Message #724

From: Melinda Green <melinda@superliminal.com>
Subject: Re: [MC4D] Announcing MagicCube4D version 4.0 beta ready for download!
Date: Sat, 24 Oct 2009 15:03:57 -0700

Dear Cubists,

Thanks everyone for testing and for your initial feedback. It’s been
very helpful. I don’t want to discourage any discussion here but please
also file all of your bugs and feature requests into our issue tracker
<http://code.google.com/p/magiccube4d/issues/list>. Just check first to
see if it is already there, and if so, then vote for it by clicking the
star, and add any additional comments you have.

Please also remember to enter your own records on the wiki
<http://wiki.superliminal.com/wiki/MC4D_Records>. I had simply linked
from my name to my solution but I /really /like the way that Klaus
entered his even better by adding his solutions to his user page and
linking to anchors on that page. I’m not going to lay down any law on
this because we’re still just figuring it out together but please see
what he did when it comes time to add your own records and consider
copying his format.

Just to add confusion at the very start, Remi sent me his shortest
{3,3,3} 2 solution with the same number of twists as Klaus’, and of
course both on the same day. I think that the claim should go to the
first solution with a digital recording but I’m not sure which one that
is. Remi’s arrived at 6:25 PDT and Klaus’ was posted at 17:04 but in
some other time zone. As Remi pointed out to me privately, this tiny
puzzle is a little bit silly in its simplicity, so maybe no shortest
record should be kept at all. What do you guys think? Personally I’m
fine with it though I’d rather see one of you guys do it in less than 5
twists so that we won’t have to figure out who gets the credit for the
first 5-twist solution. Unlike with the 2^4, I get the feeling that if
you scramble this one enough times, there’s a real chance that you’ll
find a scramble that is one twist from solved. ;-)

Now on to your feedback:

Norbert Hantos wrote:
> Amazing… Nice job, nice puzzles! Thank you all!
>
> A simple question: could I re-color a puzzle? Now I can only re-color
> the ground, the outlines and the sky… I really missed my color scheme.

There is currently no support for custom colors, as is mentioned in the
release notes
<http://wiki.superliminal.com/wiki/MC4D_4.0_Release_Notes>. That feature
needs to be redesigned with the new puzzles in mind. I’ve been thinking
about this a little and have a proposal for you guys: What if we extend
the color file to have a different number of colors on each line; then
when creating a puzzle with N faces, the program looks for the first
line in the color file with exactly N colors and maps those colors to
the faces? Any puzzle without a matching line will use the automated
coloring you see now. It might be difficult to remap the same colors to
the same faces each time because the faces may change numbers during
rotate-like twists. Roice would need to tell us if that will be a
problem, but would anyone still have a problem with custom coloring if
we can make it work that way?

thesamer@interia.pl wrote:
> Program is great! Damn! It’s brilliant! Thank you for a lot of
> entertainment (in the past, now and in the future!)
>
> 1. I suggest changing RightMouseButton and dragging for changing scale
> view insted 4D point of view (really please about this feature)
>

I think I have something even better. :-)
This has bothered me a lot too and sometimes I even found myself trying
to use the mouse scroll wheel to change the view scale like some other
applications offer, so I just hooked that up last night. Actually I
don’t have a mouse on my laptop, but the vertical scroll region on the
right edge of the touch pad does the same thing. It’s not in the current
download but will be in the next one.

> 2. Option of colouring faces needed.
>

See above and please comment on the design proposal.

> 3. I’ve tried to make macro using in "making time" another macro but I
> couldn’t.
>

Macros-within-macros should definitely be added as a feature request in
the issue tracker. Would you add that please?

> 4. Is there way to open old logs and macros? (I see new style of
> notation inside of log)
>

Not yet. That’s another feature that should be on the list. Please add
that too?

> 5. I don’t think that simplex-2 should be on an
>
> Which puzzles will have Hall of Fame? (for example simplex-2 seems to
> simple to compete with 2^4, not mentioning rest of "miracle" puzzles :)
>

As I mentioned above, I don’t really mind supporting the trivial puzzles
in the Wiki HOF because it will be no work for me. A related question
that I find more interesting is what to do with solutions to
user-invented puzzles? Some of them which are simple variations on the
existing ones definitely make sense but others are buggy or otherwise
problematic, and since all of them are officially unsupported, maybe
they don’t belong in the HOF. Thoughts anyone?

>
> Great job Melinda, Don and Roice, Jay!
>
> Solving is one thing but… Creating working and consistent things like
> your puzzles is amazing!
>
> All the best,
> Remigiusz D.
>
> Ps. I’ve already solved simplex-2 and simplex-3 :) that was nice fun!
>

I might have spent 150 hours on this update, but some of you are going
to spend that much time on a single solution!! I’m just really, really
happy that you guys are having fun!

Klaus wrote:
> Wow, these new puzzles are really astonishing. They look really good and the new freedom of rotating them in 4D is really great.
>

That part was done by Roice as inspired by the Jenn3D UI. I’d
implemented interactive 4D rotations before but resisted adding it to
MC4D because I never saw one that gave "predictable" results. When Roice
showed me how Jenn3D did it such that the center always moves in the
direction of the cursor, I was sold! Oh, and in case you didn’t notice
in the release notes, all the new transforms can be auto-rotated just
like with the 3D rotations. Very trippy!

> However there are two problems I currently have with the programme:
> First of all, please give every puzzle a default setting for "view-scale" and "eye-w-scale" because almost every time you open a new puzzle it looks really confusing and you first have to find some settings in which you can view it properly.
>

This is issue 34
<http://code.google.com/p/magiccube4d/issues/detail?id=34> and
definitely one that we should do soon.

> And secondly, the "Invent my own"-tool doesn’t work correctly, because whenever I enter {3,3,4} it just gives me a hypercube which is {4,3,3} (the dual polytope) and {3,4,3} doesn’t work at all.
>

Yes, I think that I mentioned before that the 16, 24, and 600 cell
puzzles are currently unimplemented. I expect that we will support them
someday but not in 4.0 and probably not soon. If you run out of puzzles
to solve I’ll see if I can hurry that along. ;-) In the meantime, the
invent-my-own option is there for your amusement but is unsupported.

> Despite these two facts the programme really works nice and I think I’ll do my next 3^4 in this new programme. However I have to finish my current solve first.
>
> btw: I solved the {3,3,3} 2 and it was really simple. I did it in 5 moves my first time. However, it seems like I had really luck with the scramble, because God’s algorithm can take up to 10 twists on this puzzle.

Really? How do you know that?
I suspect that there may be confusion between different possible ways of
defining a Rubik’s 4-simplex because this one is extremely easy to
solve. In fact it’s *so* easy that I suspect that its state graph is
small enough to draw. If so, then I /really /want to see that graph!!

Well, thank you to everyone who has replied or simply tested the beta.
BTW are any of you testers running on Linux or Mac operating system? I’d
really love to get some confirmation that everything is also working
well on those platforms.

Please keep up the testing; and keep the feedback, solutions, and issue
reports coming.

Happy puzzling!!
-Melinda