geekhack
geekhack Community => Keyboards => Topic started by: FunkTrooper on Thu, 06 August 2009, 17:52:59
-
Could this really be the future?
http://www.youtube.com/watch?v=PDI8eYIASf0&feature=player_embedded
Also on Ars Technica: http://arstechnica.com/hardware/news/2009/08/microsoft-hardware-unveils-pressure-sensitive-keyboard.ars
-
I dunno. Considering people's tendency to bottom out hard on keys, I don't think it will ever catch on. Likewise, if you take a half-way skilled typist, they'll learn to press the keys on their keyboard with just enough force to activate until it becomes an entirely subconscious process. If you make people stop and think about the pressure they need to apply at every key they press, it's going to slow them down immensley, probably to the stage where even Average Joe will get pissed off by it.
If it does catch on, it looks like me and my 'F are going to be stuck with eachother for the rest of our respective days =P
-
i guess it doesn't work well with buckling springs
for gaming purposes this would be nice, though
-
Again, it would be a mess, too easy to press a key with too much or too little force, causing you to do the wrong thing. It's a nice idea, but probably too imprecise for real world usage.
-
http://arstechnica.com/hardware/news/2009/08/microsoft-hardware-unveils-pressure-sensitive-keyboard.ars
says it has 8-bit pressure sensitivity? so its like a wacom digitizer for a keyboard? I don't see the point unless they plan to make the keyboard into a musical instrument, or maybe some gaming/typing tutor or maybe an rsi monitor. But either way to make it predictable a mechanical linear switch would be ideal. Honestly though it seems pointless, you don't want to change pressure or keep continuous pressure on a keyboard, an ergonomic game pad is more ideal for this (and i believe the ps2 pad initially had this).
-
I just saw this on ars as well.
Being a keyboard using fighting game fan my first thought was to try out Street Fighter with two pressure sensitive attack buttons instead of six normal ones. Could be fun.
EDIT: Also, I would like to try a keyboard where typing hard automatically holds shift for the keypress.
-
Interesting... This reminds me of some Ericson patent which talked about something similar to Matias' half-keyboard: the selected key would not depend on the spacebar but on the exerted force.
Also - and more importantly - did anyone notice that it seems to offer N-key rollover on a "simple" membrane keyboard?!?
Now mass-produce it and maybe other vendors would be forced to bring out more NKRO keyboards at a lower price.
-
I can't see this as anything more than gimmicky...considering how much my typing force can very in a day depending on the keyboards I've been using, this would seem like it would be quite difficult to use.
-
Also - and more importantly - did anyone notice that it seems to offer N-key rollover on a "simple" membrane keyboard?!?
Now mass-produce it and maybe other vendors would be forced to bring out more NKRO keyboards at a lower price.
Yes, I noticed, and I smelt a whiff of BS (and not the buckling spring type.)
I'm thinking this was a simulation.
-
Yeah, something's definitely screwy. There's no way that matrix should be able to do NKRO.
But the concept itself isn't completely BS, PS2 controllers can do this as well. For example Star Ocean III takes advantage of it.
-
(While typing away in MS Word)
Clippy: "Hi there! You seem to be typing with excessive force. You must be stressed. Perhaps you are angry about being inadequate in bed or facing a tight deadline... would you like to talk about it?"
(Minutes later...)
Clippy: "Hi there! You look like you're trying to throw your keyboard out the window..."
God save us all from Microsoft and their glorious "innovations". Their peripherals feel like flimsy garbage and their mice seem to have their buttons mounted on a hair-trigger. Unlike a Logitech mouse I can't even rest my fingers on a Microsoft mouse without accidentally triggering the switches.
-
A little time ago I was looking at this strange NKRO appearance of this keyboard to find out how many keys it could detect at the same time.
It turns out that NKRO is possible on a membrane keyboard by using resistors; Microsoft patented the following: Patent #6577250: Resistance based keyboard key discrimination (http://www.google.com/patents/about?id=HIkOAAAAEBAJ&dq=6577250).
It uses resistors in a voltage divider circuit for each row in a conventional matrix, but without the need for columns. Instead, the apply a reference voltage to the row and measure how much the voltage drops when keys are pressed. The number of keys you can discern then depends on the accuracy of your voltage sensor and the resistances your design uses.
More important is the fact that it doesn't depend on anything more than resistors in the keyboard circuit - something which is far more easy to apply to a thin membrane than your common diode. So while your controller PCB would likely need some more quality components, a cheap NKRO membrane should be possible.
I fear it ain't gonna be used in any product though, and if it ever hits the market it will probably be a rubber-dome design.
-
I was wondering about something along those lines, but I don't think it would work reliably enough for *typing*! It could work well enough for that demonstration though.
Maybe a hybrid could work - part matrix with resistive chains thrown in.
I still don't understand how capacitive matrices achieve N-key rollover, so I won't say anything is impossible.
-
Pressure sensitive keyboards exist for applications where the amount of pressure used on a key is used to control a significant variable. But they're synthesizer keyboards, for playing music rather than typing - thus, if you press down harder, the note sounds louder, as is the case with an organ.
Electronic pianos even use keys which imitate the ballistics of the hammers in a real piano.
As for rollover, I've read that typically a computer keyboard will produce ghost keys for some unusual key combinations, because they generally do not use diodes, but through a choice of which positions in the matrix of keys to leave unused, this is kept from causing significant problems. I don't know the details, though.
-
I know that, rather my point was that they don't use a "matrix" and thus achieve a higher rollover limit.
-
Ghost keys are no problem at all and a nice marketing ploy used by the manufacturers. Ghosting is easily circumvented by software. I know of no keyboard that showed ghosting. You will get ghosting if you don't program the controller correctly, but as everyone does, this is kind of a non-problem.
-huha
-
Yep. The only thing one typically sees of ghosting is the effect of a correctly implemented controller, i.e. blocking. Possibly this even is what they refer to, but then they'd have the terminology wrong.
Seems like heavy boards give me the blues today, didn't care for M or clears, now I'm on blues.
-
Yep. The only thing one typically sees of ghosting is the effect of a correctly implemented controller, i.e. blocking. Possibly this even is what they refer to, but then they'd have the terminology wrong.
If a signal is sent down row line A, and the controller chip receives signals back from column lines 3 and 7 because keys at A7, B3, and B7 are pressed, that's ghosting.
That's true even if the controller chip "remembers" that it saw signals from column lines 3 and 7 when it activated row line B the last time, and so knows that something funny is going on, and does not send "keypressed" signals for all four keys, including the ghost key, to the computer.
So the terminology is not wrong: it is the phenomenon of ghosting (a phenomenon internal to the circuitry of the keyboard itself) that prevents the keyboard from reporting to the computer, in all cases and circumstances whatsoever, the exact combination of keys currently depressed upon it, no matter which keys they are, and no matter how many are held down. The standard cure for that would be putting a diode in series with every key switch. Then you could use your typewriter keyboard to play piano music - with suitable computer software.