• 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!

Random Cubing Discussion

CrispyCubing

Member
Joined
Dec 31, 2019
Messages
200
WCA
2019YODE02
YouTube
Visit Channel
I wasn't sure where else to post this, but I had a proposal/question about cubing terminology.

As many of us know, "Lookahead" refers to the process of determining what the next step of your solve will be while still executing the current step. This is done to reduce pauses.

However, "Lookahead" is almost exclusively used to refer to the intuitive part of solves. I think there should be a similar term, "Lookinto", which refers to this same concept when applied to the algorithmic part of solves. I don't believe this has a term yet, other than "Prediction" which is often used informally when this topic is discussed.

Some examples of Lookinto include:
  • [CFOP/ZZ] Predicting AUF after PLL
  • [CFOP/ZZ] Predicting PLL (or just CP case) after OLL
  • [CFOP] Using non-standard algs for 2-look OLL in order to predict the second look (link to Jperm's video on the topic)
  • [Roux] Predicting EO before executing CMLL and/or using a different CMLL alg for a better EO case
  • [Roux] EOLR (this borders on lookahead since this is an intuitive step, so its inclusion is debatable)
While these are the more-or-less standard applications of Lookinto, there is more potential for the concept. For example, we can choose to use COLL algs which preserve a 1x2x2 block in LL in order to Lookinto EPLL. This can lead to 1LLL for many cases when edges are oriented (does not apply to H/Pi cases).

I'm interested to hear your thoughts! Does this already have a name, or do you think we shouldn't introduce this new term? What other applications of Lookinto can you think of?
I think prediction is already a pretty good term to describe this. Even if you don’t completely know what the next alg will be, you can still predict a few pieces.

Do you have other reasons to not use prediction? I don’t see how people using it informally means we need a different special term for it.
 

Waffles

Member
Joined
Apr 4, 2021
Messages
406
Location
Melbourne, Australia
Also I have something to actually discuss: does anyone actually use EOLL or something like it? I know COLL is popular because of like recognition and fast algorithms but like most people can do A perms sub 0.8. Yes there is a chance of am E perm but tbh it’s kind of like the Z perm of EOLL.

Also how many algorithms would there be and would it be a waste of time?

Just saying: I’m not actually considering learning it, I was just kind of wondering.
 

ruffleduck

Member
Joined
May 9, 2021
Messages
1,123
Location
Playing chess
YouTube
Visit Channel
Also I have something to actually discuss: does anyone actually use EOLL or something like it? I know COLL is popular because of like recognition and fast algorithms but like most people can do A perms sub 0.8. Yes there is a chance of am E perm but tbh it’s kind of like the Z perm of EOLL.

Also how many algorithms would there be and would it be a waste of time?

Just saying: I’m not actually considering learning it, I was just kind of wondering.
EP recognition is much harder than CP recognition, and A/E perms are slower than U/H/Z in both recognition and speed. I highly doubt anyone uses EOLL and if they do they should stop using it.
 

MethodNeutral

Member
Joined
Jan 4, 2017
Messages
48
I think prediction is already a pretty good term to describe this. Even if you don’t completely know what the next alg will be, you can still predict a few pieces.

Do you have other reasons to not use prediction? I don’t see how people using it informally means we need a different special term for it.

That’s true, I just think the concept could be explored more like in the way I brought up, and by giving it a name we can bring more attention to it. I think prediction is a fine term, but we could have also used “prediction” instead of “lookahead”. By giving it a term I think it warranted more focus from the community, and I think prediction/lookinto deserves more focus.
 

xyzzy

Member
Joined
Dec 24, 2015
Messages
2,876
Nobody really should be caring about this, but I do have to defend my reputation because Christopher Mowla is, somehow, really upset that I decided to spend time trying to teach him things.

See the edit at: https://www.speedsolving.com/threads/one-answer-puzzle-theory-question-thread.56699/post-1443444

This is a pattern I recognise because we're similar people. We both think we're very smart (I mean, it's true, we both are very smart), and we're both very proud of our work and like to plug it every time we have a chance to. And because I recognise it and I see myself in this, I have to point it out: he knew he wrote his original 2-gen corner permutation proof under rushed time conditions, he admitted that his "proofs" were probably incomplete, he had been told years ago that the proofs were flawed (this wasn't something I sprung on him just last Saturday), and he still plugs his Math StackExchange answer repeatedly.

(To be clear, I don't take issue with the rest of that answer. My only beef is with the 2-gen corner permutation part. Unfortunately, that's also the most interesting part and the part that deserves the most scrutiny.)

Perhaps I should have worded my initial response more cordially, but seeing incorrect proofs repeatedly being posted annoys the hell out of me. You may remember I had a link to a LessWrong article about fake explanations in my signature for about a year (before removing it somewhat more recently). This is exactly the kind of fake explanation I had in mind. (I'm not saying that Christopher is the only offender. He's not even the biggest offender! This is just the most recent incident I can think of.) You can string together a bunch of keywords and maybe give readers an impression that you know what you're talking about, but if it falls apart upon closer inspection, (i) it fails to serve the purpose of an explanation and (ii) now you've tricked less-critical readers into accepting a non-explanation. I consider preventing (ii) from happening to be my life duty.

If your objection is that "at least he tried to write a proof; you haven't!", then I should point out that I actually have, three times over. I just never posted any of them here. One of them exists somewhere in my disorganised pile of notes (this is the outer automorphism proof I sketched in another post), one of them exists on my personal blog, and one of them exists on a different forum. I vaguely had plans to turn one of them into a video lecture but real life and general laziness got in the way. (Maybe I should now, while the topic is still fresh in my mind. If there's only one good thing to come out of this petty dispute…)

ALSO FOR THE NTH TIME I WANT TO MAKE IT CLEAR: I'm trying to not hold any personal grudge against Christopher. Despite his somewhat snarky response in the puzzle theory OAQT (I'm not blind to snark; I'm snarky all the time, how could I be?), I still mentioned him in the hopes that he would find what I wrote interesting and hopefully bury the hatchet there. But nope, eyes-up reaction, for god knows what reason. And let's not forget how he referred to my suggestion of partially disabling animated avatars with a bullet list of points of how it's a bad idea; never mind that I only suggested that as a compromise between disabling animated avatars entirely and still having them on the site, in an attempt to placate as many people as possible.

I half-expect this post to get deleted, but hopefully I'll have made my point clearly enough before that happens.
 
Top