An update on my ZBF2L progress

[Jason Baum]

Hi guys,

I just thought I would give an update on my progress with learning ZBF2L for those who are interested. I learned my 228th case tonight. This puts me at the 75% mark. I have been extremely lazy at learning algs this summer though. To give you an idea, I knew about 200 algs in the beginning of May. My goal at the beginning of summer was to know 75% of ZBF2L by the US Open, and to know all of it by the end of summer. Well, I took some time off from learning algs and have just picked up on ZBF2L again a week or two ago. However, I feel like I am in the home stretch now. I have one more F2L case left to learn where the corner and edge are connected in the top layer, and after that all I have to learn are the F2L cases where the corner and edge are separated in the top layer. It is exciting that after about 8 months of learning algs, I am finally beginning to see some light at the end of the tunnel.

I also just took an average using ZBF2L + COLL/PLL. I disregarded the solves where I did not know the ZBF2L. I rarely ever time myself using ZBF2L because right now I am more concerned about recognizing/recalling the algs than I am about how fast I am. But I just took the first average with ZBF2L in a while, and was very pleasantly surprised with the results:

(12.28), 15.83, (16.20), 13.23, 14.31, 15.39, 13.75, 15.73, 13.76, 13.83, 13.28, 15.64 = 14.48 average

Keep in mind there is still plenty of room for improvement here. The only solves where I had a PLL skip using COLL (which is technically ZBLL) were the 12.28 and the 13.83. All other times were achieved using strictly ZBF2L + COLL/PLL. I am very surprised at the number of 13s I got. It's not like these were easy cases either. In fact, on one of the 13s I had a ZBF2L case where I had to orient all the edges, which are usually the worst cases. Also, I did not amnesia any algs during this session.

This should give you a good idea of what is possible with full ZB. I know it is possible to be much faster with just using ZBF2L + COLL/PLL. I see sub 14 in the near future with this method, and I'm sure even sub 13 is possible once you have mastered ZBF2L. Just imagine once some ZBLL is applied.

I don't really have a set date as to when I want to have ZBF2L learned by now. I have been busy with work all summer, and in about a month I will be busy with school again. I'm taking my time with learning these algs, and I am actually having a great time learning this method. I'm just planning to take my time, enjoy the learning process, and I'll know them all whenever it happens.

I also wanted to give my thoughts on COLL. I have been using COLL for about two years now. For a while, I was unsure on whether or not it actually saved me time or not. After two years of using these algs, I can say that COLL is AWESOME. My COLL really didn't start to get to this point until I started working on ZBF2L. But now, recognizing COLL feels just as easy as recognizing a PLL case. I can also recognize all of the U, T, H, and Air Jeff COLLs from any angle. It makes the LL so ridiculously easy, and most of the COLL algs are very fast. The only case that I don't like are the Air Jeff cases. Other than the Air Jeff itself, the other 5 algs are just terrible. This concerns me for when I get to ZBLL because if the COLL algs are this bad, how could the ZBLLs be any better? I imagine that some of them are much worse. But other than that orientation, the other 5 are really, really great. I had my doubts on COLL for a while but I can tell you now that it is worth it to know. Recognition is hard for a while and some might argue that the time saved is not worth all of the extra effort, but in my opinion it is a really nice trick to have up your sleeve.

Sorry to make this message so long, I just wanted to post my progress and my thoughts of my experience thus far with ZBF2L. Any comments or questions are welcome!

 

[joey]

Congrats on learning 75%! Thatpost was interesting to read, and I hope you keep us updated in the future!

I don't understand how you practice this recall of the algs? Do you just setup different case? Or go through all 228 case :O

 

[Dennis]

Wow, that's great learning ZBF2L is respect! And 75% is a great achivment. Try to set a date, eg 1 oct, you will know 100% of the method and 1 jan you will do your first sub14 average with the method. This not only inspires and motivated you but forces you to really try to make this goal.

Are you in the process of learning ZBLL as well?

COLL: What is an Air Jeff? How do you recall the COLL cases fast? I have difficulties seeing the permutation in a COLL case thats why I moved from COLL to OLL that way I have only the pattern to recognise instead of both the pattern and the corner permutation.

If I'm thinking to start using ZBF2L which cases should I learn first? The one's in the VH-method perhaps?

Any other that we know is learning/knowing ZBF2L and/or ZBLL?

 

[Johannes91]

What is an Air Jeff?

I think he means Bruno.

Any other that we know is learning/knowing ZBF2L and/or ZBLL?

I know around 40% of ZBLL, but I'm not learning anymore because I have better ways to waste time. There's also this guy who knows whole ZB, but he isn't a speedcuber, more into move count.

 

[Jason Baum]

joey: I don't really have a special way of practicing the cases. Usually I just solve normally and do the ZBF2L if I know it, and if I don't know it I convert it into something I don't know. The only time I set up the cases is if it's an F2L case I just learned. But I don't like this way of practicing because you know what case is coming up so you can cheat in a way and think ahead. Recognizing during an actual solve is much different because for the most part, you have no idea what case you are going to get. That's why I like practicing that way. Plus, you can actually practice multiple cases at once during a solve. When I am solving and recognizing a ZBF2L, usually my thought process goes something like this: "Okay, it's not that case, it's not that case, oh yeah! it's THAT case". Obviously it goes by so quickly that you don't really focus on it, but that's how I recognize. I see the pair and look at the edges and filter out a few similar algs until I know what one it is. During this filtration process, you can kind of remember what the algs are without actually doing them. I hope that makes sense!

Dennis: I haven't even touched ZBLL yet. I am waiting until I know ZBF2L and am comfortable with using it before I start learning ZBLL. Although, recently I have been learning how to recognize if I will have a PLL skip with the COLL case I have, so I guess that is technically ZBLL. I only know about 15 or so, though.

An Air Jeff is this alg: R U2 R2 U' R2 U' R2 U2 R. I've also heard of that orientation case being called Bruno, Pi, or Superman, but I prefer calling it Air Jeff. For recognizing, I really don't look at the corner permutation at all. I use the code recognizing system for COLL. If you're not familiar with it, basically for every COLL case there are 4 stickers you look at. You determine what the relationship of these stickers are relative to each other. For example, say you have the U orientation, the code could be this:

L R
F F

All that means is the ULB and the URB stickers are opposite of each other, and that the FLU and the FRU stickers are the same color. From here you know what alg to do. After a lot of practice, you learn to see the orientation and the code at the same time. COLL is definitely a one part recognition for me now. I think that after a lot of practice, ZBLL can also become one part recognition.

If you are interested in learning ZBF2L, then I would absolutely recommend learning all the VH cases first. Once you know these, you can very easily convert any ZBF2L case you don't know into a VH case. After these algs, I recommend learning all the cases where the edge and corner are both in the F2L but oriented incorrectly. There are only 14 of these, and a lot of them are nice. The reason I recommend learning these secondly is that these cases are a pain to convert to VH. Also, since these don't come up as often, you want to make sure you really have them locked into your brain. Once you know all of these algs, you know what I like to call Basic ZBF2L. After this point, you can learn the rest of the algs in any order you choose, but I would definitely recommend learning Basic ZBF2L first.

As far as I know, I am the only one still learning ZB. Chris Hardwick once knew all of ZBF2L and about 30% of ZBLL, but he stopped learning ZB to focus on blindfold solving. Around the time of WC05, Doug Li knew about 30% ZBLL and a decent chunk of ZBF2L, but I haven't heard from him in a while. I'm really not sure of anyone else who had made a commitment to learning it, but those two guys definitely knew the most ZB out of anyone at one point.

edit: Just saw your post, Johannes. 40% ZBLL?! That's very impressive. Why did you give up? How was your recognition/recall time when you solved with it? Were the algs fingertrick-friendly? Also, now that you mention it I do remember reading about the guy who learned full ZB. Wasn't he a memory specialist? I wonder how long it took him to learn it all.

 

[Johannes91]

edit: Just saw your post, Johannes. 40% ZBLL?! That's very impressive. Why did you give up? How was your recognition/recall time when you solved with it? Were the algs fingertrick-friendly?

Since May, I've been much more interest in programming. I don't cube much and when I do it's just for fun and a way to relax, I don't have any goals anymore.

My recall is still really good, and recognition is good for most cases but slowish for some harder ones. I still use ZBLL for OH and occasionally for two-handed, too. I can reverse and inverse algs fluently which helps a lot, and usually I can see how they work so learning them is not difficult for me. IIRC, I once learned 80+ algs in a week (but of course had a break after that to make sure I don't forget them).

I mainly optimized the algs for OH, but they are not bad for speedcubing, either. Most are taken from Bernard Helmstetter's list, there are usually enough alternatives to find at least one good alg for each case.

Also, now that you mention it I do remember reading about the guy who learned full ZB. Wasn't he a memory specialist? I wonder how long it took him to learn it all.

I don't know much about him (at least he's a programmer), here's a quote:

last year I created a simple mathematical categorization of ZBF2L and ZBLL, and from it created a handy mnemonic system which made it fairly easy to learn all those algs.

 

[pjk]

Congrats Jason. Looks like you are getting there with it, and best of all, you are motivated to do it since it is fun.

Can you do all these ZBF2L cases with one hand as well as two? I know some people (like Mitchell Stern) can do all the algs he does with two hands with one hand. I am the opposite. It would be kind of interesting if you brought those into OH like Johannes did with the ZBLL. Also, how many moves does ZBF2L take off the overall move count on average? How about just ZBLL? Or how about full ZB?

Again, congrats.

Oh yeah, and the Rockies are in Pittsburgh over the next two days. They beat the Pirates last night..... go Rockies!

-Pat

 

[Jason Baum]

Can you do all these ZBF2L cases with one hand as well as two? I know some people (like Mitchell Stern) can do all the algs he does with two hands with one hand. I am the opposite. It would be kind of interesting if you brought those into OH like Johannes did with the ZBLL.

Yeah, I can do a good chunk of the cases OH. I can see a lot of the algs intuitively now, so I can do those cases OH because they don't feel like algs. For some cases though, especially the ones where you have to flip 4 edges, I can't do them OH because I don't quite get how they work.

Also, how many moves does ZBF2L take off the overall move count on average? How about just ZBLL? Or how about full ZB?

Usually, it doesn't add much, maybe four or five moves at the very most. For some cases, it is the same amount of moves. For the really nice ones, it is less moves than normal Fridrich F2L (R U' R2 F R F').

Oh yeah, and the Rockies are in Pittsburgh over the next two days. They beat the Pirates last night..... go Rockies!

That was a heck of a game last night. We were down by 8 at one point but still came back and had a shot to win it. We were playing some good ball before the all star break, but now we've lost 4 in a row. We'll see what happens in the rest of the series!

 

[pjk]

Those seem like they would come in handy with one hand. I am suprised that it adds more moves than normal on a typical solve... I figured it would be less by a good 10%.

The Rockies just played the Brewers and only won 1 out of 3, one of the games that they lost was 1-2 in the 10th, which is a disappointment. We will see what happens the rest of the series. I guess the Pirates need the win just as much as the Rockies, probably more. But I think the Rockies can possibly get the wildcard this year, and do something in the playoffs.

 

[Jason Baum]

The Pirates need to go on like a 10 game winning streak if they are going to have any chance at competing for a wild card. Being a Pirates fan is confusing... no matter how decent the team looks on paper, things always go wrong. On nights where our hitters finally break through, our pitcher gives up 8 or 9 runs. On nights where we only give up a few runs, our hitters can't hit anything and we get shut out. It never seems to click all at once with them. It looks like the Pirates are headed for their 15th straight losing season though. And unless a miracle occurs, the Pirates will break the record next year for the longest consecutive losing season streak with 16 straight losing seasons. But hey, at least we have PNC Park... =/

 

[dolphyfan]

At least they havn't lost 10000 games like the phillies.

 

[pjk]

Sounds like the Rockies. However, I have to give some credit to the Rockies this year... they were they had the best record in the month of May in all of MLB. They won 9 series straight. Then they went on a 9 game road trip and went 1-8, 4 of the losses were due to the 4 consecutive blown saves by Fuentes. However, they are only like 5 games back from first now, so we will see what happens.

 

[dolphyfan]

The Rockies have been playing decent baseball especially against my yankees. You probably will hate me because I am a yankee fan but I am from New York.

 

[Lt-UnReaL]

ZB...adds moves?? What's the point of learning it then. :/ I thought it took off at least 12 moves.

 

[pjk]

Yeah, we are getting a little off subject. I created a thread in off-topic to continue the baseball talk so we can keep this to ZB:
http://www.speedsolving.com/showthread.php?p=11015#post11015

 

[Jason Baum]

ZB...adds moves?? What's the point of learning it then. :/ I thought it took off at least 12 moves.

ZBF2L for the most part adds moves when compared to Fridrich F2L. This is because with ZBF2L, not only are you pairing/inserting an F2L pair, you are also orienting LL edges, where as with Fridrich all you are doing is pairing/inserting an F2L pair. ZBLL is where you gain all that back and more, though. ZBLL solves the last layer in one algorithm. I believe the average move count for a ZB solve is 44 or 45 moves, compared to Fridrich (55-58).

 

[Lt-UnReaL]

Yeah I knew that. When you said it adds 4-5 moves I thought you were talking about all of ZB, because that was what pjk was asking...or at least part of what he was asking.

 

[Jason Baum]

Wow... I just did a sub 14 average using ZBF2L!

12.84, 13.92, (18.34), 14.59, 13.58, 13.16, 13.67, 15.89, 14.61, (12.39), 13.36, 13.98 = 13.96 average

The only time where I did "ZBLL" was the 13.98 at the end. All other solves used COLL/PLL. I only had to disregard one solve during this average, so I knew the ZBF2L for 12/13 solves. To be honest, I am amazed at the potential of ZB after these two averages I've done. Anyway, I didn't think sub 14 would come so quickly with this method for me!

 

[Dennis]

Congrats

 

[Lt-UnReaL]

No PLL skips? COLL makes that happen 1/12 times right? :O