Is "parabolic" a switch type?

[codemonkeymike]

Is "parabolic" a switch type? If so does it belong here? wiki/Keyboard_terminology#Switch_types

I personally think the "switch type" subgroup of Terminology is a bit weak but I am not sure if this is up for debate or not. A bit weak as in I don't think "clicky", "linear", and "tactile" really cover everything and are somewhat ambiguous

[Daniel Beardsmore]

That's a beginner's page, so I tried (albeit not entirely successfully) to keep the page as simple as possible. Parabolic is so rare that it's not worth mentioning directly within the page. Which comparatively common switches do you feel cannot be broadly classed as clicky, tactile or linear? As a beginner's guide, it's not meant to go into anything in depth.

The page does note, "other types are also described under [wiki]force[/wiki]", and you can see that both parabolic and the more common progressive rate are covered there. The force page does need more work though.

[codemonkeymike]

Does it make sense to have both a beginners page to terminology and a more complete page?
The issue I have with Clicky, Tactile and Linear is that Tactile and Linear are “feel” while Clicky is “auditory”. Also the three groups are not mutually exclusive and Clicky could be misconstrued as clackity linear or tactile switches. I think the wiki plays a part in how people ask about switch types and it may be of benefit to outline silent actuation/clicky actuation tactile/linear, just tossing around some thoughts right now. It may be as simple as having 3 groups, “Average Force”, “Change in Force”, “Auditory Action”

Where
“Average Force” - parabolic/linear
“Change in Force” - tactile/linear
“Auditory Action” - actuation click/non actuation clack/no auditory click or clack intended


Please note I am not a student of Physics, I am a computer scientist and have no idea if these are the right terms they are just placeholders for the right terms

[Daniel Beardsmore]

It's a prescriptivist vs descriptivist argument. The definitions on that page simply reflect the way the terms in question have been used for at least eight years. As a beginner's guide, it would seem logical to equip people to understand the terms as used.

Your terms need thought anyway. If there was no change in force, you'd not have any switch action. "Action" itself refers to what happens when you press the switch: does it spring back up, or remain down, and do the contacts disconnect when you release the switch or do they remain closed until the next cycle? You mean feedback, not action.

Linear switches are called linear because the force increase follows the linear plot y = mx + c (where c is the preload on the spring). The force changes, but it changes proportional to distance. This is a fairly hard concept to grasp as you don't perceive the force as changing, but it does — the oft-despised coin stacking will show this readily.

Tactile vs progressive rate is an awkward one, as we think of "tactile" as a drop in force only.

[codemonkeymike]

The terms we have been using and will continue to use (Clicky, Tactile, and Linear) should be taught but I think in the context of a higher level of understanding that these are just general terms of feedback of a much larger subject of feedback. As feedback (not action you are right on that is a better word for it) is Auditory and Physical senses together while the terms we have been using only lead us to confusion when actually trying to talk about how the switch acts when using it.

With the term linear if we were to take samples of the slope before a tactile feedback event (IE: cherry mx brown "bump") and samples of the slope after the tactile feedback event we would get the same slope. We have a bloody kludge of a word/s for it now "inherently linear" but alone tells us little about the switch feedback as well.

I am going to stop at this point and maybe make some graphs and new post before I go crazy thinking about "inherently linear tactile and clicky switches"

[Daniel Beardsmore]

I was contemplating an alternative way to broadly classify switches. The basic linear path can be thought of as an axis. This would make peaks positive, and troughs negative. You could think of four periods: lead-in, pre-actuation, post-actuation, and lead-out. "Zero level" would denote following this line (not just "linear" but following the axis directly).

Cherry MX Blue follows this axis for everywhere except pre-actuation, so it would become: 0p00 — zero-level lead-in, small positive pre-actuation, zero-level post-actuation, and zero-level lead-out.

IBM switches all broadly follow 0XXn — zero-level lead-in, no pre-actuation step, no-post actuation step, and small negative lead-out.

Alps is roughly: 0P0n — zero-level lead-in, strong positive pre-actuation, zero-level post-actuation, and small negative lead-out.

It would only be a very broad guide, and there are a lot of exceptions to consider, such as progressive rate, where the gradient itself changes. It may also be desirable to describe the shape of each stage, e.g. 0L is linear at zero-level, and nL is linear with negative c. (Imagine y=mx+c has been rotated anticlockwise.)


The point of the page you queried is certainly not to go into this level of detail. It just needs to explain the community standard terms (linear, tactile, clicky) fairly briefly and nothing more, just as a starting point — that page exists as a brief overview and a springboard to more detailed topics. There are other pages on the wiki where this can be explained in greater detail.

e.g.

[wiki]Force[/wiki]
[wiki]Clicky[/wiki]