Thread: "120-cell solved..."

Hi all,
I've just finished solving of 120-cell. In MPUlt it was not difficult at =
all. I've used 4 features of the program - macro, setup sequencies, cubie h=
ighlighting and fast temporary changing sticker color to white. With these =
features I almost didn't care about the actual colors of cells.
My method is the following: I solve 2C, 3C and then 4C. For each type I s=
elect some cell, change it to white, then move all cubies with white sticke=
rs to this cell (using setup unwinding and a couple of macros). My goal her=
e is to make all stickers of the current type on that face "white". Then I =
use another macro to sort cubies in that cell (it is a half on 3-cycle that=
moves two pieces from the cell and one outside). Non-macro twists here are=
twists of one cell, so I don't need to use F1/F2/F3 keys for this stage. T=
hen I go to other cell and so on.
At first my estimate of time for complete solving was 96 hours (I used ab=
out 2 minutes per cubie). So I didn't care about the time, and often left t=
he puzzle with running timer for a while. I think that I lost about 6 hours=
of displayed time here. But at the end of 3C stage I found that situation =
is much better than I thought about it: timer displayed 57h 05m! So if I'll=
be solving 40 pieces per hour, there will be good chances to finish at 72h=
:)
Actual time was 69h 58m 30s. And I spent last 30 minutes to resolve one f=
lipped 4C - tried to solve it like in 3^4, but coldn't remember the proper =
algorithms :)
Andrey

Andrew

Congratulations! By the way thank you again for making MPUlt. I wonder what=
will be the next challenge for you.

Nan


--- In 4D_Cubing@yahoogroups.com, "Andrey" wrote:
>
> Hi all,
> I've just finished solving of 120-cell. In MPUlt it was not difficult a=
t all. I've used 4 features of the program - macro, setup sequencies, cubie=
highlighting and fast temporary changing sticker color to white. With thes=
e features I almost didn't care about the actual colors of cells.
> My method is the following: I solve 2C, 3C and then 4C. For each type I=
select some cell, change it to white, then move all cubies with white stic=
kers to this cell (using setup unwinding and a couple of macros). My goal h=
ere is to make all stickers of the current type on that face "white". Then =
I use another macro to sort cubies in that cell (it is a half on 3-cycle th=
at moves two pieces from the cell and one outside). Non-macro twists here a=
re twists of one cell, so I don't need to use F1/F2/F3 keys for this stage.=
Then I go to other cell and so on.
> At first my estimate of time for complete solving was 96 hours (I used =
about 2 minutes per cubie). So I didn't care about the time, and often left=
the puzzle with running timer for a while. I think that I lost about 6 hou=
rs of displayed time here. But at the end of 3C stage I found that situatio=
n is much better than I thought about it: timer displayed 57h 05m! So if I'=
ll be solving 40 pieces per hour, there will be good chances to finish at 7=
2h :)
> Actual time was 69h 58m 30s. And I spent last 30 minutes to resolve one=
flipped 4C - tried to solve it like in 3^4, but coldn't remember the prope=
r algorithms :)
> Andrey
>

Nan, thank you!
I don't know about next challenge. It should be some coding - to implemen=
t more puzzles in MPUlt, to write a couple of non-puzzle games in 5D, try t=
o do something with {5,3,4} (I feel that there should be periodic patterns =
- like 22 colors for {5,3,4} and 55 for {4,3,5})... But may be I have a new=
challenge for you - try 600-cell in new version of MPUlt (uploaded today) =
;)

Good luck!
Andrey



--- In 4D_Cubing@yahoogroups.com, "schuma" wrote:
>
> Andrew
>=20
> Congratulations! By the way thank you again for making MPUlt. I wonder wh=
at will be the next challenge for you.
>=20
> Nan
>=20
>=20
> --- In 4D_Cubing@yahoogroups.com, "Andrey" wrote:
> >
> > Hi all,
> > I've just finished solving of 120-cell. In MPUlt it was not difficult=
at all. I've used 4 features of the program - macro, setup sequencies, cub=
ie highlighting and fast temporary changing sticker color to white. With th=
ese features I almost didn't care about the actual colors of cells.
> > My method is the following: I solve 2C, 3C and then 4C. For each type=
I select some cell, change it to white, then move all cubies with white st=
ickers to this cell (using setup unwinding and a couple of macros). My goal=
here is to make all stickers of the current type on that face "white". The=
n I use another macro to sort cubies in that cell (it is a half on 3-cycle =
that moves two pieces from the cell and one outside). Non-macro twists here=
are twists of one cell, so I don't need to use F1/F2/F3 keys for this stag=
e. Then I go to other cell and so on.
> > At first my estimate of time for complete solving was 96 hours (I use=
d about 2 minutes per cubie). So I didn't care about the time, and often le=
ft the puzzle with running timer for a while. I think that I lost about 6 h=
ours of displayed time here. But at the end of 3C stage I found that situat=
ion is much better than I thought about it: timer displayed 57h 05m! So if =
I'll be solving 40 pieces per hour, there will be good chances to finish at=
72h :)
> > Actual time was 69h 58m 30s. And I spent last 30 minutes to resolve o=
ne flipped 4C - tried to solve it like in 3^4, but coldn't remember the pro=
per algorithms :)
> > Andrey
> >
>

MPUlt 1.08 is reuploaded - in previous version 600-cell was a little deep-c=
ut, and it had moving centers... Proper cutting value is 0.968, not 0.96.
Note, that stickers in 600-cell have ranks 0, 1, 4 and 11+ :)

Andrey

--- In 4D_Cubing@yahoogroups.com, "Andrey" wrote:
>
> Nan, thank you!
> I don't know about next challenge. It should be some coding - to implem=
ent more puzzles in MPUlt, to write a couple of non-puzzle games in 5D, try=
to do something with {5,3,4} (I feel that there should be periodic pattern=
s - like 22 colors for {5,3,4} and 55 for {4,3,5})... But may be I have a n=
ew challenge for you - try 600-cell in new version of MPUlt (uploaded today=
) ;)
>=20
> Good luck!
> Andrey
>=20
>=20
>=20
> --- In 4D_Cubing@yahoogroups.com, "schuma" wrote:
> >
> > Andrew
> >=20
> > Congratulations! By the way thank you again for making MPUlt. I wonder =
what will be the next challenge for you.
> >=20
> > Nan
> >=20
> >=20
> > --- In 4D_Cubing@yahoogroups.com, "Andrey" wrote:
> > >
> > > Hi all,
> > > I've just finished solving of 120-cell. In MPUlt it was not difficu=
lt at all. I've used 4 features of the program - macro, setup sequencies, c=
ubie highlighting and fast temporary changing sticker color to white. With =
these features I almost didn't care about the actual colors of cells.
> > > My method is the following: I solve 2C, 3C and then 4C. For each ty=
pe I select some cell, change it to white, then move all cubies with white =
stickers to this cell (using setup unwinding and a couple of macros). My go=
al here is to make all stickers of the current type on that face "white". T=
hen I use another macro to sort cubies in that cell (it is a half on 3-cycl=
e that moves two pieces from the cell and one outside). Non-macro twists he=
re are twists of one cell, so I don't need to use F1/F2/F3 keys for this st=
age. Then I go to other cell and so on.
> > > At first my estimate of time for complete solving was 96 hours (I u=
sed about 2 minutes per cubie). So I didn't care about the time, and often =
left the puzzle with running timer for a while. I think that I lost about 6=
hours of displayed time here. But at the end of 3C stage I found that situ=
ation is much better than I thought about it: timer displayed 57h 05m! So i=
f I'll be solving 40 pieces per hour, there will be good chances to finish =
at 72h :)
> > > Actual time was 69h 58m 30s. And I spent last 30 minutes to resolve=
one flipped 4C - tried to solve it like in 3^4, but coldn't remember the p=
roper algorithms :)
> > > Andrey
> > >
> >
>

--0015175cdd1ca0937304abf8697d
Content-Type: text/plain; charset=ISO-8859-1

Very cool Andrey :)

Would you like me to add your name to the
Magic120Cellpage I have, or
is it time to move the solvers list for this puzzle to a
shared place like the wiki? (I welcome opinions on that from others too.)
Unbelievably, there are now 6 solvers of the 120-cell puzzle!

Roice

On Thu, Sep 1, 2011 at 5:48 AM, Andrey wrote:

> Hi all,
> I've just finished solving of 120-cell. In MPUlt it was not difficult at
> all. I've used 4 features of the program - macro, setup sequencies, cubie
> highlighting and fast temporary changing sticker color to white. With these
> features I almost didn't care about the actual colors of cells.
> My method is the following: I solve 2C, 3C and then 4C. For each type I
> select some cell, change it to white, then move all cubies with white
> stickers to this cell (using setup unwinding and a couple of macros). My
> goal here is to make all stickers of the current type on that face "white".
> Then I use another macro to sort cubies in that cell (it is a half on
> 3-cycle that moves two pieces from the cell and one outside). Non-macro
> twists here are twists of one cell, so I don't need to use F1/F2/F3 keys for
> this stage. Then I go to other cell and so on.
> At first my estimate of time for complete solving was 96 hours (I used
> about 2 minutes per cubie). So I didn't care about the time, and often left
> the puzzle with running timer for a while. I think that I lost about 6 hours
> of displayed time here. But at the end of 3C stage I found that situation is
> much better than I thought about it: timer displayed 57h 05m! So if I'll be
> solving 40 pieces per hour, there will be good chances to finish at 72h :)
> Actual time was 69h 58m 30s. And I spent last 30 minutes to resolve one
> flipped 4C - tried to solve it like in 3^4, but coldn't remember the proper
> algorithms :)
> Andrey
>
>
>
> ------------------------------------
>
> Yahoo! Groups Links
>
>
>
>

--0015175cdd1ca0937304abf8697d
Content-Type: text/html; charset=ISO-8859-1
Content-Transfer-Encoding: quoted-printable

Hi Andrey,

Hmm, a simplified 600-cell! That looks interesting. I noticed the keyword "=
Simplified" in the puzzle definition file. That option not only reduces the=
full 600-cell to something tractable, but also makes puzzles like 4D_ET an=
d 4D_RT enjoyable. The simplified 4D_ET and 4D_RT has a similar number of p=
ieces as 3^4. I remember I talked with Brandon about 4D_RT. I think his ini=
tial guess of 4D_RT should be essentially this simplified version. Maybe it=
's worthwhile to add a checkbox for this option in the main interface to ma=
ke it easier to toggle between the full puzzle and the simplified puzzle.

The simplified 600-cell has about 2000 pieces and is in the same level as t=
he 120-cell. But I'm not sure if I want to attempt it. I remember the VT 60=
0-cell with only trivial tips is complicated enough...

Nan

--- In 4D_Cubing@yahoogroups.com, "Andrey" wrote:
>
> MPUlt 1.08 is reuploaded - in previous version 600-cell was a little deep=
-cut, and it had moving centers... Proper cutting value is 0.968, not 0.96.
> Note, that stickers in 600-cell have ranks 0, 1, 4 and 11+ :)
>=20
> Andrey
>=20
> --- In 4D_Cubing@yahoogroups.com, "Andrey" wrote:
> >
> > Nan, thank you!
> > I don't know about next challenge. It should be some coding - to impl=
ement more puzzles in MPUlt, to write a couple of non-puzzle games in 5D, t=
ry to do something with {5,3,4} (I feel that there should be periodic patte=
rns - like 22 colors for {5,3,4} and 55 for {4,3,5})... But may be I have a=
new challenge for you - try 600-cell in new version of MPUlt (uploaded tod=
ay) ;)
> >=20
> > Good luck!
> > Andrey
> >=20