Tuesday, 27 March 2012

One puzzle masquerading as another…

I think I’ve already mentioned a particular example of this sort of beast in my description of a little example of James D’s mischievousness when he fished out a little puzzle from one of his multitude of drawers and asked the group who the entanglement whiz was – we  all pointed to Kevin and he was given a pair of bent nails to separate – when Kevin noticed that they didn’t separate the way he thought they should, James upped the ante by suggesting that we put a stopwatch on Kevin’s performance … a minute or so later James put him out of his misery by pointing out that the head on one of the nails unscrewed and that was the only way to separate the nails…beware of the puzzle that masquerades as another!
On a totally unrelated topic :-)  allow me to present a couple of Roger puzzles:
I picked up a copy of r2d2 from Wil Strijbos at Peter Hajek’s end of year puzzle party and it’s a delightfully simple-looking little puzzle. The solid cast aluminium block has a pair of channels running between opposite corners and intersecting in the centre. One of them has a small ball bearing and the other has a pair of wider sections containing larger ball bearings that effectively throttle the first channel – in fact the shape of those wider sections is such that the small ball will merrily pass unobstructed in one direction but get itself thoroughly stopped when trying to pass in the other direction … and you surmise that you will be trying to move the ball to the other end.
The clear Perspex cover stops you from using any form of direct manipulation, and using something like a magnet introduced externally simply wouldn't be sporting … it’s worth checking that the two hex bolts securing the cover don’t have any other function – they don’t and then it’s probably worth playing with it as though it were a dexterity puzzle for a while – if only to convince yourself that there isn’t a sneaky little trick or that a bit of bumping will see it right.
Then what…
I decided to explore some physics, and managed to chance upon a solution that worked fairly efficiently – except that when I mentioned my solution to Wil, he replied quite quickly that that wasn’t the solution that Roger was expecting – so I went back to the drawing board for a while and couldn’t come up with anything else – but as luck would have it I happened to be down at the Puzzle Museum a couple of weeks later and James helpfully left a copy of the official solution lying around for me (I had no idea such things existed for Rogers!) – and after reading the description, and thinking about it for a while, I think I’m going to stick with my own solution for this puzzle – I think it’s a lot more elegant :-)
Last Saturday I managed to pick up a copy of Alles Roger from a fellow blogging collector (why does that sound vaguely rude?) who was looking to thin out his collection a little – I’ve learnt not to pass up on opportunities to pick up Rogers I don’t already have as they come up for sale so seldom that they’re invariably changing hands for quite a bit on auctions.
Alles Roger has a similar conceptual layout to r2d2 – a path for a small ball bearing that needs to pass through a pair of larger ball bearings in channels that make them obstruct progress in one direction, and an initial little obstacle to pass in the form of another small ball bearing – basically you’re trying the move the small ball from the top right hand corner down to the channel along the bottom.
The first obstacle is trying to get past a small ball bearing in a parallel channel that is shaped and sloped to stop any direct or agricultural attempts at getting past it – but if you examine it carefully and experiment a little with some more subtle moves, you’ll find it can be beaten quite easily – sadly however, the hard bit is yet to come: the two larger balls obstruct the path down to the lower channel in virtually any orientation that might possible lead to the small ball going in the right direction … and applying that little bit of obvious physics is well nigh impossible because of the positioning of the down-channel off to the left …
Back at home after MPP5 I had a wee brainwave and experimented a little and found a fantastic solution – while it might not look particularly elegant and might not be the sort of thing you’d choose to do in front of another puzzler (!) it solves the puzzle, reliably, in seconds, every time – the trick is to think outside the box – WAY outside the box … but then, it’s a Roger!


  1. Those Roger puzzles looks so cool. I'm definitely looking forward to seeing more of those. Maybe even snag one eventually.

  2. Douglas Cameron28 March 2012 at 10:45

    I'm curious to know what the official solution is to r2d2 as I too solved it using some basic physics.

  3. Hi Douglas - email sent! - allard

    1. I too am curious. Our elegant solution just seems so right!

    2. ...using phsics is expected, using dexterity isn't... :-) Email me direct if you want more info.

  4. Hi Allard, I have stuck on "Alles Roger" for a long time ... Can you please give me some hints? Is it possible to solve it without using any tools ? My email is: 240388@gmail.com. Thank you.