Tim Major
Platinum Member
http://www.jaapsch.net/puzzles/pyraminx.htm7 moves would get it to less than a third of all states. Make it maybe 12-14 moves+tips.
Stefan
Legend
Before WCA had a random state scrambler, 25 moves were the standard:
http://web.archive.org/web/20080517121644/http://www.worldcubeassociation.org/regulations#scrambling
wolferber
Regular
My twitter handle is @IshIsDeep if you ever have feedback. Its probably much faster way to get int contact me with, that anything else
Lucas Garron
Premium Member
https://www.worldcubeassociation.org/regulations/history/files/regulations2008.html#scrambling
It baffles me that very few apps on mobile are doing random state, though, and replacing it with something vey inferior. There are some easy-to-adapt solutions out there (although I admit they're not as immediately ready as they could be).
Stefan
Legend
Ha, why didn't I think of that...
Judging partly by how often their scrambles include stuff like U U2 or R R', I have a feeling that many aren't made by cubers, and they might just not know that there is something better.
AvGalen
Premium Member
Random state is hard! Maybe not for you and Stefan (anymore) but it is for most everyone else. Having a "pick 25 times from R,U,F,L,D,B,R2,U2,F2,L2,D2,B2,R',U',F',L',D',B'" is quite a lot easier than understanding transformations, matrices, 2-state-generators, IDA, pruningtables, etc.
And for most people "a scramble is a scramble"
Stefan
Legend
I'm sure Lucas wasn't talking about 5x5. Pyraminx doesn't even have a million states (ignoring the tips), simple BFS should get the job done.
And Lucas was talking about re-using existing solutions.
AvGalen
Premium Member
I am willing to bet cold hard cash (or cookies) that > 50% of all people that ever wrote a scrambler don't know what BFS stands for. (burden of proof is not on me!)
It is amazing what we can now do in JavaScript that was not doable 10 years ago, but I don't know of any sourcecode for optimal scramblers except the javascript ones.
cubizh
Veteran
Perhaps a thread debating random state scrambling algorithms and their implementations in several programming languages (with links to sources) should be created.
Stefan
Legend
My point wasn't that it's easy, my point was that it's not as hard as you made it sound (with all those fancy terms you used). Also, they don't need to know what BFS stands for. I didn't know the term at all, back when I (re-)invented the algorithm (when I started at university, didn't know much, but had to guide a robot through a maze). Doesn't take a genius to come up with it.
Wait, wait, why change the subject to optimal ones?
Last edited:
wolferber
Regular
Well, I would be more than happy to incorporate random state scrambles if anyone knows where I can find the algorithms to that. From what I know about it, it is extremely complex to develop from scratch.
Carson
Premium Member
Just throwing in my 2 cents. Part of coding an app (for most of those who are not coding for their livelihood) is solving problems and the sense of accomplishment for the code you come up with. A few years ago, I created a timer/scrambler for Windows Mobile. (pre-Windows Phone) I started out with some scrambling code given to me by others. I gradually replaced it with my own code, even though my code was often not as optimized and was not random state (even for 2x2). The reason for this was that it was MY code. The scramblers did improve, and I believe I did have a random state 2x2 back in eventually before that OS died taking my app with it. If I decide to write another scrambler/timer app, it will also include my own non-random state scramblers at first. If I, at some point, learn how to code a random state scrambler, then I will write one.
AvGalen
Premium Member
So you are claiming that you did it, but that it doesn't take a genius to do it. I call lack of proof
My main point is stat providing a random scramble is easy (as long as you don't care about true randomness, let's not get into that discussion). Making a random state scrambler is A LOT harder. I would compare the first with running a few km (everyone can do it with a bit of effort) and the latter with a backflip (apparently everyone can learn it, but hardly anyone does)
Sorry for mentioning optimality, I just meant random state. But I still meant "I don't know of any sourcecode for random state scramblers except the javascript ones."
Last edited:
Similar threads
