• 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!

shadowslice e

Member
Joined
Jun 16, 2015
Messages
2,923
Location
192.168. 0.1
YouTube
Visit Channel
alrighty another roux-block type start method:
1) Roux block + centers and DB
2) 2x2x1 in BR
3) TSLE
4) 1 of 6 TTLLs (TTCLL)
5) EODF and cancel into EPLL

I know TTLL isn't the greatest but hey, only the corner cases and you don't have to worry about LL edges at all.
Isn't this 2OP from M-CELL?
I'm trying to come up with a simple beginners method that transitions into Roux rather than CFOP, right now this is what I have:

1: Solve 2 corners (DL)- Very easy for beginners to understand, completely intuitive
2: Solve FB- Use slice moves for intuitive insertion of edges
3: Solve D layer corners- Easy, intuitive
4: Solve SB- Use slice moves for easy inserts, *This and LSE are the hardest to teach*
5: CO- Teach Sune
6: CP- Teach an easy CP alg (like A/J/T perm)
7: LSE- Introduce them to EO and how to preserve it, no algs but familiarize with Arrow case *Maybe the hardest part of the method*

Here's a solve, obviously not very efficient but it should give a good idea as to what I'm going for

SCRAMBLE: U D B' R B2 U' B2 R' L' F B2 D2 R' U2 D2 L D2 R' U2 L' B2

D2 x' z' U2 F2//FB Corners (3/3)
u' R' u r' D M2 D'//FB Edges (7/10)
R' U R U' R' U R//SB Corners (7/17)
R2 U' M U r U' M U R2 U M2 U2 M2 U r'//SB edges (15/32)
U' R U R' U R U2 R'//CO (8/40)
x R' U R' D2 R U' R' D2 R2 x'//CP (9/49)
M U M U' M' U M' U2 M' U2 M' U' M U2 M U2//LSE (16/65)

A possible improvement might be to solve a 1x2x3 line and then solve the other 2 edges rather than just corners.

I feel like it might be better to do edges then corners for the FB like you would in normal beginners because I'm not sure etching weird slice moves to a beginner would go down that well.

Similar thing goes for the SB as well; solve DR and then insert corners, then solve the edges using the other stuff.

For LSE, it's not actually too bad to teach; I learned it purely frok being shown MUM' and worked everything else out from there.
 
Last edited:

Arc

Member
Joined
Jan 27, 2017
Messages
63
Location
Washington
Isn't this 2OP from M-CELL?


I feel like it might be better to do edges then corners for the FB like you would in normal beginners because I'm not sure etching weird slice moves to a beginner would go down that well.

Similar thing goes for the SB as well; solve DR and then insert corners, then solve the edges using the other stuff.

For LSE, it's not actually too bad to teach; I learned it purely frok being shown MUM' and worked everything else out from there.
I was going to reply with this method myself. +1

For EO I would try to find a pattern to cycle into arrow with just M' U M'.
 

Thermex

Member
Joined
Mar 21, 2017
Messages
188
Location
The Milky Way
So lately I've been thinking about some tyrannical caterpillar variants, most of course based off of solving the last 5 corners in some fashion while inserting the FR edge. My main gripe with classic Tyrannical Caterpillar is that it adds a lot more algs with not much efficiency gain from regular Roux, in other words TC isn't really that great of an L5C method. Here are four alternative methods I much prefer to classic TC with either way less algs (<50) or more efficient movecounts (<11 moves for the corners+the FR edge):

a. "Valiant Opossum" (VOP style orientation [~5.5 moves] + one of the 6 NLLs used in HD/VOP [~8.5 moves+the FR edge injection], total w/ AUFs is around 16 moves)
Theoretically this is only 2-3 moves more than TCMLL with about a fifth of the algs, and honestly in my opinion a lot more worth it for a beginner compared to TC. "Valiant Opossum" not only provides some very lucky singles; it also has almost the same amount of moves as standard LP+CMLL with less than half of the algs (I say "less than half" because most of the orientation cases shouldn't even qualify as algorithms, they're ultra fast 3 move inserts a beginner could learn intuitively).

b. Caterpillar. Just a plain old caterpillar. (Insert DRF corner [~4 moves] & do a CMLL plus the FR edge injection [~10 moves], mix in an AUF and you get a total of ~15 moves)
@Teoidus propsosed a method like this a while back he called "TT", but this goes a step further in utilizing FR injections to give you an LSE finish. I feel like if you're going to propose a method centered around not using a lot of algs, you might as well keep it that way for the last x edges step. Anyway, "caterpillar" is really just TCMLL without the "T" part, cutting off 80 algs while only adding about a move. The reason I prefer this over regular TC is because the moves that are lost by inserting the DFR with correct orientation are pretty much gained back by how ergonomic and short CMLLs are without having to perserve the FR edge. This also beats out traditional Roux CMLL by a move or two while maintaining slightly better lookahead since you don't have to find a full pair after BRSquare.

c. "Voracious Leech" (just ignore the name for a second, basically you do the exact same corners step I described a long time ago [1 oriented corner-in-UBL-L5C] but with a Tyrannical caterpillar style injection to skip out on L7E and instead have LSE)
Now this idea is by far the most algorithms thus far (~300!) but has some crazy efficiency. I've generated several of the VL4C cases and they're around 8 moves on average. With the FR edge injection and AUF added, this last 5 corners+FR edge technique would be around 10 moves on average, compared to the 16-17 used in normal Roux. And if this was used in conjunction with EOLR and good blockbuilding, you could probably average around 38-39 moves, just breaking that 40 move barrier, and probably being one of the most efficient/fast methods ever proposed (later I'll put up a post that includes a sub-40 solution with @mDiPalma's scramble). Obvisouly this comes at the cost of learning 300 algs, but that's definitely still a learnable amount and nearly 200 less than full ZBLL. Obviously this wouldn't be a method for everyone to learn, and I'm sure I'll get a response saying "well 42 can achieve similar movecounts with only 42 algs", but I think for someone willing to go the extra mile and learn 300 algs, this is pretty decent idea.

d. "Terrific Eagle" (Bope style 1×2×2 squares [~9 moves], Twisty EG+FR injection [~14 moves] and L7E [~16 moves, but I'm sure we could do better] giving you a grand total of just about 39-40 moves)
This idea is the weirdest of them all and also the most alg heavy; I proposed this a while back on the BOPE thread but didn't really get any responses. I don't really have too much to say about this since I really haven't experimented with it much and it would take more developments than some of the above methods to get it to sub-40, but it seems about as efficient as the last idea with around 50 more algs. The main advantage it has over some of the other TC variants here is that you can one-look six pieces as opposed to five during inspection, giving you a near 2-look solve.

I know I probably sounded overconfident and the names for these are ridiculous, but with all my ideas out of the way now, my main question remains: are Tyrannical x methods actually any better than just solving the corners normally and ending in L7E? I've personally been believing they're worth it since you're solving one extra piece by only adding a setup move+an injection while cutting down the number of algs you need to learn for the last step, but I could easily be wrong, especially if a 15 move L7E approach is on the horizon. Tell me if you think it's worth generating algs for any of these methods, or if I should just stick with developing L7E methods.

While I'm at it, I might as well throw an random bonus OH method that is basically just a fancy 2GR finish and (maybe) pretty good:

1.) EOPair (6/6)
2.) CPLine (4/10)
3.) Block (9/19)
4.) Connect the DBR corner and DR edge and put them in the d-layer (4/23)
5.) One-corner oriented L5C (~50 7.5 move 2 gen algs, 8/31)
6.) L6EP (@Neuro created an algset for this a while back, optimal is around 10 moves and 2 gen algs are 12 moves on average, so I'll say 12/43)

This is something like 50 more algs than classic 2GR while shaving off around 5 moves. Not drastically better but maybe worth it for someone (I for one might actually might use this for OH) and there's probably a more efficient way to do the first 4 steps, but I'm just too lazy to find that right now. Also @Teoidus I couldn't find the exact statistics for some of the 2GR steps so they're probably a bit off, you can correct me if I'm wrong on those.

This post took nearly a day of thinking and an two or three hours of typing (on my crappy phone) so hopefully you guys could give me some decently thought-out responses. And if you read this entire post from start to finish go ahead and give yourself a pat on the back :p
 
Last edited:

shadowslice e

Member
Joined
Jun 16, 2015
Messages
2,923
Location
192.168. 0.1
YouTube
Visit Channel
So lately I've been thinking about some tyrannical caterpillar variants, most of course based off of solving the last 5 corners in some fashion while inserting the FR edge. My main gripe with classic Tyrannical Caterpillar is that it adds a lot more algs with not much efficiency gain from regular Roux, in other words TC isn't really that great of an L5C method. Here are four alternative methods I much prefer to classic TC with either way less algs (<50) or more efficient movecounts (<11 moves for the corners+the FR edge):

a. "Valiant Opossum" (VOP style orientation [~5.5 moves] + one of the 6 NLLs used in HD/VOP [~8.5 moves+the FR edge injection], total w/ AUFs is around 16 moves)
Theoretically this is only 2-3 moves more than TCMLL with about a fifth of the algs, and honestly in my opinion a lot more worth it for a beginner compared to TC. "Valiant Opossum" not only provides some very lucky singles; it also has almost the same amount of moves as standard LP+CMLL with less than half of the algs (I say "less than half" because most of the orientation cases shouldn't even qualify as algorithms, they're ultra fast 3 move inserts a beginner could learn intuitively).

b. Caterpillar. Just a plain old caterpillar. (Insert DRF corner [~4 moves] & do a CMLL plus the FR edge injection [~10 moves], mix in an AUF and you get a total of ~15 moves)
@Teoidus propsosed a method like this a while back he called "TT", but this goes a step further in utilizing FR injections to give you an LSE finish. I feel like if you're going to propose a method centered around not using a lot of algs, you might as well keep it that way for the last x edges step. Anyway, "caterpillar" is really just TCMLL without the "T" part, cutting off 80 algs while only adding about a move. The reason I prefer this over regular TC is because the moves that are lost by inserting the DFR with correct orientation are pretty much gained back by how ergonomic and short CMLLs are without having to perserve the FR edge. This also beats out traditional Roux CMLL by a move or two while maintaining slightly better lookahead since you don't have to find a full pair after BRSquare.

c. "Voracious Leech" (just ignore the name for a second, basically you do the exact same corners step I described a long time ago [1 oriented corner-in-UBL-L5C] but with a Tyrannical caterpillar style injection to skip out on L7E and instead have LSE)
Now this idea is by far the most algorithms thus far (~300!) but has some crazy efficiency. I've generated several of the VL4C cases and they're around 8 moves on average. With the FR edge injection and AUF added, this last 5 corners+FR edge technique would be around 10 moves on average, compared to the 16-17 used in normal Roux. And if this was used in conjunction with EOLR and good blockbuilding, you could probably average around 38-39 moves, just breaking that 40 move barrier, and probably being one of the most efficient/fast methods ever proposed (later I'll put up a post that includes a sub-40 solution with @mDiPalma's scramble). Obvisouly this comes at the cost of learning 300 algs, but that's definitely still a learnable amount and nearly 200 less than full ZBLL. Obviously this wouldn't be a method for everyone to learn, and I'm sure I'll get a response saying "well 42 can achieve similar movecounts with only 42 algs", but I think for someone willing to go the extra mile and learn 300 algs, this is pretty decent idea.

d. "Terrific Eagle" (Bope style 1×2×2 squares [~9 moves], Twisty EG+FR injection [~14 moves] and L7E [~16 moves, but I'm sure we could do better] giving you a grand total of just about 39-40 moves)
This idea is the weirdest of them all and also the most alg heavy; I proposed this a while back on the BOPE thread but didn't really get any responses. I don't really have too much to say about this since I really haven't experimented with it much and it would take more developments than some of the above methods to get it to sub-40, but it seems about as efficient as the last idea with around 50 more algs. The main advantage it has over some of the other TC variants here is that you can one-look six pieces as opposed to five during inspection, giving you a near 2-look solve.

I know I probably sounded overconfident and the names for these are ridiculous, but with all my ideas out of the way now, my main question remains: are Tyrannical x methods actually any better than just solving the corners normally and ending in L7E? I've personally been believing they're worth it since you're solving one extra piece by only adding a setup move+an injection while cutting down the number of algs you need to learn for the last step, but I could easily be wrong, especially if a 15 move L7E approach is on the horizon. Tell me if you think it's worth generating algs for any of these methods, or if I should just stick with developing L7E methods.

While I'm at it, I might as well throw an random bonus OH method that is basically just a fancy 2GR finish and (maybe) pretty good:

1.) EOPair (6/6)
2.) CPLine (4/10)
3.) Block (9/19)
4.) Connect the DBR corner and DR edge and put them in the d-layer (4/23)
5.) One-corner oriented L5C (~50 7.5 move 2 gen algs, 8/31)
6.) L6EP (@Neuro created an algset for this a while back, optimal is around 10 moves and 2 gen algs are 12 moves on average, so I'll say 12/43)

This is something like 50 more algs than classic 2GR while shaving off around 5 moves. Not drastically better but maybe worth it for someone (I for one might actually might use this for OH) and there's probably a more efficient way to do the first 4 steps, but I'm just too lazy to find that right now. Also @Teoidus I couldn't find the exact statistics for some of the 2GR steps so they're probably a bit off, you can correct me if I'm wrong on those.

This post took nearly a day of thinking and an two or three hours of typing (on my crappy phone) so hopefully you guys could give me some decently thought-out responses. And if you read this entire post from start to finish go ahead and give yourself a pat on the back :p
42- could well achieve 35ish movecounts with 128 or so algs. The movecounts would be approximately
Fb+SBsquare (non-linear, NMLL): 13
BTR+: 10 (if we want injections as described in the wiki)
LSE: 13
Total: 36

There are also things metioned on the wiki which I haven't included.

I think in general these ideas are interesting though I think atm the best thing to focus on actually getting L7E really really good because if it can be made very good then it should be possible to use almost and L5C variant and still get much better/ more efficient solves.
 
Last edited:

Sue Doenim

Member
Joined
Nov 9, 2016
Messages
448
So lately I've been thinking about some tyrannical caterpillar variants, most of course based off of solving the last 5 corners in some fashion while inserting the FR edge. My main gripe with classic Tyrannical Caterpillar is that it adds a lot more algs with not much efficiency gain from regular Roux, in other words TC isn't really that great of an L5C method. Here are four alternative methods I much prefer to classic TC with either way less algs (<50) or more efficient movecounts (<11 moves for the corners+the FR edge):

a. "Valiant Opossum" (VOP style orientation [~5.5 moves] + one of the 6 NLLs used in HD/VOP [~8.5 moves+the FR edge injection], total w/ AUFs is around 16 moves)
Theoretically this is only 2-3 moves more than TCMLL with about a fifth of the algs, and honestly in my opinion a lot more worth it for a beginner compared to TC. "Valiant Opossum" not only provides some very lucky singles; it also has almost the same amount of moves as standard LP+CMLL with less than half of the algs (I say "less than half" because most of the orientation cases shouldn't even qualify as algorithms, they're ultra fast 3 move inserts a beginner could learn intuitively).

b. Caterpillar. Just a plain old caterpillar. (Insert DRF corner [~4 moves] & do a CMLL plus the FR edge injection [~10 moves], mix in an AUF and you get a total of ~15 moves)
@Teoidus propsosed a method like this a while back he called "TT", but this goes a step further in utilizing FR injections to give you an LSE finish. I feel like if you're going to propose a method centered around not using a lot of algs, you might as well keep it that way for the last x edges step. Anyway, "caterpillar" is really just TCMLL without the "T" part, cutting off 80 algs while only adding about a move. The reason I prefer this over regular TC is because the moves that are lost by inserting the DFR with correct orientation are pretty much gained back by how ergonomic and short CMLLs are without having to perserve the FR edge. This also beats out traditional Roux CMLL by a move or two while maintaining slightly better lookahead since you don't have to find a full pair after BRSquare.

c. "Voracious Leech" (just ignore the name for a second, basically you do the exact same corners step I described a long time ago [1 oriented corner-in-UBL-L5C] but with a Tyrannical caterpillar style injection to skip out on L7E and instead have LSE)
Now this idea is by far the most algorithms thus far (~300!) but has some crazy efficiency. I've generated several of the VL4C cases and they're around 8 moves on average. With the FR edge injection and AUF added, this last 5 corners+FR edge technique would be around 10 moves on average, compared to the 16-17 used in normal Roux. And if this was used in conjunction with EOLR and good blockbuilding, you could probably average around 38-39 moves, just breaking that 40 move barrier, and probably being one of the most efficient/fast methods ever proposed (later I'll put up a post that includes a sub-40 solution with @mDiPalma's scramble). Obvisouly this comes at the cost of learning 300 algs, but that's definitely still a learnable amount and nearly 200 less than full ZBLL. Obviously this wouldn't be a method for everyone to learn, and I'm sure I'll get a response saying "well 42 can achieve similar movecounts with only 42 algs", but I think for someone willing to go the extra mile and learn 300 algs, this is pretty decent idea.

d. "Terrific Eagle" (Bope style 1×2×2 squares [~9 moves], Twisty EG+FR injection [~14 moves] and L7E [~16 moves, but I'm sure we could do better] giving you a grand total of just about 39-40 moves)
This idea is the weirdest of them all and also the most alg heavy; I proposed this a while back on the BOPE thread but didn't really get any responses. I don't really have too much to say about this since I really haven't experimented with it much and it would take more developments than some of the above methods to get it to sub-40, but it seems about as efficient as the last idea with around 50 more algs. The main advantage it has over some of the other TC variants here is that you can one-look six pieces as opposed to five during inspection, giving you a near 2-look solve.

I know I probably sounded overconfident and the names for these are ridiculous, but with all my ideas out of the way now, my main question remains: are Tyrannical x methods actually any better than just solving the corners normally and ending in L7E? I've personally been believing they're worth it since you're solving one extra piece by only adding a setup move+an injection while cutting down the number of algs you need to learn for the last step, but I could easily be wrong, especially if a 15 move L7E approach is on the horizon. Tell me if you think it's worth generating algs for any of these methods, or if I should just stick with developing L7E methods.

While I'm at it, I might as well throw an random bonus OH method that is basically just a fancy 2GR finish and (maybe) pretty good:

1.) EOPair (6/6)
2.) CPLine (4/10)
3.) Block (9/19)
4.) Connect the DBR corner and DR edge and put them in the d-layer (4/23)
5.) One-corner oriented L5C (~50 7.5 move 2 gen algs, 8/31)
6.) L6EP (@Neuro created an algset for this a while back, optimal is around 10 moves and 2 gen algs are 12 moves on average, so I'll say 12/43)

This is something like 50 more algs than classic 2GR while shaving off around 5 moves. Not drastically better but maybe worth it for someone (I for one might actually might use this for OH) and there's probably a more efficient way to do the first 4 steps, but I'm just too lazy to find that right now. Also @Teoidus I couldn't find the exact statistics for some of the 2GR steps so they're probably a bit off, you can correct me if I'm wrong on those.

This post took nearly a day of thinking and an two or three hours of typing (on my crappy phone) so hopefully you guys could give me some decently thought-out responses. And if you read this entire post from start to finish go ahead and give yourself a pat on the back :p
I think that the deal is, TC methods are really not as efficient as you think they are. It looks like you're only counting the injections a 1 move, while really they're 2 or 3, because the edge could be in any of 14 places, only 3 of which take 1 move to solve. Maybe I'm missing something. Voracious Leech isn't just like 42; as I understand it, it's the same thing. 1 oriented U-layer corner+L4C? That's what 42 is, really. Terrific Eagle sounds fine for BOPE. The others are beginner's stuffs, so I don't really see the point in judging them for speed. The 2GR variant sounds interesting, maybe try it out. What I think would be really cool is using TC injections with WaterRoux. I'll try it out.
 

Thermex

Member
Joined
Mar 21, 2017
Messages
188
Location
The Milky Way
I think that the deal is, TC methods are really not as efficient as you think they are. It looks like you're only counting the injections a 1 move, while really they're 2 or 3, because the edge could be in any of 14 places, only 3 of which take 1 move to solve. Maybe I'm missing something. Voracious Leech isn't just like 42; as I understand it, it's the same thing. 1 oriented U-layer corner+L4C? That's what 42 is, really. Terrific Eagle sounds fine for BOPE. The others are beginner's stuffs, so I don't really see the point in judging them for speed. The 2GR variant sounds interesting, maybe try it out. What I think would be really cool is using TC injections with WaterRoux. I'll try it out.
I tried to include injections as two moves, but I think there might have one or two places where I only added one move. Technically 8 out of the 14 cases are one movers since the orientation of the FR edge doesn't matter (M2s are used instead of Ms when the edge is oriented) so I think 2 moves is the most reasonable estimate for the FR injections. "Voracious Leech" (I will probably change the name at some point, it just sounds awful) is not the same thing as 42, I specified that in a post a couple of pages back. With 42 you have to use a conjugated CMLL every time which gives you slightly less efficient solves than the L5C variant I made in which you solve one of 290 cases optimally. This provides some of the shortest and most ergonomic algs ever; three move inserts, sledgehammers, and altered sunes. Obviously @shadowslice e is right that 42 is a better method for a wider audience since it only has 42 algs, but in terms of pure efficiency without much regard to alg count Voracious leech is slightly better. Tyrannical eagle like I said is a WIP, there's a lot of things that would need to be developed before it could become a real method and all, but so far it looks like the most efficient variant of Bope, and one day I'll probably release it. As for the other two methods, they're not as much beginner methods as they are intermediate methods (in terms of concepts) for people who really hate learning algs. I'll probably generate and develop algs for those at some point this week.
42- could well achieve 35ish movecounts with 128 or so algs. The movecounts would be approximately
Fb+SBsquare (non-linear, NMLL): 13
BTR+: 10 (if we want injections as described in the wiki)
LSE: 13
Total: 36

There are also things metioned on the wiki which I haven't included.

I think in general these ideas are interesting though I think atm the best thing to focus on actually getting L7E really really good because if it can be made very good then it should be possible to use almost and L5C variant and still get much better/ more efficient solves.
Optimistic movecounts, but I could see that passing for a really good solver. I don't really understand what you mean by "NMLL" for FB+BRSquare, is okay if you explain that a bit more? Otherwise this method looks amazing even though I would still predict movecounts closer to 39, and I'll actually just gen algs for them today and post 'em later, I don't see why not.
 

shadowslice e

Member
Joined
Jun 16, 2015
Messages
2,923
Location
192.168. 0.1
YouTube
Visit Channel
Optimistic movecounts, but I could see that passing for a really good solver. I don't really understand what you mean by "NMLL" for FB+BRSquare, is okay if you explain that a bit more? Otherwise this method looks amazing, I'll actually just gen algs for this today and post 'em later, I don't see why not.
Basically it means you can build any of the 4 possible SBsquares.
 

Teoidus

Member
Joined
Feb 11, 2016
Messages
573
Location
Char
my main question remains: are Tyrannical x methods actually any better than just solving the corners normally and ending in L7E? I've personally been believing they're worth it since you're solving one extra piece by only adding a setup move+an injection while cutting down the number of algs you need to learn for the last step, but I could easily be wrong, especially if a 15 move L7E approach is on the horizon. Tell me if you think it's worth generating algs for any of these methods, or if I should just stick with developing L7E methods.

This was actually one of the main points behind the "TT proposal"--I'm not convinced that the injection is worth it and it complicates things. While I don't think L7E can be further optimized, I'll take an extra 1-2 moves over having to lookahead in the middle of an algorithm any day.

I did read through all of your LS/CMLL variants, but to be honest I just got confused and can't offer a good opinion on them other than that they look alright but might add needless complexity. I like (despite how 2GR seems) simple methods and any non-inspected steps that can't be described by a single "permute ___" or "orient __" make me question whether they're really necessary.

While I'm at it, I might as well throw an random bonus OH method that is basically just a fancy 2GR finish and (maybe) pretty good:

1.) EOPair (6/6)
2.) CPLine (4/10)
3.) Block (9/19)
4.) Connect the DBR corner and DR edge and put them in the d-layer (4/23)
5.) One-corner oriented L5C (~50 7.5 move 2 gen algs, 8/31)
6.) L6EP (@Neuro created an algset for this a while back, optimal is around 10 moves and 2 gen algs are 12 moves on average, so I'll say 12/43)

This is pretty cool! I'd say it saves ~2-3 moves vs F2L/2GLL (which averages around 13 + 14 = 27), and it's much easier to use tricks with this method (you can pretty much take advantage of any free pair on L, R, U, or D since R2/U2 pseudo recognition is a lot simpler vs 2GLL). How many algs are CLS and L6EP?
 

Arc

Member
Joined
Jan 27, 2017
Messages
63
Location
Washington
Okay I think I have some ideas to improve the Roux method.

1) I call this mispermuted centers. Basically the idea is that if you see a really easy FB, but the center would be wrong, just build it anyway. Before CMLL you can easily fix the centers with a couple wide u moves and an M. This adds a few moves after SB but can drop the FB move count significantly.

2) I call this liberated blocks. The idea here is that if you see a really easy SB square (or even the entire block) after you start FB, you can build that and then finish FB after. This allows some flexibility in F2B to take better advantage of lucky cases to get lower move counts.

3) This concept I call bi-color blocks. What it means is that you don't have to use the same color on bottom for both blocks. The CMLL recognition becomes slightly harder but being able to choose whichever SB is best should lower the move count a lot, and the cost is only a single R or r move after LSE.

4) S-CMLL, or Super-CMLL. This is an alternate way of doing CMLL. It requires learning only 84 algs (which is quite reasonable). The goal here is to influence EO while solving the corners to give an easier LSE case. The idea is that 42 of the algs will flip two edges while solving the corners, and 42 of them won't flip any edges while solving the corners. By adjusting the M slice to flip certain edges before performing an edge-flipping alg, or by choosing to use an EO preserving alg, you can always avoid a 6 flip and generally get better EO cases. Since the algs hardly vary in move count at all from standard CMLL (within 1 or 2 at the worst), you should get a move count reduction off of the easier LSE cases.

5) Conjugated L4EP is the idea of solving UF and UB instead of UL and UR during step 4b. Then you can offset them with a U or U' and then solve L4E with different colors (the recognition is a little harder). This should allow you to take advantage of easier 4b cases and save several moves.

6) I call this step 95, because it does step 4a and step 4b at the same time, and in hexadecimal, 4a + 4b = 95. Basically the idea is to solve EO in a specific way that puts UL and UR (or UF and UB if you combine step 95 with conjugated L4EP) into a very easily solved spot. This should shave of almost all the moves of step 4b without adding any to 4a's move count.

I think that with all these advancements, you could get sub-40 move count solves with Roux, or even sub-35, or maybe even less.
 

Thermex

Member
Joined
Mar 21, 2017
Messages
188
Location
The Milky Way
This was actually one of the main points behind the "TT proposal"--I'm not convinced that the injection is worth it and it complicates things. While I don't think L7E can be further optimized, I'll take an extra 1-2 moves over having to lookahead in the middle of an algorithm any day.

I did read through all of your LS/CMLL variants, but to be honest I just got confused and can't offer a good opinion on them other than that they look alright but might add needless complexity. I like (despite how 2GR seems) simple methods and any non-inspected steps that can't be described by a single "permute ___" or "orient __" make me question whether they're really necessary.



This is pretty cool! I'd say it saves ~2-3 moves vs F2L/2GLL (which averages around 13 + 14 = 27), and it's much easier to use tricks with this method (you can pretty much take advantage of any free pair on L, R, U, or D since R2/U2 pseudo recognition is a lot simpler vs 2GLL). How many algs are CLS and L6EP?
CLS is 50 algs and L6EP is like 65 or something.
 

Teoidus

Member
Joined
Feb 11, 2016
Messages
573
Location
Char
Okay, so it's around 30 extra algs for 3 less moves/easier pseudos. That's pretty good if your movecounts for the steps are accurate
 

Neuro

Member
Joined
Dec 23, 2016
Messages
597
Here's my L6EP document! 86 algs, 82 disregarding EPLL with an avg of ~12 moves. Fully <RU>. Some of the cases are very difficult to recognize, especially the last set (it can be easily avoided though)

Full CLS is 104 algs, so either 190 or 186 algs. Since it only saves a few moves (possibly) you may want to stick with F2L+2GLL :(
 

Hazel

Premium Member
Joined
Apr 1, 2017
Messages
1,681
Location
in your walls :3
So, here's another method idea:
1) Standard Roux FB on left
1.5) Solve the BR/BDR F2L pair.
2) Use keyhole to solve the rest of the first layer, minus DF and DFR pieces
2.5) Solve DFR while making sure FR edge is oriented - maybe 4 moves on average
3) EODF
4) CPLE - Corner permutation + last F2L edge, I'm pretty sure this has 12 algs (6 if you just mirror the others)
5) 2GLL

You could also do CPFB as the first step and adjust the other steps a bit to avoid having to recognize CP mid-solve.
 

Cale S

Member
Joined
Jan 18, 2014
Messages
2,421
Location
Iowa, USA
WCA
2014SCHO02
YouTube
Visit Channel
ZZ LSLL method

1. insert edge + phase edges

4 cases including R U R', R U' R', and R U2 R', barely even a step

2. orient corners while preserving phased edges

23 cases, 6 are normal OLLs and the rest are pretty easy

3. TTLL with phased edges

one of 24 TTLL cases


Average movecounts:
step 1: 3 or 4
step 2: 9
step 3: 13 or 14

26 average movecount with less than 50 total algs, and recog is pretty easy

also you have a 1 in 11.4 chance of skipping a step, making it 2-look
 
Last edited:

Arc

Member
Joined
Jan 27, 2017
Messages
63
Location
Washington
ZZ LSLL method

1. insert edge + phase edges

4 cases including R U R', R U' R', and R U2 R', barely even a step

2. orient corners while preserving phased edges

23 cases, 6 are normal OLLs and the rest are pretty easy

3. TTLL with phased edges

one of 24 TTLL cases


Average movecounts:
step 1: 3 or 4
step 2: 9
step 3: 13 or 14

26 average movecount with less than 50 total algs, and recog is pretty easy
This is pretty neat, but don't forget that you have 3 AUFs (before orientation, before TTLL, and after TTLL) putting you at 2.25 moves higher than just the algs alone, so up to around 28-29.
 

Hazel

Premium Member
Joined
Apr 1, 2017
Messages
1,681
Location
in your walls :3
Is this already a thing?:

Last Layer method:
1) solve EO and CP using one of 24 algs
2) 2GLL

Pretty much identical recognition to OLLCP.
 

Teoidus

Member
Joined
Feb 11, 2016
Messages
573
Location
Char
Yeah. I remember seeing a post about it a little while back. It wasn't in this thread though
 

Teoidus

Member
Joined
Feb 11, 2016
Messages
573
Location
Char
Nah. I don't think it's any more efficient than OLL/PLL.

Though, it might be an interesting 1look 2alg system for LL? I'm not sure
 
Top