Efficient Digital Cycles

I have been preparing a post about the left hand postures and fretting strategies I use in my playing, after reading through the following thread:

That promised post is still in preparation, and I will present it when I’m satisfied with it.

In preparing that post, I watched through several instructional videos and CTC interviews again.

While re-watching Shawn Lane’s Power Licks for the first time in a while, I realized something about his fretting sequences that I felt was extremely interesting, and which I’d like to share here. I have other observations regarding his picking which I’ll discuss elsewhere.

The concept I’d like to share here is something I’ve decided to call an Efficient Digital Cycle. Before I go in depth about it, I’d like to draw an analogy with something the community here is familiar with, being alternate picking and the problem of string-hopping.

An efficient alternate picking cycle requires that the downstrokes and upstrokes use antagonistic muscles. With a string-hopping movement, the upstroke and downstroke use the same muscle to escape the strings. The result is a much lower potential speed, and increased fatigue and strain in the picking hand. Making the string-hopping movement smaller doesn’t fix the problem.

I’ve long been critical of the idea of economy of movement. Making an inefficient movement smaller does not make that movement efficient, and focusing on making small movements is detrimental to learning to make efficient movements.

The idea is not an economy of movement, where we determine cost whether a movement is large or small, but an economy of movements, where we acknowledge that there is an abundance of different movements available to us, with different costs, in different currencies, and we endeavor to achieve our goals using a movement which is inexpensive to us in terms of the currencies we have available at that time. Cheap for gold is still expensive for silver.

So, with that in mind, let’s consider fretting cycles. A fretting cycle is a repeating sequence of fretting digits. The cycle length is the number of notes that it takes for the cycle of fretting digits to repeat.

For example, consider the following exercise. We play with one finger per fret (on any string) and repeat

-5-6-5-7-5-8-6-7-6-8-7-8-(repeat)   

Here, we have a cycle of length 12;

(1 2 1 3 1 4 2 3 2 4 3 4) 

As another example, we have the following A minor pentatonic scale, played with the first and third fretting fingers only

---------------------5-8-
-----------------5-8-----
-------------5-7---------
---------5-7-------------
-----5-7-----------------
-5-8---------------------

The fretting sequence here is

(1 3)(1 3)(1 3)(1 3)(1 3)(1 3) or (1 3)^6

That is, we have a cycle of length two played times. Consider the typical G major scale

-------------------------------5-7-8-
-------------------------5-7-8-------
-------------------4-5-7-------------
-------------4-5-7-------------------
-------3-5-7-------------------------
-3-5-7-------------------------------

There are multiple ways this sequence can be fretted, the two most common are

(1 2 4)(1 2 4)(1 2 4)(1 2 4)(1 3 4)(1 3 4) or (1 2 4)^6*(1 3 4)^2

and

(1 2 4)(1 2 4)(1 2 4)(1 2 4)(1 2 3)(1 2 3) or (1 2 4)^6*(1 2 3)^2

A cycle need not carry a specific starting digit, so we have equivalant fretting cycles

(1 2 3) = (2 3 1) = (3 1 2) 

and

(3 2 1) = (2 1 3) = (1 3 2) 

but

(1 2 3) =/= (1 3 2)

Now, let’s consider efficiency.

It is more expensive to transition between the 3rd and 4th fingers in a fretting sequence than between any other combination of two fingers. There are two primary anatomical reasons for this.

The first is that the 3rd and 4th fingers share common tendon and muscle attachments, and so the degree of independence possible between these two fingers is simply not at same level as the level of independence possible between the other combinations.

The other reason is that these fingers differ significantly in length, which makes it difficult to achieve a fretting posture where both fingers work together and work equally well with the 1st and 2nd fingers.

While it is possible to do this movement quite quickly, it quickly results in fatigue and strain. Transitions between the 3rd and 4th fretting fingers are to fretting cycles what string-hopping is to the alternate picking cycle.

The other class of movements which are expensive are transitions from one starting finger to another and then to fret a new note with the starting finger. For example, in the example of the pentatonic scale, the 1st finger needs to lift off from the the starting string and immediate fret a note on the next string while the 3rd finger is fretting, and vice-versa. For this reason if a, b and c represent fretting fingers, the fretting sequence

a b a 

is more costly than

a b c

unless the latter contains a transition between the 3rd and 4th fingers, as noted above. There is is a subtle point here, which I’ll go into more detail on later.

Shawn overcomes the issue of transitioning between the 3rd and 4th fingers by not doing it. Shawn almost never used the (1 3 4) or (1 4 3) fretting sequences in his playing. Instead, he re-fingered patterns to avoid these sequences entirely. These sequences appear only when it’s unfeasible to re-finger the sequence, and even then, their appearances are brief.

The other class of expensive movements is avoided in his diatonic playing by the efficient cycling of the fingers.

Consider the fretting cycle (1 2 4) repeated. After the 1st finger frets, it has two notes duration to lift from it’s original position and fret a new note, and similarly for the other two fingers. That is, in this cycle, the next finger to fret is always cheap, and so this cycle is maximally efficient.

We find that there are four such maximally efficient cycles. Namely

(1 2 3), (3 2 1), (1 2 4), (4 2 1).         

In Shawn’s playing, we can essentially consider the 3rd and 4th fingers as one, since they are almost never used in combination. So we’re left with only two such maximally efficient cycle types, (1 2 3) = (1 2 4) and (3 2 1) = (4 2 1).

For reasons I’ll discuss later, it’s wrong to consider one of these as an ascending cycle and the other as a descending cycle. Bother can be used ascending or descending, as we’ll see momentarily. Instead, I’ll call (1 2 3) & (1 2 4) the forward cycle, and (3 2 1) & (4 2 1) the back cycle. We can transition between both forward cycles cheaply and between between both back cycles cheaply. Transitioning between forward an back requires a turnaround, which has to be handled in a specific manner.

Amazingly, Shawn’s vocabulary of fast licks is almost entirely developed based on these two fundamental cycles. The notable exception is his pentatonic sequences, which he doesn’t play as fast.

What might be surprising about this economy of fretting movements is that costs are not massively altered by stretches, position shifts, string changes or even string skips. This allowed Shawn to develop an expansive vocabulary of patterns based on repetitions of these basic cycles, which he could then transition between easily.

Let’s consider some specific examples.

First, classic descending fours.

-9-7-6---7-6-----6-------|-------------------------|
-------9-----9-7---9-7-6-|-9-7-6---7-6-----6-------|
-------------------------|-------9-----9-7---9-7-6-|
-------------------------|-------------------------|
-------------------------|-------------------------|
-------------------------|-------------------------|   

The fretting sequence here (recalling that the 3rd and 4th fingers are equal) is

(4 2 1)(4 2 1)(4 2 1)(4 2 1)(4 2 1)(4 2 1)(4 2 1)(4 2 1) = (4 2 1)^8

That is, each instance of the fours pattern consists of 4 instances of the back cycle.

Consider the signature lick where Shawn alternates between groups of four and groups of five

-9-7-6---7-6-------|-------------------|
-------9-----9-7-6-|-9-7-6---7-6-------|
-------------------|-------9-----9-7-6-|
-------------------|-------------------|
-------------------|-------------------|
-------------------|-------------------|

The fretting sequence here is

(4 2 1)(4 2 1)(4 2 1)(4 2 1)(4 2 1)(4 2 1) = (4 2 1)^6

That is, each combination of four and five consists of three instances the back cycle.

For sixes, we’ll first consider the descending version

-9-7-6-------|-------------|
-------9-7-6-|-9-7-6-------|
-------------|-------9-7-6-|
-------------|-------------|
-------------|-------------|
-------------|-------------|  

The fretting sequence for each group of six is two instances of the back cycle. Ascending, we have

-------------|-------------|
-------------|-------------|
-------------|-------------|
-------------|-------6-7-9-|
-------6-7-9-|-6-7-9-------|
-6-7-9-------|-------------|

The fretting sequence is then

(1 2 4)(1 2 4)(1 2 4)(1 2 4) = (1 2 4)^4

That is, each group of ascending six is two instances of the forward cycle.

Let’s now how Shawn takes these cycles and expands upon them to attain his vocabulary. As a first example, Shawn took some descending patterns and made them ascend.
Ascending with the descending fours pattern (back cycles)

-------------------------|-------------------------|
-------------------------|-------------------------|
-------------------------|-------------------------|
-------------------------|-9-7-6---7-6-----6-------|
-9-7-6---7-6-----6-------|-------9-----9-7---9-7-6-|
-------9-----9-7---9-7-6-|-------------------------|

Ascending with the descending sixes pattern (back cycles)

-------------|-------------|
-------------|-------------|
-------------|-------------|
-------------|-9-7-6-------|
-9-7-6-------|-------9-7-6-|
-------9-7-6-|-------------|

Descending with the ascending sixes pattern (forward cycles)

-------6-7-9-|-------------|
-6-7-9-------|-------6-7-9-|
-------------|-6-7-9-------|
-------------|-------------|
-------------|-------------|
-------------|-------------|

Where this get’s really interesting however, is where Shawn changes incorporates more string changes within the cycles

-7-9-6-7---6-|-------------|
---------9---|-7-9-6-7---6-|
-------------|---------9---|
-------------|-------------|
-------------|-------------|
-------------|-------------|  

This is the forward cycle, starting with the 2nd finger. Similarly, we have the following ascending pattern

-------------|-------------|
-------------|-------------|
-------------|-------------|
-------------|---------6---|
---------6---|-7-6-9-7---9-|
-7-6-9-7---9-|-------------|

This is the back cycle, starting with the second finger.

Next, we have Shawn’s descending triads.

-6-----------------
---7---7-----------
-----9---8---6-----
-----------9---8---
-----------------9-
-------------------

The fretting sequence here is

1 2 4 1 2 3 1 2 3 

This is three instances of the forward cycle. The change from using the 4th finger in the first instance to the 3rd in the later instances is not difficult.

Let’s expand further. Let’s change string on every note within the pattern:

-9---6---7---|-------------|
---7---9---6-|-9---6---7---|
-------------|---7---9---6-|
-------------|-------------|
-------------|-------------|
-------------|-------------| 

which is the back cycle, and

-------------|-------------|
-------------|-------------|
-------------|-------------|
-------------|---7---6---9-|
---7---6---9-|-6---9---7---|
-6---9---7---|-------------|

which is the forward cycle.

There is one other highly efficient fretting movement, which is to lift a higher finger off to a lower finger which is already fretted (except 4th to 3rd). That is the sequences

1 2 1, 1 3 1, 1 4 1, 2 3 2, 2 4 2

where in each case, the lower finger is held throughout the duration that the higher finger is fretting.

We notice that the turnaround

-------6-7-9-7-6-------|
-6-7-9-----------9-7-6-|
-----------------------|
-----------------------|
-----------------------|
-----------------------|     

from the forward to the back cycle is efficient, provided that the second finger holds the fretted note throughout the movement of the 4th finger.

The opposite turnaround

-----------------------|
-----------------------|
-----------------------|
-----------------------|
-9-7-6-----------6-7-9-|
-------9-7-6-7-9-------| 

from back to forward is is less efficient, but can be performed at high speeds provided we fret the same note with the 2nd finger before and after the note fretted by the 1st finger.

The efficiency of the lift to a lower note that is already fretted explains why the classic 6 note pattern is efficient.

-9-6-7-9-7-6-

The fretting sequence is

4 1 2 4 2 1  

The note fretted by 2nd finger is maintained thoughout the duration of the following note fretted by the 4th, so the lift back to the 2nd finger is efficient.

Finally, I’ve decided to refer to the cycles as Efficient Digital Cycles rather than Efficient Fretting Cycles for a simple reason; I’ve noticed that they also appear in finger picking and hybrid picking patterns.

I hope this was interesting to read.

15 Likes

Amazing work Tom, thank you :slight_smile: :+1: :metal:

Great, great post @Tom_Gilroy. Deserving of multiple close readings!

Very interesting! Your post is thought-provoking, to be sure.

@Tom_Gilroy, I just re-watched some snippets of Shawn’s “Power Licks” video and now I can’t unsee the things you showed us - there was no 3-4 combination as far as I could tell.

I also thought about Gilbert, who seems to go a bit against the above rules as he is kind of famous for doing the whole-tone stretches as 1-3-4. But then he has a pinky that is almost as long as the other fingers, which should partially solve one of the issues connected with the 3-4 combo.

To further support your theory, I always had the impression that strong left-hand players like Tom Quayle tend to avoid 1-3-4 in their legato sequences, typically going for 1-2-3

And of course I’m looking forward to this:

What a post, I feel like I ought to drop ten quid in your tip jar.

2 Likes

Good stuff here mate! Cracked the code for the left hand :smile:

Are these Shawn Lane patterns using Alternate picking 2WPS?

The discovery of the concepts I’ve outlined above has been a major revelation for me, and it has answered some questions I’ve had for some time now.

I can play some licks/patterns from Power Licks & Power Solos at full speed (or close to it), but the ability to play those sequences at full speed did not result in the ability to play the other licks/patterns in my vocabulary at those speeds. I’ve always wondered why.

Now, I feel like I have access to the principles which will allow me to create new licks and patterns which can be played at those speeds.

It’s not there, at least not in his fast playing. I can’t unsee the structure now that I’ve discovered it either. It’s almost unbelievable, but it’s true. Shawn’s vocabulary of fast licks is almost entirely based upon these two fundamental digital cycles, and efficient turnarounds between them. Not so much Powers of Ten as Cycles of Three.

True, Paul’s pinky is unusually long and it might result in greater facility with transitions between the 3rd and 4th finger.

Also, while Paul is certainly a fast player, he’s simply not as fast as Shawn was. I can use isolated transitions between the 3rd and 4th fretting fingers in licks at Paul Gilbert type speeds reasonably comfortably, but attempting to accelerate them further quickly fatigues the extensor muscles in my forearm. Trying to quickly trill between the 3rd and 4th fretting fingers continuously causes me to fatigue in seconds.

Transitions between the 3rd and 4th fretting fingers appear in Allan Holdsworth’s and Brett Garsed’s playing. However, these transitions are again typically isolated. Also, though both fast players, Shawn was again clearly faster.

I’m not saying that transitions between the 3rd and 4th finger should not be utilized. There are situations where it would be unfeasible to use any other combination of fingers. I’m saying that those transitions are unavoidably expensive in the economy of fretting movements.

That’s not necessary, but if you’d like I can share my PayPal email address with you.

I’ll go into some detail about this later, possibly in another thread.

2 Likes

This is a really interesting post but I feel this part probably needs looking at. Unless I’m missing something. The third and fourth finger are, on the whole, independent from each other; it’s the second and third combination that aren’t. Well the second finger is independent but the third is dependent on the second.

You can test this for yourself by placing all your fingertips on a surface at the same time and checking the range of motion available to each finger by lifting them individually whilst keeping the other fingertips firmly planted. The only way you’ll get full movement from the third is by lifting the second with it.

I can see that not having fly away fingers due to the interconnectedness of 2-3 can actually help speed. Anyway, extremely interesting post.

Hi @Squeaks,

To fact check myself, I read up on this a little more. It turns out, the 3rd finger is not truly independent of either the 2nd or the 4th fingers.

Look at the following diagram of the structure of the hand:

We can clearly see that the tendon of the 3rd finger is connected to the tendons of both the 2nd and 4th fingers. However, there’s also more than just the tendons. The nervous structure is also interesting. The 3rd finger is connected to branches of both the ulnar nerve and the median nerve. The ulnar nerve branches to the 3rd and 4th fingers, and the median nerve branches to the 3rd, 2nd and 1st finger, as well as the thumb.

The result, seemingly, is that the movement of the 3rd finger is impeded to a greater or lesser degree by each of the 2nd and 4th fingers depending on their relative positions.

I start from this hand position, and try to lift each finger individually while leaving the others down:

I find that I can lift every the 1st, 2nd and 4th fingers quite high off the table, but my 3rd finger is difficult to lift and can only be lifted a small amount.

If I try to lift the 3rd finger in conjunction with either the 2nd or 4th finger, the 3rd can be lifted much further

Notice that I can lift my 3rd further when I lift with the 4th. Maybe this is different for you.

We can also try the following hand position, where the middle finger is tucked under:

Here I now find that I can lift my 1st and 4th fingers, but I can just barely move my 3rd finger. If I change the position so that my 3rd finger is tucked, I can’t move either the 2nd or 4th fingers significantly. If I tuck the 4th finger, I can move the 2nd finger but not the 3rd.

Obviously, the first posture demonstrated with all finger tips on the table is much more like a guitar playing form than the latter postures with the fingers tucked.

I think the issue with combinations of 3rd and 4th is that while playing, we must flex the 3rd while extending the 4th and vice-versa, which fights against the connected tendons and fatigues our extensors quickly. With the 2nd and 3rd, the problem is avoided because we can allow the 4th to move freely following the 3rd, maintaining the necessary slack between the 2nd and 3rd fingers throughout.

4 Likes

Great thread…sometimes I have a tendency toward thinking something “has” to be done a certain way regarding fingerings and it can be limiting. CTC has shown there’s many ways to solve the picking issue and it’s true for the fret hand as well. Have to keep an open mind.

Schenker and Gary Moore are two guys that don’t use the pinky much. Watching Schenker live most of the time he uses it just on descending scale type runs.

1 Like

Very intersting thread, great post @Tom_Gilroy. Many thanks. However, I do have one question. I seem to find playing (fingers) 1-3-4-3-1 easier (and a tad faster) than 1-2-4-2-1. Okay, I’m not talking Shawn Lane crazy speeds, but speedy. I seem to hit fatigue point quicker when repeating the 1-2-4-2-1.
Is it because my 1-3-4-3-1 finger version is unusually adept? (Relatively speaking - again, I’m no Shawn Lane) Or is the 1-2-4-2-1 lacking? (and could be worked on to significantly improve)

Just so we’re clear, Shawn Lane wasn’t playing repeated (1 2 4 2) or (1 3 4 3) cycles at Shawn Lane crazy speeds either. It’s almost all (1 2 3), (3 2 1), (1 2 4) and (4 2 1).

As for why you’re finding (1 3 4 3) cycles easier and faster than (1 2 4 2) cycles, it could be a few things. First, are you lifting your lower fingers and fretting new notes with them as you repeat the cycle, or are you continuing to fret the lower notes with lower fingers and simply lifting the higher fingers off to those lower notes? For example, are you repeating something like this:

-5-7-8-7-|
---------|
---------|
---------|
---------|
---------|

or something like this

-5-7-8-----7-----|
-------7-5---8-7-|
-----------------|
-----------------|
-----------------|
-----------------|

?

Are you using hammers and pull-offs or picking? If legato on one string, the chances are that your 2 to 4 or 4 to 2 transition is underdeveloped.

If you can play something like the latter pattern at Paul Gilbert type speeds but can’t do the equivalent (1 2 4 2) cycle easily, my guess is that there is something about your posture that is optimized for (1 3 4 3) and that you’re trying to execute (1 2 4 2) from that posture, rather that from it’s own optimal posture.

How about 2 to 4 and 3 to 4 trills?

Doing this with legato. Under closer inspection I am indeed lifting the first finger with the 1-2-4 pattern sometimes in a little ‘hop’ (or at a minimum slightly removing some of the pressure). I don’t seem to be with the 1-3-4. Very interesting

I reckon this is likely the case. Trilling is harder with 3&4, as expected.

I was thinking: The speed champions are the classical violin people. Do they avoid 3/4?

In general they would likely be the best to copy?

My girlfriend is a trained classical violinist and pianist. I discussed this with her, and she has told me that the 4th finger is rarely used at all in classical violin playing. According to her, when playing scales over multiple octaves, the fingering pattern is almost always

(1 2 3) shift (1 2 3) shift (1 2 3) shift etc

or the reverse.

I have no knowledge of violin technique beyond what she has told me. However, I’ve listened to a lot of classical violin, and I’m not really sure that fast classical violin playing is any faster than the fastest guitar playing. Actually, I’d imagine it’s markedly slower.

I love your research work here, it is amazing. I think the violin people are pretty sick, here is an example of a faster piece:

(Sheet music, https://musescore.com/musemeister/paganini-cap5)

As an ex-violinist, I wouldn’t say rarely used, but I would definitely agree that it is definitely prefered to shift and use an alternative finger for the purposes of better vibrato control. But then again the ability to utilise the shifting into different positions isn’t something that a beginner would necessarily tackle - going higher on the fingerboard of a violin requires such a high level of accuracy (due to the absence of frets). So in essence the digit use could change quite a bit based on skill level as there is more options open for finger placement and tonal outcome.

Thanks.

Classical violin players are unbelievably skilled, I just don’t think they’re faster than the fastest guitarists. This performance of Paganini’s 5th is among the most terrifying things I’ve ever seen:

Again, I can’t really comment, I don’t have the necessary knowledge of violin technique, I can only repeat something that was shared with me. That’s interesting however, thank you.

Hey Tom, one thing you have not mentioned above is the rolling of a finger to play different strings on the same fret.

However I don’t think Shawn did it at all, and my guess is that this would be among the inefficient movements. But maybe not if one can do it properly.

What do you think?
Thanks again