• 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!

Mirrored Inverse Sexy Method (MirIS)

Cride5

Premium Member
Joined
Jan 27, 2009
Messages
1,228
Location
Scotland
WCA
2009RIDE01
A beginner method optimised for speed and simplicity

EDIT: Full tutorial here.

Rationale
A while back - while trying to find the easiest possible beginner method I came up with SAM (the Single Algorithm Method).

Since then I've been teaching it (and variants of it) to a number of people, sometimes even in a foreign language. While most people seem to be able to grasp it after a while, I've found it has a number of issues:
  • Solving middle layer edges (despite being only 3 moves) turned out to be difficult to explain, mainly because the edge needs to NOT be aligned with its centre before doing the 3-move sequence. I also found that users were often confused about which side to turn in the 3-move sequence.
  • Solving orientation and permutation simultaneously in the top cross, while an elegant solution seems too much intuition for many absolute beginners. This is where a lot of them mess up and have to go to a previous step.
  • The process of solving the last edge on the top cross was probably made more difficult by sticking to a single alg, and I ended up teaching people just to use R U R' U R U R'
  • On last 5 corners users are tempted to turn the top layer since they were used to doing this while solving the first 3 on the bottom. I used to avoid this by teaching a variant which simply solves all the edges first, then all 8 corners last. This reduced the number of concepts but lead some horrible situations where the cube needs to be inverted multiple times.
  • Users often forget the last move in the 1-2-3-4 sequence because the goal (of placing the corner) is competed after 3. A little bit like the problem with ATMs that give you your cash before returning the card.

In general, I think focusing strictly on just one algorithm was detrimental to the ultimate goal of making it easier for beginners. However, I still feel that a good beginner method should minimise the number of algs, and number of moves in each alg, while avoiding making too much use of intuitive solving.

Steps
  1. Cross
  2. 3x corner-edge pairs
  3. Top cross orientation
  4. Top cross permutation
  5. Last 5 corners

Details
  • For this method the Inverse Sexy move (U R U' R') and its mirror (U' L' U L) are used.
  • We also assume that the user is starting with the white cross.
  • IS = 'Inverse Sexy'
  • MirIS = 'Mirrored Inverse Sexy'
  • C+E = 'Corner+Edge'


1. Cross
Intuitive. Use daisy if absolute beginner.

2. 3x corner edge pairs
2a
Look for a corner in the top layer, place it above its 'slot'
2b If the white sticker is on the top layer apply IS x2
2c Hold the cube so that the white sticker is facing you. If the corner piece is now on the left then apply MirIS, otherwise IS
2d Look for the adjoining slot edge in the top layer
2e Rotate the top layer so that its side lines up with its center
2f Hold the cube so that the edge is facing you, if the slot is on the left apply MirIS, otherwise IS
2g There should now be a C+E pair above the slot. Repeat 2c to insert the C+E pair

If you're looking for any pieces not in the top layer, find the slot with the piece and apply IS.

3. Top Cross Orientation
3a
Look at the edge in the last unsolved slot, hold the cube so that its yellow sticker is facing you
3b
- If the slot is now on your right:
Rotate any top layer edge without a yellow sticker facing up into right side and apply IS​
- If the slot is now on your left:
Rotate any top layer edge without a yellow sticker facing up into left side and apply MirIS​
Repeat 3a/3b until all edges in the top layer are oriented.

4. Top Cross Permutation
Attempt to align the cross pieces with their sides, if they can't be aligned then there are two cases:
4a Two adjacent pieces can be can be aligned: Rotate the top layer so that the two pieces are in the back-left corner and apply IS, then repeat step 3b
4b Two opposite pieces can be aligned: Apply IS, then step 3b, then step 4a

5. Last 5 Corners
As documented in 8355 or SAM, only using Inverse Sexy instead of plain Sexy Move


Advantages over SAM
  • C+E pair insertion has easy transition to CFOP and avoids adding another concept for separate insertion.
  • A small number of simple principles are used consistently throughout such as, "orient white/yellow sticker to face you", "place the unsolved edge into the R/L"
  • Inverse Sexy is more beautiful than Sexy ;-)
 
Last edited:

Cride5

Premium Member
Joined
Jan 27, 2009
Messages
1,228
Location
Scotland
WCA
2009RIDE01
What kind of times can you, specifically, get with this method? Just curious
37.38, 37.58, (50.44), 44.48, (35.36) = 39.81
I'm sure an experienced CFOPer with good turn speed would be able to do much better. Fancy a try?

Didn't realise you were still around, nice to see you post. Interesting method and you've clearly thought it through well, I'll probably try teaching it some time.
Hi Mat, sure has been a while. Busy as always but I try to pop in from time to time. Let me know how the teaching goes.
 
Joined
Apr 23, 2010
Messages
1,391
Location
Scotland, UK
WCA
2009SHEE01
YouTube
Visit Channel
There's some missing awkward cases which might confuse people, otherwise pretty nice tutorial and I've bookmarked it for reference :).

-For c-e pairs, doesn't mention the edge being in the wrong slot
-For top cross orientation, doesn't mention if all the yellow edges are in U but not oriented
-For L5C, there's a few awkward cases missed, here's an example which I think includes them all (also diagrams have yellow on top instead of white):
scramble: U R S U S' R' S U' S' U' z2 y'

D IS
URF is white so guide says do IS, but need to put unsolved corner at DFR first

D IS D IS*3 D' IS*3
no unsolved corner to put at DFR before fixing stuff

IS*4 D2
can't put both corners on D by rotating

D2 z IS*4 D IS*2 D' L2
 

Cride5

Premium Member
Joined
Jan 27, 2009
Messages
1,228
Location
Scotland
WCA
2009RIDE01
Thanks for the pointers. I've updated the tutorial to fix these issues, and also added some more explanation of the steps.
 

Cride5

Premium Member
Joined
Jan 27, 2009
Messages
1,228
Location
Scotland
WCA
2009RIDE01
Tips for improving speed/efficiency and transition to speed solving

These steps are roughly ordered by difficulty:

  • Improve speed/efficiency when solving C+E pairs:
    - When lifting a corner from an incorrect slot to the top layer, turn cube so that the white sticker is facing you and apply IS/MirIS to place the corner into the top layer without the white sticker facing up.
    - If lifting a corner from an incorrect slot and its white sticker is facing down, place its adjoining edge in the top layer so that its side sticker is in the same face as the matching corner sticker, then apply MirIS/IS to lift the corner and create the pair simultaneously.
    - If lifting a corner from an incorrect slot and its white sticker is not facing down, and the top sticker of its adjoining edge matches the other side sticker of the corner, rotate the cube so the white sticker is facing you, and place the adjoining edge in the back of the top layer. Apply MirIS/IS to lift the corner and create the pair simultaneously.
    - When a corner has its white sticker facing up, instead of ISx3, rotate the cube so that the corner sticker that matches its adjoining edge's side sticker is facing you, then apply MirIS/IS and use the technique above to create the C+E pair on the next application of MirIS/IS.
    - If a corner with a white sticker facing up is already joined to its edge, instead of ISx3, simply lift the C+E pair into the top layer with R U' R'.
    - Learn to recognise the R U R' (or L' U' L) insertion case. The corner is in the top layer with white facing to the side, the edge is in the back of the top layer and the top stickers of the corner and its edge are not matching.
    - Once a corner is placed in its slot, a faster pairing/insertion can be done by placing the edge in the back of the top layer and applying: F' R U R' U' R' F R or its mirror F L' U' L U L F' L'
  • Switch from Daisy to solving the full white cross from the beginning.
  • Improve speed/efficiency when solving last 5-corners by:
    - using the regular sexy move (R U R' U') to place corners that would require 5 repetitions of Inverse Sexy.
    - ensuring that when pieces are moved from D to U, the yellow sticker is not on top by choosing the correct alg (Sexy or Inverse Sexy)
    - avoiding cases that require 3-repetitions by preferring to move pieces already in D up to U using the technique above.
    - Solving orientation of corners with the correct alg (Sexy to twist clockwise, Inverse for anticlockwise).
  • Solve top cross in one step by solving orientation and permutation simultaneously, as explained here.
  • Reduce pauses between steps by looking for the next piece(s) while solving the current step.
  • Learn to solve the white cross blindfolded so that the moves for solving it can be planned during inspection.
  • Learn how to solve the C+E pairs intuitively - tutorial here.
  • Fill in the 4th slot and learn to solve the last layer with algorithms.

At this point you will now be using basic CFOP method, but you may wish to investigate other methods such as Roux or ZZ to explore other concepts and find which style suits you best.
 
Last edited:

Cride5

Premium Member
Joined
Jan 27, 2009
Messages
1,228
Location
Scotland
WCA
2009RIDE01
The tutorial on the site has been more carefully authored so I recommend following that one.

If the steps here are followed as a sequence the step 2c still works. It sets up the corner so that the white sticker is not on top, making it solvable in the next step (2c).
 

Whurm

Member
Joined
Oct 25, 2016
Messages
2
So far I'm loving this method. I've taught it to over 20 people in the nursing home where I volunteer. They LOVE it. Being able to solve it brings a smile to their faces.

They only problem I'm having is at the very end if the two last spots are diagonally opposite. For some reason everyone keeps messing up here. If it happens we usually start over. Any hints or ideas? Detailed steps would be great. After the half turn do you just solve the right bottom right corner using IS, then turn the bottom layer to solve the second? Do you every half turn back?
 

Cride5

Premium Member
Joined
Jan 27, 2009
Messages
1,228
Location
Scotland
WCA
2009RIDE01
Hi Whurm, great work volunteering at the nursing home - I'm sure they really appreciate it.

The alignment step is probably the most difficult to understand conceptually, since (in the adjacent case) you have to understand exactly which two edges are permuted correctly. For those familiar with the cube's colour scheme this is easy, but for a beginner maybe not. The important thing here is to actually attempt to align all pieces. Go though each of the four possible U-layer positions until two of the edge pieces match their side-centers. Once these matching pieces have been identified (in the adjacent case), they must be placed in B and L positions. If they are not placed there before proceeding, then you will end up back where you started.

With the correctly permuted edges in BL, just apply IS once, and you will end up with three yellow edges on top, and one in the 'slot'. Using the same approach as the previous 'yellow cross' step, rotate the U-layer until the unsolved piece is in the R-position of the U-layer. Applying IS from here will complete the top cross.

In the case that is is possible to align two top-layer edges that are opposite each other, then do IS, followed by U, then IS (to fix the yellow edge again). At this stage you will have the adjacent case, and you can follow the instructions above to solve it. When you have the opposite case, the position of the U-layer doesn't matter the first time you apply IS, since any of the two opposite edges can be considered correct relative to each-other.

The tutorial read a bit like a program, requiring you to skip to previous steps. I've updated it to be more linear, so you can just follow the alignment instructions from top to bottom. I hope that makes it clearer.

If people are still struggling with this approach to doing the top layer edges, an alternative method is the one detailed here:
http://cube.crider.co.uk/beginner.php#3te
 

gateway cuber

Member
Joined
Jan 30, 2016
Messages
468
Location
in your car driving away...
YouTube
Visit Channel
A beginner method optimised for speed and simplicity

EDIT: Full tutorial here.

Rationale
A while back - while trying to find the easiest possible beginner method I came up with SAM (the Single Algorithm Method).

Since then I've been teaching it (and variants of it) to a number of people, sometimes even in a foreign language. While most people seem to be able to grasp it after a while, I've found it has a number of issues:
  • Solving middle layer edges (despite being only 3 moves) turned out to be difficult to explain, mainly because the edge needs to NOT be aligned with its centre before doing the 3-move sequence. I also found that users were often confused about which side to turn in the 3-move sequence.
  • Solving orientation and permutation simultaneously in the top cross, while an elegant solution seems too much intuition for many absolute beginners. This is where a lot of them mess up and have to go to a previous step.
  • The process of solving the last edge on the top cross was probably made more difficult by sticking to a single alg, and I ended up teaching people just to use R U R' U R U R'
  • On last 5 corners users are tempted to turn the top layer since they were used to doing this while solving the first 3 on the bottom. I used to avoid this by teaching a variant which simply solves all the edges first, then all 8 corners last. This reduced the number of concepts but lead some horrible situations where the cube needs to be inverted multiple times.
  • Users often forget the last move in the 1-2-3-4 sequence because the goal (of placing the corner) is competed after 3. A little bit like the problem with ATMs that give you your cash before returning the card.

In general, I think focusing strictly on just one algorithm was detrimental to the ultimate goal of making it easier for beginners. However, I still feel that a good beginner method should minimise the number of algs, and number of moves in each alg, while avoiding making too much use of intuitive solving.

Steps
  1. Cross
  2. 3x corner-edge pairs
  3. Top cross orientation
  4. Top cross permutation
  5. Last 5 corners

Details
  • For this method the Inverse Sexy move (U R U' R') and its mirror (U' L' U L) are used.
  • We also assume that the user is starting with the white cross.
  • IS = 'Inverse Sexy'
  • MirIS = 'Mirrored Inverse Sexy'
  • C+E = 'Corner+Edge'


1. Cross
Intuitive. Use daisy if absolute beginner.

2. 3x corner edge pairs
2a
Look for a corner in the top layer, place it above its 'slot'
2b If the white sticker is on the top layer apply IS x2
2c Hold the cube so that the white sticker is facing you. If the corner piece is now on the left then apply MirIS, otherwise IS
2d Look for the adjoining slot edge in the top layer
2e Rotate the top layer so that its side lines up with its center
2f Hold the cube so that the edge is facing you, if the slot is on the left apply MirIS, otherwise IS
2g There should now be a C+E pair above the slot. Repeat 2c to insert the C+E pair

If you're looking for any pieces not in the top layer, find the slot with the piece and apply IS.

3. Top Cross Orientation
3a
Look at the edge in the last unsolved slot, hold the cube so that its yellow sticker is facing you
3b
- If the slot is now on your right:
Rotate any top layer edge without a yellow sticker facing up into right side and apply IS​
- If the slot is now on your left:
Rotate any top layer edge without a yellow sticker facing up into left side and apply MirIS​
Repeat 3a/3b until all edges in the top layer are oriented.

4. Top Cross Permutation
Attempt to align the cross pieces with their sides, if they can't be aligned then there are two cases:
4a Two adjacent pieces can be can be aligned: Rotate the top layer so that the two pieces are in the back-left corner and apply IS, then repeat step 3b
4b Two opposite pieces can be aligned: Apply IS, then step 3b, then step 4a

5. Last 5 Corners
As documented in 8355 or SAM, only using Inverse Sexy instead of plain Sexy Move


Advantages over SAM
  • C+E pair insertion has easy transition to CFOP and avoids adding another concept for separate insertion.
  • A small number of simple principles are used consistently throughout such as, "orient white/yellow sticker to face you", "place the unsolved edge into the R/L"
  • Inverse Sexy is more beautiful than Sexy ;-)
learning this and if I like it, I'll master this.
 

Silverback

Member
Joined
Nov 21, 2014
Messages
55
One suggestion for the top cross:
Move the edge in the slot to the top with an F' or R, so its ends up with the yellow sticker next to the yellow center (forming a part of the cross). Then rotate the top layer to put a misoriented yellow edge into that position and undo that F' or R move. This puts another yellow edge to work with in the middle layer.
Repeat until you have 3 oriented yellow edges in the top layer. Don't worry about permutation.
Before you place the fourth edge, you have to check, how many of your top edges can be put into their correct positions at the same time. If you can place all three or not more than one, you have a bad case. Then keep swapping pieces until you get a good case.
If you can place exactly two pieces, you have a good case. It doesn't matter if these edges are next to each other or opposite of each other.
Put the third edge on top of the slot, replace it with the one in the slot, rotate the top to bring the non-yellow edge on top of the slot and move it into the slot.
This will solve all 5 edges.
Hope this was clear. If not, I can make an example.
 

mitja

Member
Joined
Dec 22, 2015
Messages
232
Location
Ljubljana, Slovenia
WCA
2016POPO02
Hi. I believe this is a beautifull method to solve the cube. I teach kids regularly how to solve the cube, but i use tha regular layer by layer approach: cross, corners, 2nd layer edges, top cross, top cross perm, niklass alg for corners perm and last corners orientation. With more advanced kids I go straight to 3look OLL with cross +1-2xsune or antisune. Then 4 look PLL with A and U perm. That way I prepare them for CFOP.

My experience while teaching was:
1. The hardest is the cross. The reason is, they get to know the cube for the first time. To understand the 3D concept of solving the pieces (not the stickers colour) is hard for some of the children. It gets much easier once they move to the next stage. It also gets very easy for them to learn algs, like 2nd layer edges alg or sune, niklas alg.
If i could calculate time needed to teach them each stage, I need about 60% of the time for first 2 layers and only 40% of the time to teach them solve the rest.

What was my point? I believe this MirIS method is great, but not the complete beginners method. I would call it advanced beginners method. To understand IS move is not so easy and using kind of keyhole is also not that easy either.
Alternating between IS and inverse IS also takes some good cube awearness for a beginner.
And finally the last 5 corners are very hard for a beginner. If I wouldn't already use the simillar corner orientation and then permutation when I was first putting toogether the megaminx, I would struggle to understand it.
I think it is very nice with adult beginners, especially someone that gets scared of algs. I very much remember reactons by lots of my friends when they say:" Doing this cube is just memorising hundreds of algs. It is very boring."
They are always surprised once I show them that two thirds of the cube is done with pure intuition and never repeated twice.

Anyway, it is very tempting method and I will try it with beginners to compare how they will cope with it.
 

Le Zouave

Member
Joined
Jan 8, 2018
Messages
1
Hi, I just received my first cube 5 days ago and I must say that your method is brillant. I was scared to learn a handful of algorithms but this one only have two. The most challenging was to understand why I musn't move the 2 upper layer while doing the 3 yellow corners but I must say that I'm a vertical reader.
Once I understood what part where affected by IS and MirIS and that repeating 6 times the same algs return to the original state, it's very easy.
 
Last edited:

BirdOfChaos

Member
Joined
Aug 27, 2018
Messages
2
I love this method. Thank you very much Cride5! :)

I used to know how to solve layer by layer a long time ago, but have long since forgotten any of the algorithms. I recently came back to cubing and I didn't want to memorize too many algs before being able to solve the cube. I was able to teach myself how to solve the cube again with the MirIS method in less than an hour. Quickly my times decreased from about 15 minutes (when I still needed to read the instructions once in a while) to 3, and now I can solve in about 1m 1/2. One trick that I quickly realized I needed was to make sure that the correct edge piece for the F2L step was not in the middle layer when bringing a corner down to the white layer, or if it was, to select that slot to bring down the corner to, so that the edge would be displaced to the top, ready to use. Otherwise I'd need another separate IS move right after to bring it to the top.

My goal is to get below 1 minute, which I never achieved even back in the 90's. For that I'm looking at the tips for improving that you have listed. I find that some of the tips are clear, but for some others (like the very first one for example), I'm not sure what you mean and during which step it's supposed to be done. I think that a follow-up page with instructions on how to progressively get from MirIS to 2-look CFOP would be super useful for those of us who have "mastered" MirIS. I'm going to time myself during the various phases of the MirIS to see where I need to improve most and start there.
 
Top