The Hermode Tuning system

[seraph]

http://eceserv0.ece.wisc.edu/~sethares/paperspdf/hermode.pdf

The Hermode Tuning system is a brilliant way to dynamically alter the tuning of electronic instruments (toward Just Intonation) while retaining compatibility with 12 tone equal temperament.
This article is written by its developer, German musician Werner Mohrlok (with his computer-programmer son, Herwig) and the system is called Hermode Tuning, or HMT ( http://www.hermode.com/).

I enjoy the hermode tuning available with Logic Pro

[Kassen]

Interesting. He talks about trying to avoid audible stepping but I don't see why he couldn't *slide* notes to new values, especially as the in-harmonic elements that the new chord may have in it's attack phase might mask most of that for that period. Looking at the "developer" section (which sadly doesn't give enough info to actually develop a implementation) it seems like he's sending all his retuning info in one swoop and to line up with note-on messages. So far it looks like neither the character of the sound itself nor the speed of the piece is taken into account.

Is there a page with more in-depth info?

[seraph]

[Kassen]

Ah, but that's good! You can't file a patent without telling the patent office how it works and these days you can search many of those databases.

I don't want to copy it, I'd just like to know what he's doing, I find it quite interesting to try to make notes "aware" of the ones around them and get affected by that. Makes everything a bit more lively.

[Kassen]

Hmmmm, Niether the US patent office nor google-patents can find it :¬(

[blue hell]

The names sound German to me, so maybe it's a German patent, or EU?

[seraph]

[xjscott]

I just noticed this thread so I haven't had time to look at the documents yet, but I thought I'd summarize what my understanding is of how it works from what I remember reading about it on his site a few years ago.

You start by choosing the tuning target system you prefer. For example, you can have the 7ths in a V7 be based either on 7/4 or 15/8. There are only I think three of these you can choose in his setup?

Now, consider that 12tET is your base tuning grid. Imagine each note played then is attached to this 12 grid with a spring. When you play a V7 then, the tuning chosen is not quite just, but it is not 12 either. However, the center of gravity or inertial point or such of the system is that 12 grid, as maintained by those springs. As a result of this, you never have any comma drift happen because you are locked to the underlying 12 grid, but your chords sound almost exactly the same as in pure low limit just intonation. To my ears it sounds a little bit better that pure JI especially with wavetable synths that are purely harmonic since it avoids the extreme phase locked sound that you can get with pure JI on many synth timbres. Low limit JI to me sounds best on physical instruments and ends up sounding like a static reedy pipe organ with purely harmonic synthesized waveforms. Pure low limit JI is not a bad effect to have available as a tool, but it gets aurally fatiguing. By Hermode's never being quite being perfectly just, you get the characteristic reedyness of JI, but you avoid the phase locking and so it's less tiring to listen to.

I used to be opposed to all these autotuning algorithms since they take control of pitches away from the composer and end up sounding worse than nothing at all, but I have to admit that Hermode sounds probably as good as possible for people who want a completely automatic solution to tuning their block chords to JI dynamically as they play, without running into the comma drift problem. I think it's definitely worthy of its patent since lots of people tried to solve this problem and this solution is the least troublesome and best sounding, for people who are looking to do this sort of thing.

[Kassen]

Interesting, thanks!

[williamsharkey]

This is pretty great. I always thought that computers should do dynamic "justifying" of intervals.

I was thinking about the fundamental problem of tuning, and how it is really a game of compromise. It makes the world a more interesting place I think.


I was trying to think of a physical example of how one might solve tuning contradictions with concrete examples. So I conjured up this idea:

Consider an orchestra of 2 instruments sounding diifferent notes that are not just tuned. Well, if you are a mobile person, then if you ran in a particular direction, at a particular velocity, you could fix the tunning!***. If there are unjust 3 notes sounding, then you may need a flying object because you will need three dimensions to justify the tones. Well, once you add more and more unjust notes, you just require more dimensions of movement. So I wonder if the limits of perfect tuning are somehow limited to the degrees of movement one can exercise.

*** If the above did not make sense, think of the sound of an ambulance that passes you. Well, what if you were the ambulance. You have control over the tuning of objects by moving past them.

My idea about moving around in space to retune instruments may be half-baked, so feel free to rip it to shreds

The thing is, when you add perceptual factors into it, you have alot more leeway to solve the problem. Like how more drift can occur between notes than during notes.

I think it's cool that logic supports it, and it's sweet that they used an amiga!

[blue hell]

[seraph]

[klangumsetzer]

my virus kb va synth makes use of hmt which was added with the last updates for the b series.


best regards

eike
---