• 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 35,000+ people from around the world today!

    If you are already a member, simply login to hide this message and begin participating in the community!

Chris Tran Last Slot (CTLS) Recognition System

gyroninja

Member
Joined
Nov 19, 2015
Messages
139
WCA
2015BOLA02
Over the past few weeks I've been working on a recognition system for CTLS. You can access my third revision of my recognition system here. My previous systems either ran into flaws or were incomplete. In this document I've put some strategies at figuring out the ZBLL that you have. CTLS solves 13,804 1LLSLL cases. I think CTLS could have a lot of potential if you can get the recognition down. Even if full CTLS is not worth it 2GCTLS (CTLS but with just 2GLLs) is very good. Another interesting variant of CTLS is 2GCTOLS. 2GCTOLS (2 Gen Chris Tran Orient Last Slot) uses OCLL+1. (31 2GLL algs) After the OCLL+1 alg you will just have EPLL afterwards. While 2GCTOLS can only be done when the corners are permuted it shows up more often than normal CTLS. CTLS requires a corner edge on the U face to be solved relative to each other. 2GCTOLS on the other hand only requires a corner and edge to be oriented next to each other.

I've poured almost everything I've learned about CTLS so far into this document. If anyone has any questions or incite about CTLS don't be afraid to say something.
 
Last edited:

4Chan

Premium Member
Joined
Jun 21, 2008
Messages
2,984
Location
Lumbridge
YouTube
Visit Channel
I told some people at comps this, but the 2gen version has some sick recognition actually.
By looking at how the "odd/F2L" edge moves, you can figure out the whole edge cycle immediately. Without even tracing the whole edge cycle, just from where it needs to go. Example: UL to UR

Also, all you need for the orientation recognition is the ULB and ULF corners.
Just by looking at ULB, you can narrow down tons of cases, then after looking at ULF, you'll know the case.

It's really cool that you're looking into this!

EDIT: You did a really cool job!
 

gyroninja

Member
Joined
Nov 19, 2015
Messages
139
WCA
2015BOLA02
I told some people at comps this, but the 2gen version has some sick recognition actually.
By looking at how the "odd/F2L" edge moves, you can figure out the whole edge cycle immediately. Without even tracing the whole edge cycle, just from where it needs to go. Example: UL to UR

Also, all you need for the orientation recognition is the ULB and ULF corners.
Just by looking at ULB, you can narrow down tons of cases, then after looking at ULF, you'll know the case.

It's really cool that you're looking into this!

EDIT: You did a really cool job!
The corner orientation thing is actually really interesting. I assume you just mean looking at the normal corners and one of the corners affected by the setup move. After looking at those three corners it will be at most two different cases.

Edit: Looking back at what I said. If you look at three corners you don't have to see the last one. There would only be one case.
 
Last edited:

4Chan

Premium Member
Joined
Jun 21, 2008
Messages
2,984
Location
Lumbridge
YouTube
Visit Channel
I had a document somewhere, but I gave up on the recognition system.
I've got some free time on Friday, I'll make a short video and post in this thread :)

It's super flattering that you took the time to do this, for real! I'm mega honoured to have been mentioned!
 

megaminxwin

Current Clock NR Holder
Joined
Feb 17, 2010
Messages
575
WCA
2010MACA01
That's insane. I'd have to see an example solve or something, I still don't really get it...

*pokes jabari nuruddin* you know you want to
 

gyroninja

Member
Joined
Nov 19, 2015
Messages
139
WCA
2015BOLA02
That's insane. I'd have to see an example solve or something, I still don't really get it...

*pokes jabari nuruddin* you know you want to
Here's an example solve.

Scramble: F2 R' U' F' L' D2 B' R2 U2 D L' D2 R D2 F2 R' U2 R' D2 R2 F2

R' F (U' D) L' F D' R D // EOLine
R U' L2 // LF Block
R U R' // RF Block
L U L' U2 (L U' L' U)2 // Left Block
// I'll break down the CTLS for you
U R' // Create pseudo F2L
L' U2 R U' R' U2 L R U' R' U' // ZBLL - L Diag
R // Undo pseudo F2L
U // AUF

So as far as I understood this method can only be applied in 7/32 of all 1LLSLL cases?

Why not just learn algs for every 1LLSLL case?
This is 7/32 of 1LLSLL with edges pre oriented. CTLS only uses 493 algs (ZBLL). Algs for every 1LLSLL cases would be very hard to do as the case count is in the millions.
 
Last edited:

shadowslice e

Member
Joined
Jun 16, 2015
Messages
2,879
Location
Hampshire, England
YouTube
Visit Channel
Hmmm... Maybe this could be good for the Briggs method (v1) where you may only need to 2-gen CTLS cases. Because that would bump the movecount to sub-40 potentially.

VGJ with the recog system btw! :)

I may actually get around to learning the system if not the algs.
 

gyroninja

Member
Joined
Nov 19, 2015
Messages
139
WCA
2015BOLA02
Due to a resent update in my timer. (Well a modified version of qqtimer but whatever) You are now able to specify a custom scramble input for CTLS scrambles if you ever feel like practicing it. You can practice your CTLS here.
 
Last edited:

4Chan

Premium Member
Joined
Jun 21, 2008
Messages
2,984
Location
Lumbridge
YouTube
Visit Channel

Here are a few examples. Sorry for the low volume, I'm sleepy and I forget my ipad doesn't pick up audio really well.

For other people who want to see CTLS, check out my other video, here is is below:



If you don't want to watch the videos, i'll summarise with words:
Pretty much, you just put the corner where it needs to go, and then you ignore the bad edge and the incorrect R layer, and then you just use normal ZBLL recognition systems like Baum-Harris, edge cycles, blocks, etc.

The bad edge is really obvious and if you recognize by edge cycle, it's really straightforward easypeasy.
Orientation is silly because it's obvious too.
 
Last edited by a moderator:

gyroninja

Member
Joined
Nov 19, 2015
Messages
139
WCA
2015BOLA02
https://www.youtube.com/watch?v=_mEK1-7yrdA

Here are a few examples. Sorry for the low volume, I'm sleepy and I forget my ipad doesn't pick up audio really well.

For other people who want to see CTLS, check out my other video, here is is below:

https://www.youtube.com/watch?v=ZaEIdUJ-mZ0


If you don't want to watch the videos, i'll summarise with words:
Pretty much, you just put the corner where it needs to go, and then you ignore the bad edge and the incorrect R layer, and then you just use normal ZBLL recognition systems like Baum-Harris, edge cycles, blocks, etc.

The bad edge is really obvious and if you recognize by edge cycle, it's really straightforward easypeasy.
Orientation is silly because it's obvious too.
Wow this recognition is a lot better than what I came up with. You've been giving me some cool ideas for what to do with my 4th revision of my recognition.
 

4Chan

Premium Member
Joined
Jun 21, 2008
Messages
2,984
Location
Lumbridge
YouTube
Visit Channel
Your post and stuff actually made me think. Thank you so much for making this post!

EDIT: Nope nope nope, this is too hard to recognize for speed to be worth it, I give up again.
 
Last edited:

4Chan

Premium Member
Joined
Jun 21, 2008
Messages
2,984
Location
Lumbridge
YouTube
Visit Channel
I totally forgot one last example:

When the block is in the other direction, and especially if you do ZZ and you have a 1x1x2 in the last slot (more common than one would expect), you can just herpderp into a ZBLL and solve everything. Alternately, if you have the 1x1x2 block, just do the setup, and proceed.

Example:

Scramble: R U' R U2 R U R2 U2 R U R U' R2 U2 R2 U'

Block: U R U' R U2 R2 (NOTICE THE LAST SLOT, THIS IS THE CASE I MENTIONED)

CTLS (ZBLL+fix): y R2 B2 R U2 R U R' U R' B2 R2 + y' U2 R U R' U

The recognition for this is really simple, there's a 2x2x1 block on the left side that needs to be made, and like, the corner tells you which edges to get there in order to make it, and then you just think of which ZBLL makes that 2x2x1 block, and then you just make it, and then everything solves. The "bad" edge also gives you a hint to the edge cycle, making it pretty trivial to figure out.
I'll make a video soon, if that above example wasn't enough
 

gyroninja

Member
Joined
Nov 19, 2015
Messages
139
WCA
2015BOLA02
I totally forgot one last example:

When the block is in the other direction, and especially if you do ZZ and you have a 1x1x2 in the last slot (more common than one would expect), you can just herpderp into a ZBLL and solve everything. Alternately, if you have the 1x1x2 block, just do the setup, and proceed.

Example:

Scramble: R U' R U2 R U R2 U2 R U R U' R2 U2 R2 U'

Block: U R U' R U2 R2 (NOTICE THE LAST SLOT, THIS IS THE CASE I MENTIONED)

CTLS (ZBLL+fix): y R2 B2 R U2 R U R' U R' B2 R2 + y' U2 R U R' U

The recognition for this is really simple, there's a 2x2x1 block on the left side that needs to be made, and like, the corner tells you which edges to get there in order to make it, and then you just think of which ZBLL makes that 2x2x1 block, and then you just make it, and then everything solves. The "bad" edge also gives you a hint to the edge cycle, making it pretty trivial to figure out.
I'll make a video soon, if that above example wasn't enough
Interesting concept but if you have a 2x1x1 in the last slot like in the example why don't you just do R U R' F' to setup to do a normal ctls.
 

4Chan

Premium Member
Joined
Jun 21, 2008
Messages
2,984
Location
Lumbridge
YouTube
Visit Channel
F moves complicate things because they break orientation.

Generally speaking, you can't use a ZBLL once orientation is broken and it influences an edge that goes into the face that has broken orientation.

Unless the edge cycle doesn't move into the F/B face, then you can use a normal ZBLL, otherwise you'd have to use a 1LLL alg.
I actually thought of that too back in 2009 when I came up with this LOL

EDIT:

Here's a video example:
 
Last edited by a moderator:

gyroninja

Member
Joined
Nov 19, 2015
Messages
139
WCA
2015BOLA02
F moves complicate things because they break orientation.

Generally speaking, you can't use a ZBLL once orientation is broken and it influences an edge that goes into the face that has broken orientation.

Unless the edge cycle doesn't move into the F/B face, then you can use a normal ZBLL, otherwise you'd have to use a 1LLL alg.
I actually thought of that too back in 2009 when I came up with this LOL

EDIT:

Here's a video example:
https://www.youtube.com/watch?v=OBy_jCOENgE
If you insert the f2l edge you are now able to use any side's face turn to setup. When the left block is still in U there is one location for the f2l edge that takes a long time to setup. The other locations takes like 4 or 5 moves for the initial setup. Which isn't too bad.
 
D

Daniel Lin

Guest
wow this is really cool, I'm considering learning it.
I really don't get how it works though
so you just ignore the R layer and the bad edge, right?
But I can't recognize the case using Baum Harris/color blocks because i can't recognize ZBLLs from all angles. I have to AUF the CO case a certain way. For example, for Chameleon and Headlights I always make it face the back. After I AUF, I always look at the same pieces:FU, FRU, RU, RUF. But then some of those pieces can be R layer pieces or the bad edge. So how do I recognize the case?
 

gyroninja

Member
Joined
Nov 19, 2015
Messages
139
WCA
2015BOLA02
wow this is really cool, I'm considering learning it.
I really don't get how it works though
so you just ignore the R layer and the bad edge, right?
But I can't recognize the case using Baum Harris/color blocks because i can't recognize ZBLLs from all angles. I have to AUF the CO case a certain way. For example, for Chameleon and Headlights I always make it face the back. After I AUF, I always look at the same pieces:FU, FRU, RU, RUF. But then some of those pieces can be R layer pieces or the bad edge. So how do I recognize the case?
In your mind twist the corners that will be on the R layer in the end to be up to create a proper ocll. Once you have that ocll made in your mind you auf to where it says in the doc and recognize the cp case you have. Then you look at the pieces that are already solved and match those pieces to a diagram in the doc.
 
Top