Thread: "MagicTile v2 Preview update"

--0016e6db2e4099b3a204a3a18f26
Content-Type: text/plain; charset=ISO-8859-1

Hi All,

I've made some changes this week to the "render-to-texture" code (making the
OpenGL calls more correct), and this is allowing the program to work on a
wider variety of OS/video card setups. The latest build is available here:

www.gravitation3d.com/magictile/downloads/MagicTile_v2_Preview.zip

(I'll shortly break the old link, so I can keep updates at the date-agnostic
one above.) These changes got everybody going who reported problems, or if
there were still issues, they could be understood as limitations of the
video card. Please feel free to let me know of any further troubles you run
across.

I want to say thanks to Ed and Matt Sheerin for reports and help. And a
special *uber-thanks* to Nan and Matt Galla, who both spent significant
back-and-forth with me to understand symptoms. You guys rock!

What I seriously need at this point is some help with these {5,5} puzzles...
only 6 colors, but I joke not when I say I don't even remotely know how to
start. Every twist touches all 6 faces!

Cheers,
Roice

--0016e6db2e4099b3a204a3a18f26
Content-Type: text/html; charset=ISO-8859-1
Content-Transfer-Encoding: quoted-printable

Hi All,

> What I seriously need at this point is some help with these {5,5} puzzles=
...
> only 6 colors, but I joke not when I say I don't even remotely know how t=
o
> start. Every twist touches all 6 faces!
>=20
> Cheers,
> Roice
>

I just had another look at them, and realised the first one is even more co=
nfusing than I first thought. It's to do with the colour scheme, pieces wi=
th identical colour combinations don't have to be found between the same pa=
irs of faces. Twist the green face for instance, there is a green/yellow e=
dge between the green and red faces, and a different green/yellow edge (sam=
e piece type) between the green and white faces. That's going to make any =
solve even more complicated, and it already looks pretty difficult. I'll w=
ant to solve some of these, but I may struggle a lot! Otherwise, the progr=
am looks amazing, cheers Roice.

Matt S

--------------020503080109010201080302
Content-Type: text/plain; charset=ISO-8859-1; format=flowed
Content-Transfer-Encoding: 7bit


On 5/19/2011 7:10 AM, Roice Nelson wrote:
> [...]
> What I seriously need at this point is some help with these {5,5}
> puzzles... only 6 colors, but I joke not when I say I don't even
> remotely know how to start. Every twist touches all 6 faces!

Ha! So now we know the answer to the age-old riddle: Can Roice make a
puzzle so hard that even he can't solve it?" It sees the answer is
"Yes". The funny thing is that congratulations are in order either way!
I know that this group will not allow a new puzzle to go unsolved for
very long. I wonder who will crack this nut.

-Melinda

--------------020503080109010201080302
Content-Type: text/html; charset=ISO-8859-1
Content-Transfer-Encoding: 7bit




http-equiv="Content-Type">




On 5/19/2011 7:10 AM, Roice Nelson wrote:
cite="mid:BANLkTikcOOTf5gVECHj4boVHP8Ux0mS=ww@mail.gmail.com"
type="cite">

[...]

--000325554c7e036bdb04a3abd7fb
Content-Type: text/plain; charset=ISO-8859-1

On Thu, May 19, 2011 at 12:42 PM, Matthew wrote:

> I just had another look at them, and realised the first one is even more
> confusing than I first thought. It's to do with the colour scheme, pieces
> with identical colour combinations don't have to be found between the same
> pairs of faces. Twist the green face for instance, there is a green/yellow
> edge between the green and red faces, and a different green/yellow edge
> (same piece type) between the green and white faces. That's going to make
> any solve even more complicated, and it already looks pretty difficult.
> I'll want to solve some of these, but I may struggle a lot! Otherwise, the
> program looks amazing, cheers Roice.
>
> Matt S


Yeah, they're weird! I bet we'll see parity effects on these puzzles as a
result of what you are observing (say when you unknowingly put one of those
green/yellow pieces in the location of the other).

I'm playing with the "Petal" version, which will be significantly simpler
(doubt I'll ever be able to solve the others). After looking a little more,
I think one key to solving might be working with what is the white face by
default. That one is different than all the others because it has 5
different color combinations of edge pieces instead of 3. Doing a
commutator with it and one adjacent face will cycle just 3 edges, whereas
commutators with other adjacent faces permutes around a lot more. I just
noticed a commutator of two of the other faces that only share a vertex
alters only some of the 1C pieces, so that should be useful too.

seeya,
Roice

--000325554c7e036bdb04a3abd7fb
Content-Type: text/html; charset=ISO-8859-1
Content-Transfer-Encoding: quoted-printable