http://circuits.puzzlebaron.com/index.php
My best scores are about 80 seconds. For that to happen, I need to make some inspired guesses early so the end game is not too messed, and then efficiently clean up the end game to get to the solution.
But people are solving these in 30 seconds. That’s just clicking continuously without making any mistakes. Is this really possible?
What size? It seems like the best strategy is work from the outside in, since everything HAS to be connected. There’s not going to be anything leading off the screan.
Sorry - medium 9x9
yes, definitely. The outside squares don’t require any guessing.
1/3 of a second per square? No, it’s not possible, at least not for me. On a good day, with random two digit numbers in each square, I might be able to click on all the odd numbers in that time.
By the way, thanks loads for pointing me to yet another internet time sink.
If you want a mobile version of this puzzle and a bunch of others, check out Simon Tatham’s Puzzles (at least for Android). This one is called “Net”; there are a couple dozen others of varying styles included in the same app.
I generally prefer to solve without backtracking. The Tatham puzzles allow you to lock in a move that you know is correct, which I think is a nice feature. Otherwise you lose track of which moves were valid and which were just temporary assumptions.
I wonder if this involves hacking the computer clock
More likely, it involves writing some sort of code for solving the puzzle using a bot.
???
What numbers? The puzzle is about rotating tiles to complete circuits. There’s no numbers.
That game makes my head hurt. I prefer Linkz.
I just tried a 7x7 and got ‘Slow.’ I don’t know if I’ll persevere and attain ‘Average.’
One trick might help: In the solution to the 7x7 I tried there were no superfluous (dead-end) edges — every connection was needed. This is a big constraint on some of the cells. (For example, no edge abuts the border, but there are other applications.)
And sure enough … I applied this insight to a 9x9 just now and got … Slow!
Another possibility is that they’re pausing the timer somehow as soon as they see the initial puzzle, working it out on scratch paper or the equivalent, and then re-starting the timer and just clicking their pre-solved solution.
I looked through the source, and they don’t make it trivial to do this. The time is computed on the server, and they require submitting the solved board state (which is presumably double-checked there). Of course it would still be possible to build a bot, but it’s not just a matter of changing the timer or a bit that decides the “win” state.
I just had a very good game - no missteps, the ending just needed routine cleanup. 77 seconds. The high score was 33. I could have clicked faster and not rotated the wrong way a couple of times.
I found them here, and he designs them to work on Windows, Mac, and Linux, too. It’s a great collection, with more features, like the ability to undo back to where you got stuck and see where you went wrong. I agree the ability to lock in squares makes it much, much easier.
You might want to try the version (“netgame”) in the link I posted above. It has an option that allows links to wrap from side to side and top to bottom, so the outside squares can’t be done at a glance.
I have been playing this game way too much these past couple of weeks. Thanks! Anyway, I just got a 68-second time, with virtually no mistakes. My next best time is like 90some seconds. I can see someone who plays way more than I do, and maybe uses a touchscreen, getting a crazy-fast time. It definitely has patters and acquired logic in its play.