Author |
Topic: physical CS toy puzzles (Read 1351 times) |
|
amichail
Senior Riddler
Posts: 450
|
 |
physical CS toy puzzles
« on: Jun 23rd, 2005, 6:26pm » |
 Quote  Modify
|
Can we build physical toy puzzles to demonstrate CS ideas?
I think this would be one way to generate interest in the field.
How would you build a physical puzzle to demonstrate various sorting
algorithms?
Here's a simple idea for a computer game that perhaps could be turned
into a physical board game.
You have two players, one player playing the role of a sorting
algorithm while the other player plays the role of the adversary.
Initially, you start with n points on a horizontal line representing the inputs.
One player picks two points to compare. The other player determines
the result of the comparison. The points then get linked, with the
larger element ending up above the smaller element.
So the players play with a partial order that eventually ends up being
a sorted sequence represented by a vertical line of linked points.
One player (the algorithm) aims to sort as quickly as possible while
the other player (the adversary) tries to slow down the sort as much
as possible.
|
|
 IP Logged |
DropZap - a new kind of block elimination game
|
|
|
towr
wu::riddles Moderator
Uberpuzzler
Some people are average, some are just mean.
Gender: 
Posts: 13730
|
|
Re: physical CS toy puzzles
« Reply #1 on: Jun 24th, 2005, 12:14am » |
Quote Modify
|
on Jun 23rd, 2005, 6:26pm, amichail wrote:| Can we build physical toy puzzles to demonstrate CS ideas? |
|
Mazes? Mastermind? Chess 
Quote:How would you build a physical puzzle to demonstrate various sorting
algorithms? |
|
I'd simply hand over a shuffled deck of cards and suggest different ways to sort it and to find out which is better.
|
|
IP Logged |
Wikipedia, Google, Mathworld, Integer sequence DB
|
|
|
Noke Lieu
Uberpuzzler
pen... paper... let's go! (and bit of plastic)
Gender:
Posts: 1884
|
|
Re: physical CS toy puzzles
« Reply #2 on: Jun 26th, 2005, 11:26pm » |
Quote Modify
|
nice. A little dry though? Although thats part of why we get computers to do it for us...
To spice it up though, you could remove one card. Or have two "five of spades"?
A friend of mine occassionally talks about something like this with lengths of spaghetti. and you have too find the longest one. OR the middlest one- different lengths reqire different techniques.
|
|
IP Logged |
a shade of wit and the art of farce.
|
|
|
towr
wu::riddles Moderator
Uberpuzzler
Some people are average, some are just mean.
Gender:
Posts: 13730
|
|
Re: physical CS toy puzzles
« Reply #3 on: Jun 27th, 2005, 12:35am » |
Quote Modify
|
on Jun 26th, 2005, 11:26pm, Noke Lieu wrote:nice. A little dry though? Although thats part of why we get computers to do it for us...
To spice it up though, you could remove one card. Or have two "five of spades"? |
|
Or shuffle two decks, and only take part (some cards will be missing, some will be double).
Quote:| A friend of mine occassionally talks about something like this with lengths of spaghetti. and you have too find the longest one. OR the middlest one- different lengths reqire different techniques. |
|
You can very easily find the longest one, by standing the spagetti on end, and let gravity 'sort' them (then just pick out the one sticking out). That's a trick you can't do in CS.
|
|
IP Logged |
Wikipedia, Google, Mathworld, Integer sequence DB
|
|
|
rmsgrey
Uberpuzzler
    
Gender:
Posts: 2874
|
|
Re: physical CS toy puzzles
« Reply #4 on: Jun 27th, 2005, 3:42am » |
Quote Modify
|
on Jun 27th, 2005, 12:35am, towr wrote:You can very easily find the longest one, by standing the spagetti on end, and let gravity 'sort' them (then just pick out the one sticking out). That's a trick you can't do in CS.
|
|
Unless you allow parallel processing...
|
|
IP Logged |
|
|
|
towr
wu::riddles Moderator
Uberpuzzler
Some people are average, some are just mean.
Gender:
Posts: 13730
|
|
Re: physical CS toy puzzles
« Reply #5 on: Jun 27th, 2005, 4:24am » |
Quote Modify
|
on Jun 27th, 2005, 3:42am, rmsgrey wrote:| Unless you allow parallel processing... |
|
You can certainly speed up sorting, but I don't see a way to properly do the equivalent of 'sort by gravity'.
|
|
IP Logged |
Wikipedia, Google, Mathworld, Integer sequence DB
|
|
|
rmsgrey
Uberpuzzler
Gender:
Posts: 2874
|
|
Re: physical CS toy puzzles
« Reply #6 on: Jun 28th, 2005, 6:07am » |
Quote Modify
|
on Jun 27th, 2005, 4:24am, towr wrote:
You can certainly speed up sorting, but I don't see a way to properly do the equivalent of 'sort by gravity'.
|
|
One way is to start a bunch of processes counting simultaneously, and see which one reaches its value first.
|
|
IP Logged |
|
|
|
towr
wu::riddles Moderator
Uberpuzzler
Some people are average, some are just mean.
Gender:
Posts: 13730
|
|
Re: physical CS toy puzzles
« Reply #7 on: Jun 28th, 2005, 6:09am » |
Quote Modify
|
But they all have to communicate with each other to decide who reached the maximum/minimum value. And you'd need log n interactions for that, wouldn't you?
|
|
IP Logged |
Wikipedia, Google, Mathworld, Integer sequence DB
|
|
|
Grimbal
wu::riddles Moderator
Uberpuzzler
Gender:
Posts: 7527
|
|
Re: physical CS toy puzzles
« Reply #8 on: Jun 28th, 2005, 6:16am » |
Quote Modify
|
on Jun 27th, 2005, 3:42am, rmsgrey wrote:
Unless you allow parallel processing... |
|
Just like spaghetti, it is much harder if you don't allow parallel spaghetti... 
|
|
IP Logged |
|
|
|
rmsgrey
Uberpuzzler
Gender:
Posts: 2874
|
|
Re: physical CS toy puzzles
« Reply #9 on: Jun 28th, 2005, 6:22am » |
Quote Modify
|
on Jun 28th, 2005, 6:09am, towr wrote:| But they all have to communicate with each other to decide who reached the maximum/minimum value. And you'd need log n interactions for that, wouldn't you? |
|
Depends on your architecture. I'm well beyond what I comfortably know, so I'm stabbing wildly in the dark, but knowing how many processes are running at any given time would be one way of tracking.
|
|
IP Logged |
|
|
|
|
RIDDLES SITE
WRITE MATH!
Home 
Help 
Search 
Members 
Login 
Register
Reply
Notify of replies
Send Topic
Print