Daniel Beardsmore wrote: ↑Hypersphere wrote: ↑Cherry mx switches are inherently linear …
People say this, but what do they mean?
What precisely is and is not allowed in a tactile switch?
There's a lot of variation in force curves. For example, Marquardt designed a switch where the force drops then hits a wall instead of rising and falling:
https://plot.ly/~haata/237
Or it can start out being forceless, before the wall, as with Tokai:
https://plot.ly/~haata/tokai
I would say that the issue with Cherry—if there even is one—is more that the magnitude of the tactile peak is comparatively low. Nobody calls Alps switches "inherently linear", but they are: Alps switches are linear, with a piece of metal stuck in to obstruct the switch. Not many switches are truly tactile, and those that are, are often less tactile than inherently linear ones. Mitsumi tactile switches for example, typically use a design with a horizontal spring that serves as both the return spring and the source of tactility, but these are widely panned for not being tactile! (Yet, if you press a loose switch, it's extremely tactile!) Membrane buckling spring feels pretty linear to me, and it is: most of the force curve is a straight line, with a drop near the end.
It's a much more complex than the Good Guys versus Bad Guys approach that people like to take against Cherry, which is why I've been contemplating the idea of an ASDR-like analysis of force curves to classify switches based on their fundamental principles.
Cherry switches are just more gentle, and honestly that's not a bad thing, as increased tactility is like your fingers engaging in a constant argument against keys that are trying their hardest not to be pressed. MX Blue offers just enough feedback, although it does seem that they take some time to break in and soften — I found them very harsh and jagged feeling for a long time. MX Brown is more targeted to the subconscious — I find that no matter how tactile the switch, nothing registers in my head from it (that is, I still don't know if a key registered from touch alone).
I completely agree with this. If you look at the force curves from Haata on plotly, Alps switches are just as "inherently linear" in that you can extrapolate the the beginning portion of their force curves and it is generally the same line you would get if you started at the end and did the same, ignoring bottom out. Those SKCM blues that we all want, same deal there. Actually, the only exception I can see is SKCM brown, which seems to have a pretty absurd force curve, but even then after all that tactility and that little baby bump every Alps fan knows and loves, a good ol' linear section brings it all home. So I would agree that alps are more
tactile than Cherry MX, whether you measure that by length of the bump, sharpness of the bump, highest peak of the bump, or area under the bump
. I would say that they are just as inherently linear or artificially tactile because the tactile leaf is as much of an evolutionary quirk as changing the slider shape. We can
remove it and the switch doesn't care. Alps and Cherry MX are much closer to each other than either is to buckling spring in this sense. Springs with some pizzazz in the middle, not really a bad thing.
While I'm at it.. The force curve for topre is kind of interesting. A good portion of it is linear, but it's in the middle of the travel and it's probably the longest negatively sloped stretch of force curve I've seen. So I could see how some people say they feel linear, and I could see how others might think they're very tactile, as the whole curve looks like you zoomed in on the tactile bump of some other switch. Ok, maybe an MX brown, but liberties must be taken sometimes
glad to bring you today's potshot at MX browns, the foremost tactile-linear switch.
