[sdiy] WTB: Paia EKx40

[harrybissell]

Hah... you have walked into my TRAP!!!  :^P

Indeed... if we want to have pure waveforms we have to look to the ideal.

Now lets look at the real... "sawtooth" waveforms with a flat spot at the top ?
(TB-303)
Triangles with glitches in the middle ?  (various)
Sawtooths with glitches everywhere ??? (MS-20, ARP Little Brother)

We don't really need perfection in these waves... we need perceptual perfection.

Can we really hear the rise time difference... or the non-linearity of a rise and
fall time ?
In a lot of synth cases, yes we can... because the manufacturer's standard of
'good enough
for rock and roll' isn't good enough for those who actually listen to their
instruments.

Everyone who is familiar with the sound of the lab instruments KNOWS that the VCOs

we use do not even come from the same planet in terms of quality (linearity,
harmonics
correct - like none for a sine wave - f + 1/2 2f + 1/3 3f... for a sawtooth - and
noise / jitter)

Also our VCOs don't cost ten thousand dollars each, either...   :^P

there is also a perceptual difference in different people... some people (used to
lab oscillators)
might prefer the 'pure' sound... while others prefer more 'random' (or dirty,
distorted,
sound with some character, or flavor etc..).

some... I have heard... even like the sound of a BBD   :^P

H^) harry

Magnus Danielson wrote:

> From: "john mahoney" <jmahoney at gate.net>
> Subject: Re: [sdiy] WTB: Paia EKx40
> Date: Tue, 9 Mar 2004 19:42:10 -0500
> Message-ID: <032f01c40638$85c17580$6400a8c0 at BABYUTEST>
>
> > > ----- Original Message -----
> > > From: "Harry Bissell Jr" <harrybissell at prodigy.net>
> > > |
> > > | IMHO a sine wave has NO harmonics, a VCO has no drift
> > > | or jitter, sawtooth and pulse waves have infinite
> > > | (fast) rise (and or fall) times.
> > >
> > > That is not opinion, humble or not. That is the definition ;-)
> >
> > Well, we need infinite rise times and *linear* fall times on the sawtooth
> > waves.
> > --
> > john  <--(being extremely picky, and hoping that linear is the right term)
>
> Linear in what respect? :->
>
> You mean to say that there is a constant (positive) slope between the
> infinit fall-times.
>
> BUT! Being picky like hell... you will *NEVER* have that, since even if you
> have say 50 GHz of bandwidth, you have a fall-time of about 7,5 ns. For 40 GHz
> you would be patching with semirigid cable and SMA connectors.
>
> So... "infinit fall and rise-times" is just down the trash before you got up.
>
> ... and jitter/wander free? BHA! Phase-noise of various sorts is a matter of
> life, or maybe you have borrowed my mystery magical -35 K cooler-spray? ;O)
>
> In state of the art research atomic clocks they laser-cool their atoms down to
> a few uK in order to lower the temperatur based phase-noise (doppler/relative
> effects included!).
>
> So, being picky means realizing just which deficiencies we must learn to live
> with, the issue is really how grave they are. What we needs to learn is when
> lowering stuff under some limit means there is nothing more to do. Thus, how
> much nearer the "ideal" it is meaningfull to go. We can't do much better than
> such rough engineering rule of thumbs it seems.
>
> Cheers,
> Magnus