TC Electronics Polytune

Buco · June 3

It's a good tuner in any case but polyphonic tuning is what wins it for me.

B25Gib · June 3

Also: The Polytune has a strong metal & spring clamping design that is very durable and doesn't fatigue and break after a couple years like the composite plastic of the SNARK design that two of mine failed!

rudolfochrist · June 3

I also use the (same) Polytune for the last eight years or so. Over that time it lost its ability for polyphonic tuning. I guess I dropped it way to much. Still a good tuner 😅.

Buco · June 3

I actually didn't realize that currently they are the only polyphonic tuner out there. So they literally changed the game. It got voted as best overall by Guitar World.

But I think people are missing the point. Advantage of polyphonic tuning isn't the convenience. It's the compromise it's making to get the instrument to intonate better across the fretboard. At least I think that's what it's doing. It's either that or it's inaccurate lol

https://www.guitarworld.com/features/the-best-guitar-tuners#section-best-overall

paulmcevoy75 · June 3

It shouldn't change the intonation different from a regular tuner. There are "sweetened" tunings, the Peterson tuners can be set for them, but I wouldn't probably spend a lot of time thinking about them.

Buco if you find that your strings are out of tune up the neck, your bridge might be a little out of place. If you Google "how to intonate a guitar" it will show you which direction to move the bridge. Sometimes it's just a millimeter or so.

Guitars are always a bit out of tune, somewhere. Mostly, anyway. You can spend a lot of time tracking down little intonation errors but in my experience the strings are rarely perfectly in tune after a few minutes so unfortunately you have to sort of deal with the same intonation issues Django and Wes and everyone else dealt with :)

I was putting compensated nuts on my guitars for a while but ultimately I ain't got time for that. Selmer style bridges could be better updated for consistent intonation but I haven't gotten there yet. The zero fret is a good idea intonation wise. Not perfect but better than a regular nut.

My complaints above about D'Addario tuners aren't about their intonation but that they burn through batteries, don't work all that well and they clips break. The TC one is pretty solidly made.

wim · June 3

It's the compromise it's making to get the instrument to intonate better across the fretboard.

Can you offer any evidence to back this up?

As I understand it, there is exactly one pitch that each string should be tuned to (e.g. A string at 440 Hz fundamental) assuming we’re going for equal temperament. I don't see how polyphonic or single string tuning would change the ideal target result.

Buco · June 3

https://www.djangobooks.com/forum/discussion/comment/121713#Comment_121713

That's unitune, it doesn't have polyphonic tuning feature.

Buco · June 3

@paulmcevoy75 I know how to check the intonation and it was a little off so I tweaked the bridge. But that's not what I was going after. And to reply to @wim at the same time, I don't have any proof. But the thing led me to think it employs some of sort of algorithm to get you more in the just tuning territory.

When we tune each string to a tuner, we do in equal temperament. But then when you do that and play root and a third, say D and F#, it won't be in tune. F# will be sharp. So you flatten a D string a touch if you want them to be closer to perfect harmony. And so on.

So I thought maybe this tuner does that for you. When you strum all strings at once, it figures out and it gives you that compromise right away. Maybe foolishly. I wanted to email their customer support and ask that question but the link to customer support directs me to the forum. Maybe I'll try anyway but some other time.

Before this tuner, what I would do is I would tune the open strings. Then I'd check octave vs perfect 5th, tweak. Check root plus third, tweak. Check octave between D and B string, usually 7 and 10th fret, tweak. And so on. With this tuner it seemed like I didn't have to do any of that stuff. Just strum all open strings, tune what needs to be tuned and it's done. Whatever I play then it sounds good. Where before I would cringe at some of the above mentioned spots right after open strings are tuned, before I'd go through my tweaking process. I had a same experience with the Martin I have.

flaco · June 4

Maybe a little off topic. I have huge ethical concerns with AI that are beyond the scope of this forum, but this would be an interesting use case for AI. Someone smarter than me could develop a program with AI, or a tuner with AI, that would analyze the intonation, nut height, placement of every fret, etc. of a specific guitar and then suggest the perfect compensation so it plays as in tune as possible across the neck.

flaco · June 4

@adrian - possible Soundslice feature?

"You're absolutely right—achieving perfect intonation across the entire fretboard of a guitar is fundamentally impossible due to the compromises inherent in equal temperament tuning, string tension physics, and fret placement. However, your idea of developing a program to analyze a specific guitar and suggest an optimized tuning compromise is both fascinating and feasible with the right tools.

Here's what it would take to build such a system:

1. Data Collection

You'd need to record the pitch of:

All open strings
All fretted notes (typically frets 1–12 or 1–24)

This could be done using:

A microphone and pitch detection algorithm (like YIN, autocorrelation, or FFT-based methods)
A mobile app or desktop software interface to guide the user through playing each note

2. Pitch Analysis

For each note:

Measure the actual frequency
Compare it to the ideal equal-tempered frequency
Calculate the cents deviation (how sharp or flat it is)

3. Optimization Algorithm

This is the core of your idea. The program would:

Use the deviation data to model the intonation curve of each string
Apply an optimization algorithm (e.g., least squares, genetic algorithms, or gradient descent) to find the best tuning offsets for each open string
The goal: minimize the average or maximum deviation across all fretted notes

4. Output

The program would suggest:

A custom tuning for each string (e.g., A = 439.3 Hz instead of 440 Hz)
Possibly even alternate tunings for specific keys or chord voicings

5. Optional Enhancements

Machine learning to improve tuning suggestions based on user feedback
Integration with digital tuners or DAWs
Visualization tools to show intonation maps across the fretboard

Would you like me to sketch out a prototype of how this program might work in Python, or would you prefer a conceptual diagram of the system architecture first?

You said:

Yes, please sketch a Python prototype.

Copilot said:

Here's a prototype that simulates pitch deviations across the fretboard and uses least squares optimization to suggest the best open string tuning offsets for minimizing overall intonation error.

🎯 Optimized Tuning Offsets (in cents)

These are the suggested adjustments to the standard tuning for each string:

String 1 (Low E):   -0.61 cents
String 2 (A):        +1.19 cents
String 3 (D):        -1.92 cents
String 4 (G):        +2.69 cents
String 5 (B):        +1.69 cents
String 6 (High E):   -0.73 cents

These small adjustments help reduce the average deviation across all frets.