Older cubers discussions

[mark49152]

They are about things around speedcubing.
- Introducing myself
- blindfolded solving, memory etc.
- going back from a competiotion
- my job - philosophy and connection with speesolving (I'm a fellow of philosophy on the university)
- what exists? world and words
- Euro 2016 in Prague.

I'm too lazy to make subtitles.

I wish I could understand Polish - the philosophy one sounds interesting. And your thoughts on blind and memory too.

 

[h2f]

@mark49152 Thank you for your kind words. I can see few similiarities between philosphy and speedsolving, for example thinking in a different systems. For example: Roux, CFOP or ZZ are such systems and one can move between them changing its point of view. It's hard for me to describe it in English but I find it to be one of philosophical issues. In my job I'm in the Polish analitycal philosphy which is connected with logic (semantics).

 

[teacher77]

Welcome @teacher77! This is definitely the place to be. I'm about 10 months ahead on getting into speed cubing. It's hard for me to keep track of everyone's average in here, partly because people always improve so fast. But yes there should be plenty to learn from this thread, and even those who are faster now will have good tips for where you are at. Plus you'll get lots of feedback if you post a video unlike the video area of the forum (from my experience). So looks like you are currently doing F2L? Do you do two look PLL and OLL?

Hi Jason and thanks for the warm welcome !

Yes I use 2-look OLL and 2-look PLL (that's what I meant by 4LLL - 4 look last layer). The only alg that I still don't use is the E-perm. All E-perm algs that I saw are either hard to learn or slow to perform.

And yes, I started using Fridrich-like F2L 3 weeks ago. Although my times had worsen during 2 weeks, they are now down by 6 seconds compared to what they were before F2L. See the graph I posted. I'm really proud of that 6 seconds gain in a single week.

 

[teacher77]

Welcome @teacher77 , you're in the right place

For E perm I use R' U' R' D' R U' R' D R U R' D' R U R' D R2. It helps if you get used to pushing D' moves with your left ring finger, as well as pulling D moves. Once you get the hang of it, it flows beautifully.

That's 17 (rather hard to remember) moves while using A-perm twice is 18 (easy) moves. How can it be much faster ?

 

[Jason Green]

Hi Jason and thanks for the warm welcome !

Yes I use 2-look OLL and 2-look PLL (that's what I meant by 4LLL - 4 look last layer). The only alg that I still don't use is the E-perm. All E-perm algs that I saw are either hard to learn or slow to perform.

And yes, I started using Fridrich-like F2L 3 weeks ago. Although my times had worsen during 2 weeks, they are now down by 6 seconds compared to what they were before F2L. See the graph I posted. I'm really proud of that 6 seconds gain in a single week.

I saw later, I missed the part under your graph somehow before.

As far as algs being hard to remember, as you probably know once they get into your muscle memory there is really nothing to remember at all. As someone mentioned, algdb is a great way to try how different options feel. I still have a few algs I might consider changing some day.

 

[teacher77]

The logical next step for you would be to learn full PLL. I just finished that a few weeks ago. It might seem like a massive undertaking, but it's really not that bad. I started by learning the cases that have one side already completely solved, the J perms and F perm. They are easy to recognize during solves and have nice algs. The other option, if you want to reduce the amounts of E perms you need to do, is to start by learning the PLL cases with diagonal corner swaps. N perms are rare, but the algs are among the easiest to learn (not necessarily fastest to execute). The Y perm is really nice and easy to recognize, V perm is also easy to recognize, though the alg is still messing up my head sometimes...

Thanks, this is great advice ! What source are you using for algs ? I use the ones on badmephisto's site (except the Eperm, which I can't stand). Do you recommend them ?

Also, I was wondering if you had tips to help recognize the different cases ? How many sides do you need to look at ?

That said, you probably still have a lot of room for improvement also by improving F2L and look ahead. That's an area where you never stop to improve. Slow solves where you concentrate on trying to find the next pair while solving the previous pair are very good to do. I think it's good to do it mostly intuitively, but if there is some case that always takes you a lot of moves to solve, you can look up some alg for how it could be done more efficiently. algdb.net is a good source for algs.

Yes, this is what I'm concentrating on right now. In fact, I find it hard to look ahead because I'm not always looking at the top of the cube (I like solving F2L cases where a corner is on the D layer and I need to look under to identify it ; I'll post a video where it shows ; advice on this would be great). I might be silly to say, but I had a much easier time at looking ahead when I used a layer by layer approach for F2L.

 

[teacher77]

For the E-perm watch cyotheking's video. When I first learned it I did D' moves with my right ring finger and it was terrible. If you do all D and D' moves with your left ring finger the standard alg is really a very nice alg.

Which alg do you call the "standard" alg for Eperm ?

 

[h2f]

Which alg do you call the "standard" alg for Eperm ?

I think badmephisto's is a standard alg. It's very fast and when you know standard OLLs you can see that it's made of 2 OLL's.

 

[teacher77]

Here's a video showing what I said about solving pairs with their corner on the D layer. It's not a PB or anything but if you care to comment, I'd gladly listen

EDIT : seems like the vid was too large. Lemme find a way to shrink it.

 

[Jason Green]

I used pretty much entirely bad mephisto's cheat sheet for OLL and PLL. I think the only one I changed was the N Perm that the guys here told me about.

 

[JanW]

That's 17 (rather hard to remember) moves while using A-perm twice is 18 (easy) moves. How can it be much faster ?

You also have to account for recognition time and cube rotations. If you do the E perm as 2 A perms, there's a stop in between.

I also disliked the E perm at first, it takes a while to get used to. I'm still not very fast performing it, but it keeps getting better. On youtube you can look up cyotheking's "E perm like a pro" video and see that it can be done in <1 second (though probably very few people are capable of this...). His like a pro -videos are very good for learning how to finger trick PLL algs.

Thanks, this is great advice ! What source are you using for algs ? I use the ones on badmephisto's site (except the Eperm, which I can't stand). Do you recommend them ?

Also, I was wondering if you had tips to help recognize the different cases ? How many sides do you need to look at ?

I'd recommend algdb.net. There are several different algorithms for every case to choose from. Try many of them and find the ones you like the most. I got my algs from there, but mostly I ended up picking the same algs that are on badmephisto's cheat sheet. It seems I ended up using different algs than badmephisto's for one of the R perms, F perm and a couple of G perms.

In theory it's possible to recognize PLL from any 2 sides. I can't. I can recognize most of them from 2 sides, as long as I'm looking at the correct 2 sides. To recognize the ones I mentioned with one side solved, J perms and F perm, you see one solved side, then you need to look at one other side, any other side, to recognize which case it is. For some of the others it's about locating the headlights, then locating a pair and seeing where it is in relation to the headlights. The only cases I can recognize from any 2 sides are the ones used for edge permutation in 2 look OLL, H perm, Z perm and U perms.

 

[mark49152]

That's 17 (rather hard to remember) moves while using A-perm twice is 18 (easy) moves. How can it be much faster ?

It flows nicely and is all quarter turns apart from the last. The moves are not hard to remember once you see the patterns in them. The R face oscillates up and down and the U/D moves alternate back and forth into it.

The alg I think of as the standard E perm is sometimes called the Rowe E perm after Rowe Hessler who popularised it. It's very similar to the algorithm I gave except it starts with an x rotation and does not have the trailing R2. Personally I prefer to trade the rotation for the move.

I find A perm faster than E perm but I could not do two of them with an AUF faster than an E perm. If two A perms works for you then fair enough, stick with that until the minor speed increase makes it worthwhile changing.

 

[One Wheel]

I've tried to do this, but I've never been able to figure out how to do it so that it would become "really nice"... I now tried it again and can see the benefit of performing the first D' move with left ring finger. I'll try to get used to that. The last D' move, however, seems to be perfect to execute with right ring finger...

How does that work? When I try that my right ring finger is out of position from the last R'. I might be doing my R moves all wrong, I suppose.

 

[One Wheel]

I used pretty much entirely bad mephisto's cheat sheet for OLL and PLL. I think the only one I changed was the N Perm that the guys here told me about.

I also like a different U-perm than badmephisto's, also from Chris Olson: M2 U M U2 M' U M2. Switch out U for U' (push with right index finger) to rotate the other way. It just flows very nicely, and I could never do the R U one without a regrip, although I use it for bigger cubes (5+)

 

[JanW]

How does that work? When I try that my right ring finger is out of position from the last R'. I might be doing my R moves all wrong, I suppose.

Or then I am doing my R moves wrong... I tried gripping R layer differently, in such a way that I can't do D' moves with right ring finger, and now it feels a lot more stable. I must try this a bit more, I think it can be faster. Thanks!

Edit: Yup, I was doing the R moves wrong. Tried the new version about 20 times and it's already faster than my old way of doing it.

 

[newtonbase]

I also like a different U-perm than badmephisto's, also from Chris Olson: M2 U M U2 M' U M2. Switch out U for U' (push with right index finger) to rotate the other way. It just flows very nicely, and I could never do the R U one without a regrip, although I use it for bigger cubes (5+)

I use a similar U perm when the bar is at the front so it's M2 U M' U2 M U M2 and the usual R U one when it's at the back. I worked it out from a blindsolving comm but then found out it was already well known.

 

[newtonbase]

Anyway, I came here to say that I tried a rare timed 3BLD solve tonight. Thought I'd blast it to see what sort of time I could potentially get without worrying about accuracy. Good news was that I was only a twisted corner off. Bad news was that it was only marginally quicker than my sole official solve despite an easy scramble and nearly 3 months more practice. I have 6 days to fix this. PBs are going to be scarce in Macclesfield.

 

[mark49152]

Mark, what's your split of memo versus execution?

 

[newtonbase]

Mark, what's your split of memo versus execution?

I have no idea. The opportunity for timed solves is so hard to come by that I've never looked at splits. If I get a chance I'll have a look at the videos from North London

 

[teacher77]

I think badmephisto's is a standard alg. It's very fast and when you know standard OLLs you can see that it's made of 2 OLL's.

I just noticed that badphemisto actually has several E-perm algs.

On his "2-look PLL cheatsheet", it says x' [ ( R U' R') D ( R U R') ] u 2 [ ( R' U R ) D ( R' U' R ) ].

THAT's the one I can't stand.

On some youtube video, I found this one : z U2 R2 F (R U R' U')3 F' R2 U2 z'.

THAT's the one that I find slower than two A-perms.

However, on badmephisto's regular PLL page, it says : x' (R U' R') D (R U R') D' (R U R') D (R U' R') D'.

NOW this one is great. MUCH faster than two A-perms. In fact it's nearly as fast as a single A-perm.

A special thanks to @One Wheel for pointing out the existence of cyotheking's video, where he uses this last alg.