# Finger trick notation

#### mark49152

##### Super Moderator
Staff member
Is there a notation for describing fingering for algorithms? Like the way sheet piano music is annotated with fingering?

My notes are littered with annotations like "(g-rt-u)" embedded into algs (= regrip right thumb on U). These help me remember fingerings I've seen in videos or worked out myself when studying algs. At the moment though, it seems there's no concise and precise way for cubers to communicate fingerings, only by description or video.

If no such notation exists and anyone else cares, I'd be interested in helping to work one out here, and that's why I created this as a new thread rather than a OAQ.

#### MarcelP

##### Member
That is very interesting. When learning new algs I seem to have a problem with finding the best finger tricks. It would be nice if these where described in a uniform way.

#### kunparekh18

##### Member
I thought of posting a fingertrick notation I thought of some time ago, but didn't, here's how I would "describe" fingertricks:

1. Hand - R or L.

2. Finger - 1, 2, 3, 4, 5 for thumb,index,middle,ring and pinky.

3. A or B, A for the front of the finger, B for the back.

4. Sticker - Conventional scheme or your BLD scheme (if you use Speffz, whatever suits you really), example, DFR would be F in my BLD scheme, so I would use F.

5. -Ps or -Pl for Push or Pull. Here, push means any turn away from you, while pull means any turn towards you. For example, a U' with the left hand would be a pull while a U' with the right hand (the way right-handed OHers do it) would be a push, I hope this is clear.

r for regrips (in the middle of an alg).
If you want to describe how you grip the cube while doing an alg instead of how you fingertrick it, you could use the same notation except the Ps or Pl part.

Examples:

1. How I would do the last F' of the common T-perm:

R1AF-Ps (here I call F' a "push" as if you would do an x U' (which is F') it would be a push).

2. How I do the first F' of the Jb perm (R U R' F' R U R' U' R' F R2 U' R' U')

R2BF-Ps (kind of like a Japanese U')

3. D's in the E perm (x' R U' R' D R U R D' R U R' D R U' R' D' )

L4BL-Ps (in my BLD lettering scheme LBD is L)

The Ps - Pl can be confusing sometimes, in this case D would usually be called a push and D' would be a pull, but if we go by the "towards-you away-from-you" logic D' would be a push and D would be a pull.

Last edited:

#### TDM

##### Member
r for regrips (in the middle of an alg).
r could be confused with Rw or slicing on big cubes. I sometimes use . to show regrips, e.g. R' U R' U' R D' R' D . R' y R2 U' R2' Dw R2
I don't think the notation should be included in the alg, unless it's something small that won't make the algorithm difficult to read. Because of this, any 'notation' should be done separate to the alg, so we don't really need a notation as we could just describe it in words, as someone has done here.

#### Lucas Garron

It would be nice to have a notation that is general enough to describe any algorithm accurately using finger tricks, but no one has made a proposal that clearly stands out as a good convention. (As a good example, the Heise keyboard layout for computer cubing was very well designed and caught on fast).

Make sure to take a look at Gripper.

#### kunparekh18

##### Member
r could be confused with Rw or slicing on big cubes. I sometimes use . to show regrips, e.g. R' U R' U' R D' R' D . R' y R2 U' R2' Dw R2
I don't think the notation should be included in the alg, unless it's something small that won't make the algorithm difficult to read. Because of this, any 'notation' should be done separate to the alg, so we don't really need a notation as we could just describe it in words, as someone has done here.

If that's the problem, rg could be better. I too don't think each and every finger trick should be included with every alg. If someone just asks a question on how to execute a particular alg one could explain using the above notation.

I agree this notation could make an alg difficult to read. This notation is just for clarification

#### mark49152

##### Super Moderator
Staff member
I was thinking of notation that would be embedded in the alg, to avoid supplementary description while making it absolutely clear how the execution is intended. It should therefore be compact and minimal. It should be easy to read, although would be longer of course. It wouldn't be an essential part of the alg, just an unambiguous and uniform convention as an option for describing an execution of that alg where that's what someone wants to do. For example, on an alg page someone may write "algorithm XXX, suggested execution YYY". The YYY would be the alg with embedded fingering notation, in order to describe the fingering accurately, but the original alg XXX is still the most important information and the reader might choose to execute it differently anyway.

Thanks for the interesting comments so far. I'll see what comes in over the next few days, then try to combine the inputs into a proposal.

#### sneaklyfox

##### Member
This would be good, but I can't think of a neat, concise way to do it. If you come up with one, that would be nice.

#### kunparekh18

##### Member
Update the alg.garron applet (RubikPlayer iirc), making it show the fingers on the cube as well?

#### cubernya

Update the alg.garron applet (RubikPlayer iirc), making it show the fingers on the cube as well?
The beta one is Javascript, so it'd be much more difficult to tell it where to put the fingers in a concise manner.

#### speedpicker

##### Member
I have spent the last year writing a guide to fingertricking and OLL, in which I devised my own notation system, which proceeds vertically. Here for example is

Sune:

[Picture of case goes here]

(R U R' U) R U2’ R'

[Description of alg in words goes here]

Initial Grip: F RIU (grip the cube in F grip, that means thumb on F face, fingers opposite, and place the right index finger on U)
R R (R means with whole right hand)
U RI (RI means right index finger)
R’ R (R means with whole right hand)
U RM (RM means right middle finger)
R R (R means with whole right hand)
U2’ L IM DF (L IM DF means Left Index Middle finger Double Flick)
R’ R (R means with whole right hand)

Obviously the stuff in brackets is for the benefit of this post. Once you get the general idea it flows very well.

From a learning point of view it is clear precisely how to execute each individual movement as the abbreviations translate into intuitively understandable plain language, and annotations are provided where there are alternatives or an extra level of clarification is required. All algorithms are also described in words for further assistance in learning (things like: "eject front right pair using a standard ejection so the corner is above the front right slot, then the sledgehammer pair into the slot." I find this plain and simple explanation of an alg of tremendous benefit in initially learning the movement).

Progress on the book (its currently at 144 pages) has stalled since I got a new job which has taken up huge amounts of my free time (practice has stalled a bit too, grrr) but I hope to release the whole thing with detailed fingertrick suggestions for each popular OLL alg, including executions from top cubers and all from multiple angles and initial grips. This is in addition to a 35 page essay/guide on executing and selecting fingertricks for algorithms, an in depth discussion of nuanced rotation technique, an entirely new grip is introduced, as is a wholly original (and reaaaallly fast) fingertrick which can be applied to dozens of standard algs. If anyone wants to ask me anything about it, I will be at the British Champs, with a copy to hand. When I finally release it it will be as a kindle compatible ebook and as a PDF, so you can learn on the go with your phone/tablet/laptop/whatever.

#### speedpicker

##### Member
The problem with videos is the mucking around trying to nail exactly what the guy did. Bear in mind what is second nature to the expert may take the beginner many passes to thoroughly absorb. Take Breandans OLL video: to be certain of every single subtlety, you have to watch it repeatedly, rewind certain parts again and again, and perhaps take notes during the process for later recall. You may ask "what was that finger doing whilst that move was going on? Exactly how did that regrip happen, I know it did, but when exactly, how did he do that R2 move and remain with his thumb still on the front? Not to mention the added bonus of having the information at your fingertips at any time, rather that having to load a video, go to 2:16, watch 5 seconds of footage and then rewind it. The benefits of having that all notated and written down should be clear. You then only need to refer to the video to clarify a point or to prove to yourself that a thing is possible. In an ideal learning situation, videos should supplement book learning, not the other way around.

#### mark49152

##### Super Moderator
Staff member
Using videos as a way of noting and communicating algorithm fingering makes about as much sense as using videos to note and communicate the algs themselves, IMHO.

#### KongShou

##### Member

Why don't we number the fingers 0-9

So starting with the right hand, we have 1 for the thumb, 2 for index, 3 for middle, 4 for ring and 5 for little. This ways seems logical and intuitive. For the left hand we do a similar thing and have 6 for thumb, 7 for index, 8 for middle, 9 for ring and 0 for little. This is done so that we only need one digit.
We can put this in front of the actual move in the algorithm, e.g. 2U, and move which require two fingers like U2 can have two numbers before it, in order, like 23U2 for the U2 double flick. There is cases where you have to push with your finger, for example I like doing U with my left index finger(7) sometimes. This can be denoted 7-U, using the - for the push. This is also useful for OH i imagine.

Now wrist turns can be written in small r and l, for right and left hand respectively. The inverse can be r' and l' for anticlockwise. An 180 degree turn is of course r2 or l2

We can also note where to put the fingers to start an algorithm, although this require more thought to invent a quick conventional way to note all the stickers. For now i suggest we use the old way of three letters to denote a corner, i.e. DFR or FUR, and two letters to denote an edge.. So my T perm might start with:

1DR 2UFR 3UR 4UBR (leave blank for 5 as it does not touch the cube) 6FL 7BUL 8BL 9BLD 0DBL

However this way is quite tedious and I'm sure someone else could come up with a better way.

For a turn we do not need to note the stickers because it is pretty clear where to press on the cube to turn it. However, if you say require the back of a finger to flick a side, then we can easily have 2' for the back of right index finger.

We can also have regrip in the form of a /. We can put either r or l in front of the dash to make it clear which hand it is. We can also have clockwise regrip and and anticlockwise regrip by putting a ' behind the letter. For example: r/ or l'/ There is also of course 180 degree regrip and that can be represented like r2/.

Using this notation my T perm would look like:

rR 2U r/r'R' 7U' r'R' 2F r2R2 7U' r'R' 7U' rR 7-U r'R' 1F'

Please give me some feedback. The only problem I can think of is that it looks confusing. But you can probable get over it. After all, there is only 4 types of notation: numbers for fingers, lower case letters for wrist turns, dash for regrip and - for a push move.

#### Lucas Garron

The beta one is Javascript, so it'd be much more difficult to tell it where to put the fingers in a concise manner.

Actually, it would be *easier* to modify. I'd totally be interested in such a feature, but it wouldn't be a top priority.

Please give me some feedback. The only problem I can think of is that it looks confusing. But you can probable get over it. After all, there is only 4 types of notation: numbers for fingers, lower case letters for wrist turns, dash for regrip and - for a push move.

Not a bad idea. My main concern isn't that it's confusing, it's that it is completely ambiguous. If something like this catches on, it would be really annoying to read and alg and not know immediately what it means. Something compatible with SiGN would be awesome, though. I could probably add it to the parser for http://alg.garron.us/ some day.

Last edited:

#### KongShou

##### Member
Actually, it would be *easier* to modify. I'd totally be interested in such a feature, but it wouldn't be a top priority.

Not a bad idea. My main concern isn't that it's confusing, it's that it is completely ambiguous. If something like this catches on, it would be really annoying to read and alg and not know immediately what it means. Something compatible with SiGN would be awesome, though. I could probably add it to the parser for http://alg.garron.us/ some day.

Thanks for the feedback! Im glad that it is useful!

I guess to make it comatable with SiGN you can probably put it in brackets?
So:

R(r) U(2) R'(r/r') U'(7) R'(r') F(2) R2(r2) U'(7) R'(r') U'(7) R(r) U(7-) R'(r') F'(1)

And i agree that it is ambiguous, but there isnt much to learn so i guess you could get used to it quickly?

#### elrog

##### Member
I thought of posting a fingertrick notation I thought of some time ago, but didn't, here's how I would "describe" fingertricks:

1. Hand - R or L.

2. Finger - 1, 2, 3, 4, 5 for thumb,index,middle,ring and pinky.

3. A or B, A for the front of the finger, B for the back.

4. Sticker - Conventional scheme or your BLD scheme (if you use Speffz, whatever suits you really), example, DFR would be F in my BLD scheme, so I would use F.

5. -Ps or -Pl for Push or Pull. Here, push means any turn away from you, while pull means any turn towards you. For example, a U' with the left hand would be a pull while a U' with the right hand (the way right-handed OHers do it) would be a push, I hope this is clear.

r for regrips (in the middle of an alg).
If you want to describe how you grip the cube while doing an alg instead of how you fingertrick it, you could use the same notation except the Ps or Pl part.

Examples:

1. How I would do the last F' of the common T-perm:

R1AF-Ps (here I call F' a "push" as if you would do an x U' (which is F') it would be a push).

2. How I do the first F' of the Jb perm (R U R' F' R U R' U' R' F R2 U' R' U')

R2BF-Ps (kind of like a Japanese U')

3. D's in the E perm (x' R U' R' D R U R D' R U R' D R U' R' D' )

L4BL-Ps (in my BLD lettering scheme LBD is L)

The Ps - Pl can be confusing sometimes, in this case D would usually be called a push and D' would be a pull, but if we go by the "towards-you away-from-you" logic D' would be a push and D would be a pull.

This is an ok system, but there's no way to know how to hold the cube. You tell which finger does the pushing, flicking, or pulling, but you don't mention if you are griping the DF and DB edges like in ZZ, the left block like in Roux, or anything else. There really is no better way to get the information than a slowed down video shot from different angles of someone preforming the algorithm very quickly.

Last edited:

#### Lucas Garron

I would propose the following changes:

• Use braces instead of parentheses, to avoid a conflict of notation.
• Keep the grip notation before the move (since you need to know how to grip before you do the move).
• Consistently use ' instead of -.
• Have a letter for each finger. Ideally, I'd like the hands to be treated symmetrically, so the best idea I have is to append the lowercase letter of the hand.
• Tr/Tl (Thumb)
• Ir/Il (Index finger)
• Mr/Ml (Middle finger)
• Gr/Gl (rinG finger, to avoid having too many uses for "R")
• Pr/Pl (Pinky finger)
• Use the notation {#:sticker} to denote the condition that the numbered finger is on the sticker.
• Define a standard grip: {1:FR} {2:BUR} {3:BR} {4}BDR {5:X} {6}FL {7}BUL {8}BL {9}BDL {10:X}, or {Tr:FR} {Ir:BUR} {Mr:BR} {Gr}BDR {Pr:X} {Tl}FL {Il}BUL {Ml}BL {Gl}BDL {Pl:X} with my finger suggestions.
• Have defaults.
• In standard grip, U is {2}U, U' is {2}U, R is {R}R, D' is {4}D', etc.
• If {2:UFR}, then F is {2}F.
• Standard followthroughs: Ir is followed by Mr (e.g. for U2), Gr is followed by Mr (e.g. for M2')
• Use uppercase letters for L/R regrips.
• Write regrips as separate moves.
• Use / to denote resetting to the standard grip before the regrip
• Specify that placing a finger over a sticker moves the existing finger (and doesn't specify where exactly that finger goes).
• Implicitly assume that a finger on a sticker on an adjacent slice will be move over for a move, e.g. {4}M = {4:BD} {4}M

T-perm: {/R'} R U R' U' {R} R' F R2 U' R' U' {R'} R U R' {Tr}F'
U-perm: {/} R2' {Gr:FU} r2 U' M' U2' r' R U' {Ir:BR} {Gr}M2'
J-perm: {/R'} R U R' {Il:UFL} R U R' U' {R} R' F R2 U R'
G-perm: {/R'} R U R' y' R2 {Ml}u' {/} R U' R' U R' {Pr}u R2
G-perm: {R} R2' {Ml}u' R U' R {Tl}U R' {Ir}u R2 y' {/L'} L U' L'

I'm sure this isn't perfect, but I think it's relatively practical. Opinions?

Last edited:

#### KongShou

##### Member
• Use braces instead of parentheses, to avoid a conflict of notation.
• Keep the grip notation before the move (since you need to know how to grip before you do the move).
• Consistently use ' instead of -.
• Use the notation {#:sticker} to denote the condition that the numbered finger is on the sticker.
• If {2:UFR}, then F is {2}F.
• Use uppercase letters for L/R regrips.
• Write regrips as separate moves.
• Specify that placing a finger over a sticker moves the existing finger (and doesn't specify where exactly that finger goes).
• Implicitly assume that a finger on a sticker on an adjacent slice will be move over for a move, e.g. {4}M = {4:BD} {4}M

I agree with all these, in fact, some of these ideas are brilliant.

• Have a letter for each finger. Ideally, I'd like the hands to be treated symmetrically, so the best idea I have is to append the lowercase letter of the hand.
• Tr/Tl (Thumb)
• Ir/Il (Index finger)
• Mr/Ml (Middle finger)
• Gr/Gl (rinG finger, to avoid having too many uses for "R")
• Pr/Pl (Pinky finger)

These im not sure about. The reason why i used numbers, especially all one digit long, is because i wanted to keep it short. Now i dont know about you, but It takes me much longer to find the right finger using your notation, even through it should supposedly make sense and intuitive. perhaps we can name it 1-5 from thumb to pinky, and have l or r after it. So 1r/1l 2r/2l 3r/3l 4r/4l 5r/5l. I know this contradict my previous statement about keeping it short, but i guess it is better to have it symmetrical.

• Define a standard grip: {1:FR} {2:BUR} {3:BR} {4}BDR {5:X} {6}FL {7}BUL {8}BL {9}BDL {10:X}, or {Tr:FR} {Ir:BUR} {Mr:BR} {Gr}BDR {Pr:X} {Tl}FL {Il}BUL {Ml}BL {Gl}BDL {Pl:X} with my finger suggestions.
• Have defaults.
• In standard grip, U is {2}U, U' is {2}U, R is {R}R, D' is {4}D', etc.

I am really against the idea of a standard grip. I think everyone has their own turning style and so the notation should be kept as liberal as possible to allow the individual to develop their own "standard grip". Although this grip is probably common and most people use it, I still think it is a bad idea to standardise turning style.

However i think the defaults could be developed further. They will save a lot of time and make everything easier. The obvious and intuitive stuff like U is {2}U shouldnt need to be noted down.

Standard followthroughs: Ir is followed by Mr (e.g. for U2), Gr is followed by Mr (e.g. for M2')

I would agree with this as default, but for the M'2 doubleflick some people also like using Gr followed by Pr(or 4r 5r). This will then create ambiguity.

Thanks for the feedback and suggestions!