[Help Thread] Megaminx Discussion

[googlebleh]

Well, every turn is a double wide turn, so it wouldn't make sense if we used R- instead if R--. It's like using R instead of R2. And for the r' d r d' thing, I'm sure it's partly because of what I just typed, and partly because what we use now is easier to read.

Not 2R+ 2D- or 2R++ 2R--? Nothing else uses +/- so I questioned the necessity for two when it's always a double turn.

Maybe I'll file under "just because" in my head.

 

[qqwref]

Yeah, I don't like how the scrambler works for a number of reasons.

Spoiler

- Using R and D to denote all-but-one-layer turns, as opposed to more normal notation
- Using + and - for clockwise and counterclockwise, as opposed to more normal notation
- Using a moveset that is very unlike normal Megaminx solving moves
- Only using double turns, thus decreasing the number of possible scrambles
- Only having two possibilities for each turn, meaning it would take >200 turns for it to even be possible to generate every scramble
Not to mention TNoodle is a dumb idea...

But it's probably best to keep it the way it is, since so many people are already familiar with it. And yes, all of the design decisions are "just because" aka tradition.

 

[Divineskulls]

Re: Megaminx Thread

Not 2R+ 2D- or 2R++ 2R--? Nothing else uses +/- so I questioned the necessity for two when it's always a double turn.

Maybe I'll file under "just because" in my head.

If we used r d' r' d, someone would question why it is written like that if we always use double turns. Something that makes sense in one person's mind might not make sense in another's. Although, in this case, Stefan is 'one person,' and he's always right, soooooo... ;P

 

[JianhanC]

He doesn't just do the first one that pops up in front of him? It would take lots of time to look all around the cube for the easiest one.

I guess that would be more accurate. Sorta like spotting the first/easiest one to solve, like the first F2L pair you solve.

 

[googlebleh]

If we used r d' r' d, someone would question why it is written like that if we always use double turns. Something that makes sense in one person's mind might not make sense in another's. Although, in this case, Stefan is 'one person,' and he's always right, soooooo... ;P

Didn't address my 2R+ 2D- (although now that I think about it, it should be R2+ and D2-), but that's okay. I got the reason why, logical or not

Using a moveset that is very unlike normal Megaminx solving moves

I can understand that because it makes the Megaminx much easier/faster to scramble. However, I agree that it reduces the efficacy of scrambling:

Only using double turns, thus decreasing the number of possible scrambles
Only having two possibilities for each turn, meaning it would take >200 turns for it to even be possible to generate every scramble

Maybe one day when we think we've exhausted all the possible 'minx scrambles with our current scrambling moveset we can change it.

 

[qqwref]

Maybe one day when we think we've exhausted all the possible 'minx scrambles with our current scrambling moveset we can change it.

It's not that, it's more that we aren't using enough moves to get everywhere - basically the scrambler can only access the positions that are the closest to solved with this scrambling metric. So although the puzzle always looks scrambled there are probably some biases that make the puzzle easier to solve than it would be with a totally random state. For instance, maybe it's easier to make pairs or stars than it should be. Simply allowing single turns as well as double turns would roughly halve the number of moves needed to get to every position, making the scrambles less likely to be noticeably biased.

 

[googlebleh]

It's not that, it's more that we aren't using enough moves to get everywhere - basically the scrambler can only access the positions that are the closest to solved with this scrambling metric. So although the puzzle always looks scrambled there are probably some biases that make the puzzle easier to solve than it would be with a totally random state. For instance, maybe it's easier to make pairs or stars than it should be. Simply allowing single turns as well as double turns would roughly halve the number of moves needed to get to every position, making the scrambles less likely to be noticeably biased.

Oh, so we actually can get every position, it's just we would need longer scrambles. And some positions have a smaller chance of appearing with the current method of scrambling, so it is less likely to be "more scrambled", so the current scrambles are, on average, easier than random-state scramblers?

 

[Carrot]

Oh, so we actually can get every position, it's just we would need longer scrambles. And some positions have a smaller chance of appearing with the current method of scrambling, so it is less likely to be "more scrambled", so the current scrambles are, on average, easier than random-state scramblers?

Official scrambles feels easy So I always practise handscrambles so I can get surprised at all those delicious blocks everywhere when I go to comps.

EDIT: lol, I posted something earlier that made NO sense at all, so just changed it to something more senseful

 

[antoineccantin]

Official scrambles feels easy So I always practise handscrambles so I can get surprised at all those delicious blocks everywhere when I go to comps.

EDIT: lol, I posted something earlier that made NO sense at all, so just changed it to something more senseful

I've always seemed to get good times in comp at megaminx. Maybe it's because I'm used to handscrambleing them?

 

[MrRubiksUFO]

Is there any list with good CP algs (not the 5 corner cycles)?

 

[Carrot]

Is there any list with good CP algs (not the 5 corner cycles)?

https://sites.google.com/site/onelookpyra/megaminx/cp

That's what I use

 

[sokoban]

Colour Recognition for Colour Neutral Solver?

Hi, I've been solving the megaminx now for a while with the beginners method, and I decided to move onto the more advanced last layer method (full oll, pll) from https://sites.google.com/site/permuteramera/

All of the cases have the yellow side on top (not oriented), and since it shows the colours on the side, I'm assuming it is a ZBLL-esque kind of thing. It's also the same with CLL (not that I'm learning that or ZB right now). It only shows colours for the yellow side, assuming that the cross or first layer is solved on white. Let me cut to the chase. Is there a way to do CLL or ZB on any side that's not yellow? I haven't seen it being documented.

P.S if this is a copy and I just haven't searched hard enough, let me know and I'll take this down

 

[MaeLSTRoM]

Umm, the cases are shown like that because most people have yellow as the U face when doing LL.

The colours on the side shouldn't be used for recognition, but you should rather use the patterns you see to be able to recognise the cases, as in finding bars of pieces etc.

Also, since you're moving from the beginnners method, I would recommend that you stick to 4LLL. You don't really need more than that to even be sub-1 with Megaminx, and its a lot of effort if you give up half-way through. What I mean by this is learn how to orient edges quickly, and then learn the OCLL, CPLL and EPLL subsets.

 

[sokoban]

Thanks!

 

[MaeLSTRoM]

Thanks! But is there a way of doing it for CLL - per say?

Ahaha, CLL is an even bigger set iirc. Again, by using patterns of what colour stickers are in what positions, you can recognise on any LL colour. In theory, its just the same as learning CLL recognition on 3x3 but with many more cases.

 

[sokoban]

Thanks so much! Do you know 2LLL for mega or...?

 

[MaeLSTRoM]

Thanks so much! Do you know 2LLL for mega or...?

I'm currently in the process of learning PLL for Megaminx, and I have been for over a year now >_> I'm hoping to finish it before the end of the year, but it takes quite a bit of effort and time to learn cases. After I finish PLL, I'm going to start learning some OLLs at least for easy cases and also for some of the awkward ones. Like I said, I'm only doing it because I really love mega, I wouldn't recommend doing it to anyone else unless they're already pretty quick and are having trouble improving.

 

[sokoban]

Wow I didn't realize the commitment! Better stick to the 3x3! once again thanks so much and hopefully going to first competition this year down in England

 

[qqwref]

It's not a matter of megaminx requiring crazy commitment. Full OLL/PLL on megaminx is just far, far more complicated than on 3x3x3, and IMO not even worth learning unless you already average under a minute. 4-look last layer for megaminx is more than enough to get great times.

 

[MaeLSTRoM]

It's not a matter of megaminx requiring crazy commitment. Full OLL/PLL on megaminx is just far, far more complicated than on 3x3x3, and IMO not even worth learning unless you already average under a minute. 4-look last layer for megaminx is more than enough to get great times.

Yeah definitely. I'm now getting to the point that the cases I have left are really crappy to recognise. Although, with most of those cases the 2-look solution is probably just as quick by the time you've recog'd it. The most annoying case is probably that weird 5cycle of corners that just looks like a mess. The alg is so good but I can't recognise it at all, its on my list of cases to learn recog for atm actually XD