Older cubers discussions

[defhacks]

I'll see if I can get this on camera in the next week or two, I'm already convinced I can make this work. Adjusting to this new stuff is going to seriously slow me down for as much as a couple months but once the new habits are instinctive it's gonna be a huge boon for my solves.

Sounds similar to:

It's both inspiring and frustrating how easy he makes it look..

Must be that amazing Aolong V2 hardware he's using

 

[JohnnyReggae]

After a bunch of untimed blind solves and a few timed DNF's I finally managed to get a timed blind success. I'm still struggling with memo and always have to go back to corners after doing edges just to remind myself what I memo'd during corners in the first place. It is a little frustrating because if I eliminated the double memo I could instantly drop a couple minutes off the times. I did manage to follow it up with another success albeit a little slower, still a success.

Still chuffed non the less

 

[pglewis]

Sounds similar to:

It's both inspiring and frustrating how easy he makes it look..

Must be that amazing Aolong V2 hardware he's using

Always so fun watching him and just shaking my head.

In a quick watch it seems like he's mostly using thumb over the top for F/F'. Also, being Collin Burns, he's faster doing weird things than I am with regripless 2-gen . I'm comfortable with thumb from the top or bottom but there are lots of cases I would like to be able to do F face turns while leaving the thumbs in place, thus this attempt to use the index fingers.

Grip: thumbs front, ring fingers on the back, pinkies bracing the S layer; index and middle finger of both hands floating and free. At first I didn't trust stability on the S layer and the turns were tentative, awkward, and not very promising but with a little conscious attention to bracing with the opposite-hand pinky I was able to convince myself to just push down with the index and all will be ok. The exact grip is new and I don't have any muscle memory built up yet but if I'm in that position I definitely have workable, regripless sledge and hedgeslammer for front slots with it that land right back in neutral grip.

I already added a lefty U2' to my toolbelt for the first time in over 3 years of speedsolving so my solves are already slightly broken for a little while; this is the perfect time to see if I can get the index F/F' turns into muscle memory as well.

After a bunch of untimed blind solves and a few timed DNF's I finally managed to get a timed blind success. I'm still struggling with memo and always have to go back to corners after doing edges just to remind myself what I memo'd during corners in the first place. It is a little frustrating because if I eliminated the double memo I could instantly drop a couple minutes off the times. I did manage to follow it up with another success albeit a little slower, still a success.

Still chuffed non the less

Congrats!

If your sticker-to-speffz translation (or whatever scheme) is slow then you'll trace slower and have to retain memo longer. One drill I've done is tracing as fast as I can, reciting the letters without any attempt to memorize. Corners are slower for me, that extra sticker pauses my brain. It comes together quickly with practice though, you'll surpass me at 3bld in no time with a little work.

 

[JohnnyReggae]

Congrats!

If your sticker-to-speffz translation (or whatever scheme) is slow then you'll trace slower and have to retain memo longer. One drill I've done is tracing as fast as I can, reciting the letters without any attempt to memorize. Corners are slower for me, that extra sticker pauses my brain. It comes together quickly with practice though, you'll surpass me at 3bld in no time with a little work.

I should give this a try. My sticker translation is getting faster so hopefully with some more practice it will flow a lot quicker.

 

[mark49152]

After a bunch of untimed blind solves and a few timed DNF's I finally managed to get a timed blind success. I'm still struggling with memo and always have to go back to corners after doing edges just to remind myself what I memo'd during corners in the first place. It is a little frustrating because if I eliminated the double memo I could instantly drop a couple minutes off the times. I did manage to follow it up with another success albeit a little slower, still a success.

Going over memo again is fine. Generally it's better to revise it later, than spend ages on it in the first place. An initial memo plus 1-2 reviews is good (memo corners, memo edges, revise corners, revise edges, then optionally revise both again). Sounds a lot, but you'll spend less time on each step. Your speed will naturally improve.

The tracing practice method Phil suggested works well. It's good to break it down and understand where your bottlenecks are. If your bottleneck is tracing, sighted solving helps too. You can memo & solve 2-4 pieces at a time, so you get to practise tracing and your solving method without taxing your memory too hard.

 

[JohnnyReggae]

Going over memo again is fine. Generally it's better to revise it later, than spend ages on it in the first place. An initial memo plus 1-2 reviews is good (memo corners, memo edges, revise corners, revise edges, then optionally revise both again). Sounds a lot, but you'll spend less time on each step. Your speed will naturally improve.

The tracing practice method Phil suggested works well. It's good to break it down and understand where your bottlenecks are. If your bottleneck is tracing, sighted solving helps too. You can memo & solve 2-4 pieces at a time, so you get to practise tracing and your solving method without taxing your memory too hard.

Good to know that I'm actually on the right track then I do think my corner memo recall is hampered by how long it takes to memo edges, so hopefully the faster that gets the less problem I'll have. I've been using word pairs for corners and audio for edges which seems to be working for me for now. As for execution I seem to get that right most times when doing untimed solves.

 

[Tom Joad]

As I mentioned in the FB group my latest 3x3 fad has been VHLS in order to get some more advanced EO control in my tool belt. I know it's not going to provide a huge benefit on its own but I decided to take it on anyway because 1) there aren't many algs, 2) the algs are all fairly short, and 3) recognition is very very easy.

The bigger story is the deeper rabbit hole it has sent me down on fundamentals: grip and finger tricks. I had already started to pay attention to keeping in neutral grip as much as possible during F2L in slow solves and drilling the VHLS algs seemed like a perfect exercise to work on that as well.

The first problem to overcome is the lack of a U2' move with my left. I learned a lot of my algs with a right hand re-grip at the start to accommodate the fact that I could only do U2 with my right. Slowly examining what I do for U2 with my right made me realize I've been gripping too low on the back with my left ring finger. With smallish hands that makes it difficult or impossible to get a good flick with my middle finger for the 2nd turn. And just like that, now I have a usable U2' for the first time and a bazillion algs with a bad re-grip habit to break.

Next up is the hope of smooth F/F' moves without a re-grip, I'd definitely like to get a workable sledgehammer without re-grips. It's a beautiful thing when the whole last slot + OLL flow nicely without a single re-grip.

Regardless of whether or not I end up actually using VHLS much my F2L is going to be so much better once i break some of my instinctive right hand re-grips and am comfortable with the finger-trick alterations.

See latest jperm vid on which alg sets to bother with... only his opinion , of course.

I don’t even know what VHLS is and I still do no edge orientation when I insert my last pair... edit, I just watched the video and now I know what it is! Thought it was just an alternative to beetamax

 

[mark49152]

Good to know that I'm actually on the right track then I do think my corner memo recall is hampered by how long it takes to memo edges, so hopefully the faster that gets the less problem I'll have. I've been using word pairs for corners and audio for edges which seems to be working for me for now. As for execution I seem to get that right most times when doing untimed solves.

I would recommend using images for both. Audio is short term, and at the times you're talking about, especially since you're having to review corners after edges, you'd probably fare better using longer term memory for edges too.

 

[pglewis]

See latest jperm vid on which alg sets to bother with... only his opinion , of course.

I don't know if he's scanning this thread or just has a crystal ball. Me: "I need to work on regrips"... 3 days later...

I think one reason VHLS is discouraged, and what JPerm seems to emphasize, is some people want to use it in lieu of learning full OLL and I wouldn't suggest that. There are already plenty of other documented and respected opinions on why not to bother with it, my main reasons to go ahead and swim upstream:

• The alg count is tiny. For connected pair there are 7 unsolved cases. One is standard insert, one is sledgehammer; that leaves just 5 more to cover all connected pair cases (if mirroring). R U R' cases are just 6 more (plus mirrors), since normal insert is one case. I had it learned in two days of drilling, it wasn't a huge investment if I drop it and never use it again.
• I think this is highly overlooked: it's not only useful on last slot, it has value for general edge control. If I notice a misoriented F2L edge in the target slot I have options to flip it during the insert. I have no real metrics but I've had a "fair number" of solves where I was able to use it to fix up EO on the 2nd or 3rd pair and thus rotationless for the rest of the solve.
• OCLLs (all edges oriented) are among my fastest OLLs. What's the worst of the bunch, Pi? Maybe H? It would be rare to twiddle my way into a worse case than a normal insert would have given me, bypassing 6 move T and P OLLs are the only ones that come to mind. OTOH: my OCLL recognition is very good compared to OLL in general, I can do S/AS/H from any angle without AUF and can mirror all the rest, almost all of them are 2-gen, and I can do all of them regripless. The payoff on last slot isn't all that great but it's not awful either.
• It's a four-case subset of ZBLS, so everyone that uses ZBLS in order to use more ZBLL had to learn 'em.
• This should probably be listed first: it's fun for me at the moment

I think the best argument against it is it's contrary to the CFOP philosophy, as I see it: leveraging the fact that brute force is often faster for one-time problems vs. more fancy, elegant approaches. VHLS recognition is nearly as easy as 2-look OLL and only requires looking at the edges on top... but it does add an extra look, slight as it might be.

Thought it was just an alternative to beetamax

 

[JohnnyReggae]

I would recommend using images for both. Audio is short term, and at the times you're talking about, especially since you're having to review corners after edges, you'd probably fare better using longer term memory for edges too.

To be honest I seem to have better recall with the audio than with the images. Typically I come across 3 to 4 word pairs with corners and 5 to 7 for edges and my edge recall is always better than corners. I've thought about using audio for corners as well, but I'm going to persevere with the images for now.

 

[pglewis]

To be honest I seem to have better recall with the audio than with the images. Typically I come across 3 to 4 word pairs with corners and 5 to 7 for edges and my edge recall is always better than corners. I've thought about using audio for corners as well, but I'm going to persevere with the images for now.

My experiments with audio memo have been... let's just call it "less than promising" lol. I think if you're accurate enough with audio for edges already it's awesome, conjuring images is just one more thing to slow my memo personally. At least for 3bld. If you have an interest in mbld you'll need to use images to retain it unless you're some sort of audio memory savant.

 

[JohnnyReggae]

My experiments with audio memo have been... let's just call it "less than promising" lol. I think if you're accurate enough with audio for edges already it's awesome, conjuring images is just one more thing to slow my memo personally. At least for 3bld. If you have an interest in mbld you'll need to use images to retain it unless you're some sort of audio memory savant.

Multiblind is something to keep in mind and is something I'd like to pursue like doing 4bld as well. I struggle with mental images and recalling them tends to be tricky. It could also just need more practice just like everything

I'm keen to start exploring 4bld just because I think it is a lot harder than 3bld but something that is attainable. I mean Stanley Chapel just got a new WR 4bld of 1:06 and some change ... so how hard can it be

 

[mark49152]

Multiblind is something to keep in mind and is something I'd like to pursue like doing 4bld as well. I struggle with mental images and recalling them tends to be tricky. It could also just need more practice just like everything

I'm keen to start exploring 4bld just because I think it is a lot harder than 3bld but something that is attainable. I mean Stanley Chapel just got a new WR 4bld of 1:06 and some change ... so how hard can it be

4BLD is great fun and IMHO just the right size for an oldie like me . It's not as long-winded and tiring as 5BLD, and is more forgiving than 3BLD where even a small pause or fumble makes the difference between a good and bad solve.

Regarding audio versus images, do whatever works for you, I guess - each to their own. I use "audio" for 4BLD centres, which is typically 15-18 letters, a bit more than 3BLD edges, and my tracing is a bit slower than edges. For those reasons it's rare that I can remember it without a review, and whenever I review audio, I tend to visualise the letters/spelling, and it ceases to be pure audio memo - which is why I put it in quotes.

If you're struggling with image-based memo, you could try something like memoryleague.com. That has a training mode where you're presented with 30 images and have to memorize their order in 30 seconds. It's great for getting the brain used to quickly associating and ordering fairly random things. I use my MBLD rooms for it .

 

[SpartanSailor]

I guess all I can see on the list who are on this thread often are you, me, and Jae. It would be nice to actually spend some real time talking to you guys - I hope we can work something out.

It was nice to see you and Jae Friday evening! Glad dinner worked out so we could chat.

I’m home now and trying to remember our discussion about comms for 4BLD centres.

Is there somewhere to find a reference with the comms listed? I think it’s going to take some time to understand the interchange and stuff, but suspect after that it’ll come easier. I’m going to try to learn centres the way you and Jae described... I’m sold with solving two at a time. Although, I’ll have to learn how to deal with parity later. Right now it’s a lot to process, so I’m going to just learn centres first and from there.

 

[openseas]

It was nice to see you and Jae Friday evening! Glad dinner worked out so we could chat.

I’m home joe and trying to remember our discussion about comms for 4BLD centres.

Is there somewhere to find a reference with the comms listed? I think it’s going to take some time to understand the interchange and stuff, but suspect after that it’ll come easier. I’m going to try to learn centres the way you and Jae described... I’m sold with solving two at a time. Although, I’ll have to learn how to deal with parity later. Right now it’s a lot to process, so I’m going to just learn centres first and from there.

It was a blast for me. I broke #SeniorNAR for 3BLD single with 55.1, a PR for 4BLD with 8:46.55. Could have been a mean and good PR but first two attempts were 1 center cycle off (inverse comm, I guess), especially the first one with 6:46.

US Nats organization was super, ran efficiently and smooth. Running a huge competition itself was more organized than Worlds (both of them were great in almost every aspects but the way Nats organized staffs were more efficient. You can't just buy the experience )

The most memorable thing was being able to meet our fellow senior cubers, @Mike Hughey & @SpartanSailor . Hope we can do more frequently in near future.

Regarding 4BLD, below link is the one we recommended: Daniel's 4BLD center. You don't need a list - go through this video and try to come up with your own for center. Comms for centers are too many if you treat them as algs but much easier if you come up with intuitively then optimize later.

 

[SpartanSailor]

4BLD centres question:
What are “floating buffers”? I guess that’s more of a general blind question, but let me narrow the scope of the response I’m looking to get.
I’ve still got a LOT to do with learning centre comms for 4BLD (having just started yesterday), but already notice a potential problem with my buffer (Ubl). I can do setup moves to get the sides and event he back (and bottom in some cases). But what about when my buffer is solved? Is it okay to just switch to another buffer and use different interchanges/insertions? Or is it better to do U/U’ as part of the setup and stick with the same buffer throughout?

 

[openseas]

4BLD centres question:
What are “floating buffers”? I guess that’s more of a general blind question, but let me narrow the scope of the response I’m looking to get.
I’ve still got a LOT to do with learning centre comms for 4BLD (having just started yesterday), but already notice a potential problem with my buffer (Ubl). I can do setup moves to get the sides and event he back (and bottom in some cases). But what about when my buffer is solved? Is it okay to just switch to another buffer and use different interchanges/insertions? Or is it better to do U/U’ as part of the setup and stick with the same buffer throughout?

Let's say your UBL/UBR/UFR/UFL is A/B/C/D, your original buffer is UBL.
1) If it is solved, treat like a normal 3BLD, meaning start a new cycle break. Whatever target you started a new cycle, you'll get there when the cycle break ends.
2) You can float, meaning make a U', then your original B becomes your UBL. Then, you can start a normal cycle. But when that cycle ends, you have to do "U" to return to your original position.

I started #2 option but realized that I often forgot to unsetup floating (in the above example, U). So, I'm doing #1 now. If you use commutators and run into U face color, you can shoot to B/C/D instead of starting a new cycle. If you shoot all U faces, you rarely need to start a new cycle, mostly centers can be solved with one long cycle.

 

[SpartanSailor]

Let's say your UBL/UBR/UFR/UFL is A/B/C/D, your original buffer is UBL.
1) If it is solved, treat like a normal 3BLD, meaning start a new cycle break. Whatever target you started a new cycle, you'll get there when the cycle break ends.
2) You can float, meaning make a U', then your original B becomes your UBL. Then, you can start a normal cycle. But when that cycle ends, you have to do "U" to return to your original position.

I started #2 option but realized that I often forgot to unsetup floating (in the above example, U). So, I'm doing #1 now. If you use commutators and run into U face color, you can shoot to B/C/D instead of starting a new cycle. If you shoot all U faces, you rarely need to start a new cycle, mostly centers can be solved with one long cycle.

Makes sense. My preference would be #1. But I am not sure how to do a U-face commutator. I’ve figured out how to D-face exchanges, but sometimes it seems like I just solved 1 piece, rather than the two I had in mind—Obviously, I’m making a mistake with the interchange or the insertion.

 

[mark49152]

4BLD centres question:
What are “floating buffers”? I guess that’s more of a general blind question, but let me narrow the scope of the response I’m looking to get.
I’ve still got a LOT to do with learning centre comms for 4BLD (having just started yesterday), but already notice a potential problem with my buffer (Ubl). I can do setup moves to get the sides and event he back (and bottom in some cases). But what about when my buffer is solved? Is it okay to just switch to another buffer and use different interchanges/insertions? Or is it better to do U/U’ as part of the setup and stick with the same buffer throughout?

I try to avoid ever having a cycle break on centres, by solving the U face pieces last on every other side. If I do end up solving all U pieces before all other faces are solved, I'll insert those other pieces into the memo somewhere rather than start a new cycle. If I have a solved buffer at the start, I just solve the diagonally opposite piece first (U2).

 

[SpartanSailor]

Case in point... how would I do this? I’m pointing to my buffer.