A few questions

[fleeting_penguin]

I have a few questions, and I didn't want to clutter up the speedcubing forum anymore, so here goes:

1) How do I get better at cross and F2L? I'm really annoyed how slow I am, and how I've seen youtube videos of people with keyhole beating me at F2L. I waste maybe 5 to 10 seconds on the cross, and sometimes take as long as a minute to finish the F2L if I have to separate and re-orient c/e pairs a lot. I've been working slow and trying to be smoother, but I don't think my ability to find c/e pairs has improved and if anything I have practiced slow, so I perform slow. Do I just need to work harder and look ahead more?

2) On Macky's page, it talks about learning COLL as a way to get faster. How does this help a cuber's average if COLL applies in a small number of cases?

3) What other modifications to Fridrich are useful, while not requiring quite the work ZB does? I've heard VH is not worth learning except as a step to ZB; does it have any independent value as a speedcubing system (i.e. would working to learn some of it improve my average at all)? How is MGLS? Other LL modifications? What about extended cross (are there any other Fridrich F2L modifications?) I'm very interested in exploring all of the exotic alternatives, if for nothing more than to satisfy my curiosity.

4) I've tried the Pochmann method for blindfolding, but it takes a long time without having a speedcube for fingertricks on the PLL algs. What's a faster but still accessible blindfold system that would allow me to do blindsolving for friends without taking too long and boring them (because this is my primary motivation:rolleyes?

5) What's a good system for FMC?

 

[ExoCorsair]

1. Practice. Build crosses in as few moves as possible. That comes from practice. Look-ahead also comes from practice. You can try learn the algorithms instead of trying to do F2L intuitively.

2. Not really. OLL + PLL still achieve very fast times. I personally don't think it is worth learning the 40 COLL algorithms unless you are going to pick up ZB or VH. Remember, you save time by executing fewer moves, but seeing your F2L times, your recognition probably will not be very good (at least when starting), and you'll potentially lose more time recognizing cases.

3. Pure Fridrich is fast as it is. I think that ZB is the only really worthwhile system to pick up out of those you listed. The only person I know who uses MGLS is Lucas Garron, and I think you'd be better off asking him how it is. Based on what I know of that MGLS, I think that learning VH + COLL (fewer algorithms) is a better alternative.

4. There are plenty of threads and posts regarding different BLD methods.

5. From what I've seen, you can just do a Fridrich/Petrus solve with optimal F2L/block building, with some LL control. To get below say, 35, you'd be better off PM'ing someone who is good at it.

 

[fleeting_penguin]

3. Pure Fridrich is fast as it is. I think that ZB is the only really worthwhile system to pick up out of those you listed. The only person I know who uses MGLS is Lucas Garron, and I think you'd be better off asking him how it is. Based on what I know of that MGLS, I think that learning VH + COLL (fewer algorithms) is a better alternative.

4. There are plenty of threads and posts regarding different BLD methods.

5. From what I've seen, you can just do a Fridrich/Petrus solve with optimal F2L/block building, with some LL control. To get below say, 35, you'd be better off PM'ing someone who is good at it.

3) I get the impression VH really isn't worth learning if a person already knows Fridrich and doesn't plan on ZB. How many moves are usually shaved off by using VH?


5) There's a lot of terminology being tossed around on the Yahoo group I am having trouble understanding. I'm under the impression there is some method that is comprised on a Petrus style solve and lots and lots of specialized tricks.

 

[ExoCorsair]

You asked what CFOP extensions are useful. So I answered with VH being better than MGLS.
In VH, you will add about 3-4 moves to your last F2L pair (negligible time difference), then get one of your 40 COLL cases (about +0.5s), and then an easy edge PLL (U, H, Z), which would reduce average times quite substantially if your other PLLs are slow. All this for much fewer algorithms than MGLS.

I forgot about extended cross. Not worth doing unless you want to block-build and become good at it, in my opinion. Which, again, requires practice to obtain optimal 2x2x1 squares and 2x2x2 blocks (something you apparently don't really want to do, eh? )

Do a search before you ask, or browse the appropriate sub-forum (BLD solving). Seriously... (http://www.speedsolving.com/showthread.php?t=1445).
I'd say if you already know Pochmann, go to M2 if you want to get faster. 3-cycle

You could look at the weekly competitions to see how other people solve the fewest moves event too, you know. And there are no 'tricks' when cubing... >___>
But yes, many people do Petrus F2L or X-cross to start.

 

[fleeting_penguin]

You asked what CFOP extensions are useful. So I answered with VH being better than MGLS.
In VH, you will add about 3-4 moves to your last F2L pair (negligible time difference), then get one of your 40 COLL cases (about +0.5s), and then an easy edge PLL (U, H, Z), which would reduce average times quite substantially if your other PLLs are slow. All this for much fewer algorithms than MGLS.

Does an easy PLL really shave off 0.5s?

 

[CorwinShiu]

This really depends on your PLL Speed. For me, if I magically knew the COLL, it would shave off 1.5 if I would of gotten an N or E perm otherwise. Most of the time you don't get N or E, and I would shave off anywhere from 0.5-1 second.

I don't think it's worth learning yet for you. Also, COLL only works when you have all the edges orientated, and if you orientate them then COLL, it's going to be slower then pure Fridrich.

 

[Johannes91]

COLL also loses time, because recognition and execution is slower than OLL. Certainly not worth it, if you use Fridrich.

 

[AvGalen]

5. Read this: http://www.speedsolving.com/showthread.php?t=1566

 

[adragast]

1) I think you should try to learn the F2L cases that you have problems with (when you have to separate and reorient corner/edge). 5-10 seconds for the cross seems OK to me especially if this means that you can continue smoothly the F2L but it should not take you up to one minute to finish the F2L (40-50sec max).