Junior Arithmancer

ifcomp2018-game
#1

A great game for those who love mathematical puzzles. I liked it better than A Beauty Cold and Austere - chiefly because it’s more focused.

I got 101 points - 100 for doing everything I was told to do and 1 for

visiting Infinity.

Is it possible to get more?

#2

I haven’t run into any more either, but:

I was surprised that there wasn’t anything for causing an overflow (though it was noted by one of the examiners).

#3

Another terrific game about math by Mike Spivey. I’ve posted a review of it on my blog.

#4

Hmm…

[spoiler]I got a bonus point, and I’m pretty sure it was for causing an overflow. (So then I had to work out how to exceed 2 million without hitting overflow.)

How did you hit infinity? Divide by zero? Maybe there’s only one bonus point available that can be acquired at either overflow or infinity? Although infinity is really also overflow…[/spoiler]

#5

I too got 101 points, but for the same reason you did. I had a hunch that the number −900 would give another bonus point. But it didn’t (though it’s a nice challenge to try to calculate it):

It only resulted in an ‘inky blackness’ colour (which the game didn’t recognise in any special way).

The game placed low on my personal randomised list of games to play, so I only got a chance to play it after the competition had ended. But I’m very glad I did play it. It’s an excellent game. While it’s basically just a bunch of arithmetic puzzles, it had lots of nice little touches, and the running commentary made it so much fun! I liked that the examining committee even commented on meta-commands, like ‘save’ and ‘undo’ (and do try the ‘sing’ command!).

And the general design of the game was excellent too, e.g. that the various spells were slowly introduced, one at a time, and that you didn’t have to complete all the numbers to get the committee to accept you. Stuff like that makes the game much more accessible for non-die-hard number crunchers. And the addition of extra challenges, for us that like that sort of thing, was a very nice bonus.

(Doug Orleans) #6

Club Floyd also got the same bonus point as Sobol, and no others. We were hoping that 69,105 would get one, but no such luck. (Still a fun challenge to get there, though.)

In case you didn’t see the postmortem, Spike just posted version 2 (linked on IFDB), which adds more bonus points. I’m guessing our save file won’t work, though, and I’m not ready to start over from scratch just yet…

#7

I was hoping to get one by

creating as many consecutive elements of the Fibonacci sequence as possible

but that didn’t work either.

#8

I got a different bonus point by:

Hitting the rainbow colors in order. I actually didn’t notice the “in any order” in the instructions until I ran the sequence and then realized I’d gotten a bonus. I don’t remember what sequence I used, but I did use negative numbers to do it.

(Doug Orleans) #9

Oh, forgot to post here. DavidW on ifMUD discovered that one (I think by disassembling the bytecode?), and also negative 2 billion.

#10

I really enjoyed this game. So with some time off for Thanksgiving, I thought it would be fun to write a brute force solver for the numbers. It’s not tested thoroughly, but I think I did it right.

In case people are interested, this is what I learned (format is the number followed by total number of solutions, then 3/4/5/6 spell solutions) -

3.14159 (1488) 0/6/558/924
2.71828 (656) 0/35/105/516
1.41421 (1290) 0/84/466/740
1.73205 (248) 0/0/70/178
2.23606 (7607) 0/0/65/7542
1.61803 (176) 0/0/176/0
0.69314 (420) 0/0/56/364
1.64493 (228) 0/0/12/216
1.20205 (6469) 1/26/498/5844
0.57721 (186) 0/0/162/24

#11

Thanks for the hint. It’s a very nice challenge!

Note that it only works if you …

… haven’t previously completed the rainbow challenge. If you have, your solution isn’t acknowledged at all, and you don’t get a bonus point. I think this should be considered a bug. (I’ve only tested this in version 1 of the game, though, so it might be fixed in version 2.)

#12

(Author update here: I had Huftis send me the sequence he used to produce this bug, and I confirmed it in the Comp release of JA. I couldn’t track down exactly what part of the code caused the bug, but other sequences I tried that accomplish the same overall goal don’t produce the bug. So the problem is something specific having to do with Huftis’s sequence. Huftis’s sequence doesn’t produce the bug in Release 2, though, so I’m cautiously optimistic that the bug doesn’t appear in any form in Release 2. Author out.)