SV: Re: [sdiy] Korg MS20 good deal?

[Rasmus Nyåker]

Decided to post a post I made at the vintagesynth.com forum as it is  
quite close to this discussion. The original thread that rendered this  
reply from me can be found here:
http://vintagesynth.org/phpBB2/viewtopic.php? 
t=10390&postdays=0&postorder=asc&start=0  (for those who find interest  
in the subject).
Here you go:



Another problem with this controller-time we live in is that the  
controllers are mainly used for controlling a few parameters - you give  
the software some physical control, but you still have to point and  
click in order to create new sounds from the base.

  As Polardark mentioned earlier about the sliders and being intuitive  
through common standard I wanna talk about the restrictions about  
trying to bend this kind of approach. An example is the Alesis Micron  
which main controller and mean of navigation is an endless rotary-knob  
- these are common and most of use are used to use them. The problem  
with the Micron is that it is sized like a regular kind of knob, but  
works like most endless rotarys - i.e. You have to turn it multiple  
turns around it self to go from minimum value to maximum value. This  
multiple revolution turning is something that most of us are used to  
with endless-rotarys that are big in diameter and with a small negative  
bump to place the finger in while rotating the endless knob. Turning a  
knob sized endless rotary ten revolutions takes alot more time and  
takes away the good points with enless rotaries. Simple programming  
could make this design mistake disappear - just set the minimum and  
maximum value in the same 360 degree space. Due to low resolution on  
the endless rotary they chose to do it the original way. A much smarter  
approach would be to use the mentioned max 360 space but have a button  
for fine tuning each parameter.

  This leads us into the regular knobs vs. endless rotary-knobs  
discussion. A regular knob is high-resolution and is instantaniously in  
its visualisation once it has been used to alter the sound - an endless  
rotary can never on its own be used as a mean to visualize data. Once  
again I have to hail the genius of Clavia when designing these  
LED-collars for the nordlead3 and G2, also seen on the Behringer  
Midi-controllers and lots of digital mixers. This little collar of  
LED's makes the endless rotary in the form of a regular knob the  
perfect input and mean of value visualisation, much like the knobs on a  
non-programmable analog synthesizer. Moog Music have chosen an approach  
that is very interesting with the LP - the use of regular knobs and  
LED-collars - this IS a revolutionary approach when it comes to  
programmable synthesizers using regular knobs as you get a visual of  
the data programmed. I like this, simple things that shows alot of  
thought has been invested in the visualisation design.

  Back to the softsynths:
  Another problem with "knobs on screen" is that the instant visual  
status of 30 or more knobs is often intimidating for the human  
perception and we lack the ability to sort out how a sound is  
constructed instantly thereby making the high number of knobs more of a  
waste of screen-real estate instead of making a good use of it. If we  
instead decide to visualize the parameter-data at a small part of the  
screen-real estate just as visuals and use a flexible interface that  
allows us to edit each part separately (I guess this what was you were  
trying to show me WilsonTheRocker). The number of data that is  
instantly visualized should be managable by the user either he is  
professional or a beginner - once again - flexible interface is the  
key. You could get a first view of overview parameters - as an example  
try to thing about the layout of a Yamaha CS5 - but with an ability to  
zoom in on each sector and get offered a very extended set of  
parameters. I think flexible leveled interfaces may be a key here. At  
all time though we could use a rectangular area at the top or bottom of  
the window (think 600px wide and 20px high) to visualize all of the  
parameters in forms of bars, colors or whatever for the hardcore users.

  The key with controllers is the same - flexability. Having 100 knobs  
won't make your sounds made faster than 15 knobs that are used in a  
clever way. Once again - think about the previous example. Let the  
knobs in there initial state form an easily overviewable architectural  
view - once again think about the CS5. Then you should have the ability  
to change view and get the fifthteen knobs to control for example just  
the oscillator settings. We would need at least one screen telling you  
what parameters you change or even better one for each knob (like the  
new Novation and Korg controllers) - combine this with the LED-collar  
and you are getting close to something usable. The problem however is  
the human mind - we get used to one thing looking in one way have one  
function - we are not as flexible as this kind of human interface. We  
need a more flexible control surface visually - and you guessed it -  
this is were the Lemur and its likes come in. With a multi-touch-screen  
we are able to let the interface change in a visual way and this is  
really the final frontier in the interation between the human and the  
machine in our line of interest! The design can be flexible,  
overviewable and complex at once - the means of interaction is ready to  
not only visualize the status of the parameters, but also to let us  
interact with them in a direct fashion. To sum it up, we need to see  
the Lemur and its likes as more than just live-control surfaces, but  
more as a new way of Human-Computer Interaction.

  Hope I didn't loose you along the way!
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: text/enriched
Size: 5966 bytes
Desc: not available
URL: <http://synth-diy.org/pipermail/synth-diy/attachments/20060415/8bcafef3/attachment.bin>