4mm travel - how did this become standard?

IanM

22 Mar 2011, 13:58

and is it critical for good feeling keys?

I've been thinking about the length of the keystroke on our favourite keyboards, and how approximately 3.5-4mm travel with the actuation point around 2mm is typical. How did we get to this point? is it an accident of history? I can imagine the how key travel of a typewriter was originally defined by mechanical advantage, and not necessarily for the luxury of the ideal ergonomic comfort.

Of those key switch types that do have lower travel, none are as highly regarded as the best full travel switches. Why did this status quo occur? Arguably there hasn't been a real incentive for keyboard manufacturers to choose or develop a more pleasing short travel keyswitch. Some users may prefer laptop style keys due to the short travel, but they are happy anyway with what's available.

Other ubiquitous devices like mice, game controllers and mobile phones have buttons that operate very comfortably with only 0.25mm-1.5mm travel. So why are we still using and preferring to use 20yr-30yr old switch designs that may be based on even older limitations, instead of seeing an evolution toward designs that are potentially more ergonomic, less fatiguing and faster to operate?

It's certainly not easy to throw away decades of ubiquitous typing skills, and we're conservative by nature so that's obviously a major factor. However Apple are now supplying short travel keyboards as standard, laptops are everywhere, tablets are altering the way we use our PC's, so perhaps the next generation will utterly reject full travel keys. Would they be right to do so? Is the full travel key provably the best? or do we just need a better short travel switch that is the equal of the best Topre and Cherry MX?

User avatar
webwit
Wild Duck

22 Mar 2011, 14:56

I think the main driver behind short travel is compactness, flatness and presentation, not comfort or ergonomics. Then they might have improved the short travel switches to make them feel crisp etc. before they get worn out, but that doesn't take away some inherent problems. It works with very light keys and no hand travel like the datahand. Because you won't bottom out or if you do, there is not much force behind it. But with a classic layout keyboard, it means bottoming out, and with little travel like that, it is not ergonomic at all. Nice to type on for 10 minutes, painful for 2 hours.

ripster

22 Mar 2011, 15:58


User avatar
daedalus
Buckler Of Springs

22 Mar 2011, 17:30

Doesn't the Selectric have 6mm of travel?

I always lol at IBM material describing the Model F as a low-profile keyboard.

User avatar
kps

22 Mar 2011, 22:29

daedalus wrote:I always lol at IBM material describing the Model F as a low-profile keyboard.
In the early '80s, the German bureaucracy proclaimed in DIN 66-234 (later ISO 9241-4) that a keyboard shall be no thicker than 35mm at the home row. No IBM Selectrics or beam springs need apply.

IanM

23 Mar 2011, 08:26

webwit wrote:I think the main driver behind short travel is compactness, flatness and presentation, not comfort or ergonomics.
I don't doubt it, but low profile is also claimed to be better for RSI prevention (although it would be nice to see some real proof if this is the case)
webwit wrote:inherent problems. It works with very light keys and no hand travel like the datahand. Because you won't bottom out or if you do, there is not much force behind it. But with a classic layout keyboard, it means bottoming out, and with little travel like that, it is not ergonomic at all.
keys bottoming out with a harsh impact is unpleasant, but if the switch is a more optimal resistance, and the bottom out can be controlled so that it is cushioned and not harsh, can the short travel become truly ergonomic? Apple et al with scissor switches actually require more force than the lighter MX and Topres. To me the issue looks like a lack of thorough engineering on the keyboard, rather than a limitation of the human hand to cope with short travel.

User avatar
webwit
Wild Duck

23 Mar 2011, 13:58

IMO having a desk of the right height is better for RSI prevention. A keyboard which stands 1cm less in height and claims it can thus better prevent RSI, is just marketing talk on the label with the idea to get your money.

I don't think light, short travel keys can be ergonomic without introducing problems. In fact, it is already hard to get light classic mechanical key switches of any travel which are ergonomic in terms of preventing bottoming out. You lose tactile feeling at some point, see Cherry MX Brown and MX Red which are bordering the limit. It is an inherent limit of these designs, and with short travel keys, those problems get bigger.

I also think this because of the Datahand and how it developed from the earliest patents. At one point they concluded classic mechanical switches were not working, and they designed and developed a magnetic switch for the purpose instead, so you have light, tactile, short throw switches. Since it is also designed for a situation where your hands don't move and where you are lightly tapping with your fingers, unfortunately you can't transfer this design to a full-size keyboard.

One could of course build a classic keyboard with microswitches and adjust your typing style by putting less force in the keyboard (to prevent bottoming out and feeling like you're typing on a plate). But commercial vendors will never do it, all their keyboards will hover around 55g activation force. If they make it 30g with microswitches, people will start to hurt themselves. They won't adjust their style for the same reason people didn't massively start typing in Dvorak many decades ago.

Findecanor

23 Mar 2011, 16:11

I remember reading some study which advocated key travel of around 4 mm .. but I think that it could have been influenced by the status quo. If the majority of people are used to 4 mm, and have adjusted their style to it, then it is likely that they will say that 4 mm is the best for them.

User avatar
webwit
Wild Duck

23 Mar 2011, 17:33

Meh, I bet the study had a better argument and didn't just plainly dismiss things in this way. Likewise I could dismiss your argument by saying a large number of people are used to 2mm, have adjusted their style to it, and wrongly claim that 2mm is best for them. I think it works better by supplying some hard facts. Such as the problem of light switches, low travel, and consequently bottoming out. If you take away the keyboard and start typing on your desk, and do it for a few hours, can you imagine your fingertips will start to hurt? So, no, I don't think 4mm is some random travel, I think it is the travel that works best while allowing for a better landing. Low profile switches, just like the tactile touchscreens of the future, are for amateurs, for consumers who type little.

ripster

23 Mar 2011, 18:08

So why is ALPS 3.5mm?

Because it's ALPS!
Image

User avatar
Daniel Beaver

23 Mar 2011, 18:52

I did not know that about ALPS. Do they activate at 2mm, or is that not a consistent figure among mechanical switches?

User avatar
sixty
Gasbag Guru

23 Mar 2011, 18:59

The blue NEC switches activate at what feels like less than 2mm, then the click comes at like 4. Its quite a confusing switch. I find them quite lovely, but its not easy to get used to them.

User avatar
kps

23 Mar 2011, 19:01

ISO 9241-4 says "The key displacement shall be between 1,5mm and 6,0mm. The preferred key displacement should be between 2,0mm and 4,0mm." The ISO key travel recommendations seem to be derived from a 1969 paper that recommended travel between 0.05 inches (1.3mm) and 0.25 inches (6.4mm): Kinkead, R. and B. Gonzales, Human Factors Design Recommendations for Touch-operated Keyboards. Final Report (Document 12091-FR), Minneapolis Honeywell, Inc., 1969.

I think it's a mistake to consider travel out of the context of force profile and activation point. It's pre-travel and post-travel that matter, not total travel. Pre-travel should be short (1.5 - 2mm, unless you count the ultra-light resting displacement of a Selectric); post-travel should be long, but have a force curve that ensures it is never fully used.

ripster

23 Mar 2011, 19:21

I'm voting for Pianos. 9mm but better tone than your typical computer keyboard. Plus tickling dem ivories beats those nasty lasered bumpy things.
Image

Findecanor

23 Mar 2011, 21:21

On pianos, the total travel is different from different manufacturers and different depending on where on the key you strike.
The activation point is also somewhere in the middle..

ripster

23 Mar 2011, 22:10

Yep. Different weights too. Mine measures kinda stiff on the RipOmeter. I should complain to the piano tuner.

User avatar
kps

23 Mar 2011, 23:46

Image
Early keyboards, pt. 2 by Marcin Wichary, on Flickr

… but seriously, I doubt pianos were a direct influence on many computer keyboards.

ripster

24 Mar 2011, 01:15

I'm just glad I don't have to ponder any more why keyboards should be purple and mice should be green.

User avatar
sixty
Gasbag Guru

24 Mar 2011, 01:23

Seems they didn't always agree on that. I have a few boards from the mid 90s that have orange PS2 plugs!

neo

25 Mar 2011, 02:13

I prefer short travel low profile keys. If only there were any high quality keyboards with such keys. I use Filco for high quality of construction and perfect layout.

User avatar
sixty
Gasbag Guru

25 Mar 2011, 02:22

neo wrote:I prefer short travel low profile keys. If only there were any high quality keyboards with such keys. I use Filco for high quality of construction and perfect layout.
http://geekhack.org/showthread.php?p=269396

:D

User avatar
webwit
Wild Duck

31 Mar 2011, 22:10

I have another theory which is not only related to travel but also to required force to activate a key.

I remember when I was a kid I sometimes used my father's IBM typewriters, and it was not about speed, it was about accuracy. Because first tipp-ex and later correction tape were finite and cost money. You couldn't just keep using the delete key and make many typos as you can with computers. Now if you compare this to a Cherry Brown board, they are pretty light and easy to press, but I also make more typos. It wouldn't have been an option at the time. The required force and travel of typewriters could be the optimal balance between speed and accuracy for the conditions of those times. Then later keyboards like the beam spring keyboards, the Model F and the Model M tried to simulate the feel of the typewriters, because that is what people were used to.

IanM

01 Apr 2011, 18:43

webwit wrote:I have another theory which is not only related to travel but also to required force to activate a key...
That's why I mentioned mechanical advantage when I started the thread. I was still curious about this so I dug out on old typewriter I knew we had somewhere. It's an old machine: 1950s or earlier, made by Underwood. Condition is not great as it's been stored for decades, and it probably hasn't seen oil in that time. However, the keys are actually very light in feel, much more so than I expected.


The typebars are very slender, so I guess they are very light and don't require much effort to lift. The full travel is approximately 17mm, I only measured roughly but I think by design it is intended to be 2/3 of an inch.

Post Reply

Return to “Keyboards”