Kenny Hill's new performance intonation "wiggly" fretting

Construction and repair of Classical Guitar and related instruments
amezcua
Posts: 545
Joined: Tue Oct 04, 2016 9:50 pm

Re: Kenny Hill's new performance intonation "wiggly" fretting

Post by amezcua » Mon Feb 19, 2018 6:44 pm

Thanks for adding that information Rasputin. It seems to be messy the way they have changed the name . If it`s Lehman`s version of what Bach intended they should stay with the same name . I thought about the squiggles on the page by Bach and imagined him thinking "A nod`s as good as a wink to a blind horse ". Maybe Bach had a sense of humour like that .
The key colours was mentioned in lessons at my school but the teacher sitting at a grand piano would probably have been using ET so it all seems a bit daft when the same anomaly is taught in music colleges to this day . None of that key colours effect applies to ET .
But aside from that I would say hearing a good example is the best way to decide if you like a temperament so I can recommend the Lehman version of the first Bach Prelude . The youtube video compares it to ET. In the ET part the sound of notes resonating just after they are played hangs like a cloud in the air. The ET (cloud ) sounds harsh and bitter. The Lehman sounds very beautiful . To me it`s a clear cut distinction . See if you agree with that example .
It`s a pity the Korg does not have that in the programme . Now I have to work it out the hard way .

amezcua
Posts: 545
Joined: Tue Oct 04, 2016 9:50 pm

Re: Kenny Hill's new performance intonation "wiggly" fretting

Post by amezcua » Mon Feb 19, 2018 9:00 pm

I checked out the video of the Turkish gentleman with his adjustable fret guitar .He plays Kirnberger III , ET, and Werkmeister to see which you prefer. But when it shows the fretboard for Kirnberger his frets are going different ways to mine and I`ve checked mine carefully after fitting . Maybe there`s a Kirnberger lookalike .
The video at the ET section shows one wiggly fret very out of line with the others .
Last edited by amezcua on Tue Feb 20, 2018 9:03 am, edited 1 time in total.

Rasputin
Posts: 959
Joined: Fri May 12, 2017 12:25 pm

Re: Kenny Hill's new performance intonation "wiggly" fretting

Post by Rasputin » Mon Feb 19, 2018 9:33 pm

I have no idea what video you are on about.

There are a million comparison videos on YT but it's 99.9% piano. I agree that other systems sound a lot better if you stick to the same key.

Werkmeister / Kirnberger III etc were designed for keyboard instruments so they needed to cater for a wider range of keys than we commonly encounter in guitar music. If we were starting again and focusing on classical guitar, we would be able to do better than any of the existing piano temperaments.

It is not difficult to envisage a system where the fretboard (or even just the frets, if you linked them down the side) clips in and out so that you can have two or three configurations for the same neck. This would also enable me to invent Rasputin I and then, if I didn't like it, just keep the ET board on while I developed Rasputin II. All TT, of course. I am half-serious about commissioning such a beast when the time comes for a new guitar, although that will not be for a while.

What range does a guitar temperament really need to cover?

Gb Db Ab Eb Bb F C G D A E B F# C# G# D# A#
*****************

Just the bit with the asterisks? You get the relative natural minors for free but have to think about those raised sixths and sevenths.

Edit: the asterisks moved when I pressed submit - they are supposed to start under Bb and finish under E

amezcua
Posts: 545
Joined: Tue Oct 04, 2016 9:50 pm

Re: Kenny Hill's new performance intonation "wiggly" fretting

Post by amezcua » Tue Feb 20, 2018 9:08 am

Temperaments were designed to work with all or most keys . The bonus is that they create different emotional reactions . Bach wrote Well Temperaments music in 24 different keys . Think why he did that . Not one universal key . 24 different keys .
Let me see if I can spell his name without looking .Tolgohan Cogolu. No. Tolgahan Cogulu.That`s it .
Microtonal Bach Experiment. Comparing 3 Temperaments .Kirnberger III .ET , and Werkmeister .

Rasputin
Posts: 959
Joined: Fri May 12, 2017 12:25 pm

Re: Kenny Hill's new performance intonation "wiggly" fretting

Post by Rasputin » Tue Feb 20, 2018 11:15 am

Yes, interesting video, but really it just demonstrates that temperament is a compromise, which we already knew. The virtue of ET is supposed to be that although it's not great, at least it is the same in all keys - but my point was that we don't actually use all they keys so it is probably not a sensible compromise for our particular instrument.

What I was saying is that even if you believe in key colours (I do) and even if you believe that the reason for key colours is that using a few notes which are consistently out of tune - but only slightly - is a good thing (I don't - this is where we differ) you still reach a point where the note is so far off that it really is out of tune, and not just a pleasing imperfection like the variations in the grain of a piece of wood or (well, I was thinking of Lucy Liu's eyes, but I don't want to be accused of objectifying women, so come up with your own example). In a keyboard temperament that point has to be a long way from the home key, because keyboard music can and therefore does modulate a lot. For reasons that we're all familiar with, guitar music isn't like that - so it should be possible to construct a temperament that does stray a little from just temperament in the home key and neighbouring keys (this theory of key colours is a bit like homeopathy) but not as much as the keyboard temperaments do. IOW, even on your view of key colours, you would still ideally want to use a temperament designed specifically for guitar, rather than borrowing one of the keyboard temperaments.

The more I think about it, the more the classical guitar seems an obvious candidate for its own temperament. It is only rarely played with other instruments and in all but the most skilled hands it tends to come off second best to the instrument it is paired with - so there isn't really a problem in terms of clashing with another tuning system. Also, it tends to stick to a fairly small range of keys, so it is possible in effect to put most of the wolf notes where they will never be played.

I also can't help thinking how useful it would be to have a partial capo that sat on the headstock rather than going around the neck and let you leave the A string fretted at C if you are in C or the low E string F if you are in F or Bb, etc. Imagine it was designed with a rocker so you could knock it down a fret with almost the same action as fretting a note - then you would have your leading tone in the bass when you wanted it.

Combining those things - and throwing in the TT system - would result in an instrument which was able to play more music in more keys, had a more pleasing sound in those keys than ET and - if you are right about how they work - featured key colours, i.e. made a virtue of the fact that it is impossible to have every note perfectly in tune.

amezcua
Posts: 545
Joined: Tue Oct 04, 2016 9:50 pm

Re: Kenny Hill's new performance intonation "wiggly" fretting

Post by amezcua » Tue Feb 20, 2018 7:57 pm

It`s coming down to a definition of "in Tune". As Prince Charles might say "Whatever that means ". (That`s a Princess DI reference ).

Rasputin
Posts: 959
Joined: Fri May 12, 2017 12:25 pm

Re: Kenny Hill's new performance intonation "wiggly" fretting

Post by Rasputin » Tue Feb 20, 2018 10:01 pm

Just for once in this area, I don't think it is - but never mind.

amezcua
Posts: 545
Joined: Tue Oct 04, 2016 9:50 pm

Re: Kenny Hill's new performance intonation "wiggly" fretting

Post by amezcua » Thu Feb 22, 2018 12:18 am

For "in tune " think of the many violinists that play everything sharper to stand out against an orchestra . Then listen to some Fritz Kreisler recordings and you hear very different frequencies compared to other fiddle players . It`s expressive . But listen to the same piece by another famous player and you hear the difference. An old film about Stephane Grapelli with some jazz guitarists amuses me . He starts the tune and it`s miles off the exact tuning but you know the music and it makes you smile. Relaxing jazz playing .

hillguitar
Posts: 148
Joined: Thu Aug 03, 2006 7:29 pm
Location: Ben Lomond, CA

Re: Kenny Hill's new performance intonation "wiggly" fretting

Post by hillguitar » Fri Feb 23, 2018 7:14 pm

My understanding is that the True Temperament Fingerboard uses common equal temperament as its basis, so I assume that’s what we use. Ever since I discovered Bach’s Well Tempered Clavier and the concept behind it, back when I was in high school, I have assumed there was just one system of “equal temperament”, whatever it is that contemporary piano tuners use on keyboards, whether by using an oscilloscope or by counting beats in 5ths, 4ths and 3rds. Kirnberger and Lehman are new names to me so I’ve had some catching up to do. I do find it very insightful to characterize the TT system as more of a compensation process than a temperament process. This is correct and well said. The variations in fret locations are really just trying to dial in the precise pitches to a more accurate rendition of "equal temperament".

Individual string compensation at the bridge and the saddle is certainly less invasive and less work on an existing instrument. Changing a fingerboard on an existing instrument is no fun. But building an instrument from scratch with the TT fingerboard is not too much extra work. Fretting is harder but do-able, and there is the added cost of purchasing the fingerboard and frets, also do-able. Back in the1990s I built a few instruments following to Greg Byers nut and saddle intonation system. I felt it was a good thing but didn’t adopt it permanently, ironically because normal guitar intonation didn’t much bother me, and I didn’t want the added complication. And now, 25 years later, this is definitely a complication, and I do catch the irony.

With individual string compensation of the nut it most directly affects the intonation of the notes at the first fret compared to the open string. Individual string compensation of the bridge can certainly make the 12th fret octaves perfect, all of which seems like a good thing. These refinements would affect intervals and harmonies all over the fingerboard to some varying extent, though maybe not as comprehensively. It would be worthwhile to do two similar guitars side by side to hear whats the same and what’s different. I might try that sometime later. I have no idea how math could either prescribe or describe the differences, that’s just not my expertise. It would be worthwhile to use an accurate tuner and read the beats-per-second of every note on the fingerboard and compare it with the specific frequencies they’re supposed to be. That is basic research that I still haven’t done. I gearing up to do that soon, now that I’m becoming convinced this line of effort has worthwhile results. I still really love and rely on the way my ears flirt with the moods and flavors of sound. It’s the sound results of the TT that pulls me in. A studio guy I met at the Namm show described it as Fiji water compared to LA tap water. Not a bad description. If I can come up with some arithmetic that helps explain it, all the better. I don’t expect any of this to yield perfection, but there seems to be some real nice satisfaction, at least for a while.

By the way, barre chords are easy, no difference at all. That may be counterintuitive, but it’s true. Left hand accuracy is not a problem. Resale? That’s another question. I’ve currently got two new orders. We’ll see.
Kenny Hill
Hill Guitar Company
Ben Lomond, CA 95005

SteveL123
Posts: 874
Joined: Thu Feb 09, 2017 5:05 pm

Re: Kenny Hill's new performance intonation "wiggly" fretting

Post by SteveL123 » Fri Feb 23, 2018 8:10 pm

Mr Kenny Hill,

How about build a test mule guitar with 2 different magnetically attached/ removable fret boards. One fret board is equal temperament, the other true temperament. One can switch back and forth within seconds to evaluate the 2 different temperaments on the same guitar,

amezcua
Posts: 545
Joined: Tue Oct 04, 2016 9:50 pm

Re: Kenny Hill's new performance intonation "wiggly" fretting

Post by amezcua » Sat Feb 24, 2018 12:31 am

Here is a fine solution for Rasputin I just happened to look up Joni Mitchell on Wikipedia and to me she was jjust one of my favourite singers. But she used over 50 different tunings on her guitar . It was simplified at that stage for her by using a Virtual Guitar which took the standard guitar tuning from the guitar she was playing ------and digitally converted the notes into whatever tuning she had devised . That`s a brain bender for me .What a clever lady . And born the same year as me too .
If you all look up the TT site (one of them at least ) where it states the temperament used on one bendy fret guitar , it tells you without any smoke or mirrors that it`s the Bradley Lehman /Bach 1722 Temperament . In black and white .
The Kirnberger Temperament comes from Johann Phillipp Kirnberger who studied under Bach .The man himself . Hard to get closer to him than that .

User avatar
Trevor Gore
Posts: 975
Joined: Sun Aug 07, 2011 12:57 pm

Re: Kenny Hill's new performance intonation "wiggly" fretting

Post by Trevor Gore » Sat Feb 24, 2018 5:04 am

Rasputin wrote:
Mon Feb 19, 2018 9:33 pm
What range does a guitar temperament really need to cover?

Gb Db Ab Eb Bb F C G D A E B F# C# G# D# A#
*****************

Just the bit with the asterisks? You get the relative natural minors for free but have to think about those raised sixths and sevenths.

Edit: the asterisks moved when I pressed submit - they are supposed to start under Bb and finish under E
Well, I thought I'd have a look at designing a guitar temperament that mathematically optimised the scales Bb thro' E to Just temperament (JT). i.e. how close you could get the intervals of each of those 7 tone scales to match JT (at the expense, of course, of the other keys which would be way out). It's interesting what happens when you let a computer optimise things without any preconceived bias. For the base key, for which I chose A, to pin it to A440, all the middle intervals (major third to major sixth) came out pretty close (within 2 cents) of JT as did the major 7th, except for the 5th and minor 6th which were 6 cents flat. The octaves were forced to pure. The big errors were on minor seconds (36 cents sharp, which sounds a lot, except that equal temperament minor seconds are nearly 30 cents sharp of JT and are not exactly a well worn interval anyway) the major 2nd and minor 3rd were about 12 cents flat of JT and the minor 7th around 20 cents flat. The keys will all sound different to each other, of course, because of the differently sized intervals, so there will be "key colours".

Just for laughs, I thought I'd see how close I could get to this temperament using straight frets and only nut and saddle compensation. The answer to that was nowhere near close enough! So this would definitely need some sort of wiggly frets implementation. The TT guys may well have done this already.

I think I'll find someone with a programmable temperament keyboard and listen to what the various keys sounds like before I go any further!

Another thing that doing exercises like this demonstrates is that accurate ET is a pretty good all round compromise. (It's a still shame about those 3rds, though).
Trevor Gore: Classical Guitar Design and Build

Rasputin
Posts: 959
Joined: Fri May 12, 2017 12:25 pm

Re: Kenny Hill's new performance intonation "wiggly" fretting

Post by Rasputin » Sat Feb 24, 2018 9:21 am

All very interesting stuff. That test mule guitar would be good, although if you are interested just in the tuning aspect you can achieve the same thing with a midi guitar and some software, or (this is what I had in mind to do) by finding an existing MIDI perfomance of a guitar piece and changing the tuning in the software. I have not found any MIDI files though, and if I create one electronically and let the computer play it, I'm not sure that is a brilliant test. I found a piano competition site where contestants upload their performances as MIDI files (so presumably we are talking about real performances and pretty good ones as well) that might be the best source of files.

I suppose it's possible that the best compromise for a seven (major) key instrument is actually no different from the best compromise for a twelve key instrument, which (according to taste) could be one of the Kirnberger temperaments.
Trevor Gore wrote:
Sat Feb 24, 2018 5:04 am
Well, I thought I'd have a look at designing a guitar temperament that mathematically optimised the scales Bb thro' E to Just temperament (JT). i.e. how close you could get the intervals of each of those 7 tone scales to match JT (at the expense, of course, of the other keys which would be way out). It's interesting what happens when you let a computer optimise things without any preconceived bias. For the base key, for which I chose A, to pin it to A440, all the middle intervals (major third to major sixth) came out pretty close (within 2 cents) of JT as did the major 7th, except for the 5th and minor 6th which were 6 cents flat. The octaves were forced to pure. The big errors were on minor seconds (36 cents sharp, which sounds a lot, except that equal temperament minor seconds are nearly 30 cents sharp of JT and are not exactly a well worn interval anyway) the major 2nd and minor 3rd were about 12 cents flat of JT and the minor 7th around 20 cents flat.
Those are some interesting results Trevor. When you say you picked A as the base key, do you mean that getting A major accurate was prioritised, or that, since A itself is bang on the conventional frequency, that is the one you are giving us the info for?

I am intending to give G and D equal priority, because they are in the middle of the range, and let the inaccuracies grow from there. This approach may imply (even) more use of the keys F and Bb - and maybe C - than we are used to (think new possibilities for arranging old music). This in turn underlines the need for a device that keeps the low F / Bb fretted but does not block the LH at any point. I was thinking of one that sat on the headstock and levered down. Typically when you invent something you discover that you are not the first, but I can't find anything like this out there.

Rasputin
Posts: 959
Joined: Fri May 12, 2017 12:25 pm

Re: Kenny Hill's new performance intonation "wiggly" fretting

Post by Rasputin » Sat Feb 24, 2018 9:44 am

Rasputin wrote:
Sat Feb 24, 2018 9:21 am
I found a piano competition site where contestants upload their performances as MIDI files (so presumably we are talking about real performances and pretty good ones as well) that might be the best source of files.
It turns out that the entrants are of a high standard but there are quantisation problems with the files that spoil them. Maybe there is a way to switch quantisation off altogether.

User avatar
Trevor Gore
Posts: 975
Joined: Sun Aug 07, 2011 12:57 pm

Re: Kenny Hill's new performance intonation "wiggly" fretting

Post by Trevor Gore » Sun Feb 25, 2018 12:42 am

Rasputin wrote:
Sat Feb 24, 2018 9:21 am
When you say you picked A as the base key, do you mean that getting A major accurate was prioritised...
No. All selected keys had equal priority.
Rasputin wrote:
Sat Feb 24, 2018 9:21 am
...or that, since A itself is bang on the conventional frequency, that is the one you are giving us the info for?
Yes. The only constraints I put on were that A=440 Hz (have to pin it somewhere) and that all notes with the same name are pure unisons or pure octaves. None of the keys was bang on pure JT and all of the keys varied from JT in different ways, but were as close to JT as that selection of keys would allow. Whether or not they "work" is a different matter. The proof of that pudding is in the listening.

Not a classical guitar story, but worth a mention, I think. I was reading a piece about the Rolling Bones. I thought most of their stuff was in G (which Mick likes to sing in and Keef open tunes his guitar to). This piece had a couple of set lists which included the keys of songs for those sets. The same key was used for consecutive songs only once in each set (of 19 and 20 songs respectively). The keys used (both sets) were: A, B, C, C#, D E, F, G, Am, C#m, Em. So it may only be rock'n roll, but still a pretty comprehensive coverage of the fretboard and another argument for (accurate) ET.
Trevor Gore: Classical Guitar Design and Build

Return to “Luthiers”