The cube challenge revisited

[potwnc]

I am amazed that some people accepted my Rubik's cube challenge. And after I saw what Peter has achieved I stopped working on my own challenge! Peter, congratulations!

When I set the challenge I did not think that it could be animated like it has been so well by Peter (and others). Instead what I had in my mind was a computer version of the actual Rubik's cube - a cube that would start with a random configuration and that the user could manipulate by mouse clicks to solve the puzzle.

This is the path I took and, not surprisingly, I soon ran into problems!

But the exercise was worth it because it showed me what features could be added to PTE in the future to make such a "slide show" much more simple.

To demonstrate the kinds of problems I faced and to demonstrate how far I got I have posted a new, 2-dimensional demo of the "cube" because it is much simpler for everyone to understand (including me!) in two dimensions. It can be downloaded from here.

It starts by showing the solution then, after a few seconds, it generates a "random" version of the puzzle that you must solve. In this demo you can only move numbers 14, 13 and 12 and you do that by clicking on the bottom of a number to move it down, on the right of a number to move it right and on the left of a number to move it left. Each move for these three numbers can be undone so that you end up where you started.

Comments as always are appreciated.

Ray

[fh1805]

Ray,

I apologise for having mis-understood what your original challenge was really about. But many thanks for setting it. I've enjoyed working on my variant of Rubik's Cube over the past two-three weeks and have undoubtedly learned an immense amount about 3D animation in PTE as a result. This I might never have done had it not been for your challenge.

regards,

Peter

[Lin Evans]

Hi Ray,

What you had in mind can't be realistically done with PTE. There are a total of six sections of the Rubik's Cube which can be rotated, but only three sections can be rotated on any one slide with PTE. To rotate the other three requires a different construct and the two constructs can't occupy the same space at the same time without setting all individual (27 cubes) to zero opacity on all six sides. That would mean each time one of the alternate set of three sections containing nine "cubes" were to be rotated, 162 separate opacity setting would need to be keyframed out then back in. Since opacity is not "inherited" in the Parent/Child settings, it would be terribly complex to try to do this in real time - hence the necessity for doing this operation on separate slides where no opacity settings need be changed.

Perhaps some day in the future if an option for parent/child inheritance of opacity becomes available, this might be feasible, but not today.

Best regards,

Lin

[potwnc]

I will post a new demo of the 2-dimensional version this weekend as well as some observations about what I learned when I worked on it.

Stay tuned for more cube discussions and a new cube challenge!!!

Ray

[potwnc]

Here's a working version of the 2-D puzzle that you can download and solve (skill level = very easy). Not all possible movements are programmed. I may get around to finishing it one day but not in the near future.

I know that PTE was not intended to allow us to make puzzles like this but I'm sure those who have been following my challenge realize, it's a great exercise to push PTE to its limits and learn how it works - and doesn't - at the same time.

The most important thing I learned working on this is that a very useful new feature would be to allow clicks on objects not only to go to slide x but to go to keypoint y in slide x. Of course this would require the introduction of numbered keypoints. A second useful feature would be to go to a random keypoint in a random slide when the user clicks on an object. With these two features I believe I could produce the full, working solution of the 2-D puzzle in a fairly short amount of time.

As for my Rubik's cube challenge, these same features would help PTE in being able to meet the version of the challenge that I originally had in mind.

Here's a new version of my challenge that I know can be done in PTE 6. Maybe Peter would like to accept this version of the challenge: When the slideshow first starts the user can click in one of 3 (or 4 or 5) different places and each click leads to the Rubik's cube starting with a different, unsolved version of the cube. Then from that starting point, and without the user clicking anywhere, the moves to the solution from that starting point are animated as the user watches the cube solve itself.

I also found the following which I believe is a defect in 6 - and probabbly earlier versions: if slides have been set up to go to slide number x when the user clicks an object and then a slide that is not the last slide is deleted, all references to target slides should have the number of the traget slide automatically updated but that doesn't happen.

Have fun with my 2-D "cube" and post your feedback.

Ray

[JRR]

Ray:

You might want to check the link again

It didn't take me to the Puzzle, took me elsewhere

[potwnc]

Thans for reporting the problem. I've updated the link.

[fh1805]

...Maybe Peter would like to accept this version of the challenge...

I'm all cubed out! Sick of the sight of the little devils!!!!

regards,

Peter