JTLE - does it have potential?

My friend told me some interesting stuff few minutes ago, he told me JTLE (John Tamanas Last Edge). I found it very interesting with all 2-gen algorithms, it can be very fast I think.

Here's the link:
http://algobase.110mb.com/JTLE.html

anyone using it? Well, I'm not very fast cuber. I wonder if someone get some sub-15 with it.

Sorry for my mispellings or grammar mistakes, I'm not native english
*ouch that should not become my excuse* :fp

Honestly, I was one of the few people encouraging him to get the algs and make this method when he first thought of it, but frankly, no.
It does has little potential.

This method is not meant to be used exclusively. Let me show you some example solves.

A solve where you should use it:

scramble: R2 U' R U2 R2 U2 R' U R U' R' U' R' U R2 U' R' U' R' U2
JTLE
Start: skip
Finish: U2 R' U' R
JTLE: R' U2 R2 U2 R U' R U' R'

0 + 4 + 9 = 13 moves

Normal
First: U2 R U R2 U' R'
Finish: U' R' U2 R U R' U' R
OLL: R U2 R' U' R U' R'

6 +12 + 7 = 21 moves[/QUOTE]

A solve where it doesn't make much of a difference:

scramble:R' U' R2 U' R U R2 U' R U2 R U' R U2 R U2 R' U' R' U
JTLE
Start: R2 U R U2 R2
Finish: U' R' U' R
JTLE: (U) R' U2 R2 U R U R2 U2 R'

5 + 4 + 10 = 19 moves

Normal
Start: R2 U2 R2
Finish: U' R U2 R' U R U' R'
OLL: (U2) R U R' U R U2 R'

3 + 8 + 8 = 19 moves


A solve where JTLE should not be used:

scramble:R2 U' R' U' R2 U R' U R' U' R' U R' U R' U' R' U' R U
JTLE
Start: U2 R
Finish: R U R' U R U R' U2 R U R'
JTLE: (U') R U2 R2 U2 R2 U2 R

2 + 11 + 8 = 21 moves

Normal
Start: R U R U R
Finish: U2 R U' R'
OLL: (U') R U R' U R U2 R'

5 + 4 + 8 = 17 moves

If you use either Petrus or ZZ you should learn these algorithms. It cannot harm your solves, and it may save a lot of moves (The first solve saved 8 moves!) Also, I'll put your name on the page if you learn it.

jtjogobonito said:

This method is not meant to be used exclusively. Let me show you some example solves.

A solve where you should use it:

scramble: R2 U' R U2 R2 U2 R' U R U' R' U' R' U R2 U' R' U' R' U2
JTLE
Start: skip
Finish: U2 R' U' R
JTLE: R' U2 R2 U2 R U' R U' R'

0 + 4 + 9 = 13 moves

Normal
First: U2 R U R2 U' R'
Finish: U' R' U' R U R' U2 R U2 R' U' R
OLL: R' U2 R U R' U R

6 +12 + 7 = 25 moves

Click to expand...

What about doing U' R' U2 R U R' U' R?
Still, with JTLE it's shorter. Just saying that the difference shouldn't be 12, but 8 moves.

Stachu,
why does it only has little potential?

John,
Nice example
In speedsolving, how can you decide whether you need to use JTLE or not? maybe in FMC you can try both and get 12 moves that saves a lot. But, in speedsolving, there almost no time to decide..

Well, I'm going to learn it after my first competition (Indonesian Open) next week maybe, and working more practice to speed up my petrus solve..

I think the first thing you should do to recognize it is to look for the edge and try to find a pair. If the pair and the edge seem like they would work well together, use the normal approach, but if the edge seems like it would be out of the way use JTLE. Also, if the c/e pair is already made use JTLE, unless you can tell the second c/e pair will be horrible (as shown in the third example solve).

I see how this would be useful now. On average it doesn't seem like it'd save any moves at all, but if inserting the DR edge is really ugly, it makes sense that you could save time by doing it during OLL.

ardi4nto said:

Stachu,
why does it only has little potential?

John,
Nice example
In speedsolving, how can you decide whether you need to use JTLE or not? maybe in FMC you can try both and get 12 moves that saves a lot. But, in speedsolving, there almost no time to decide..

Well, I'm going to learn it after my first competition (Indonesian Open) next week maybe, and working more practice to speed up my petrus solve..

Click to expand...

I'm just saying that in most cases, one might as well place that edge before, unless a 'convenient' case comes up.

But when the case is good, it's good.

I think the R and U series can make a fast execution. And instead you have one choice of edge, you have 5 that can be inserted on the F2L. So you can make a 'whatever edge' move that doesn't requiered to much thinking and moves.

Of course if you have an easy normal petrus you dont need to force the JTLE.

That' what i thought after trying this. But what if JTLE compared with MGLS?

trying-to-speedcube... said:

jtjogobonito said:

A solve where you should use it:

scramble: R2 U' R U2 R2 U2 R' U R U' R' U' R' U R2 U' R' U' R' U2
JTLE
Start: skip
Finish: U2 R' U' R
JTLE: R' U2 R2 U2 R U' R U' R'

0 + 4 + 9 = 13 moves

Normal
First: U2 R U R2 U' R'
Finish: U' R' U' R U R' U2 R U2 R' U' R
OLL: R' U2 R U R' U R

6 +12 + 7 = 25 moves

Click to expand...

What about doing U' R' U2 R U R' U' R?
Still, with JTLE it's shorter. Just saying that the difference shouldn't be 12, but 8 moves.

Click to expand...

I think that solution is misrepresenting Petrus. A real Petrus user wouldn't let you get away with that. One could also easily have seen R U2 R' U2 R2 U R. This leads to a less pleasant OLL but still.

This method compliments my insane version of ZZ-Fridrich excellently.

Dirk BerGuRK said:

I think that solution is misrepresenting Petrus. A real Petrus user wouldn't let you get away with that. One could also easily have seen R U2 R' U2 R2 U R. This leads to a less pleasant OLL but still.

Click to expand...

Sorry about that. I didn't see that when I was solving, but the resulting OLL still makes JTLE a better approache in that solve.

I'm going to start learning them tonight for OH. I really can't see myself using them much for regular solving because I have to try to leave the edge out of step 4. I'm just so used to doing it. the only reason I like them for OH is because they are 2 gen.

It may just be me being ignorant and missing it, but nobody seems to have mentioned that in Petrus you could save some moves by using the U layer edge in the 2x2x2 or 2x2x3.
It seems like everybody has just mentioned using the U layer edge in the <R, U> part.

Escher said:

It may just be me being ignorant and missing it, but nobody seems to have mentioned that in Petrus you could save some moves by using the U layer edge in the 2x2x2 or 2x2x3.
It seems like everybody has just mentioned using the U layer edge in the <R, U> part.

Click to expand...

You definitely can, but, in my opinion, it ruins the "flow" of it. I would much rather use two or three extra moves to have a pure 2 gen execution rather than a fast 2 gen slowed down by a D turn or y turn, 2 gen alg, then D turn again.

jtjogobonito said:

Escher said:

It may just be me being ignorant and missing it, but nobody seems to have mentioned that in Petrus you could save some moves by using the U layer edge in the 2x2x2 or 2x2x3.
It seems like everybody has just mentioned using the U layer edge in the <R, U> part.

Click to expand...

You definitely can, but, in my opinion, it ruins the "flow" of it. I would much rather use two or three extra moves to have a pure 2 gen execution rather than a fast 2 gen slowed down by a D turn or y turn, 2 gen alg, then D turn again.

Click to expand...

You also have to make sure that the edge is oriented properly when put into the 2x2x2/2x2x3 which takes some extra thinking and time.

Escher said:

It may just be me being ignorant and missing it, but nobody seems to have mentioned that in Petrus you could save some moves by using the U layer edge in the 2x2x2 or 2x2x3.
It seems like everybody has just mentioned using the U layer edge in the <R, U> part.

Click to expand...

That's actually a really good idea.
A good approach might even by making 2x2x1 blocks on 2 opposite sides, then using M slice for EO and inserting last edge to make the 2x2x3. y and finish F2L, z and do JTLE.

Another alternative is to also generate another set of algs with JTLE edges and EJF2L corners combined. (DFR twisted, DR at UR).

There isn't anything bad about JTLE, but is it worth it learning this set of algs? Does it have more chances to skip steps than EJF2L? Does it require less moves?

If using ZZ-Fridrich with JTLE, remember to NOT leave your last-slot-edge in the DR position, otherwise your last F2L pair will not be possible without disturbing another pair.

I think I may be the only person who does ZZ-Fridrich, so this probably won't affect anyone, but me.

For anyone thinking of doing ZZ-Fridrich, EO-Cross is much easier with JTLE because you don't have to worry about the DR piece.

Oh boy, JTLE AND EJF2L? Nah....lol. That's too much thought for such a simple, fast step. Shortcuts I'm looking for are in the 2x2x3, like the sort of full CLS "MGF2L" that I told you about JT. Really though if you come up with some kind of psuedo block stuff that makes 2x2x2 blocks easier so I can focus on 2x2x3 expansion transition, without making me learn 108 algs and becoming full CN, I'd love you.