Yeah it's pricey for an app but well worth the money imo. I can't see my TU-2 getting any use from now on (other than on stage).I still can't bring myself to pay for an iPhone app!
$12.99 over here...
I believe it is although I'm not able to verify it myslef other than to say it works very well to my ears.it its as accurate as a strobe tuner....
Whats more your ears can hear down to 100000th of a semitone and tonight I'm going to use a laser cutter to trim my steak with.I'm a VSAM owner - trust me, if it's as good as a VSAM, $12.99 is a STEAL. :lol:
Is it actually as accurate as a virtual strobe? sweetened tunings, metronome, and pitch settings aside, the VSAM is accurate to 1/1000th of a semitone. I'd be surprised if an iPhone could pick up that sort of change...
EDIT - why, so it is. Crazy.
I'd be surprised if an iPhone could pick up that sort of change...
Why? I believe the iPhone is capable of recording at 16/44.1. Without going deeply into detail, if that sample rate is sufficient to record music, then it follows that it must also be able to record out-of-tune music, no? And if it can be recorded at that resolution, then also that data set can be analyzed.
The complicated answer involves FFT windowing, so there is a minimum time interval that is required for a given resolution of frequency, but that's not too taxing. What is taxing is that the pitch of the instrument under test is likely to vary more greatly than that resolution. In theory there should be some point where the two phenomena intersect.
I believe, but could be wrong, that the accuracy of these tuners is based on the processors clock speed.
FFT does require a giant amount of DSP, so it is probable they would have to limit the number of bins, which limits resolution. I don't think that's the only way to do a tuner though, obviously the electronic tuners aren't doing that. But then their resolution is not as good . . .
FFT is a non-issue for a desktop but I dunno how much power these iPhones are running.
Anyway, I can think of two ways to do it, one is the brute-force FFT with a large sample size, then just look for the bin with the highest amplitude. That takes a lot of DSP.
Less expensive method is a small sample size to simply identify the frequency you are trying to hit without regard to how in-tune it is. Then do an upsample (if required for higher frequencies) and count off sample intervals (approx. 100 samples for 440Hz, for example), and analyze the change in amplitude between samples. If it's perfectly in tune, each sample interval will be the same amplitude.
It's cheaper if you avoid the upsample, but that limits resolution.
Also, the small bin size to ID the frequency of interest means that your detection of low frequency notes will be poor. The only solution for that is to throw lots more DSP at the signal.
There's probably a still easier solution involving higher math that I don't know.
Really odd that you just posted this. I just bought StroboSoft 2.0 for my Mac last night. At $100, it's a bit more than the iPhone app--but I love it. Intonated 3 of my guits last night with better results than I've ever gotten by ear. If I start playing out more, I'll likely get the iPhone app too.
You paid $100 for tuner software? I use this free software and it works good enough for me. http://www.nch.com.au/tuner/index.html