[sdiy] frequency counter

[WeAreAs1 at aol.com]

In a message dated 6/26/04 3:10:04 PM, ijfritz at earthlink.net writes:

<< No disagreement with any of this.  However, it doesn't sound like you are 
relying on your ear alone -- there is always something you are beating 
against, no?  Certainly the piano tuner starts with a tuning fork and then 
uses beats of one note against another.  I knew one person once who had 
perfect pitch within 1/8 of a semitone (12 cents), but that is quite rare, 
I think. >>

Hello Ian,

Yes, there's always some kind of tuning reference.  I personally would not 
hire a piano tuner who relied on perfect pitch and did not also use a tuning 
fork as a reference.  The typical piano tuning routine is to tune middle C or 
middle A to the reference fork, then put away the fork and go through the circle 
of 5ths, taking into account that you can't tune 5ths to a perfect, no-beating 
state (that wouldn't work on a, umm, well-tempered clavier, as they say).  
You get that middle octave of notes in tune and sounding good for all the 
different intervals (5ths, 4ths, 3rds, etc), then move outward from there, simply 
tuning all the octaves of those middle notes (yes, I'm oversimplifying).  Keep 
in mind that the piano, like the Minimoog, has three oscillators per note.  
Oops, I mean three strings per note.  Except on the lowest notes, which have just 
two oscillators.  Strings, that is.  So there's a lot of deadening and 
opening of each of those three strings while you are tuning and comparing them to 
each other (using felt strips or these little rubber "string stopper" tools.  
It's fascinating process, kind of like putting together a big puzzle, and I love 
to sit and watch and listen to them do it.  (I have tried, and failed, many 
times to do it myself.  I have all the tools and most of the knowledge, but 
lack the patience, I guess)

The point here is that the tuner guy is always using something else as a 
reference -- the tuning fork; the most adjacent 5th, the nearest octave, the other 
two oscill---strings of that key.  This is directly analogous to the basic 
system I use when calibrating a synth, although I sometimes use a Yamaha DX7 
instead of a tuning fork, and I use a BOSS guitar tuner to get that first VCO 
tracking correctly.  It can certainly be done without the BOSS tuner, but it's 
rough on the ears for those very high octaves (where it really counts).  If I 
was tuning a synth without access to a guitar tuner, I would use the DX7 as a 
pitch reference for the scaling of the first VCO.  Just as the piano tuner puts 
away his tuning fork after he tunes the first middle A or C, I put away the 
DX7 after scaling and tuning the first VCO.  I then use that first VCO as the 
tuning reference for the other VCO's.  

The logic is simple: In the real world, these VCO's need to play in tune with 
each other -- not with a tuning fork or a DX7 or whatever -- so their best 
reference is each other (as long as the first VCO scaling is perfect, or very 
close to perfect.  With synths, as with pianos, it's more important that the 
VCO's track each other than it is for them to track the outside world (other 
"in-tune" instruments).  We will notice the internal beating way before we will 
notice that the overall pitch is off (beating with the outside world is 
otherwise known as "warmth").  As a matter of practicality, you'll never be perfectly 
in tune with the outside world.  The best you can hope for is to be very 
close, and to make sure that all your VCO's track together without beating.  It's 
the beating that is the mark of an out-of-tune instrument, not the actual pitch 
that the note plays.  Of course, we hope that both the beating and the actual 
scaling are as close as we can get to the rest of the world, but that's just 
not going to happen in life.  

As an example, try this sometime:  Get two Yamaha DX7's, initialize the patch 
to that one-operator sine wave organ sound, tune their master tunings 
together, then play them together.  You will notice that they never really play 
perfectly in tune with each other -- there's always some kind of beating, sometimes 
it's very obvious, especially on high notes.  I have no idea why this 
happens, but it does.  Now, if you can't get two exactly identical digital synths to 
be in tune with each other, how do we expect to ever be in tune with other 
instruments that are made out of wood and steel (or made with finicky discrete 
transistors)?  It's not a big problem -- it's actually what makes orchestras 
sound so rich and full.  The same goes for guitars -- the physical nature of how 
guitars are constructed prevents them from ever being fully in tune, even with 
themselves (Google "Buzzy Feiten Tuning" for more on this).  It's not a big 
problem in the real world, though.  All those slightly differently tuned 
guitars sound good when played together (this is why we like to double-track or 
triple-track rhythm guitar parts on records - especially nice in stereo).

Perfect pitch can be useful to a musician, but more often, I think it's kind 
of a curse.  It can distract the musician if they hear one note in an ensemble 
that's slightly out.  Worse yet, the perfect pitch guy often relies on his 
magical skill to identify notes, not by their musical relationship to other 
notes (5th, 3rd, 7th, etc.), but by the actual note name of the note.  This, 
without the relationship understanding, is really a useless trick, and can be very 
limiting.  A guy I know who has perfect pitch can instantly identify all the 
notes of any complex chord, but he cannot (without some effort and time) 
identify the name of the chord, the chord type, or the harmonic relationship of the 
notes.  For example, he'll instantly tell you that you're playing F#, A, C, 
and E, but can't immediately tell you the most important thing -- that the chord 
is an F#m7b5.  Oh, eventually he'll figure it out -- but you're already at 
the bridge by then.  Another perfect pitch guy I know is a high school choir 
director.  Ever listen to a high school chior?  His remarkable ability to hear 
exact pitch just drives him crazy when listening to those kids, um, "sing".

I have very highly developed relative pitch skills, better than almost anyone 
I have ever met.  However, I definitely need some kind of reference pitch to 
get started.  In recent years, I have discovered that I am starting to develop 
a semi-reliable ability to pick a reference pitch out of the air.  Not a 
perfect reference pitch, but a close-enough reference pitch.  That is, not a 
perfect A440, but close, and definitely not Bb or B.  I do this mainly by mentally 
conjuring up the sound of certain songs or records whose key is already known 
to me.  For instance, Led Zepplin's "Kashmir", which is in the key of D, seems 
to be very easy to remember.  The Beatles "Day Tripper", in E, or "Hey Jude" 
(F) are another couple of reliable songs.  Believe it or not, I have also 
memorized the "bhooooop" sound of the Minimoog's internal A440 reference tone 
(Minimoog owners will smile about that -- It's such a classic, characteristic 
sound, and nothing else sounds like it, especially at that instant when you first 
switch it on).  Using this method to find notes out of silence, I'm rarely off 
by more than a half step.  However, I'm never right on -- except if I 
randomly get lucky.  

It's a marginally useful trick that lets me tune a guitar close to standard 
pitch even if there's no tuner or reference pitch available, and it allows me 
to identify the key of any music I hear on the car radio or television without 
a reference pitch (by using my relative pitch skills, comparing to my mentally 
conjured reference pitch).  My relative pitch skills came strongly into play 
very early in life and have steadily improved eavery year, but this weird 
"imperfect pitch" thing really only started sometime in my late thirties (I am 
almost 50 years old).  Most perfect pitch people seemed to learn their skill 
almost immediately, as if it was like the ability to identify color or something 
like that -- "you either have it or you don't" they often say.  I'm pleased 
(and somewhat surprised) that I have been able to develop this little trick of 
mine, and I notice that it gets better and better with practice, but I know it 
will never turn into perfect, "A440-not-A441-not-A439" pitch - and I'm thankful 
for that!

Michael Bacich

P.S. - Has anybody noticed how some records these days are not in tune with 
standard pitch?  On older, pre-digital, "stoner rock" records, this might be 
understandable, but on modern records??  At first, this seems kind of 
incredible, almost impossible, considering how we rely so heavily on tuners and digital 
technology in the studio.  It almost always has to do with the song being 
slowed down or sped up slightly after the mix has been done, before mastering.  
This often happens at the behest of the record company.  Some twerpy 
pony-tailed, goateed, latte-slurping wannabe musician behind a walnut desk thinking the 
song needs to be "just a little bit faster for our market".  It's also a reason 
(along with the dreaded Autotune plug-in) why vocals often sound rather 
unnatural on those records (a slight "chipmonk" effect).