[sdiy] Pots vs Encoders, was Re: [sdiy] dave smith *instruments*

[cheater cheater]

Scott Jason and Eric,
not to hijack your FPGA thread, but I thought I'd chime in with some
on-topic. :P

> There's an implication in this thread that encoders cant be used as
> performance controls and that's silly.

They can, for some definitions of 'performance'. You can put two cds
in two cd players, with recorded tracks at the exact same beat rate,
press play on one of them, slowly move the crossfader, and call that
an amazing performance. But I can't.

> They can but with their inherent
> limitations.  Positional feedback isn't always necessary - I want a slower
> attack, I want more resonance, I want.... - that's just turn a dedicated or
> assignable encoder left or right.

This is only until you start thinking of it a bit harder and trying to
up the ante, as they say. You know, a violinist doesn't think whether
the next note he wants to play is 'higher' or 'lower'. If this were so
the performances would be full of pretty bad glissando and terribly
articulated notes. This is the same with any performance control on
any instrument, be it a string, a mouthpiece, the tongue, the open end
of a trombone, or a key. When you want to play a specific pitch on a
piano, do you think the following? 'oh, hmm, the key that my finger is
depressing is higher than the pitch I want to play next. Let's try
this key a few steps to the left. Is that the right key? Oh, no it
isn't. Let's try a bit higher. Ugh, still not the right one. Let's try
the black one a bit lower. Oh crap, I was pressing keys too fast and
now my hand has randomly moved to the highest pitch on the keyboard! I
hate being epileptic!'

This is the kind of experience you get with encoders. And, sure, if
you turn veeeryyy sloooowwlllyyyy you can get them to do almost what
you want them to. But I play more than one note per second on my
keyboard and I expect the same kind of controllability from the
resonant filter with which I want to change the leading overtone more
often than once per second.

I'm not trying to knock down on you or call you a newbie. I'm pretty
sure you're great at what you're doing. But the fact that having to
turn your performance controls real slow and really carefully doesn't
mean it's not a problem, it just means it's not a problem for you,
because you never needed anything more. Maybe your kind of music does
not require this, I'm sure mine and a lot of other people's does.

I do not want to feel like I'm being epileptic and therefore encoders
are definitely bad for me.

> 3. Testing the 'Osc Fine Freq' setting that has a range of -50 to +50, if
> you spin the knob 360 degrees in about 2 seconds or less the display will
> smoothly increment and show a change of 24 ticks

This is not good enough for me. Often I need to adjust a knob 6-10
times per second. To give a simple example, it needs to go from a to b
back to a and so on. An encoder can in no way provide this. This is
not the kind of thing that an encoder was ever meant for. If someone
never wanted to do this sort of thing, he either plays reeeaaalll
sloooowww music (good for him, I like slow music too) or isn't a synth
player, but much rather a piano player who wants some new sounds (and
plays a polysynth by just using the keyboard and maybe the mod/pitch).

> 4. If you spin the knob in about 1 second the display will change from -50
> to +50, skipping most or all intermediate values

What if I want to make an adjustment from -50 to +25 in 0.3 seconds?
Funny that I can quantify this, but this sort of thing happens quite
often when I'm playing.

> 5. Spinning the knob as quickly as possible ~360 degres results in
> unpredictable response, either no change, max change, or anything in
> between.

And you say that's not a problem? If this happens during performance,
everyone leaves, unless they're your friends, but friends forgive
anything.

> 7. With a little practice you can get a 'click feel' and could probably get
> quite skilled at turning a knob and getting exactly so many 'clicks'.

No, this is terrible. You are hurting yourself and your skill by
learning this sort of thing.

D.

On Tue, Feb 2, 2010 at 17:32, Dave Manley <dlmanley at sonic.net> wrote:
> Tom Wiltshire wrote:
>>
>> I accept that a panel full of encoders and a single LCD screen isn't much
>> competition for pots, since the pot positions sometimes show you something
>> useful, and only the encoder whose value is on the LCD tells you anything.
>> But this isn't really comparing like with like, although that is the
>> situation on the P08 or Evolver, which is where we came in.
>
> There's an implication in this thread that encoders cant be used as
> performance controls and that's silly.  They can but with their inherent
> limitations.  Positional feedback isn't always necessary - I want a slower
> attack, I want more resonance, I want.... - that's just turn a dedicated or
> assignable encoder left or right.  The Mopho/Tetra for example has dedicated
> encoders for common performance parameters, the other encoders are patch
> assignable to what ever you want.  So pick your preset, listen to the sound,
> and tweak away.  The Mopho/Tetra interface also makes it really easy to
> reassign an encoder in real time, but you do have to scroll through the
> parameter list, so it does take attention.
>
> I was just testing a Mopho looking at the overall quality and the rotary
> encoder issues you describe.  A few comments:
>
> 1. The rotary encoders are 24-step.  I believe this is the part:
> http://search.digikey.com/scripts/DkSearch/dksus.dll?Detail&name=P10860-ND
> 2. They have a very solid feel, The data sheet says they're rated for 30K
> operations for what's that worth.
> 3. Testing the 'Osc Fine Freq' setting that has a range of -50 to +50, if
> you spin the knob 360 degrees in about 2 seconds or less the display will
> smoothly increment and show a change of 24 ticks
> 4. If you spin the knob in about 1 second the display will change from -50
> to +50, skipping most or all intermediate values
> 5. Spinning the knob as quickly as possible ~360 degres results in
> unpredictable response, either no change, max change, or anything in
> between.
> 6. If I put the knob at min value, and then turn counter-clockwise as fast
> as possible occasionally the display will take a value between -50 and -45,
> indicating some misreads.
> 7. With a little practice you can get a 'click feel' and could probably get
> quite skilled at turning a knob and getting exactly so many 'clicks'.
>
> I don't see any glitchy behavior.  Not denying that others see it, or I
> won't see it in a year on this Mopho, just reporting what it currently does.
>
> -Dave
>
>
> _______________________________________________
> Synth-diy mailing list
> Synth-diy at dropmix.xs4all.nl
> http://dropmix.xs4all.nl/mailman/listinfo/synth-diy
>