Thoughts on tuning curve weighting ⚖️

Henry

Hello all,

I hope the pianoscopers are well.

I have a general question which tuners might weigh in on:

Any opinions on curve weighting? Between precise and idealised.

Any qualitative descriptions on how one end of the spectrum sounds to your ears compared to the other?

Do you ever change the setting depending on type/age of piano?

Thanks! 🙏

[Edited post to be more general]

Frank Illenberger

The weighting parameter is essentially an artifact of pianoscope’s inharmonicity model. I originally introduced it to make the app more robust against measurement errors. In the meantime, the algorithms have become very precise, so that when working with a single instrument, a value of 100% could actually be a good future default (it is currently 85%). I’m still a bit reluctant to change it, though, because I’ve had good results with the current default and want to gain more experience with the new setting first.
There are situations where a lower value could still make sense. For example, if you want a typical tuning curve for a family of instruments and don’t want to bother measuring the inharmonicity of each individual instrument, then a more abstracted tuning curve that smooths out the individual deviations of a single measurement could be beneficial.

MilanDka

Frank Illenberger Yes, that makes sense. Typically in the factory, where I have the individual models already recorded, I don’t have to record each one again. For example, I have the Hoffmann P162 recorded only once, and I use it to tune all instruments of that model. I only remeasure the inharmonicity from time to time — due to climatic conditions, seasonal changes, variations in string dimensions, and so on.

Henry

Thanks so much, that’s really helpful. I’ll keep that in mind for generic tuning files.