One rather un-obvious property of most mechanical keyboard switches made today is that the tactile feedback actually does not match the actuation point - meaning, the point at which the switch "clicks" is not the same point where the PC starts registering the keypress. In fact, in e.g. Cherry MX switches, these two are handled by separate mechanisms! The only switch type where actuation == tactile feedback that I know of is buckling-spring, as in the IBM model M.
This realization was a big moment of disillusionment in (most) mechanical keyboards for me :)
Wow, thanks for posting this. I always felt like something was off in a subtle way with most mechanical keyboards I've used, but your explanation nails it.
Totally explains why I ended up settling on HHKB, which actually feels like it registers the button press right when you feel that "click" (given it is much closer to IBM Model M than to the usual mechanical switches).
Same here... I'm not too sure how the Topre switches in the HHKB do work but I went from years and years on Model Ms to Topre. I'm using HHKB since so long I don't even remember when I started using them. I tried all the Cherry out there but Topre is my favorite switch by very far.
As someone who likes linear black cherry switches this is not an issue. They have no bump.
Why do I like those linear switches? Because you can mash them hard (then you need no bump because the key bottoms out and that is your definite bump) or you can type them close to inaudibly silent. Which is kind of important if you e.g. do audio recordings.
Why black ones? I am a bass player, I have strong fingers and most other linear switches feel to light to me.
There are beam spring based switches that also have this property. There are even some MX keycap compatible ones available now but I think they are all require a hall effect PCB.
I guess some clicky optical switches may also behave this way?
with my blue switch keyboard, I cannot get a key to register before the "click" and cannot get a click without a key showing up (using my finger) so in a practical way, this really isn't a problem
Personally, I like this. I've since gone linear, but, when I was a fan of tactile keys, my favorites were the ones that placed actuation just before tactile feedback. That allowed me to type much more lightly, because I only had to press enough to get to the leading edge of the tactile bump.
I did once love my Model M, but it's more tiring to use than newer keyswitch designs are. Pushing enough to actuate the key means building up enough force to buckle the spring, at which point there's not really anything I can do to avoid bottoming out. It's sort of the double IPA of keyswitches: really, really, really good in small doses, but not necessarily something you want to be experiencing for 8 hours straight.
As far as I can tell, this is an intentional part of the design of the Cherry switches and is desired by many fans. It personally has never bothered me, even with Reds which don’t really have a tactical feedback point.
Topre has the same fundamental issue. The rubber dome buckles at some point, but the switch is activated by the spring deforming enough to change the capacitance above the pads.
Then again, in my experience, it's not an issue. The resistance curve of the rubber domes makes it hard to feel the buckle yet not activate the switch. Once it buckles, it's hard to avoid the "thock".
I too like my Topre boards. But I also haven't gotten a chance to try any of the newer Cherry-style third-party switches.
I think that it’s not really possible to conclude anything from the force charts. Switch feel is so much more than amount of resistance spring has. I’ve been testing switches that should be very similar according to charts, yet they felt totally different.
I think there are plenty factors at play here. Spring resistance, wobbliness of the switch, activation distance, feel of the tactile bump/click…
Recently I've been using Otemu brown, violet, and orange and Gateron browns. All should be quite similar, yet in the end they feel totally diferent. I love Otemu oranges and dislike brown and violet. Gateron browns feel very good, but a bit wobbly for my taste.
I have tested more than 20 switch types, and most of my guesses based on someone else opinion or specification were usually wrong. The only thing I know is that I want tactile switches, with rather low activation force. Except of that, I have no idea if I'll like the switch or not.
I like the way that author approaches this, but I'm not sure we can conclude anything from a scientific approach here. Switches are just like art or music. Sometimes you like a song you think you should hate. Yet it makes something to you and you just can't stop listening ;)
What’s worse, you need to swap the switches for the whole keyboard (or at least a significant portion) to effectively test them. Just getting single sample switches doesn’t really tell you how a whole keyboard of them will feel.
Did anyone do a blinded mechanical switch ABX test? I bet most would fail to differentiate except between the major categories - linear, buckle, dome, ...
I've been using mechanical keyboards since 1992. I can definitely tell the difference between major switch types, and I can also tell the differences between switches I'm familiar with. For the most part though, you're probably right that if you just sat down at an unknown keyboard and started typing you'd probably only be able to identify it's major type and whether or not you like the feel of it.
Feel is subjective, but it does matter, and it's something worth exploring for people who spend 8+ hours every day using a keyboard. It can also have important impacts on ergonomics.
> Feel is subjective, but it does matter, and it's something worth exploring for people who spend 8+ hours every day using a keyboard. It can also have important impacts on ergonomics.
Does this level of detail regarding feel really matter? I mean, all keyboards are usable and people do adapt to new/unfamiliar input devices, and it's not like laptops ship with high-action mechanical keyboards.
And regarding ergonomics, the standard keyboard layout is renowned for not being stellar ergonomics-wise, so I'm skeptical regarding how relevant ergonomics is in this field. In fact, isn't a standard Microsoft Natural ergonomics keyboard way better than your typical mechanical switch keyboard?
How noticeably have you found that switches age? I've been bashing away on this keyboard for about 7 years now, and I just got the latest model for work. Both apparently use Cherry Brown switches but the new one feels subtly 'squishy' compared to the old one, and I'm wondering if they've changed the formula or if I just need to 'run it in' and then they'll feel the same.
Also I wonder how much of the preference is like old running shoes - I don't care how ratty my old shoes are or how nice the new ones are, the new ones will feel wrong compared with the old ones for a while.
This is closer to musician instrument preferences. The audio component of switch mechanics isn't as important to me as the tactile experience. Tiny changes in the springs can be felt, your keys will feel too weak or stiff. The mechanics of the switch will feel broken, wobbly, or tight.
Check out high end electric pianos at your local music shop, and compare to $100 Wal-Mart kid keyboards. The resistance and smoothness and feedback from the keys integrate with muscle memory better (and in the case of pianos, graduated activation helps determine the volume of a note and sustain.)
I don't know if typing with good switches helps with rsi, but it certainly helps with comfort and satisfaction.
A good clarinet or trumpet or saxophone will have similar features to the feel of different types of mechanical keyboard switches. Sure, you can play decent music on cheap mass produced elementary school instruments, but it's going to objectively feel better playing a high end bespoke instrument. No jarring or disruptive aspects in the motion to interfere with the process of putting characters on the screen.
I don't think the comparison to digital piano keyboards is apt. The velocity sensitivity on a digital piano is of utmost importance, being able to "enter the MIDI velocities" accurately and consistently, often at very high speeds and correct timing and tempo is of utmost importance. And with cheap digital pianos this is borderline impossible to do well.
While a for a PC keyboard, a basic membrane keyboard will do the job outside of maybe some very niche highest levels of competitive esports
Mech keyboards and obsession with them are purely a luxury, hobby thing
I mean they are different in terms of force curve and all. Sure. I wouldn't tell cherry red and silent red apart just by their feel, and silver vs red might be hard as well, but all other types I had at home are pretty distinct.
And I am not a mechanical keyboard afficionado. I own one mechanical keyboard for well over a decade. But I had to do with cherry switches on a non-keyboard related electronics project.
Each switch does have a different feel based on the spring constant, even within the same category. It's often hard to tell from one keypress, but fatigue becomes a key differentiator when typing all day. I found that out the hard way when a switch felt perfect in the tester kit, but rendered my fingers dead tired after a full day's work.
In general, from the whole "mechanical switches" industry.
OP is fine, he's measuring things. The audiophile industry also always includes "frequency response" graphs and all kinds of other measurements in their reviews, yet the whole field is still a bit of a sham.
Constant subtle horn-tooting in his prose as well that's irritating. But from what I've seen on Reddit, there is a huge keyboard-phile community so I can understand the desire to clearly define the space. Although I don't know what they call themselves. Keyheads? Vectophiles?
I have had exposure to multiple different switch types over the years, not as many as enthusiasts however.
Personal preference being expressed here, while I can and have used buckling springs at length, they are not my favourite for two reasons. One, they click twice. Once when actuating and once again on the way back up so they are 2x as noisy. Two, they have a fair bit of resistance. Really, likely the best way to type on a buckling is to push down and let the key push your finger back up. It can be tiring, especially if you're not imprinted on this way of typing.
My favourites are complicated ALPS. Yes, I know they are no longer made with the exception of simplified whites (from Matias), but I have a lot of exposure to ALPS and am actually typing this on complicated orange ALPS right now. You can look at force curve graphs, but the typing experience is very different compared to buckling. When you press the key (the travel needed to actuate is not as far either compared to buckling), you get one click not two. The key does not need to push your finger back up because it's lower force, so when you hear or feel the click your finger can already be swinging away to another key.
MX is not as good as ALPS, but it has a large market share. Just like VHS wasn't better than Beta, but won anyway.
I haven't much good experience with buckling springs: my PC122 (battlecruiser) after just an year have most used keys usable only hitting them veeeery strongly, also unpleasantly feeling the second click on the finger, the old one same issue. It's not the switch per se but the arch...
Cherry, Gator, Outemu etc so far have not shown such short life behavior...
I miss ALPS switches. They were absolutely the best. MX is no better than nice rubber dome in my mind (which I use today) while being much less reliable.
So why not get some Matias switches (3 to choose from), or a keyboard then? You can buy a switch tester from them and see if you like the feel of this implementation of it. It's too bad they don't send it out with some random keys attached as an option. https://matias.store/
If you don't want to bother with the tester, the grey quiet-click are roughly equivalent to cream dampeneds and are good for use in an office environment so that neighbours don't burst into tears at your acoustic assault in the open-concept floor plan. If this isn't an issue and noisier/bouncier is more fun, then white is what you want.
Yes it's a nice illustration why you should not use a SaaS like plotly, especially considering that there are many options out there which could have put the plot into the post directly.
I want to know "will I like typing on this keyboard" without buying it first. I can't tell that from reading a dozen reviews, because people love that clacky IBM keyboard that drives me up the wall. Keyboard enthusiasts are not to be trusted. Graphs give me a much better idea of what I'd actually be getting.
> Graphs give me a much better idea of what I'd actually be getting
The issue with that is that you "have to have the feel" for the graph. Except for extreme circumstances, if you're not familiar with how to approximate the feel from the graph, it won't really help you.
I guess my point is that you still have to have some kind of actual, real-world experience.
In my case, I kind of deduced that I liked "bumpy" switches, so I went the "brown" route. I was happy with my Gateron Browns. When I wanted a second keyboard to not haul the first around, I figured Cherry Brown would be "close enough". They supposedly have the same specs and everything. Nope. They feel much "harder" to press.
I do like them enough, so I didn't send the KB back. But I do tend to miss some keys when I switch from Gateron to Cherry, and I do tend to press like crazy on the Gateron after switching from the Cherry.
I think the point is to be able to predict the answer without actually trying it. If I know I like X and Y, and I can see that Z has a very similar force curve, then it's likely to satisfy one of my preferences. I might have others - e.g. sound and key spacing - but that just suggests other measurements that should also be made. Having more detailed/accurate force measurements still seems like a good thing.
Readit News
This realization was a big moment of disillusionment in (most) mechanical keyboards for me :)
Totally explains why I ended up settling on HHKB, which actually feels like it registers the button press right when you feel that "click" (given it is much closer to IBM Model M than to the usual mechanical switches).
Why do I like those linear switches? Because you can mash them hard (then you need no bump because the key bottoms out and that is your definite bump) or you can type them close to inaudibly silent. Which is kind of important if you e.g. do audio recordings.
Why black ones? I am a bass player, I have strong fingers and most other linear switches feel to light to me.
I guess some clicky optical switches may also behave this way?
I did once love my Model M, but it's more tiring to use than newer keyswitch designs are. Pushing enough to actuate the key means building up enough force to buckle the spring, at which point there's not really anything I can do to avoid bottoming out. It's sort of the double IPA of keyswitches: really, really, really good in small doses, but not necessarily something you want to be experiencing for 8 hours straight.
Then again, in my experience, it's not an issue. The resistance curve of the rubber domes makes it hard to feel the buckle yet not activate the switch. Once it buckles, it's hard to avoid the "thock".
I too like my Topre boards. But I also haven't gotten a chance to try any of the newer Cherry-style third-party switches.
I think there are plenty factors at play here. Spring resistance, wobbliness of the switch, activation distance, feel of the tactile bump/click…
Recently I've been using Otemu brown, violet, and orange and Gateron browns. All should be quite similar, yet in the end they feel totally diferent. I love Otemu oranges and dislike brown and violet. Gateron browns feel very good, but a bit wobbly for my taste.
I have tested more than 20 switch types, and most of my guesses based on someone else opinion or specification were usually wrong. The only thing I know is that I want tactile switches, with rather low activation force. Except of that, I have no idea if I'll like the switch or not.
I like the way that author approaches this, but I'm not sure we can conclude anything from a scientific approach here. Switches are just like art or music. Sometimes you like a song you think you should hate. Yet it makes something to you and you just can't stop listening ;)
Did anyone do a blinded mechanical switch ABX test? I bet most would fail to differentiate except between the major categories - linear, buckle, dome, ...
Feel is subjective, but it does matter, and it's something worth exploring for people who spend 8+ hours every day using a keyboard. It can also have important impacts on ergonomics.
Does this level of detail regarding feel really matter? I mean, all keyboards are usable and people do adapt to new/unfamiliar input devices, and it's not like laptops ship with high-action mechanical keyboards.
And regarding ergonomics, the standard keyboard layout is renowned for not being stellar ergonomics-wise, so I'm skeptical regarding how relevant ergonomics is in this field. In fact, isn't a standard Microsoft Natural ergonomics keyboard way better than your typical mechanical switch keyboard?
Also I wonder how much of the preference is like old running shoes - I don't care how ratty my old shoes are or how nice the new ones are, the new ones will feel wrong compared with the old ones for a while.
Check out high end electric pianos at your local music shop, and compare to $100 Wal-Mart kid keyboards. The resistance and smoothness and feedback from the keys integrate with muscle memory better (and in the case of pianos, graduated activation helps determine the volume of a note and sustain.)
I don't know if typing with good switches helps with rsi, but it certainly helps with comfort and satisfaction.
A good clarinet or trumpet or saxophone will have similar features to the feel of different types of mechanical keyboard switches. Sure, you can play decent music on cheap mass produced elementary school instruments, but it's going to objectively feel better playing a high end bespoke instrument. No jarring or disruptive aspects in the motion to interfere with the process of putting characters on the screen.
While a for a PC keyboard, a basic membrane keyboard will do the job outside of maybe some very niche highest levels of competitive esports
Mech keyboards and obsession with them are purely a luxury, hobby thing
Deleted Comment
And I am not a mechanical keyboard afficionado. I own one mechanical keyboard for well over a decade. But I had to do with cherry switches on a non-keyboard related electronics project.
From the OP, or from the discussion here? Serious question. Because it's genuinely not clear which you mean (for me it's the latter).
OP is fine, he's measuring things. The audiophile industry also always includes "frequency response" graphs and all kinds of other measurements in their reviews, yet the whole field is still a bit of a sham.
Personal preference being expressed here, while I can and have used buckling springs at length, they are not my favourite for two reasons. One, they click twice. Once when actuating and once again on the way back up so they are 2x as noisy. Two, they have a fair bit of resistance. Really, likely the best way to type on a buckling is to push down and let the key push your finger back up. It can be tiring, especially if you're not imprinted on this way of typing.
My favourites are complicated ALPS. Yes, I know they are no longer made with the exception of simplified whites (from Matias), but I have a lot of exposure to ALPS and am actually typing this on complicated orange ALPS right now. You can look at force curve graphs, but the typing experience is very different compared to buckling. When you press the key (the travel needed to actuate is not as far either compared to buckling), you get one click not two. The key does not need to push your finger back up because it's lower force, so when you hear or feel the click your finger can already be swinging away to another key.
MX is not as good as ALPS, but it has a large market share. Just like VHS wasn't better than Beta, but won anyway.
Cherry, Gator, Outemu etc so far have not shown such short life behavior...
If you don't want to bother with the tester, the grey quiet-click are roughly equivalent to cream dampeneds and are good for use in an office environment so that neighbours don't burst into tears at your acoustic assault in the open-concept floor plan. If this isn't an issue and noisier/bouncier is more fun, then white is what you want.
Tai-Hao also sometimes makes ALPS keycaps.
The Problem with Mechanical Switch Reviews - https://news.ycombinator.com/item?id=13006608 - Nov 2016 (5 comments)
> Graphs give me a much better idea of what I'd actually be getting
The issue with that is that you "have to have the feel" for the graph. Except for extreme circumstances, if you're not familiar with how to approximate the feel from the graph, it won't really help you.
I guess my point is that you still have to have some kind of actual, real-world experience.
In my case, I kind of deduced that I liked "bumpy" switches, so I went the "brown" route. I was happy with my Gateron Browns. When I wanted a second keyboard to not haul the first around, I figured Cherry Brown would be "close enough". They supposedly have the same specs and everything. Nope. They feel much "harder" to press.
I do like them enough, so I didn't send the KB back. But I do tend to miss some keys when I switch from Gateron to Cherry, and I do tend to press like crazy on the Gateron after switching from the Cherry.