Would You Pay $300 For A Modern Capacitive Buckling Spring Keyboard?

[iri]

...and parak explained, why.

[Krogenar]

And like I say a lot, it comes down to the economies of scale. We're what, maybe a hundred to two hundred units sold? Call it 500 units throwing in with the rest of the world, maybe 750. And these are one-time sales, since if they're true to the F, they'll last 30 years.

...

So I have little hope we'll see it any time soon, unfortunately. It's just too damn expensive a proposition for pretty much anyone to undertake.

This, unfortunately.

Maybe someday the 3D printing will become more affordable; if it's like other printing technologies, it will. Interesting discussion of what makes a Model M or F 'feel' as it does -- the barrels were something I had not considered.

[fohat.digs]

The barrels have got to be it.

An F has 3 metal plates! (2 on the AT) and the one at the top is directly visible while you type.

Then think of a flimsy plastic sheet with "chimneys" coming up for the M, and the single metal plate is all the way at the bottom, under a rubber mat.

[wcass]

I’m working on replacement membranes now. Development is going well and it looks to be doable on the cheap but an hour or so of labor; this for a 60% board with a Model M barrel frame. But after that I want to use what I learn to finish the XTant which will feel like an F because it will have F barrels and steel plates. The cost of a kit to convert an old XT to a 10keyless should be < $100

[sth]

I would absolutely buy a small capacitive buckling spring board, but it would be really nice if there was a choice of weights for my dainty fingers.

Maybe it's just my imagination but I thought my F felt like it actuated easier than my M. (goes to check)

yeah i do remember that now.

[dante]

If fohat.digs and rootwyrm give it their stamp of approval then the answer would be: YES.

[fohat.digs]

If fohat.digs and rootwyrm give it their stamp of approval then the answer would be: YES.

Ha Ha Ha

I don't have $300 in my Paypal account just now, but I am not worried. This will never happen.

(definition of never = somewhere between 2 weeks and 2 decades)

But I doubt that a Model F style contraption will surface again. It is just too cumbersome and old-fashioned. There are a lot of finicky mechanical and electrical parts that must be married harmoniously by trial and error. It reminds me of Thomas Edison's old laboratory in New Jersey where all the machines were run off of leather belts running around pulleys on shafts at the ceiling.

The world operates on supply and demand. Demand here is small, and there is a reasonably good supply.

Since I am not flush with money right now, I cringe at a 3-digit price, but the fact that you can, with patience (a small number of weeks or months), buy very nice (if used, but hey) Model F keyboards for less than what "premium" new keyboards list for, shows clearly that there is adequate supply, albeit not abundant.

Since I am a "tinkerer" then buying cast-off pieces at bargain prices and refurbishing them, for me, is both a project and a "hobby" in itself.

If you just want to buy something "off the shelf" then you are out of luck. Get out your tool set, go on ebay, and wait it out.

[Parak]

I don't have $300 in my Paypal account just now, but I am not worried. This will never happen.

(definition of never = somewhere between 2 weeks and 2 decades)

But I doubt that a Model F style contraption will surface again.

I think you should be a little bit more optimistic! Although that timespan is reasonable :p

[rootwyrm]

I don't have $300 in my Paypal account just now, but I am not worried. This will never happen.

(definition of never = somewhere between 2 weeks and 2 decades)

But I doubt that a Model F style contraption will surface again.

I think you should be a little bit more optimistic! Although that timespan is reasonable

Nope. As I covered in exquisite detail above, it is just not a realistic proposition. It's too damn expensive. I mean, seriously, how many PhD level EEs do we have that will give away the time and money (and yes there is a lot of money involved) just to create new silicon for the capacitive sensing? And of course, it's a sensitive enough thing that any work they did would likely have zero reuse and require substantial modifications for any changes made.
You're also ignoring the fact that the size and assembly of the F is a critical element of the barrel plate (the upper) and just ripping the number pad off is going to change the feel. As well as missing the metal plates. The only upside to the metal plates is that they're simple so the tooling is easy to reproduce. Except you still need a high accuracy stamping setup - those machines cost upwards of $10K used and tend not to fit in your garage. Any element you change is going to change the feel of the board, period. If you even change the thickness of the metal frame, you've changed the feel. (It's not psychosomatic. It would affect the flex and X axis movement weight.)

And even if we had access to absolutely everything as is, do you have any idea how many prototype phases it'd take? I don't even want to guess beyond "a whole hell of a lot." And to properly evaluate any given prototype requires you have all of the previously detailed elements A-Z one step short of production ready. Which means they have to be same manufacturing process, same materials, same assembly method. And we're talking probably a dozen plus prototype runs. Every one requiring tweaks before the next.

So as much as I would love one, and would buy one, as a small scale project it is not economically possible. And as a commercial product, it's not financially viable. Even when I factor in prototyping equipment being used for multiple products and diversified offerings, the quantities required to get a return on investment are far higher than the number of keyboards likely to be sold or the price of the keyboard goes through the roof (which means less sales, which craters financial viability.)

[Parak]

Nope.

<snip>

I must respectfully disagree! For one, as mentioned, a proof of concept capacitive controller replacement already exists (not by me) and works, without requiring any custom silicon and uses generically available discretes. It just needs a fair amount more work to make it a bit more stable, and have it work with custom PCBs as well. This is where a capable EE/code monkey would be able to help.

Secondly, barrels have been duplicated (by __red__) via DIY molding. I do have a plan to getting barrels produced in larger quantities if needed, but that is quite far off since spares can still be had relatively cheap. This goes for springs and flippies as well. This is where a materials engineer would help a little.

Thirdly, the prototyping for the kind of board that the aforementioned PCB (by me) is designed for will be dirt cheap in material costs. Yes, this includes hold down plate and the back plate. I can't speak for the feel (that's what prototyping is for), but I'm ensuring that all barrels will be uniformly held down to the PCB, which as I recall the biggest issue in both Fs and Ms, prompting the need for bolt mods.

So, while not being viable as a commercial product, this can and will definitely get done... eventually.

[wcass]

the steel barrel frame could be plasma cut like one of these:
http://stores.hilltopwelding.com/-strse-31/HTW%2C-Hill-Top-Welding%2C/Detail.bok
that is $28 for 18 gauge 32"x12".

and if you can get the FEEL of an F using a well developed completely customizable conductive controller like TMK, why bother going capacitive?

[Parak]

and if you can get the FEEL of an F using a well developed completely customizable conductive controller like TMK, why bother going capacitive?

Well, a custom capacitive pcb should be relatively easier and cheaper to make than a custom membrane, for one, since you can use any run of the mill pcb fab. There's also the slight side benefit of NKRO, which actually helps me while gaming once in a while. But you'll probably be done with yours faster at this rate

[wcass]

i have figured out how to solve both issues. bottom membrane is a PCB. diode arrays are at the top of the columns (not covered by top membrane). the top membrane cost me about $2 and about 20 minutes of time. the controller is currently a teensy, but should be able to replace with Atmel chip and circutry as part of bottom membrane PCB fab. check out my "custom buckling spring membranes" thread in "building things together"