Pyramid Puzzle lost and wanted
Dave Whiteley
Several years ago my wife and I visited San Fransisco's Pier 39, and found a
puzzle shop. We bought a fiendishly difficult puzzle which consisted a three
sided pyramid formed from (yellow) plastic spheres on a black plastic base.
The side of the pyramid was (if I remember correctly) 5 balls long.
It dismantled into 8 or 9 pieces. I remember that the nasty pieces had the
balls in a row of 3 or 4 with one at a right angle at one end.
I loaned this puzzle to someone, and I have not seen it since.
EITHER
Is this puzzle still available anywhere, and can I buy it by mail order.
OR
Does anyone else have the puzzle, and can they post or send me the pattern
of the pieces so that I can make myself a version of it.
Thanks in advance,
Dave Whiteley
_
Dave Whiteley d.l.wh...@elec-eng.leeds.ac.uk
Department of Elec Eng, The University of Leeds,
Woodhouse Lane, Leeds. LS2 9JT
Phone 0113 2332059 Messages 0113 2332000
Fax 0113 2332032
My opinions are mine - probably not theirs, whoever they are.
HAVE FUN and BE KIND TO BATS
Tim Udall
I have a version of this puzzle. I purchased it about 3 years ago,
disassembled it without paying much attention and have not been able to
reassemble it yet (of course I have not been trying continuously). On
the base, is a label with this info:
Giant Pyramid - a very difficult puzzle.
Item No 186.
Gordon Brothers
P.O. Box 4614
Chico, CA 95927
Here's what the pieces look like:
There are 9 pieces of equal sized attached spheres.
Each will lie flat, that is, all the pieces are "planar".
They assemble into a tetrahedral shape with an edge of 5 spheres.
4 of these. The angle is 90 deg.
O O O
O
And 1 each of these. All angles are 60 deg.
O O O O O O O O O O O O O
O O O O O O
Dave Whiteley
In article <3590F5EC...@openroute.com>,
Yes, thats the puzzle I was seeking.
Next question.... Has anyone solved this puzzle, or do the pieces sneakily
re-configure themselves when you first dismantle it.
I now have a pile of glass marbles and epoxy resin on my desk, but not in a
pyramid.
Dave
Johan Myrberger
I too have this puzzle in pieces on my desk... If anyone has the
solution please let me know...
Or better.. Has anyone written some software to solve this kind of
sphere puzzles in a generic way?
Best regards
/Johan Myrberger
mailto:Johan.M...@ebc.ericsson.se
http://home.bip.net/johan.myrberger
Chris Hartman
Johan Myrberger wrote:
>
> I too have this puzzle in pieces on my desk... If anyone has the
> solution please let me know...
See http://www.gamepuzzles.com/prpuzzls.htm for what I _think_
is a place to buy this puzzle. (I didn't buy mine there, so
I can't be sure it's the same pieces, but I assume it is.)
Great site for other puzzles, too.
Here's my solution. Extremely easy Obpuzzle: figure out the
coordinate system and how I named the pieces. (See below
for the original description if you missed the original thread.)
[No spoiler space necessary, as the solution needs "decoding"
anyway.]
[0,0,0,4] is P.
[0,0,1,3] is P.
[0,0,2,2] is L1.
[0,0,3,1] is L4.
[0,0,4,0] is J.
[0,1,0,3] is P.
[0,1,1,2] is L2.
[0,1,2,1] is L1.
[0,1,3,0] is L4.
[0,2,0,2] is P.
[0,2,1,1] is L2.
[0,2,2,0] is L1.
[0,3,0,1] is C.
[0,3,1,0] is L2.
[0,4,0,0] is C.
[1,0,0,3] is S.
[1,0,1,2] is S.
[1,0,2,1] is L4.
[1,0,3,0] is J.
[1,1,0,2] is L3.
[1,1,1,1] is S.
[1,1,2,0] is S.
[1,2,0,1] is C.
[1,2,1,0] is L1.
[1,3,0,0] is L2.
[2,0,0,2] is I.
[2,0,1,1] is L4.
[2,0,2,0] is J.
[2,1,0,1] is L3.
[2,1,1,0] is J.
[2,2,0,0] is C.
[3,0,0,1] is I.
[3,0,1,0] is L3.
[3,1,0,0] is L3.
[4,0,0,0] is I.
> Or better.. Has anyone written some software to solve this kind of
> sphere puzzles in a generic way?
Well, if enough people bug me about it, I'll give out the program I
used to find this solution. It isn't generic, though. It makes
heavy use of the fact that the "L" pieces cannot go into the corners
of the tetrahedron. Actually, here's the description of how I
went about it:
Backtracking search (obviously). First, place the four "L"s. Then,
starting from each corner of the tetrahedron, try each of the remaining
pieces.
(This is better than trying each of the remaining pieces everywhere.
The reason is (as is quickly found when playing with the real puzzle)
that most of the "L" configurations have corners where no piece will
fit. We cut this branch of the search tree much quicker using this
method.) Then, see if the last piece matches the remaining space.
The solution above took just over a minute to compute on my R10000
SGI O2. (That's a 150Mhz MIPS chip).
When I wrote this code as an undergraduate eight years ago it took
about 45 minutes on the school VAX.
-Chris
> Tim Udall wrote:
> >
> > I have a version of this puzzle. I purchased it about 3 years ago,
> > disassembled it without paying much attention and have not been able to
> > reassemble it yet (of course I have not been trying continuously). On
> > the base, is a label with this info:
> >
> > Giant Pyramid - a very difficult puzzle.
> > Item No 186.
> > Gordon Brothers
> > P.O. Box 4614
> > Chico, CA 95927
> >
> > Here's what the pieces look like:
> >
> > There are 9 pieces of equal sized attached spheres.
> > Each will lie flat, that is, all the pieces are "planar".
> > They assemble into a tetrahedral shape with an edge of 5 spheres.
> >
> > 4 of these. The angle is 90 deg.
> >
> > O O O
> > O
> >
> > And 1 each of these. All angles are 60 deg.
> >
> > O O O O O O O O O O O O O
> > O O O O O O
--
Prof. Chris Hartman http://tigger.cs.uaf.edu/hartman
har...@arsc.edu
University of Alaska, Fairbanks and Arctic Region Supercomputing
Center
Dave Whiteley
Previous stuff snippy snip snip....
Many thanks.... I now have a pyramid of balls on my desk.
I have made two versions, one using glass marbles and one using pulp balls
(sometimes known as cotton balls) glued together with PVA glue.
Sods law now decrees that I will find my original lost puzzle.
Thanks all.
(Who can I inflict the puzzle on now?)