirmed :)

Pairwise Results:
Evolved CopyPasta
Evolved Imp
Evolved Janitor
Evolved Dwarf
Leaderboard:
    8 - Evolved
    5 - Janitor
    4 - CopyPasta
    2 - Dwarf
    1 - Imp
The tounrnament took 341.223506214 seconds, or 5.6870584369 minutes.

PS I'm fixing that typo in "tounrnament"

:)

I profiled your code a little bit and found that most of the time is spent in Game.run()

Well, yes.

That's exactly what I would expect.

Do you have the exact results of the profiling?

Also, does that thing about Game.run() include the time spent in Game.execute()?

No to both

Well, Game.run() only really does two things: copy the code of the contestants into the core, and then do a while loop of executing commands.

So maybe the copying is taking all of the time?

dropbox.com/s/k22rmx85d1d8cq5/profiled.nps?dl=0

It's an .nps document.

How do I open that?

VisualVM load snapshot (I think)

okay

Are ArrayLists substantially slower than arrays?

Like, much slower?

The player's code is stored as an ArrayList because of how it is constructed one line at a time.

The core is an array.

I think they only make a tiny difference.