• Welcome to the Speedsolving.com, home of the web's largest puzzle community!
    You are currently viewing our forum as a guest which gives you limited access to join discussions and access our other features.

    Registration is fast, simple and absolutely free so please, join our community of 40,000+ people from around the world today!

    If you are already a member, simply login to hide this message and begin participating in the community!

What Should I Learn Before I Stop Learning?

Status
Not open for further replies.

jskyler91

Member
Joined
Mar 23, 2011
Messages
676
Location
Berkeley CA
YouTube
Visit Channel
READ THIS BEFORE YOU VOTE!!!!!!

I have been debating what I should undertake learning next for a while now and I have decided to ask you guys for your opinion. I feel like I am close to stopping the bulk of my learning and I want to know what you guys think are the most important things I should learn assuming I want to be the best that I can be, before I stop learning large things and just increase my TPS/ Lookahead. Before you vote and or leave your feedback here, you should know a few things about the way that I cube/ what I already know. Also, just so you know, I hope for this poll to give me an idea of whether or not people think I should or should not, considering who I am/ my traits, learn certain things at all.

Things about how I cube:

1. I use CFOP as my template, I know that other methods have more potential, and i am up for making some form of a CFOP/ Petrus/ Roux method, but I am not giving up on CFOP completely. This might not be my smartest move, but it is something I am sticking to.

2. I don't like to regrip very much unless absolutely necessary

3. I plan on learning a few more things/ sets of algs and then I am probably going to stop learning soon and just increase TPS.

4. Therefore, increasing my TPS now is NOT something I want to consider yet, so don't even bother trying to say that. I will do this when I have learnt all that I want to learn.

5. I prefer to keep my cubing relatively symmetrical and uniform. This means that I like to do relatively the same thing for f2l cases fromt he right and the left, and at least the same thing/ concept for the front and maybe something different from the back.I believe this will make me faster in the long run because I will be keeping things relatively simple so I will be able to do them really fast later on in my cubing career. This also means I don't like learning a bunch of specialized cases that are only used 1 in 100 or 200 solves

6. I cube not only because I enjoy it, but also because I want to be one of the best cubers in the world. MINIMUM, I want to max out my own potential. This means I will do something, no matter how hard, if it will make me faster in the long run. Keep in mind, however, that I don't want to learn a bunch of random algs for super unique cases which might make me faster, but only if I can remember this alg which I never use quickly. This will not make me actually faster IMO

7. I have pretty good algs recall, but not the best. This means that I will probably never learn ZBLL because I would just die trying to recall and I don't think I would EVER become fast at them. This also means that I would prefer smaller, higher use algs sets. I am not completely against 200 count alg sets, but that would be my limit so no ZBLL.

8. I prefer not to learn a last layer method that doesn't use OLL/PLL at least some of the time. I have spent a lot of time perfectly those algs from multiple angles and I do not want to waste that time. I realize this might limit my potential slightly, but this is one of the few limits I willingly accept. So saying "Switch to ROUX " won't really be helpful.

9. I am fine with some blockbuilding, but not a crazy amount. Yet again, not switching to ROUX now, I am a CFOPer in method, although I would totally accept subsets/tricks like x-cross or free fopping as well.

10. This is key: "I PLAN ON STOPPING MY LEARNING AT SOME POINT AND JUST INCREASING TPS/ LOOKAHEAD". I will only do this, however, when I think I have learnt all that i need to be as fast as I can be considering all of my previous points.

11. I also don't want to learn too much, so if I have learnt enough to max out my potential then I will stop learning. Too much in my head will likely just slow me down.

What I Already Know How to Do:

- Full CFOP

- WV for the most part, although my recall is bad because I don't practice it

- A small group of easy OLL Skips like the tripple sledgehammer and stuff or where you have all of OLL oriented and you have something like F2L case 26 left.

- COLL for Pi and H cases

- Some OLLCP

- 2 side PLL Recog

- A substantial amount of F2L algs to the point that I am pretty darn confident I don't need to learn much more on F2L in general, except maybe in approach or new things to think about/Multislotting

- Basic Multislotting: http://www.youtube.com/watch?v=cqEu1_GC4ug

- Some R-OLL, for the all edges oriented OLL cases that is. I still need to expand this to all cases.

- OLL/PLL from at least 2 angles each (Excluding the ones that act as natural mirrors like E V H, Z perms of course) Except for Y Perm and G (A), G(C), & RA/B, I wouldn't mind a new f perm or a few mirrors for the cases listed above where I don't have one.

- Basic/ Intermediate x-cross, I do them about 1 in 5 or so solves.

- How to orient all edges for things like EO and Roux, also, I am pretty good at recognizing if an edge piece is or is not oriented during f2l.

- Edge control with last and sometimes second to last slot.

- A substantial amount of fingertricks including thumbs and flicks/pulls

- Pretty darn good algs in general

- How to avoid regrips/ cube rotations, and when to do them as well.

I think thats it, I may have missed a few things, but those are the major alg sets/ learnings I have undertaken. Please let me know what you guys think I should learn next/ before I stop learning and go for TPS/ Lookahead. Just to clarify, I won't stop completely learning, just stop learning major changes/ alg sets. Thanks!!!

Explanation of the POLL:

1. RLS stuff- I don't think all of the RLS stuff is worth it as the algs really suck for many and just doing OLL is better in those cases.

2. Full COLL- Learning the rest besides Pi and H cases

3. More OLLCP, I only now a few for the edges oriented cases

4. Blockbuilding- Spend more time working on doing it more often and quicker than I already do it

5. ZZLL- Phasing/ ZZLL algs

6. ZBF2L- Always having edges oriented for ZZ or something, I really doubt I will do this, unless someone makes a really good argument

7. ZBLL for H case- I thought this would be an easy case to do this for and it comes up fairly often

8. R-OLL- Using it for all cases with 2 side PLL would be really awesome PLL recognition

9. How to Avoid ENVY- Creating an R-OLL alg list that made it so I always avoided ENVY

10. Other- Othter things you think are really important for being at my max.

YOU MAY VOTE MORE THAN ONCE, SO CHOOSE ALL OF THE THINGS YOU THINK I SHOULD WORK ON.
 
Last edited:

Dacuba

Member
Joined
Aug 14, 2010
Messages
342
Location
Germany
Aren't you considering learning parts of ZBLL, like the easiest cases both in recognition and execution?
I have seen especially the cases where edges are correctly permutated used by very fast cubers.

Also going through reconstructions might get you more ideas than you expect.

edit: I'm sorry I've read your post and replied before the poll was there, sorry if my post is inappropiate now
 
Last edited:

jskyler91

Member
Joined
Mar 23, 2011
Messages
676
Location
Berkeley CA
YouTube
Visit Channel
Aren't you considering learning parts of ZBLL, like the easiest cases both in recognition and execution?
I have seen especially the cases where edges are correctly permutated used by very fast cubers.

Also going through reconstructions might get you more ideas than you expect.

edit: I'm sorry I've read your post and replied before the poll was there, sorry if my post is inappropiate now

It is, but NP, I messed up and posted the thread too soon
 

Divineskulls

Member
Joined
Jun 16, 2011
Messages
853
Location
Levittown, PA
WCA
2011RECH01
YouTube
Visit Channel
If you want to keep using your oll/pll, I would suggest experimenting with Heise or ZZF2L. Both will improve your blockbuilding exponentially when practiced enough, and if you already have knowledge of EO, EOline should be easy. While I know Heise isn't useful for speedsolving, it is useful for efficiency, and in my opinion, when you are more efficient, lookahead is much easier. So you would be killing two birds with one stone, i.e. working on lookahead and learning/working on something new. I don't see a point in you learning a whole new set of algs; you know a lot of algs already, from the look of it. But if you insist on learning alg sets, I would suggest Rowe's RLS, its the most beneficial compared to it's alg count, in my opinion. Hope I helped. :)
 

jskyler91

Member
Joined
Mar 23, 2011
Messages
676
Location
Berkeley CA
YouTube
Visit Channel
If you want to keep using your oll/pll, I would suggest experimenting with Heise or ZZF2L. Both will improve your blockbuilding exponentially when practiced enough, and if you already have knowledge of EO, EOline should be easy. While I know Heise isn't useful for speedsolving, it is useful for efficiency, and in my opinion, when you are more efficient, lookahead is much easier. So you would be killing two birds with one stone, i.e. working on lookahead and learning/working on something new. I don't see a point in you learning a whole new set of algs; you know a lot of algs already, from the look of it. But if you insist on learning alg sets, I would suggest Rowe's RLS, its the most beneficial compared to it's alg count, in my opinion. Hope I helped. :)

That is helpful, although I just wonder how often RLS comes up unless you actually mirror it to all angles. I have been debating this for a while. It seems like I would learn about 80-90 algs to be able to use them 1 in 6-7 solves, maybe less since many of my f2l cases don't end with a block formed and then inserted. This is my only real qualm with RLS, that and the annoyance of mirroring. I feel like COLL or OLLCP is better because it doesn't really involve mirroring and can be used any time you run into that case. I am not sure, I guess I will see how many people vote similarly to you. :)

EDIT: I guess the chances of getting a COLL or a OLLCP are less than 1 in 7, but then if I do R-OLL and avoid ENVY I can do this every solve practically.
 
Last edited:

A Leman

Member
Joined
Jan 22, 2012
Messages
631
Location
New Jersey
After you stop learning algs, I doubt you can practice tps nonstop. You can use block-building like petrus,heise,roux to recover between fast speed days. They will help improve your cross and inspection.You could also try to maximize what you see in inspection by trying to plan out most of the f2l. But before you stop learning new algs, Dacuba made a great point-study reconstructions and find a more advanced last layer subset to learn(like parts of rls or the 2gll for H and PI).
 

jskyler91

Member
Joined
Mar 23, 2011
Messages
676
Location
Berkeley CA
YouTube
Visit Channel
After you stop learning algs, I doubt you can practice tps nonstop. You can use block-building like petrus,heise,roux to recover between fast speed days. They will help improve your cross and inspection.You could also try to maximize what you see in inspection by trying to plan out most of the f2l. But before you stop learning new algs, Dacuba made a great point-study reconstructions and find a more advanced last layer subset to learn(like parts of rls or the 2gll for H and PI).

Ya, I put some examples in the Poll, but so is 2GLL basically ZBLL, or what is the difference, I didn't find the wiki very helpful for this. Could you explain it to me?
 

A Leman

Member
Joined
Jan 22, 2012
Messages
631
Location
New Jersey
2gll is when you get the zbll case and the corners are correctly oriented. They can all be solved with RU turns and recognized by their coll case. I guessed you learned coll for H and Pi because they are easy to recognize and might be interested in just learning those 20 zbll cases (8 for H and 12 for pi). Basically 2gll are the fingertrick friendly Zbll.
 

jskyler91

Member
Joined
Mar 23, 2011
Messages
676
Location
Berkeley CA
YouTube
Visit Channel
2gll is when you get the zbll case and the corners are correctly oriented. They can all be solved with RU turns and recognized by their coll case. I guessed you learned coll for H and Pi because they are easy to recognize and might be interested in just learning those 20 zbll cases (8 for H and 12 for pi). Basically 2gll are the fingertrick friendly Zbll.

So this woud only be for Pi and H and only when the edges can be permuted with 2 gen right? How would one go about recognizing if the could or could not solve the whole cube when they get either of these cases? It seems to me that I would recog COLL and then ask are the edges in a state that I can permute them from; this would be great for when I could, but a waste of a time when I couldn't. I don't like wasting time checking for something only to find that I can't do it, this is literally wasting time and achieving nothing for it. I like the idea of 2 gen, but not if it has the potential to waste time. I might just learn ZBLL for Pi and H though. Let me know if I am misunderstanding you. :)
 

A Leman

Member
Joined
Jan 22, 2012
Messages
631
Location
New Jersey
You said you already use coll for H and Pi which means you would already look at the hyperorientation every time to force epll.when you get the right corner case you would just use zbll instead of the coll you see.And since they are 2gen and some are a sune combos, your fingers will still do them quickly.
 

jskyler91

Member
Joined
Mar 23, 2011
Messages
676
Location
Berkeley CA
YouTube
Visit Channel
You said you already use coll for H and Pi which means you would already look at the hyperorientation every time to force epll.when you get the right corner case you would just use zbll instead of the coll you see.And since they are 2gen and some are a sune combos, your fingers will still do them quickly.

Ah, so this is for particular corner permutation cases and so then I would just, when I got that case, use full ZBLL, I see, that sounds much better, which ones is it for Pi and H?
 

jskyler91

Member
Joined
Mar 23, 2011
Messages
676
Location
Berkeley CA
YouTube
Visit Channel
Isn't COLL a subset of OLLCP?
I keep forgetting to learn R-OLL... :/

R-OLL is knowing what OLL algs do to permutation of pieces, right?

Basically, it allows you to know which corner permutation case you will have and if you use it how I plan it will allow you to avoid ENVY cases completely. This with 2 side PLL means excellent PLL Recog and speed.
 

jskyler91

Member
Joined
Mar 23, 2011
Messages
676
Location
Berkeley CA
YouTube
Visit Channel
These are the algs I use(there is a slight chance the pictures won't match the case because I did not check)

View attachment 2067
View attachment 2066

I hate Y perms
Thanks
You don't like Y-perms?!?! What's wrong with you?!

Oh, and thanks

I don't mind y perms or even E perms, but N and V are two of my worst algs and I would love to never have to do them. Also, Y perms are ok like sub 1.3 in general, but other algs are faster so I would gladly take them over ENVY.
 

jskyler91

Member
Joined
Mar 23, 2011
Messages
676
Location
Berkeley CA
YouTube
Visit Channel
voted for everything

never stop learning

I wont, but I also know that I have limits and if I have too much info in my head then I will likely be slowed down.

EDIT: This thread is mainly to see what people think I should learn before i start Speeding up everything.
 
Last edited:

PandaCuber

Member
Joined
Jun 24, 2011
Messages
1,689
Location
Uruguay
YouTube
Visit Channel
I voted for Blockbuilding xcross.

I think you shouldnt switch to another method, I just think you should take concepts from other methods and apply them to CFOP.
So take a petrus block and apply it to your cross (xcross)
Play with roux and figure out how to recognize cmll faster (coll help) Roux also helps with no re-grip. Sure you could do longer F2l algs, but whatever.

Another thing you could work on is learning more PLLs. Im sure if you knew atleast 2 algs for each pll, execution would be a lot faster. Like elimination of U2s and such.

If you want to be fast, you have to make your eyes faster. Lookahead > TPS. No pauses + High TPS = Greatness.

To resume what I said : Learn other methods, because they will give you another perspective of the cube. I not saying make Roux/Petrus your main method, but instead learn how to use them to your advantage.
 

jskyler91

Member
Joined
Mar 23, 2011
Messages
676
Location
Berkeley CA
YouTube
Visit Channel
I voted for Blockbuilding xcross.

I think you shouldnt switch to another method, I just think you should take concepts from other methods and apply them to CFOP.
So take a petrus block and apply it to your cross (xcross)
Play with roux and figure out how to recognize cmll faster (coll help) Roux also helps with no re-grip. Sure you could do longer F2l algs, but whatever.

Another thing you could work on is learning more PLLs. Im sure if you knew atleast 2 algs for each pll, execution would be a lot faster. Like elimination of U2s and such.

If you want to be fast, you have to make your eyes faster. Lookahead > TPS. No pauses + High TPS = Greatness.

To resume what I said : Learn other methods, because they will give you another perspective of the cube. I not saying make Roux/Petrus your main method, but instead learn how to use them to your advantage.

Thanks for the tips, although I already said that I do PLLS from multiple angles and some x-cross, but I agree I should do more. You guys can vote for more than one things BTW, so vote for everything that you think I should work on.
 
Status
Not open for further replies.
Top