Peterson 490ST Strobe tuner....good?

[Farrell]

Hi Richard. Indeed, the Verituner calculates tunings very differently from the RCT, SAT or TuneLab. One can program in what partials to look at, which partials from which notes to compare, it can compare various designated partials from several notes to the target note, and it can weight the various partials from the various notes. Does that make any sense?

Ron Koval has done much work on programming the way the Verituner works. For example, the following is how the lowest octave is calculated in one of his setups:

The Verituner will look at 10:5 octaves, 4:1 octaves, and 8:1 octaves, and will weight them 30%, 35%, and 35%, respectively. In the middle of the keyboard, he is looking at 4:2 octaves which are weighted at 60% and 6:3 octaves weighted at 40%. And on and on. You can change the rules for each octave. The Verituner measures up to like 8 or 12 partials on each note, and then uses whatever partials you inputted rules demand.

What I find really fun to do with it is done with a crummy piano, like a low end spinet or whatever. You start out measuring all the plain wire A notes to start calculating a tuning (this is just so that it has something to start with). Next you go to A2 and strike the note. Watch immediately at the Verituner display. Initially, for the first second or so, will show a target pitch that is based solely on the plain wire section (at this point the tuning calculation is going to be similar to what a SAT might calculate - it assumes the bass strings have nice inharmonicity). Let's say that A2 is right at the initial target pitch. After a second or two, when the Verituner has had a chance to sample the partials for A2, it will recalculate the tuning and fit A2 to a better pitch. So when you first strike A2 the Verituner displays zero in pitch deviation, but after a second or so - after it has sampled and calculated, it will change easily five cents or more from that initial curve. And then, of course, as one plays all the notes, the tuning is refined even further.

Awesome machine!

Terry Farrell
  
----- Original Message ----- 
From: "Richard Brekne" <richard.brekne@grieg.uib.no>
To: "PTG" <pianotech@ptg.org>
Sent: Friday, June 28, 2002 5:35 PM
Subject: Re: Peterson 490ST Strobe tuner....good?


> 
> 
>  Bradley M. Snook wrote:
> 
> >Oh, I see, you are not unconvinced about tuning curves in general, just
> >the
> >ways in which they are currently being calculated. 
> 
> Yes... except perhaps the Verituner...but I am not sure enough about how it
> goes about doing what it does. The other thing is the fact that the other ETD's
>  guess at too many notes (82 or more of them). I know the "guesswork" is pretty
> darn good... but as I said in my last..... the results seem less even as I look
> closer at what they really are.  
> 
> To do the job << right >> you simply need to know the real partials ratio or
> ladder for every note, and you have to have an algorithm that makes all the
> right compromises to achieve the closest to whatever tuning temperament you are
> trying to employ with this information.
> 
> >Yes, I completely agree.
> >Even though I prefer TunLab, I really dislike the way in which it
> >calculates
> >the inharmonicity constant. After you measure the inharmonicity of a note,
> >it condenses all that data into a single number. It is this number that is
> >used to reconstruct where any given partial should be located. I really
> >think that this is a serious flaw, because where the partial is, and where
> >the software thinks it is, are almost always slightly different. This
> >naturally gives a slight difference when one confirms the tuning aurally.
> 
> Didnt know that.... glad to tho :)
> >
> >It is often overlooked that it is possible to do a 'perfect' tuning curve
> >manually if one is willing to invest the multitude of hours. Once every
> >partial is measured, it is very easy to use a spread sheet to adjust the
> >tuning curve to what ever you feel is most appropriate. Naturally no one
> >has
> >the time to spend on every piano, but if there is a piano that you tune
> >regularly for important concert situations, it might be worth some time.
> 
> I think this is basically what Verituner claims to be able to do real time, and
> why users keep reporting that its best tuning on a piano is figured on the
> second, or ensueing passes. And that part of how it accomplishes this is by
> relying on this ratio function you briefly point at below. It quickly nails the
> partial ratios down on the fly as you tune, and uses these to continually
> update the tuning for any piano note for note. Tho I do know know this for
> sure.
> 
> >One
> >more thing: as long as you know the fundamental/partial ratio, you can use
> >any partial to tune any given note. For example if I know that the 2nd
> >partial is 2.00271895424837:1 and the 3rd partial is 3.00965728900256:1,
> >tuning either of these partials will produce the same results for the
> >fundamental and all the other partials.
> >
> 
> 
> 
> >> I agree its interesting, but I am not sure I agree with your second
> >point.
> >To
> >> be sure there is more information provided. That doesn't at the same time
> >mean
> >> that less information is actually being used.
> >
> >I was trying to say, that software, in general, allows for the utilization
> >of more information. For example, if you were to measure all the relevant
> >partials of the piano, software could assist you in the positioning of a
> >tuning curve based on many different approaches. You could monitor the
> >progression of 3rds, or whatever kind of interval you think is important.
> >
> 
> I'll go along with that alright.... its just that I wonder how much of this is
> actually being used by tuners... Present ETD's make it very very easy .... not
> to say outright seductive... to just follow the dial...and perhaps make a
> couple aural adjustments (that may or may not be justifiable) along the way.
> Software that does what you describe above is not yet out there.. (or what ?)
> tho some of us have been asking for it for some time now. 
> 
> >
> >Bradley M. Snook
> 
> Cheers
> 
> RicB
> Richard Brekne
> RPT NPTF
> Griegakadamiet UiB
> 
>