2 Jul
2014
2 Jul
'14
11:33 p.m.
I am wondering how those who like tactile feel in a fader would feel about
a sideways rotary control. I am thinking of something like the pitch or
mod. wheel on a synth. But no middle detent. Rotary encoders are great for
control surfaces, but the feel of a fader as compared to a rotary knob
makes faders easier to think with. I am thinking if I can use the same
idea with a fader feel where the control is endless, it would work well.
The box would have to be thicker of course and may be less use for stage.
The box could be made thinner by making it touch sensitive and self
centering like the pitch wheel, but I think that starts to make the
motorfader look better. The up side of the sideway encoder is that the
midi messages are just direction/number of ticks.
My first thought was a touch strip like the old guitar style stage
keyboards, have it inside a slot so it is easy to feel without seeing it.
Moving the finger up just sends up tics and moving it down gives down
ticks, so the finger can start anywhere on the strip which will be where
the (virtual) fader is now and move from there. I would have to use a adc
and take ticks off the LSB after doing a greater or less than calc. The
encoder has this out directly.
Also, ticks and direction happen to work well with the control board I
have on hand... the software can change that to a hard value if required
though.
--
Len Ovens
www.ovenwerks.net
3 Jul
3 Jul
12:57 a.m.
On Wed, 2 Jul 2014 14:29:48 -0700 (PDT)
Len Ovens <len(a)ovenwerks.net> wrote:
I am wondering how those who like tactile feel in a
fader would feel about
a sideways rotary control. I am thinking of something like the pitch or
mod. wheel on a synth. But no middle detent. Rotary encoders are great for
control surfaces, but the feel of a fader as compared to a rotary knob
makes faders easier to think with. I am thinking if I can use the same
idea with a fader feel where the control is endless, it would work well.
The box would have to be thicker of course and may be less use for stage.
The box could be made thinner by making it touch sensitive and self
centering like the pitch wheel, but I think that starts to make the
motorfader look better. The up side of the sideway encoder is that the
midi messages are just direction/number of ticks.
My first thought was a touch strip like the old guitar style stage
keyboards, have it inside a slot so it is easy to feel without seeing it.
Moving the finger up just sends up tics and moving it down gives down
ticks, so the finger can start anywhere on the strip which will be where
the (virtual) fader is now and move from there. I would have to use a adc
and take ticks off the LSB after doing a greater or less than calc. The
encoder has this out directly.
Also, ticks and direction happen to work well with the control board I
have on hand... the software can change that to a hard value if required
though.
--
Len Ovens
www.ovenwerks.net
Sounds interesting.
A bit more work, but it's possible to get miniature toothed belts and pulleys.
I wonder if this would give a thinner structure and a nicer feel as you would
be pushing the belt up and down as your 'slider'.
--
Will J Godfrey
http://www.musically.me.uk
Say you have a poem and I have a tune.
Exchange them and we can both have a poem, a tune, and a song.
1:48 a.m.
On Wed, 2 Jul 2014, Will Godfrey wrote:
On Wed, 2 Jul 2014 14:29:48 -0700 (PDT)
Len Ovens <len(a)ovenwerks.net> wrote:
> I am wondering how those who like tactile feel in a fader would feel about
> a sideways rotary control. I am thinking of something like the pitch or
> mod. wheel on a synth. But no middle detent. Rotary encoders are great for
> My first thought was a touch strip like the old
guitar style stage
> keyboards, have it inside a slot so it is easy to feel without seeing it.
>
> Also, ticks and direction happen to work well with the control board I
> have on hand... the software can change that to a hard value if required
> though.
Sounds interesting.
A bit more work, but it's possible to get miniature toothed belts and pulleys.
I wonder if this would give a thinner structure and a nicer feel as you would
be pushing the belt up and down as your 'slider'.
I had thought of that too. I was looking for simpler... maybe for version
2 (or 3 :) ) The belt would be teeth up I would guess. The belt itself
would be in a groove to keep it from moving sideways with just the teeth
sticking up. The encoder would be moved by the flat side of the belt. It
would be a good question how much side force the encoder could stand
without additional support for the shaft... and how much the whole thing
ends up costing as compared to motorized faders. Which ever way I go, I
may need to have some friction added to make sure the disk/belt stops when
the user removes their finger. or maybe the cheap encoders used are not as
free spinning as the shaft encoders I have worked with in the past.
--
Len Ovens
www.ovenwerks.net
8:22 a.m.
You would use belts to get "endless" faders, to avoid motorized restore?
I guess people who like it tactile, prefer to feel the position of a
fader knob and btw. we likely want that it appeals to all senses, IOW we
also want to see the fader knob positions.
Regarding to wheels also keep in mind, that faders could be very long
but for an "endless" wheel you perhaps can't do the whole fade by
keeping the finger on the wheel, you perhaps need to set back the finger
to continue turning the wheel.
Btw. for faders knobs you could use fader bridges, if you want to use
group of faders ganged.
JFTR you could optimize such a mixer to your needs, that is the
advantage of DIY, but DIY likely is more expensive and a lot could go
wrong. OTOH if restoring the position shouldn't be needed, replacing the
short faders of a nanoKONTROL with long faders might be possible and
would cost less. I wonder if longer faders would be useful for the 128
MIDI steps.
3:31 p.m.
On Thu, 3 Jul 2014, Ralf Mardorf wrote:
You would use belts to get "endless" faders,
to avoid motorized restore?
I guess people who like it tactile, prefer to feel the position of a
fader knob and btw. we likely want that it appeals to all senses, IOW we
also want to see the fader knob positions.
I have been trying to think what things are about tactile feel are helpful
to doing the job and what parts are pure resistance to change. One of the
things I have seen more than once on this list is "use your ears not your
eyes". Fader positions can be represented as LEDs or on a screen. I note
that while control surfaces generally use motor faders, they do not use
motor pots, but rather encoders. Some control surfaces have no indicator
where the virtual pot is, but rather show the effects on a screen beside
all the pots, for example a picture of the frequency responce for a group
of pots used for eq. Others have a light ring, but the light ring even for
something with 128 or more values only has 12 LEDs to show position. Some
highend surfaces have a dial shaped led display that might have 20 or so
LEDs right above the encoder. It seems we can only tell aproximatly where
things are with our eyes anyway.
Regarding to wheels also keep in mind, that faders
could be very long
but for an "endless" wheel you perhaps can't do the whole fade by
keeping the finger on the wheel, you perhaps need to set back the finger
to continue turning the wheel.
Design and training. I would actually think it would be easier to do a
full fade with a wheel or belt because the user is not restricted to where
the fader knob is, they can actually put their finger above where the
virtual fader position is.
would cost less. I wonder if longer faders would be
useful for the 128
MIDI steps.
1024 actually, 14 bits assigned, though not all are used.
--
Len Ovens
www.ovenwerks.net
5:21 p.m.
On Thu, 3 Jul 2014 06:27:25 -0700 (PDT)
Len Ovens <len(a)ovenwerks.net> wrote:
On Thu, 3 Jul 2014, Ralf Mardorf wrote:
Actually I was going to suggest the idea of a 'general purpose'
belt encoder generating MSB & LSB. Even though most kit can't handle both, it
gives you much finer resolution for anything that does.
As you say, have the belt toothed side up for grabability (tm) and simple
smooth pulleys top and bottom, but use a toothed pulley for the encoder, mounted
inside the casing *below* the belt. Mount it on a slotted bracket so you can
adjust tension.
I don't think you'd need anything either side of the belt. The pulleys will
keep it in line, and it would actually be easier to feel for the belt without
having to look at it.
As for not being able to get the full range in one sweep, you'd pretty quickly
learn how to 'walk' you fingers along the belt while moving it in the opposite
direction.
Damn! I wish *I* was doing this now :)
--
Will J Godfrey
http://www.musically.me.uk
Say you have a poem and I have a tune.
Exchange them and we can both have a poem, a tune, and a song.
You would use belts to get "endless"
faders, to avoid motorized restore?
I guess people who like it tactile, prefer to feel the position of a
fader knob and btw. we likely want that it appeals to all senses, IOW we
also want to see the fader knob positions.
I have been trying to think what things are about tactile feel are helpful
to doing the job and what parts are pure resistance to change. One of the
things I have seen more than once on this list is "use your ears not your
eyes". Fader positions can be represented as LEDs or on a screen. I note
that while control surfaces generally use motor faders, they do not use
motor pots, but rather encoders. Some control surfaces have no indicator
where the virtual pot is, but rather show the effects on a screen beside
all the pots, for example a picture of the frequency responce for a group
of pots used for eq. Others have a light ring, but the light ring even for
something with 128 or more values only has 12 LEDs to show position. Some
highend surfaces have a dial shaped led display that might have 20 or so
LEDs right above the encoder. It seems we can only tell aproximatly where
things are with our eyes anyway.
Regarding to wheels also keep in mind, that
faders could be very long
but for an "endless" wheel you perhaps can't do the whole fade by
keeping the finger on the wheel, you perhaps need to set back the finger
to continue turning the wheel.
Design and training. I would actually think it would be easier to do a
full fade with a wheel or belt because the user is not restricted to where
the fader knob is, they can actually put their finger above where the
virtual fader position is.
would cost less. I wonder if longer faders would
be useful for the 128
MIDI steps.
1024 actually, 14 bits assigned, though not all are used. 
9:29 p.m.
On Thu, Jul 03, 2014 at 04:21:38PM +0100, Will Godfrey wrote:
Actually I was going to suggest the idea of a
'general purpose'
belt encoder generating MSB & LSB. Even though most kit can't handle both, it
gives you much finer resolution for anything that does.
As you say, have the belt toothed side up for grabability (tm) and simple
smooth pulleys top and bottom, but use a toothed pulley for the encoder, mounted
inside the casing *below* the belt. Mount it on a slotted bracket so you can
adjust tension.
I don't think you'd need anything either side of the belt. The pulleys will
keep it in line, and it would actually be easier to feel for the belt without
having to look at it.
Unless the belt is tensioned quite a bit you'd need something
to support it - without causing too much friction.
It all seems a nice idea, but OTOH you can get a Yamaha motorised
fader for 20 Euro, see
<http://www.thomann.de/it/yamaha_ls_901vmotorised_fader.htm>.
I wonder if you can get and assemble all the required parts for
a usable 'belt fader' for that price.
I once (and only once) had the occasion to use a mixer with
quadrant faders - they feel really perfect, I'd prefer them
over linear ones any time.
Ciao,
--
FA
A world of exhaustive, reliable metadata would be an utopia.
It's also a pipe-dream, founded on self-delusion, nerd hubris
and hysterically inflated market opportunities. (Cory Doctorow)
11:01 p.m.
On Thu, 3 Jul 2014, Fons Adriaensen wrote:
It all seems a nice idea, but OTOH you can get a
Yamaha motorised
fader for 20 Euro, see
<http://www.thomann.de/it/yamaha_ls_901vmotorised_fader.htm>.
I wonder if you can get and assemble all the required parts for
a usable 'belt fader' for that price.
That is about the same price I was looking at for motor faders to. But...
that is not the whole story... You need also the driver for the motor, a
touch sensor so it knows you are controling it and don't burn the motor
out trying to go against your movement and you also need an analog in
port. Also needed is the circuitry to receive the virtual fader position
from where ever and apply it to the fader logic. SO the fader is $20, but
the whole price is closer to $50 (for diy, manufactures can design for way
cheaper with custom chips, etc. as they are already making a board of some
sort.) An encoder is pretty much direct digital read and only one
direction data flow. So the cost of belt and pullies may be still less
than the motor fader in a diy situation. In a manufactured product, motor
faders give the user what is "expected" which is a way lower risk than the
cost of re-engineering and a possible product flop. I am sure using a
large touchscreen could be done for a reasonable price if the expensive
projects had worked out well, but that doesn't seem to have happened.
People want faders they can feel... we want to be able to feel them and
operate them without looking either because we are watching something
else, or even because it is dark. In fact too many screens may make the
room to light for some people or in a live situation.
I haven't even started ordering parts, let alone building, and I have
already learned a lot. :) I am almost thinking about doing more than one
kind of fader just to see what the differnce in complexity, cost and usage
is. I will have to think of a reasonable way of using more than one kind
with out it being confusing.
Because of my usage, where the computer is the other part of the mix desk,
There are differences from what a stand alone control surface would be
like. I am relying on the computer screen(s) for things like fader
position and may have a small, skinny, monitor strip for the bottom of the
screen to give other feedback. I am not even sure that my USB interface
will be fast enough to make any kind of fader work, it will give me a pile
of buttons though, I will at least have a way of controlling transport and
some other things that is not focus dependant (does not require the
application I am controlling to be the window with focus).
--
Len Ovens
www.ovenwerks.net
4 Jul
4 Jul
8:55 a.m.
On Thu, Jul 03, 2014 at 01:58:02PM -0700, Len Ovens wrote:
On Thu, 3 Jul 2014, Fons Adriaensen wrote:
I don't think so. You need an 'H-bridge' for each motor, an LN293
(IIRC) has two of those and costs around 2 Euro. Then you need an
AD converter which can be multiplexed and doesn't need to be very
high speed. Something with 8 channels, 12 bits will be less than
1 Euro per channel. Then the touch sensor, again this can be done
for around 1 Euro per channel. All the rest you need anyway, or
can be SW. So a pessimistic guesstimate would be 25 Euro per fader.
Ciao,
--
FA
A world of exhaustive, reliable metadata would be an utopia.
It's also a pipe-dream, founded on self-delusion, nerd hubris
and hysterically inflated market opportunities. (Cory Doctorow)
It all seems a nice idea, but OTOH you can get a
Yamaha motorised
fader for 20 Euro, see
<http://www.thomann.de/it/yamaha_ls_901vmotorised_fader.htm>.
I wonder if you can get and assemble all the required parts for
a usable 'belt fader' for that price.
That is about the same price I was looking at for motor faders to.
But... that is not the whole story... You need also the driver for
the motor, a touch sensor so it knows you are controling it and
don't burn the motor out trying to go against your movement and you
also need an analog in port. Also needed is the circuitry to receive
the virtual fader position from where ever and apply it to the fader
logic. SO the fader is $20, but the whole price is closer to $50
9:15 p.m.
On Fri, 4 Jul 2014 06:55:45 +0000
Fons Adriaensen <fons(a)linuxaudio.org> wrote:
On Thu, Jul 03, 2014 at 01:58:02PM -0700, Len Ovens
wrote:
Not forgetting of course that doing it different is fun :)
You can't put a price on that, and I'll say right now, there is a lot of kit
I've built that is totally uneconomic, inefficient and unrepeatable. However, I
use some bits every day and am quite proud how well it has stood up over time.
Back to the belts.
I'd cut slots in the top plate (you'd have to anyway to get the belts and
pulleys in), then put a strip of (say) 6mm thick nylon or PTFE on the underside
of the top plate. The belts will hide it and will slide smoothly over it. Using
the toothed side against encoders with toothed pulleys means the belt can be
quite slack without slippage.
--
Will J Godfrey
http://www.musically.me.uk
Say you have a poem and I have a tune.
Exchange them and we can both have a poem, a tune, and a song.
On Thu, 3 Jul 2014, Fons Adriaensen wrote:
I don't think so. You need an 'H-bridge' for each motor, an LN293
(IIRC) has two of those and costs around 2 Euro. Then you need an
AD converter which can be multiplexed and doesn't need to be very
high speed. Something with 8 channels, 12 bits will be less than
1 Euro per channel. Then the touch sensor, again this can be done
for around 1 Euro per channel. All the rest you need anyway, or
can be SW. So a pessimistic guesstimate would be 25 Euro per fader.
Ciao, It all seems a nice idea, but OTOH you can get a
Yamaha motorised
fader for 20 Euro, see
<http://www.thomann.de/it/yamaha_ls_901vmotorised_fader.htm>.
I wonder if you can get and assemble all the required parts for
a usable 'belt fader' for that price.
That is about the same price I was looking at for motor faders to.
But... that is not the whole story... You need also the driver for
the motor, a touch sensor so it knows you are controling it and
don't burn the motor out trying to go against your movement and you
also need an analog in port. Also needed is the circuitry to receive
the virtual fader position from where ever and apply it to the fader
logic. SO the fader is $20, but the whole price is closer to $50
8:50 a.m.
On Thu, 2014-07-03 at 06:27 -0700, Len Ovens wrote:
14 bit (LSB/MSB) are 16384 steps, IOW -8192 to 0 (also 1 step) to +8191
two's complement, however often only 128 steps are provided.
would cost
less. I wonder if longer faders would be useful for the
128 MIDI steps.
1024 actually, 14 bits assigned, though not all are used. 
11:35 a.m.
On 03/07/14 23:27, Len Ovens wrote:
On Thu, 3 Jul 2014, Ralf Mardorf wrote:
The old fashioned very large round knobs with a reasonable amount of resistance
as faders were very good for positioning by feel ... angular position,
especially during crossfades between different sources but also in terms of
adjustment by a few degrees, is nicer to work with in many ways than linear
faders ... the main advantage that a bank of linear faders has is the 'graph'
they draw with the position of the handles, and that you can easily find the
ones that are up by feel without looking. But as soon as you don't have motors
(or dedicated faders) you lose that advantage. Then probably rotary knobs, if
large and solid enough, are better. A ring of leds isn't expensive or difficult
for feedback and going a bit further so there are maybe two or three on at once,
with brightness controlled to give more than just discrete positional info, and
perhaps different patterns like fill or spread for different purposes can really
communicate a lot if you want to get detailed ... so channels used for pan, eq,
level etc get different patterns assigned.
Simon
You would use belts to get "endless"
faders, to avoid motorized restore?
I guess people who like it tactile, prefer to feel the position of a
fader knob and btw. we likely want that it appeals to all senses, IOW we
also want to see the fader knob positions.
I have been trying to think what things are about tactile feel are helpful to
doing the job and what parts are pure resistance to change. One of the things I
have seen more than once on this list is "use your ears not your eyes". Fader
positions can be represented as LEDs or on a screen. I note that while control
surfaces generally use motor faders, they do not use motor pots, but rather
encoders. Some control surfaces have no indicator where the virtual pot is, but
rather show the effects on a screen beside all the pots, for example a picture
of the frequency responce for a group of pots used for eq. Others have a light
ring, but the light ring even for something with 128 or more values only has 12
LEDs to show position. Some highend surfaces have a dial shaped led display that
might have 20 or so LEDs right above the encoder. It seems we can only tell
aproximatly where things are with our eyes anyway.
3824
days inactive
3826
days old
linux-audio-user@lists.linuxaudio.org
12 comments
5 participants
participants (5)
-
Fons Adriaensen -
Len Ovens -
Ralf Mardorf -
Simon Wise -
Will Godfrey