I just got that app and sampled it with the two bikes that are here at our vacation house. Looks pretty cool and simple to use. It was able to detect slight differences in tension. It has suggestions to get more accurate results. And over versus under shouldn't matter, because it says to use the length (and diameter) or the spoke only from bottom of nipple to the first crossing, which makes sense from an engineering perspective.
I'm not planning to use it to build wheels from scratch (been doing that by feel, ear, and no tension meter for more than 20 years with great results). But I see no reason it wouldn't work just fine.
And over versus under shouldn't matter, because it says to use the length (and diameter) or the spoke only from bottom of nipple to the first crossing, which makes sense from an engineering perspective.
oh do you have to input the length of the spoke? (to the crossing I mean). How accurately do you have to measure?
I know that when i pluck spokes on a 3x wheel, the tensions can be identical and I get at least 3 distinct tones on a front wheel, and more on a back wheel. I'm sure it's "good enough" for a general checkup though.
Hearing different "tones" may be actually hearing different timbres, that is different distributions of audio power in the overtones. I believe the app uses an FFT (Fast Fourier Transform algorithm) to sort thorough this and identify the fundamental.
As for accuracy in measuring length, the tension determination is proportional to length so a 5% error in length will give 5% error in tension, to first order anyway.
I played around with this app last night with several wheels, measuring different sounding spokes on the same wheel, and trying different input values for length and diameter on the same spoke. First, the output tension values were quite consistent with the same spoke and input values - typically repeatable within 1 kgf. And it easily indicated a different tension value for spokes that sounded slightly different, as it should, but repeatably showed the same tension for same-sounding spokes. Changing length by 2-3mm didn't make much difference, but the wrong spoke diameter changed the indicated force by 5-10 kgf.
At one point I was doing this as my wife was using electric toothbrush in a nearby bathroom - that changed the base level quite a lot, and added some error. So the directions to use in a quiet place are appropriate. It didn't seem to matter much with my wheels if I intentionally damped the other crossing spoke.
I don't have a tension gauge to compare with this (and with this app, I'm not likely to get one), but the outputs it showed seemed quite reasonable compared to the recommended tension values I've read. BTW, all the wheels that I was checking have proven to be extremely reliable, and I don't have any "bad wheels" to try. But it did show the quite high DS tension I expected on my most recently built rear wheel that has about 300 trouble and tweak free miles.