been doing more work on a fixed install at church.
Running frequency sweeps to tune the system and comparing android/iOS and QU RTA tools.
The iOS one was rubbish, frequented on android was pretty good - nice spike on the selected frequency. The QU followed the frequency, but with a really wide bell (5-6 bars either side of the frequency generated)
I’m intrigued as to why the QU RTA was showing a very wide signal, when the input was basically a pure tone (same for QU and M300 generated tones)
For tuning a room you shouldn’t use a 31band RTA! FFT analyzer is the tool to use! iOS app “Audiotools” has a quite good FFT analyzer but for professionell use you should consider a professional tool (software as well as meassurment-microphone!)
I have been using a 1/3oct RTA for 30+ years. I admit it is not the perfect tool but I can make it work. Past few years I have started using a custom CD with 1/3oct. filtered noise bands. Use SPL meter. Works better than anything I have ever tried. I use a combination or PEQ and 1/3oct EQ for filtering.
The tools I have might not be ideal, but them is the tools I have 
It was the comparison between frequensee on android and the 31 band RTA on the QU that was interesting. Looks like each band on the QU is a lot wider than I’d have thought, a pure tone should only trigger one of the meters - or is that where I am making my mistake…
Ok… It’s better than nothing 
just sometimes you can missunderstand a RTA especially for narrow frequency obligeration at the meassure point due speaker crossover! You might think you need to raise the GEQ band but nothing happens and your phase alignment is gonna be way more worse!
A pure tone should only affect one band - but!!!:
You can generate a pure tone in the mixer yes, but how your speaker and the room you are in gonna reflect it - that’s what the tone finally is - and believe me that will not be exactly one narrow band!
Ou yeah and of course how your meassurement mic is calibrated! So at least 3 sources where misstakes can happen!
Absolutely, but…
The tablet was showing a very clear, and very narrow, peak. The RTA looked roughly Gaussian.
It was a decent mic, and the effect was consistent throughout the sweep.
An RTA doesn’t necessarily imply that an FFT was used to translate the data. A series of resonant filters would achieve the response we see on the Qu.
Yes, I suspect that it’s a set of filters rather than an FFT…