slider instead of rotary knobs / input devices separate to the electric element

It would be great also to be able to use sliders instead of the rotary knobs for the variable passive electric elements (e.g poti).

At a first glimpse it looks "easy", but maybe a broader generalization could be also interesting: I mean the separation of the mechanic input device and the variable passive electric element. So, the input devices "rotary knob" or "slider" would be moveable, scaleable and interchangeable, separately of the electic element.

It would be great also to be able to use sliders instead of the rotary knobs for the variable passive electric elements (e.g poti).

This has been requested several times.
There are some advantages of knobs over sliders and I don't see many advantages of sliders.
So this is in the "low priority" list of things to do: it doesn't add functionality and the usability is lower imo.
But this can be discussed, maybe there are things I can't see.

At a first glimpse it looks "easy", but maybe a broader generalization could be also interesting: I mean the separation of the mechanic input device and the variable passive electric element. So, the input devices "rotary knob" or "slider" would be moveable, scaleable and interchangeable, separately of the electic element.

This sounds interesting: you could have all the "controls" together in some place instead of having them all around the circuit.
How to implement this feature is the challenge, there is no way to do it in the current implementation of components.
Implementing "detachable" parts from a component can add lots of possibilities, but it's not straight fordward.

Some things seem easy but are indeed very complex to implement.
For example grouping components: I started to implement it thinking that it should not be too hard, but ended up being a nightmare.

My request stems from an situation, where a poti was used as an input device for manipulating things on a display.
Here, a slider input would be interesting in order to "emulate" more realistic inputs for games or scrollbars.

For the generalization, some solutions come into my mind (but all have some disadvantages)

1. Similar type of wire: the variable components have voltage inputs in order to change the variable input (like the temperature).
   Problems: users might confuse the wires and component connectors; more cable spaghetti, strange component appearance (voltage controlling a temperature).
   
2. Additional type of wire: The controls have (hideable) wires, which do not conduct a current and do not have a voltage, but only provide the connection between the controller and the component. Could also be used in order to connect e.g. a motor shaft to a poti.
   Problems: users might confuse the wires; more cable spaghetti; can a motor shaft also be connected to a LDR? Where is the end of the physical interpretation (so: does a normal resistor also provide a temperature or heat flow output? Shall there be "pure" mechanical/thermal components like a gear or a thermal capacitor, too?)
   
3. controller moveable like a shown label or value: Similar to other losely connected things of a component, the controller has its own context menu and interactions.
   Problems: no visible connection between controller and component; shall the controller move linked to the component?
   
4. LabView / CircuitJS like controllers: The controllers have a completely separate space in SimulIDE (e.g. separate window)
   Problems: where? how avoiding visual overflow? no visible connection between controller and component.
   

I'm not sure which of them is more practical and not mind-blowing in implementation - and whether it is necessary to have it as far away as solution 4.
As a physicist, it would be great to have number 2 - as a system engineer, it would blow up the number of cable and component types, in order to systematise (thermal, mechanical, informational, etc.) - as a coder, I would ask whether it is really as important for a microcontroller simulation... But maybe one could ignore the systematisation and/or start low.

Anyway.. at least scalable knobs (and sliders) would be nice.

My request stems from an situation, where a poti was used as an input device for manipulating things on a display.
Here, a slider input would be interesting in order to "emulate" more realistic inputs for games or scrollbars.

Ok, I see the use case, specially in the context of "Boards" (graphical subcircuits).

I see two parts, the first one could be implemented and used first to configure controls, later the second part could be added:

1- A set of configurable and reusable controls.
Currently there are some controls only used in one component like the "Joystick" used in KY23.
These could be available for other components.
Some other use voltage as input: SR04, these could use some of the controls.

This needs the implementation of some kind of "Control factory".
So any component can say: give me a control of this type with these characteristics.
Or an independent control can be created and later linked to other components.

These controls could provide generic information: 0% to 100%.
Then components should do whatever they need with this information: convert to resistance, temperature or whatever.


2- A system to link controls to devices.
Maybe the best aproach is some kind of "label", in a similar way than Tunnels.
So you create an independent control, you assign a label to it and any component can connect to it using the same label.

This could be used to connect one control to several components, for example to solve dual potentiometer:
https://simulide.forumotion.com/t764-dual-potentiometer-control

If a component is using a "control label" then it hides  it's own control and shows a label:

These are great ideas :-)

I like the control factory and the control providing a generic 0% to 100% (or better min to max, in order to cope e.g. with neg. temperatures).

For the second step "linking controls to components" In my honest system engineering opinion, it might be better to have separate tunnels (e.g. with different color and/or visual representation). This would avoid mixing volts with information. On the other hand, using the same might boost flexibility, since the control signal could also be amplified, filtered, stored etc. similar to voltages. I my honest coder opinion this would provide a great "hacking" tool to tweak the simulation.

That was a very rough idea, but it would work like this:

Those labels are not independent, they are part of the control or part of the component.
The labels have no pins, I used tunnels just to create an image, but they will not look the same.
So it is not possible to mix with voltages.

Controls exist as a new component, you can grab a control and add to the circuit, configure it and set a "Name" for it.

Then you have for example 2 potentiometers with their own controls.
There is an option in the potentiometer to connect to an independent control where you set a "Name".
When you set a name in the pot then it's own control disappears, a label appears and it gets linked to the control with same name.

That's a good idea - and more intuitive compared to (ab)using available tunnels. ;-)

This new feature can solve this issue:
https://simulide.forumotion.com/t1556-linking-components#6837