Why would you argue *not* to learn something?

[brunson]

This was posted in a different thread, but pushed me ask a question that's been bugging me for a long time.

You mean OCLL. What I don't get is, why would you want to learn 100+ CLS algorithms when you already know all 20 F2L cases (with the edge oriented) and all 7 OCLL algorithms?

I've seen this sort of position in other discussions and it confuses me.

I'd like to turn it around and ask: Why would you *not* want to know 100+ CLS algorithms? Why would you not want to know Full OLL, COLL, OCLL, CMLL, PLL, MGLS and all of ZB for that matter? How can you possibly imagine that knowing *more* about anything is a negative in any way?

The only thing I can think that you're arguing against is the effort that it would take to learn the algorithms, but then why even bother learning to solve a cube at all? You could just watch TV instead.

People please chime in if you see this pattern as I have:

Someone asks "should I learn X?" Most who already know X respond, "Sure, you may as well, what's the harm?" Meanwhile, a horde of people that haven't put forth the effort to learn said X chime in with "no, it's a waste of time, don't bother" and offer a crap load of analysis on why it's useless. Rarely (ever?) have I seen someone exclaim, "You know, I learned all 200 algorithms for this method and I really wish I hadn't, it was a monstrous waste of time, and I recommend don't go near it!"

Is it just me out on a limb? Anyone else calling "sour grapes"?

Curious.

 

[IamWEB]

lol chime in.

anyway...

Yes, you could just go watch TV, but the cube obviously interests the person. They feel like putting in the effort required to learn how to solve it, but not the effort to learn 200 algorithms (at least when all 200 are required for one method/step/system).

It's all about how far you're willing to go.

 

[JTW2007]

I agree. I'm not really sure what the negative is. Who cares what the process is like? I'm interested in the end product.

 

[nitrocan]

The harm: More cases to recognize and more time to waste. Putting effort into something isn't something humans love to do.

 

[Johannes91]

Why would you not want to know Full OLL, COLL, OCLL, CMLL, PLL, MGLS and all of ZB for that matter? How can you possibly imagine that knowing *more* about anything is a negative in any way?

learning ≠ knowing

The only thing I can think that you're arguing against is the effort that it would take to learn the algorithms

This. As an extreme example, I wouldn't recommend memorizing thousands of "cross algs". It's simply not worth the effort it would take, and there are much better ways to practise if you want to get better at cubing.

People please chime in if you see this pattern as I have:

[..]

Anyone else calling "sour grapes"?

In many cases, definitely. Like all the color neutrality discussions. But not always; some things *are* useless (or at least significantly less useful than some other things), and you should do them only if you enjoy the process itself (like memorizing algs), not because you hope it will be useful in the future.

 

[Si1v3rx51ay3r]

recognition would take practice, of course. larger scale=more time spent.

but if you are dedicated enough then go for it =D

 

[JTW2007]

Since Johannes brought up color neutrality, I'll revise my opinion to cover a case that I had neglected to consider before. I think that if the skill in question is undeniably faster than what you're using, there's no reason not to learn it. However, if there is some debate about what's faster, use what you're most comfortable with. Full OLL is faster than 2-look, and there's not much of an argument against that. So full OLL is worth learning. Fridrich is faster than the Dan Knights LBL method, so Fridrich is worth learning. Something like color neutrality does have equally legitimate argument from either side, and evidence backs this up. There are incredibly fast color neutral solvers, and there are incredibly fast single-color (monochromatic?) solvers. So that is up to the cuber to decide.

 

[Johannes91]

I think that if the skill in question is undeniably faster than what you're using, there's no reason not to learn it.

Consider memorizing an optimal solution to all cross positions. Or all extended cross positions. Knowing them would certainly be faster than not, but I don't think that's the only thing that matters. If something takes an unreasonable amount of effort to learn and there are other things you could practise to improve much more, that's a very valid reason to not learn it.

Something like color neutrality does have equally legitimate argument from either side, and evidence backs this up. There are incredibly fast color neutral solvers, and there are incredibly fast single-color (monochromatic?) solvers.

That's not convincing at all. There aren't enough world class cubers averaging very near each other to draw general conclusions like that. In my opinion, it makes more sense to compare a color neutral cuber's results with and without color neutrality. Many single-color solvers have always been that way, and you just can't know what their results would be like if they were color neutral.

 

[Sin-H]

I think that if the skill in question is undeniably faster than what you're using, there's no reason not to learn it.

Consider memorizing an optimal solution to all cross positions. Or all extended cross positions. Knowing them would certainly be faster than not, but I don't think that's the only thing that matters. If something takes an unreasonable amount of effort to learn and there are other things you could practise to improve much more, that's a very valid reason to not learn it.

Something like color neutrality does have equally legitimate argument from either side, and evidence backs this up. There are incredibly fast color neutral solvers, and there are incredibly fast single-color (monochromatic?) solvers.

That's not convincing at all. There aren't enough world class cubers averaging very near each other to draw general conclusions like that. In my opinion, it makes more sense to compare a color neutral cuber's results with and without color neutrality. Many single-color solvers have always been that way, and you just can't know what their results would be like if they were color neutral.

seconded.

 

[blah]

Maintenance cost. That's what I call it.

@brunson, the guy you quoted is me. I've mentioned in another thread that I *have* finished learning all of MGLS at different points in time. I learned different CLS cases when I was exploring different alternative solutions. But never at any single point in time did I know all of CLS, and as of now, I practically know none of CLS. I also keep forgetting half my COLL algorithms and have to relearn them again and again and again everytime I decide to do OH or big cubes.

It's not that there's anything bad about additional knowledge. But the fact is, nothing really stays in your long term memory unless you practice it over and over and over again. I'm pretty certain that I'll forget some of my OLLs if I don't touch my cubes at all for say, 6 months.

It's the maintenance cost.

 

[Cride5]

It's not that there's anything bad about additional knowledge. But the fact is, nothing really stays in your long term memory unless you practice it over and over and over again. I'm pretty certain that I'll forget some of my OLLs if I don't touch my cubes at all for say, 6 months.

Seconded... As an example ZBLL has an obvious move-count advantage over OCLL/PLL. However, more algs = less practice on each one. My OCLL algs are generally faster than my PLL algs not just because they're slightly less moves, but also because I use them 3x as much and the small number of cases have lightning-fast recognition. There's deffo a trade off between number of algs and speed of recognition, recall and execution. The move-count advantage gained by using more algs isn't an advantage unless they can be executed at similar or higher speed (assuming they're being used for speedsolving).

 

[cookingfat]

What about if someone said "why would I want to learn the OLLs with no edges oriented? I use partial edge control to make sure that I never come across these cases"

 

[rachmaninovian]

r4/r5 algs are pretty forgettable for me. even if i practise big cubes like crap i might just forget some algs here and there :O i know about ~100 algs in total for big cubes...
so yea i have to agree with blah's maintenance cost thingy.

 

[Ewks]

The only thing I can think that you're arguing against is the effort that it would take to learn the algorithms, but then why even bother learning to solve a cube at all? You could just watch TV instead.

Don't you guys get that cubing isn't all about algs and constatly learning new methods etc. I for one am really lazy at learning new algs I might have a puzzle unsolved for months because I just can't get myself to learn that single alg, but still I enjoy cubing and I think I could say that I'm rather fast at it too.

There are things like looking ahead etc. which take high amount of practise to get good at and I would put practising those ahead of learning new algs.

I would say that think about when learning new algs/methods/shortcuts. If you're at 30 or 20 second average don't learn too much because if you only learn algs and practise them you'll never get good at looking ahead etc. When you're sub-15 or maybe more like sub-13 and you have shaved off every single second you can just by practising execution, look ahead etc. then learn new short cuts or what ever you want to learn.

I'm ending my post by saying once again that there's more in cubing than just learning algs. You don't have to make the effort of learning 200+ algs to get fast.

 

[CuBeOrDiE]

Well...

first off, it depends how lazy you are. Not every one wants to learn 100 plus algorithms

then, It would be hard to remember them all. And, in order to remember all those algs, you must constantly practice them

lastly, sometimes, it isn't neccessary to learn that many things. Helpful, but not essential

That's my opinion.

 

[Kit Clement]

I look at it this way: when I sit down to start cubing, I can do one of two things to get faster:

1. Practice solving
2. Learn algorithms

One of these being much more fun than the other. I'm not saying I don't learn new algorithms ever, but I do it at a much less frequency than normal practicing.

 

[brunson]

There are several reasonable opinions on why some things are not worth the time. Johannes, that was a good reductio ad absurdum, so you're right there are definitely things that don't make sense. Jason Baum has commented on the recognition problems with ZBLL. You're also correct in pointing out that I use learning and knowing interchangeably, but I mean "knowing" not "cramming" (see below).

I'll comment on a couple of points I disagree (minorly) with.

First that comes to mind is "there's more to cubing than learning algs". I think I'd counter that with "there's more to cubing than just getting fast". I happen to enjoy algorithms, not necessarily the learning part, but the decomposition and analysis of them. My constant algorithm study probably slows me down since I'm always introducing new things to think about, but I've also come to the realization that I'll never be as fast as you young 'uns, so that's not such a big deal.

@blah, I removed your name from the quote tag because I didn't mean this to be directed at you, it's more of a pattern I've seen and yours was the last comment I read before I posted. However, "learning" all of MGLS but not all at the same time would, to me, be analogous to learning all of PLL, but only remembering half of them at any point. I wouldn't consider that "learned", I'd more likely call it "crammed", like staying up all night studying for a calc test, then forgetting all you studied after a week. MGLS works well for Lucas and he used it in his 10.00s record.

Still on blah's post, but a little off topic. I forgot how to do a J-perm last week, an algorithm I've used daily for two and a half years. I was solving before bed and hit a J and could not for the life of me remember how to do it. I refused to look it up, but couldn't get it out of my head and kept going over and over it in my mind until I fell asleep. I woke up at about 2am and it was in my head: R U R' F' and I was like, "F', F', F', that's what I was missing!" I crept quietly out of bed, took my cube to the bathroom and then did about 50 J-perms in a row before going back to sleep. That's not really what you were talking about, but it was a funny story to me and I thought I'd share.

I'm not going to harp any more on the color neutrality issue, it's been done to death, I'll just reiterate a couple of my positions from other postings. If being color neutral slows you down, you're not really color neutral. I have yet to come across anyone that used to be color neutral and went back, once they have it it seems to be a skill no one wants to give up. Finally, if you have never been color neutral, then you are unequipped to offer a valid discussion on its merits or lack therof.

@cooking fat, I've made this point before: What if you insert your third corner/edge pair using an open slot technique, knowing you'll orient two edges on the final insertion, then you discover you've "accidentally" made your last pair and you're looking at OLL #1? You're in for a two look OLL or else removing and reinserting then corner. I only do partial edge control on the last corner, so if you're better at it than me and can start orienting edges earlier, then that may never happen.

Everyone made excellent points, and though I may not agree with all of them, there's a good chance not everyone agrees with me, so that's fine. I'll also go ahead and point out that pretty much everyone that responded is faster than I am.

Please continue discussing if you're interested, I'm really enjoying reading all your responses.

 

[cookingfat]

@cooking fat, I've made this point before: What if you insert your third corner/edge pair using an open slot technique, knowing you'll orient two edges on the final insertion, then you discover you've "accidentally" made your last pair and you're looking at OLL #1? You're in for a two look OLL or else removing and reinserting then corner. I only do partial edge control on the last corner, so if you're better at it than me and can start orienting edges earlier, then that may never happen.

Well I'm not saying that you're wrong, you probably have much more experience than I do, I'm not even sub-25 yet and I don't know all OLLs. I don't do partial edge control myself either, I just saw that someone on this forum (can't remember who) has stated that they have intentionally not learnt these OLLs because they can always orient 2+ edges during the last pair. I just wanted to know what you thought of that and you've answered really well.

I actually did the complete opposite and learnt the 'no edges' OLLs first to avoid a 3-look OLL.

 

[Jokerman5656]

Still on blah's post, but a little off topic. I forgot how to do a J-perm last week, an algorithm I've used daily for two and a half years. I was solving before bed and hit a J and could not for the life of me remember how to do it. I refused to look it up, but couldn't get it out of my head and kept going over and over it in my mind until I fell asleep. I woke up at about 2am and it was in my head: R U R' F' and I was like, "F', F', F', that's what I was missing!" I crept quietly out of bed, took my cube to the bathroom and then did about 50 J-perms in a row before going back to sleep. That's not really what you were talking about, but it was a funny story to me and I thought I'd share.

Haha! I've done that exact same thing but with a big L OLL.

 

[Kian]

Well, my thoughts are basically this...

There are a finite number of things you could learn/be able to execute effectively on a cube. Thusly, you might want to be reasonably judicious about the things you spend your efforts learning. That's the only real reason I can consider.