Status
Not open for further replies.
Why do people solve rubiks magic's from one solved state to the other? It is just completely based on memory, why do that when it's possible to completely scramble the puzzle and try to solve it from that?

Might be an obvious reason, I've never had it so I don't know.
 
You're about five minutes faster than me so I'm not sure what you do already since I'm doing pretty basic things for last 4 edges and last 2 centers. For edges I just solve the first eight that standard way breaking the centers into lines. Then for the last four I solve the two inner edges right next to the midge and solve parity if I have it. Then I solve the outer edges and solve parity if I have it. It's not the most efficient when it comes to dealing with parity but I think the alg I use won't solve inner edge parity if everything else is solved. Haven't really experimented with single slices of the 3rd innermost layer though. That could work.

For Centers I solve almost always with 1x5 lines unless I see a 2x2 or 2x3 block in which case I'll build that real quick and insert it. Last two centers I get as many solved as I can and then use the commutator. I usually end up using the comm 7-10 times. On a good solve it will only be 4-ish. I really need to work on doing lines though. You can always get it to where you only need to use the commutator once to solve the last two centers apparently.

My first two centers take me about 2:30 right now and total for centers is around 6 minutes-ish. It all depends on when I get to the last two. It takes me at least a minute to finish up the last two centers once I start using the commutator. My 3x3 phase is about 1:00-1:15 assuming I don't make a mistake.


That's pretty bad lol any advice?

If you're using commutators 7-10 times, that is definitely a huge problem...

Try this: http://www.youtube.com/watch?v=Xz4ei39dI3E
 
How would I post a picture on a thread?

[noparse]
[/noparse]
 
The recognition and recall comes with practice. My recognition and recalls on OLLs I 1-look is currently awful. Instead of auto-piloting 2-look I pause to see if it's a dot case, then if it is I pause to recognize it. If it's a case that has another similar one I pause to differentiate them, and then I pause to recall the alg. Then I execute the alg, slowly. It's all part of the learning process. I average about 8 seconds for last F2L pair+LL using 2-look OLL and full PLL but when I start adding in the new OLLs it adds nearly 2.5 seconds to my LL time.
 
It doesn't matter if you're slow right now unless you have a competition coming up within like a week or two. Might as well incorporate it into solves right now.

In the long run you'd wish you used the PLLs you know right now in solves. Recognition for you is going to be hard no matter if you start now, or when you finish full PLL.

If you start now you can begin learning how to recognize, that way when you finish up, it's not a massive overhaul of stuff you have to learn to recognize.
 
Imo you should do what you feel most comfortable with. If you think you're going to waste time, don't use that until you're more comfortable with it.
 
Imo you should do what you feel most comfortable with. If you think you're going to waste time, don't use that until you're more comfortable with it.

Why? This will only hurt you in the long run. If I know how to one look an OLL but 2-look it instead because it's currently faster it's not helping at all. You need to practice recognition and recall in solves. If you don't want to do it in solves use qqTimer's LL scrambler. You learn new things to get faster so if you learn it and then don't use it, why bother learning it? Incorporate everything you know into your practice solves. If you have a competition coming up, like Ryan said, don't change anything.
 
Status
Not open for further replies.
Back
Top