Problème en mode Poly

The low E-string "problem" was actually something that puzzled us for a long time during development. We did a lot of automated tests based on sound files where the tuning on both the polyphonic- and chromatic tuner was spot on, but got feedback from a lot of beta testers saying there was a difference between the polyphonic tuner and chromatic tuner.

What we found -- which was since verified by several users -- was that it was the polyphonic tuner that was "correct" and the chromatic tuner was of by 3-10 cents. This puzzled us even more as we've been doing accurate chromatic tuners in a lot of our multi-fx processors for a long time without any problems. Another thing that puzzled us was the offset wasn't the same everytime, but tended to vary between perfect and up to 8-10 cents off.

So after a lot of testing and hair scratching we found out that the problem is not related to the tuner algorithm, but is actually user-related:

From observing our betatesters we learned that when tuning polyphonically you don't know which strings will be out of tune, which means that you don't rest your hand on the tuning pegs or neck. But when you tune monophonically you know exactly which string you'll be tuning, so you automatically rest your hand on the corresponding tuning peg.

What we found was the pressure from the hand on the peg is enough to take the string off by up to 5-6 cents depending on how heavy handed you are... we actually saw a few testers take a single string out by more than 10 cents!

So what you're seeing is not a difference between the two tuners in PolyTune, but a difference between resting your hand on the neck/tuning peg or not and is a very real offset of pitch caused by the sheer pressure of your hand. The reason you never saw it before is because, no other tuner has the ability to "double check" the tuning.

Pour les pressés ou les non anglophones: ils se sont arrachés les cheveux sur ce problème, puis ils sont arrivés à la conclusion que lors de ce pb c'était le mode polyphonique qui était juste, et le mi du monophonique légèrement faux. Ils auraient alors conclu que, lors d'un accordage polyphonique le guitariste ne touchait à aucune clef, puisqu'il ne sait pas laquelle il va devoir actionner, alors que lors d'un accordage monophonique il tenait la clef pendant la mesure, et que ce seul fait suffisait à désaccorder légèrement la corde.

Il ne vous reste plus qu'a faire l'expérience sans toucher à la clef pendant la mesure!