Best Project or Innovation

matt3o · 28 Nov 2018, 11:09

Best Project or Innovation

Sponsored by

Elecplus

Vote for a chance to win a vintage keyboard from Elecplus.

Final vote!
This is the final vote for Best Project or Innovation. What is the best keyboard released this year by the community or a company? What's the most sophisticated switch that has been developed? What the most interesting project the community has conceived?

You can change your vote at any time up until 2018-12-03 19:00:00. Contrary to previous DTA editions, this is the final vote, there is no "vote for nominee" phase.

The nominees are

Afferscheim's homemade keypad

Afferscheim's homemade keypad. It's not every day that rotating knobs are integrated in such an elegant manner to a numpad or a keyboard

-----------------------

The DMA's Common Sense Controller

The DMA's Common Sense Controller. While already being nominated last year I think it deserves another nomination. The inclusion of bluetooth (bluetooth model F and beamspring) and the topre controller makes it worth it for this year. Also DMA's dedication to the project is exceptional.

-----------------------

Hyper 7 build by jae-3soteric

Hyper 7 build by jae-3soteric... because it's... huge.

-----------------------

Bucis' gaming keypad

Bucis' gaming keypad. Because it's a very nicely polished, innovative project that's more of a glove-like input device rather than yet another keypad.

-----------------------

UHKB

UHKB It has always been a cool project, and it turned into a super professional product.

-----------------------

sealclubber's 3d printed mammoth (sb147)

sealclubber for his 3d printed mammoth (sb147). Everything was designed from scratch and opensourced. A huge undertaking and a truly unique project!

scottc · 28 Nov 2018, 12:53

DMA 100%. A flexible and super cheap capacitive controller - what's not to love?!

seebart · 28 Nov 2018, 13:03

scottc wrote: ↑DMA 100%. A flexible and super cheap capacitive controller - what's not to love?!

Agreed and voted!

ZedTheMan · 28 Nov 2018, 18:11

Though I think I prefer the quality of the Hyper 7, sealclubber's work is truly outstanding, considering it was built from scratch. Insane dedication in that thing.

sb147, more like SCP147.

Myoth · 28 Nov 2018, 18:13

DMA's controller makes me sweaty in the pants ... Topre custom when ?

Chyros · 28 Nov 2018, 20:42

Gotta go for the Hyper 7. What a board!

jae-3soteric · 29 Nov 2018, 01:15

Chyros wrote: ↑Gotta go for the Hyper 7. What a board!

Thanks for the kind words - I'm super stoked to see it here!

Muirium · 30 Nov 2018, 13:05

Common sense guided my decision. DMA’s the new Xwhatsit. Both doing things some ways beyond my understanding, but just what so many of us need in real use.

Menuhin · 30 Nov 2018, 16:23

What DMA's Common Sense Controller does?

After reading the 1st post following the link, I still cannot get it: "I'm using the keyboard as capacitance to voltage converter... (now) to be analog in and digital out at the same time"

First of all, I cannot find the background of why this is created, i.e. what problem common keyboard controllers have and why it is bugging some people, and how this creation has solved such a problem?
Sorry, I am just not an electronic person to comprehend that quickly.

Menuhin · 30 Nov 2018, 16:29

I still have high hope for the UHKB, however, when I click on "Trackpoint module", it is written in its description:
"This item is under development, and we don't have a solid ETA yet. Please subscribe to our blog or newsletter, and be part of the development process."

But it allows me to add that to the purchase cart and have my money taken with that option.
What is going on? Really need to some update about it.

If it sports the function keys on the left and a good metal case, I bet some people venting about the Korean VE.A version 2 now may hop onto this one.

hansichen · 30 Nov 2018, 16:47

Menuhin wrote: ↑What DMA's Common Sense Controller does?

After reading the 1st post following the link, I still cannot get it: "I'm using the keyboard as capacitance to voltage converter... (now) to be analog in and digital out at the same time"

First of all, I cannot find the background of why this is created, i.e. what problem common keyboard controllers have and why it is bugging some people, and how this creation has solved such a problem?
Sorry, I am just not an electronic person to comprehend that quickly.

I'm a noob in this regard as well so if I write bullshit other people should correct me

With a normal keyboard (eg Cherry, Alps, normal membrane boards) you have a normal circuit. If you press the key electricity can go through the switch and the controller recognizes that this key is pressed. With a capacitive system you have an ongoing level of electricity on the pcb and when pushing down a key the amount of power that goes through this pad changes. So instead of "either way there is power or not" you have to scan the pcb for a change of electricity. This is of course a more complex system.

These days mainly Model F/Beamspring and Topre uses this technology. With model F and beamspring keyboard the xwhatsit controller got famous cause it made many boards usable. With dma capsense this is pushed onto the next level. DMA's controller works on cheap hardware (15$ programming kits which should be available worldwide, while an xwhatsit controller can be up to 78$ shipped), it can be used universally for many different boards and it has a lot of new features like bluetooth, big beamspring support and more. It pretty much opens a new world to custom model F projects (I think that's where the project started), custom topre projects and many other things.

Menuhin · 30 Nov 2018, 18:25

@hansichen

Thanks for the elaboration.
In any fund raising pitch, there's also the why question. I can see the DMA being unique and powerful to turn any board into Model F or Topre like, i.e. capacitive, and even potentially like having analogue electronic piano keys; on the other hand, it seems to have a narrow niche because it is opening a world to only to builders who fancy capacitive mechanism. See layman like me can't even see was it does and what the big difference it makes to the keyboard world.

Muirium · 30 Nov 2018, 18:32

People also use Xwhatsit’s controller as a quick and easily programmable custom keyboard controller for regular non-capsense switches. It’s meant to be capsense only but I’ve seen it used to drive straight up contact switches, like Alps and MX. Sure, it wasn’t designed for it, but sensing binary switches instead of analogue is a piece of cake.

CommonSense could do the same work, I assume. But I haven’t seen it used this way, because Xwhatsit’s real power was its setup utility and programming GUI. For those of us who think graphically, Xwhatsit sure made a wonder.

CommonSense is the more capable controller. It’s just not as plug in and go, from what I gather.

Hypersphere · 30 Nov 2018, 21:45

Was torn, but had to go with the Hyper 7.

wcass · 02 Dec 2018, 04:43

With CommonSense, you can do capacitive buckling spring in any layout. I am trying to decide what to do next; classic SSK, tenkey pad, 60%, 40%, or ergonomic.

Hak Foo · 02 Dec 2018, 04:52

I like the Hyper7 because it seems like a joke brought to life. (The same reason I now have a replacement PCB in my AT101W)

JBert · 02 Dec 2018, 15:22

Menuhin wrote: ↑I still have high hope for the UHKB, however, when I click on "Trackpoint module", it is written in its description:
"This item is under development, and we don't have a solid ETA yet. Please subscribe to our blog or newsletter, and be part of the development process."

But it allows me to add that to the purchase cart and have my money taken with that option.
What is going on? Really need to some update about it.

If it sports the function keys on the left and a good metal case, I bet some people venting about the Korean VE.A version 2 now may hop onto this one.

The keyboard is just out of its crowdfunding development stage but the extra modules aren't. No idea why they are pre-orderable from a store when the UHK team haven't finished developing them yet, seems like they could get in hot water should anyone start a court case for not having received the goods they ordered.

JBert · 02 Dec 2018, 15:28

Still voting for the UHK team, even if it won't matter so much anymore.

The keyboard and its firmware are brilliant, so far the only complaints I have is that the extra mouse modules aren't there yet, and that it lacks an extra key on the bottom right where now a big 2.25u button sits.

__red__ · 02 Dec 2018, 22:46

I have to stick my nose in here and explain why the Common Sense controller is such a big deal.

In all the previous controllers PCB designers had to be very, VERY careful about how they did board layout because all the previous controllers only had one "threshold" for all the keys.

This meant, that as someone who designs these types of PCBs before capsense it would take me 2,3, sometimes even four revisions to get a PCB that would actually work. With each revision costing a minimum of $50 and several weeks for delivery this controller was a GAME CHANGER.

It's a game-changer because you can set individual thresholds PER KEY which means you don't need perfection.

In fact - look at this:

: faxe.jpg (1.3 MiB) Viewed 91108 times

You couldn't get a more irregular and less predictable keyboard than that, yet Common Sense instrumented it without an issue.

GAME

CHANGER

(and it's off the shelf hardware)

Muirium · 03 Dec 2018, 09:31

Exactly. Why give the award to a spoof when there’s such a quality contender which benefits so many real projects? Common Sense isn’t just a great piece of work, it’s an enabler of so many more.

Wodan · 03 Dec 2018, 10:02

Would really love to see DMA win this. The project always reminds me of the bright side of the Model MF project and that amazing projects still happen on DT.

Khers · 03 Dec 2018, 10:51

DMA all the way!

Tobbbles · 03 Dec 2018, 13:35

H7 takes the biscuit here - it's so otherworldly.

__red__ · 03 Dec 2018, 17:02

Tobbbles wrote: ↑H7 takes the biscuit here - it's so otherworldly.

Any other year I would agree with you...

But the amount of time / effort / development that DMA has selflessly put into Common Sense is insane. I mean, the guy has this working on a devkit that it sold directly from the manufacturer - he explicitly chose to empower the community over making a business out of it.

He easily has thousands of hours in this project and he directly works with and supports us PCB developers, sometimes live on irc to help us get shit done.

"On the shoulder's of giants".

He's a giant.

__red__ · 03 Dec 2018, 17:04

__red__ wrote: ↑ In fact - look at this:

Needless to say I probably outed my secret-santa gift here so let's keep it between us girls in the members-only section eh?

__red__ · 03 Dec 2018, 17:07

(Also - I just signed up for this year so if you like the idea of getting something completely unique in the world you have a one in six chance of being paired up with me)

hansichen · 03 Dec 2018, 17:34

The whole idea of a Hyper 7 board is rather old, here are two old builds, one by 7bit himself:

The Hyper 7 by Jae is a good evolution of this idea and a nice project. But imo it can't compete with the possibilities and dedication of the common sense controller. Your numpad/macropad looks really nice __red__, I think even without this hint we would have noticed that it was made by you

It's nice to see this project getting more and more used, keep up the work and post some of your future builds

Menuhin · 03 Dec 2018, 17:42

__red__ wrote: ↑I have to stick my nose in here and explain why the Common Sense controller is such a big deal.

In all the previous controllers PCB designers had to be very, VERY careful about how they did board layout because all the previous controllers only had one "threshold" for all the keys.

This meant, that as someone who designs these types of PCBs before capsense it would take me 2,3, sometimes even four revisions to get a PCB that would actually work. With each revision costing a minimum of $50 and several weeks for delivery this controller was a GAME CHANGER.

It's a game-changer because you can set individual thresholds PER KEY which means you don't need perfection.

...

You couldn't get a more irregular and less predictable keyboard than that, yet Common Sense instrumented it without an issue.

...

I know the vote is going to end soon but for the sake of getting some better understanding of how keyboard projects work, let me try to rephrase __red__ statement to see if I understand the significance of this controller:

In the design of (e.g. the matrix of) a keyboard circuit, the designer has to be careful to match the rows and columns, parallel and series so that each key has that "one threshold" for the signal of key closing.

By (somehow) turning all switches to capsense, it is not necessary to carefully come up with calculations and testings to that "one threshold" design - one can have more freedom and/or can just do a crappy joke and things can be programmed post-hoc afterwards using capsense.

How the DMA's controller while only connect to only the columns and rows yet be able to program for the characteristics of each key is not yet fully understood for me - probably with a column and a row fixed, the controller can already locate a specific key.

Freedom of keyboard layout design... hmm...

Spoiler:

jae-3soteric · 03 Dec 2018, 18:00

hansichen wrote: ↑The whole idea of a Hyper 7 board is rather old, here are two old builds, one by 7bit himself:

The Hyper 7 by Jae is a good evolution of this idea and a nice project. But imo it can't compete with the possibilities and dedication of the common sense controller. Your numpad/macropad looks really nice __red__, I think even without this hint we would have noticed that it was made by you It's nice to see this project getting more and more used, keep up the work and post some of your future builds

Whist I agree with your sentiment - I'm not sure I can agree with your terminology. For me the other two examples were just caps on a board, skipping some of the Hyper 7 caps and layout choices.

My build is the only one that has the true layout, with custom PCB, QMK Lisp port and the case / plate etc. Everyone just sees the large board and don't realise it actually functions like the LISP boards did, which is the major differentiation for me.

__red__ · 03 Dec 2018, 18:39

jae-3soteric wrote: ↑My build is the only one that has the true layout, with custom PCB, QMK Lisp port and the case / plate etc. Everyone just sees the large board and don't realise it actually functions like the LISP boards did, which is the major differentiation for me.

I freaking LOVE this board, and as someone who has done a whole bunch of fab and development I am in awe of this work too. I'm an old UNIX graybeard who remembers when half of these modifiers were still in common use on X11 terminals

The craftsmanship here is clear, as well as both the dedication and the research required to pull this off in a historical LISP way and you're right - anyone who sees this as just 'keycaps on a board' doesn't recognize the scale and scope here to pull this off.

This build is absolutely epic - no doubt.

The primary reason I'm voting for DMA is because it facilitates a whole new class of keyboard. I really, REALLY wish these two projects were in different awards as they are both very much deserving of community recognition.