Metanet Forums

squibbles wrote:More often then not we ended up just writing programs to solve them for us, though.

OneSevenNine wrote:I'd join the race except for the fact that I can't find a single problem that I wouldn't just write a program to brute-force, other than possibly the first one which I could probably manage. Unless brute-forcing is how you're supposed to do it, which I doubt it is.

projecteuler.net wrote:I've written my program but should it take days to get to the answer?

Absolutely not! Each problem has been designed according to a "one-minute rule", which means that although it may take several hours to design a successful algorithm with more difficult problems, an efficient implementation will allow a solution to be obtained on a modestly powered computer in less than one minute.

Scrivener wrote:my calculator... the only thing i can do programming of any real significance on.

1. I need it to run extremely quickly.
2. It's a totally non-intensive doodad that I want for convenience.

Tsukatu wrote: You're free to get the answer by any means you want, including brute-forcing. But many of the later problems introduce subtle complications that would make the brute force take ages if not optimized in some clever way.
If you want to get technical, the way I've solved most of these is through a brute-force approach, but very few of my solutions have taken more than a few seconds because I reduced the problem space dynamically based on the ongoing results of the brute-force search, and that is not something a common clueless code monkey would think or be able to do.

Tsukatu wrote: Regarding Java, I am beside myself with joy to know that every CS professor I've had at UCSC despises Java. Fuck it, deep and in the skull, while its children watch.

OneSevenNine wrote:I've been thinking about trying teach myself python or something over the summer.

Scrivener wrote:BASIC programming.

Tsukatu wrote:

Scrivener wrote:BASIC programming.

Fine, there is one thing worse than Java.

incluye wrote:Incidentally, are there any better tactics than brute-forcing with Problem 1? (As in a for loop, starting at 3, incrementing by 1 each step, if divisible by 3 or 5 add to total sum)

((1..333).map { |i| i*3 } | (1..199).map { |i| i*5 }).inject(0) { |sum, n| sum + n }

threes = (1..333).map { |i| i*3 }
fives = (1..199).map { |i| i*5 }
union = threes | fives
union.inject(0) { |sum, n| sum + n }

taaveti wrote:Personally, I would add three times the sum of the natural numbers <= 333 (which is the sum of the multiples of three < 1000) and five times the sum of the natural numbers < 200 (which is the sum of the multiples of five < 1000), then subtract fifteen times the sum of the natural numbers <= 66 (which is the sum of the multiples of fifteen < 1000, i.e. those which were double-counted).
Each of these sums is the sum of natural numbers 1..N for some finite N, which is to say there is a closed form: N * (N+1) / 2. Net result: three increments, six multiplications, three right shifts (which could actually be done at the end, for only one right shift), an add, and a subtract.

Tsukatu wrote:http://www.projecteuler.net

I know some of you must have heard of this, because it was here that I first heard of it and started on it. I didn't get very far before I stopped and forgot my login information.
Semi-recently, a computer science-oriented club I'm affiliated with has been discussing some of these problems, so I've gotten back in the game.

I've only really worked on them during biweekly club meetings, and I finished problem #25 earlier today at one of those meetings. (I've been doing them strictly in order, without skipping.)

Has anyone else worked on this / is anyone working on these problems actively?

Zephyr wrote:Okay I made an account, I can only do a little bit of programming, but I guess I'll try.
Would WolframAlpha help?