Author Topic: A symmetric stagger, ergo design  (Read 946 times)

0 Members and 1 Guest are viewing this topic.

Offline Gorbon

  • Thread Starter
  • Posts: 17
  • Location: Greece
A symmetric stagger, ergo design
« on: Mon, 22 February 2021, 05:03:09 »
I've been having an itch lately for a new ergo design, so I thought why not give it a try.

I took the following ingredients:
  • The number of keys from a TKL (87)
  • An inexpensive and easy to find extended keycap set (e.g. enjoyPBT 117 key set)
  • A comfortable home-row hand placement with:
    • arms angled and floating
    • straight wrists
    • palms slightly rotated to face each other
    • fingers curled and relaxed

I tried to bring all the above together by placing the keys in comfortable positions, while keeping the keyboard as compact as possible (lengthwise at least). This is what I came up with (don't pay too much attention to the Mac layout and the layers):

262752-0
  • Fully symmetric design that allows for better use of the thumbs and doesn't overload the right pinkie (brackets, braces, etc, are under a thumb activated layer)
  • A width of 16u (a right-hand mouse would be 8u away from the center, whereas in a 60% it's 8.25u away, due to its asymmetry)
  • A curved home-row (a,s,d,v - m,k,l,;,) seems optimal here. Perhaps a non-qwerty layout might suit it better (Colemak?)
  • There is room for a 0.91" OLED display near the top (the ghosted key there)
  • The "tab" and "return" keys at the arrow cluster, are rotated 1.25u bottom row keys (not sure if this is a good idea, otherwise they would be replaced by 1u keys)
  • There's 2.25u of separation between the two halves
  • Needs 5 extra keycaps over a 104-key set (4 x 1.5u bottom-row and 1 x 1.75u home-row). It should be easy to find these in an extended set
  • No need for stabilizers (yay!)
  • All the legends wouldn't be correct (unless you go for blank keys), but at least the key profiles would be
So what do you think? Should I go ahead and build this thing? Is there anything obvious I'm missing?

Thanks for reading.

PS: I think I'm gonna call it Syster87 (Symmetric Stagger Ergo, 87 keys). Keyboards need fancy names these days.

Offline nevin

  • Posts: 1125
  • Location: US
Re: A symmetric stagger, ergo design
« Reply #1 on: Mon, 22 February 2021, 07:05:42 »
very interesting. very few 14 column+ ergos out there.
couple suggestions....
- in the programming, slide the numpad over to the left one column
- offer a variant without dedicated F-keys
- i'd take (3) 1u rorated thumb keys over a 1u & a 1.5u (i like to chord modifiers, putting shift in between ctl & alt (or cmd & opt on mac) so you can easily hold two keys with one thumb
- the next version, you could do a truly split version
Keeb.io Viterbi, Apple m0110, Apple m0120, Apple m0110a, Apple 658-4081, Apple M1242, Apple AEK II, MK96, GH60/Pure, Cherry g84-4100, Adesso AKP-220B, Magicforce 68

Offline Gorbon

  • Thread Starter
  • Posts: 17
  • Location: Greece
Re: A symmetric stagger, ergo design
« Reply #2 on: Mon, 22 February 2021, 08:57:33 »
Hi nevin and thank you for the feedback.

- in the programming, slide the numpad over to the left one column
Yes, I had it like that initially, but it's harder to hit "0" with a thumb that way, so I moved it to the right.

- offer a variant without dedicated F-keys
Yeah sure; I'd have to get around building this one first though.

- i'd take (3) 1u rorated thumb keys over a 1u & a 1.5u (i like to chord modifiers, putting shift in between ctl & alt (or cmd & opt on mac) so you can easily hold two keys with one thumb
I've been going back and forth about the thumb keys. My thumbs fall right in the center of those 1.5u keys and to fit 3 keys, I'd either have to remove the arrow cluster or move everything to the outside, which isn't ideal. Also, I've found that a 1u key for space is too small for my thumbs to hit accurately.

- the next version, you could do a truly split version
There are so many well designed split keyboards out there that I don't think I could come up with anything better. I'd like to think that in this design, the increased hand angle that the row and column staggering allow, combined with the 2.25u hand separation, offer some of the benefits of a split keyboard in a one-piece unit.
« Last Edit: Mon, 22 February 2021, 09:16:00 by Gorbon »

Offline jamster

  • Posts: 916
  • Location: Asia
Re: A symmetric stagger, ergo design
« Reply #3 on: Mon, 22 February 2021, 09:54:15 »
Just throwing this out there- you can pretty much prototype this, and a whole load of different layouts, with a couple of Dumang DK6 boards placed next to each other (if you wanted a single board instead of a split). Though the up/down arrows and F7 would have to be slightly off to either side.

Offline Gorbon

  • Thread Starter
  • Posts: 17
  • Location: Greece
Re: A symmetric stagger, ergo design
« Reply #4 on: Tue, 23 February 2021, 03:43:13 »
Just throwing this out there- you can pretty much prototype this, and a whole load of different layouts, with a couple of Dumang DK6 boards placed next to each other (if you wanted a single board instead of a split). Though the up/down arrows and F7 would have to be slightly off to either side.
Yes, I've seen that and it's a great idea and implementation, but availability, pricing, shipping/importing can be problematic, especially in my corner of the world.

But in order to prototype a design you don't really need to get too fancy. You can export the image from KLE, print it in 100% scale, glue it on a foam board, stick the switches on it and type away. I would encourage anyone even a little interested in ergonomics or/and keyboard design to give it try. Float your arms in front of you in a comfortable position, keep your wrists straight and fingers relaxed and track the natural movement of your fingers onto a board. You'd be surprised at what you'd be able to discover and figure out how even popular "ergo" boards (there's one that slips my mind right now, starting with "AL" and ending with "ICE") are actually pretty poor designs. You could try the same on a small one-piece ortholinear keyboard and you'd probably arrive at similar conclusions.

Also, high-speed video recording seems to be a common feature now on phones. You can point the camera on a specific area of the keyboard, type for a little while and watch in slow motion how you press the keys. This is why I added the gaps to the keys pressed by the ring and pinkie fingers. I noticed that I was consistently pressing those keys way off-center.

Finally, to go a bit on a tangent, some years ago when I started learning about keyboard design, I always assumed that row staggering is bad and ortholinear is good, but that's not universally true of every design. If you don't want to rotate the two halves (split angle) you can, symmetrically, row stagger the keys and that should get you part (most?) of the way there.

For instance on a standard keyboard, the row staggering on the right side helps align your hand with the keys, but of course it does the opposite on the left side. You need to twist your left hand unnaturally outwards in order to hit the reverse staggered top row and this is why touch typing lessons teach pressing "c" with the middle finger instead of the index. This is also why stepped caps-lock keys were introduced. With a twisted wrist, an overextended pinkie fans out further to left, causing mistypes. A stepped caps-lock is really a band-aid solution to serious design flaw.

My inspiration for this design were the Esrille, the NEC PC-9801 and the NEC Ergofit and only when I started testing their design choices, I appreciated and understood their elegance.

262874-0262876-1262878-2

Offline jamster

  • Posts: 916
  • Location: Asia
Re: A symmetric stagger, ergo design
« Reply #5 on: Tue, 23 February 2021, 05:45:04 »
Nice idea on the slightly spaced pinky keys... I just tried and see what you mean about hitting some keys off centre, and will try modifying key spacing.

Yeah, the Alice... imo it'd be better off not being called ergonomic at all. I don't even get the cosmetic appeal. It's really just overly twee, but at the same time clunky-looking.

Offline tp4tissue

  • * Destiny Supporter
  • Posts: 13172
  • Location: Official Geekhack Public Defender..
  • OmniExpert of: Rice, Top-Ramen, Ergodox, n Females
Re: A symmetric stagger, ergo design
« Reply #6 on: Tue, 23 February 2021, 06:04:48 »
The layout's fine,  but Flat is no go.  Gotta Tent.. 40-70* degrees

Offline Gorbon

  • Thread Starter
  • Posts: 17
  • Location: Greece
Re: A symmetric stagger, ergo design
« Reply #7 on: Tue, 23 February 2021, 06:57:27 »
The layout's fine,  but Flat is no go.  Gotta Tent.. 40-70* degrees
I agree; tenting* is best, but that would make the design significantly more involved (3 separate PCBs, complicated case) and decided not to dive that deep yet. Maybe on version 2.0 when my skills have improved.

*: But 70°? Damn! I'd probably need climbing holds instead of palm rests to hang on.

Offline Findecanor

  • Posts: 4690
  • Location: Koriko
Re: A symmetric stagger, ergo design
« Reply #8 on: Tue, 23 February 2021, 08:10:35 »
... (3 separate PCBs, complicated case)  ...
One solution could perhaps be to make it as one PCB in three sections, but with diagonal routed channels and "mouse bites" keeping them together. Then the builder could choose between building it flat or angled.

If you do that, then you might have to sacrifice the F-key in the top middle, and number the F-keys to the right one less, but then you'd also get F8-F12 in the traditional group that people are used to.

There are a bunch of existing DIY keyboards that have optional break-away sections with mouse bites. For example, on the Corne, the outermost 1u columns are optional and break away.

When you lay out the matrix, to save on wiring between separate parts, the keys of each sub-PCB should best be on a sub-rectangle in the logical matrix, with the rectangle shaped as close as possible to a square.

My personal opinions on the layout:
* I think that you should move the middle-finger columns up to the line of the ring finger columns: because that is the longest finger.
* Then move the Tab row in on both sides to get a symmetric uniform 0.25u stagger: It would make it more traditional, and the offset between middle, ring and pinky columns would actually be closer to that of a columnar keyboard with proper offsets, only rotated. (I have measured and compared against (pictures of) the NEC PC-8801-KI)
(I have actually obsessed for years about a layout based on similar ideas, but being two-part and lacking the centre clusters.)
Man must shape his tools lest they shape him
-- Arthur Miller

Offline Gorbon

  • Thread Starter
  • Posts: 17
  • Location: Greece
Re: A symmetric stagger, ergo design
« Reply #9 on: Tue, 23 February 2021, 17:39:54 »
* I think that you should move the middle-finger columns up to the line of the ring finger columns: because that is the longest finger.
* Then move the Tab row in on both sides to get a symmetric uniform 0.25u stagger
From a comfortable hand position (straight wrists, palms slightly facing each other) and starting from the curved home-row without moving my hands (the red dots are where my fingers naturally rest), I've tracked the paths of my fingers on the keys.

262906-0

Due to the palm rotation relative to the keyboard (supination):
  • the finger paths form arcs
  • the keys are pressed with the side of the fingertips
  • The index finger path is shorter, because it hovers higher and doesn't have the same reach without moving the hand
  • The pinkie has a large reach, because it sits very close to the keys
So it doesn't look like a uniform 0.25u top row stagger would work well here. But I think your suggestion would work nicely on a tented keyboard like the μTRON at a low (or no) split angle, because the tenting brings the key plane mostly parallel to the palms and now the finger paths form straight lines which would align with the row staggerring.

262908-1

One solution could perhaps be to make it as one PCB in three sections, but with diagonal routed channels and "mouse bites" keeping them together. Then the builder could choose between building it flat or angled.
Thank you for the suggestions, I didn't know all the different options and intricacies and will definitely have a look at that. Although I have to admit that it'd be cleaner/simpler/better to have two separate PCB designs (tented and flat). Also, as I mentioned above, a tented keyboard would probably need a different layout to perform optimally.

In any case, thanks again for the feedback and suggestions.
« Last Edit: Tue, 23 February 2021, 17:46:28 by Gorbon »

Offline wolverine92

  • Posts: 26
Re: A symmetric stagger, ergo design
« Reply #10 on: Tue, 23 February 2021, 22:20:34 »
I would suggest looking at the XBows. I don't use mine any longer since I moved to Split, but it has some design aspects similar to what you are going for here.

Sent from my Pixel 5 using Tapatalk


Offline Gorbon

  • Thread Starter
  • Posts: 17
  • Location: Greece
Re: A symmetric stagger, ergo design
« Reply #11 on: Wed, 24 February 2021, 05:24:41 »
I would suggest looking at the XBows. I don't use mine any longer since I moved to Split, but it has some design aspects similar to what you are going for here.
Yes, I like their offset pinkie columns and thumb keys, as well as the solid construction. I was initially considering getting one, but several other things put me off.

How did you find yours? Did you get used to it easily? Was it comfortable over longer typing sessions?

Some of the things I find problematic with its design are:
  • It's pretty wide at 18u+, impacting right-hand mouse ergonomics
  • It needs software to run on your computer for programmability (there is no Linux support)
  • Custom shaped keys, so no upgrades
  • Last but not least, this radial layout
262984-0

They took this concept of "your fingers fan-out when you stretch them" and run with it, but I don't think that this is very prevalent while typing (apart perhaps for the pinkie) and certainly not on a flat keyboard.

For this fanning-out to manifest, you have to type with your palms parallel to the keyboard, while overextending your fingers to press the keys without moving your hands (you'd need to have pretty small hands for this to happen). I don't think that this is comfortable and haven't seen people type like that (at least I don't). Also, fanning-out the index columns and especially the inner index columns, makes no sense to me and pushes the keys unnecessarily too far.

On my design, I did find that fanning-out just the pinkie columns helps ("q" and "p" are offset to the outside) and on the Esrille (tented, columnar keyboard) they also fanned-out the ring and pinkie keys. The ring finger probably didn't really need it, but I guess with the way they rotated their keys, it fit better in the design and also if you have small hands it might help.

262982-1

Anyway talking and theorizing is easy; I need to build a prototype and start testing it extensively, to make sure that this layout actually works.

Offline vvp

  • Posts: 830
Re: A symmetric stagger, ergo design
« Reply #12 on: Wed, 24 February 2021, 13:19:08 »
The best fan-out is no fan-out.
Fan-out only makes keys more far away from the home position.

Offline kurplop

  • THE HERO WE DON'T DESERVE
  • Posts: 985
  • "Losing the digital battle one digit at a time."
Re: A symmetric stagger, ergo design
« Reply #13 on: Wed, 24 February 2021, 15:11:40 »
The fanning never made sense to me either, for the same reason.

Offline wolverine92

  • Posts: 26
Re: A symmetric stagger, ergo design
« Reply #14 on: Wed, 24 February 2021, 19:27:46 »
I would suggest looking at the XBows. I don't use mine any longer since I moved to Split, but it has some design aspects similar to what you are going for here.
Yes, I like their offset pinkie columns and thumb keys, as well as the solid construction. I was initially considering getting one, but several other things put me off.

How did you find yours? Did you get used to it easily? Was it comfortable over longer typing sessions?

Some of the things I find problematic with its design are:
  • It's pretty wide at 18u+, impacting right-hand mouse ergonomics
  • It needs software to run on your computer for programmability (there is no Linux support)
  • Custom shaped keys, so no upgrades
  • Last but not least, this radial layout
(Attachment Link)

They took this concept of "your fingers fan-out when you stretch them" and run with it, but I don't think that this is very prevalent while typing (apart perhaps for the pinkie) and certainly not on a flat keyboard.

For this fanning-out to manifest, you have to type with your palms parallel to the keyboard, while overextending your fingers to press the keys without moving your hands (you'd need to have pretty small hands for this to happen). I don't think that this is comfortable and haven't seen people type like that (at least I don't). Also, fanning-out the index columns and especially the inner index columns, makes no sense to me and pushes the keys unnecessarily too far.

On my design, I did find that fanning-out just the pinkie columns helps ("q" and "p" are offset to the outside) and on the Esrille (tented, columnar keyboard) they also fanned-out the ring and pinkie keys. The ring finger probably didn't really need it, but I guess with the way they rotated their keys, it fit better in the design and also if you have small hands it might help.

(Attachment Link)

Anyway talking and theorizing is easy; I need to build a prototype and start testing it extensively, to make sure that this layout actually works.
If I had the skills to design my own, I absolutely would do that over buying one. I liked the XBows when I used it. The fan out wasnt really an issue and I didn't get finger fatigue. I just decided I wanted fewer keys and transitioned to a Lily58 and now a Corne. I don't see myself ever picking up the XBows. But it was a good starter for me.


Sent from my Pixel 5 using Tapatalk


Offline Gorbon

  • Thread Starter
  • Posts: 17
  • Location: Greece
Re: A symmetric stagger, ergo design
« Reply #15 on: Thu, 25 February 2021, 01:55:25 »
The best fan-out is no fan-out.
Fan-out only makes keys more far away from the home position.
On a standard keyboard, if I place my fingers on [ a s d c ] and [ , l ; ' ] (this curved, spread-out home position is comfortable for me) and extend/curl my fingers with straight wrists, "q" -with its offset position- is the only key that is right in the center of where a finger lands.

If I had the skills to design my own, I absolutely would do that over buying one.
Learning the tools (KiCad, FreeCAD, Fusion 360, etc) well enough to make a keyboard PCB/Case is actually not that hard if you are determined. It just takes time/effort.
« Last Edit: Thu, 25 February 2021, 09:45:43 by Gorbon »

Offline Gorbon

  • Thread Starter
  • Posts: 17
  • Location: Greece
Re: A symmetric stagger, ergo design
« Reply #16 on: Fri, 26 February 2021, 15:48:23 »
Found another recent project with a similar approach of designing around the finger arcs of a tented palm, taken to its logical conclusion. It's the Willow64.

263139-0

I think the concept is sound, but when you start rotating rectangular keys, the distances between the keys increase and it's hard to contain the size and shape of the keyboard, which obviously has happened here.

Nevertheless, it's very interesting and unique and the Japanese creator has a nice blog writeup. There's also some discussion in the creator's posts on reddit.

Offline tp4tissue

  • * Destiny Supporter
  • Posts: 13172
  • Location: Official Geekhack Public Defender..
  • OmniExpert of: Rice, Top-Ramen, Ergodox, n Females
Re: A symmetric stagger, ergo design
« Reply #17 on: Sat, 27 February 2021, 12:54:36 »
Found another recent project with a similar approach of designing around the finger arcs of a tented palm, taken to its logical conclusion. It's the Willow64.

(Attachment Link)

I think the concept is sound, but when you start rotating rectangular keys, the distances between the keys increase and it's hard to contain the size and shape of the keyboard, which obviously has happened here.

Nevertheless, it's very interesting and unique and the Japanese creator has a nice blog writeup. There's also some discussion in the creator's posts on reddit.


The layouts is impactful on FLAT keyboards because of how deeply the wrist has to rotate inwards.

With Proper Tenting, 45* Degrees and up,  you don't need these complicated arcs, it makes no difference.

This is the same principal which applies to the discussions on the ergodox's thumb section, where people complain it's too far from the core.

It's not too far,  it's only too far, if your wrist has to rotate inwards to hit it.  TENTED, it feels at the perfect distance.

Offline nevin

  • Posts: 1125
  • Location: US
Re: A symmetric stagger, ergo design
« Reply #18 on: Sat, 27 February 2021, 14:47:41 »
Found another recent project with a similar approach of designing around the finger arcs of a tented palm, taken to its logical conclusion. It's the Willow64.

(Attachment Link)

I think the concept is sound, but when you start rotating rectangular keys, the distances between the keys increase and it's hard to contain the size and shape of the keyboard, which obviously has happened here.

Nevertheless, it's very interesting and unique and the Japanese creator has a nice blog writeup. There's also some discussion in the creator's posts on reddit.

very interesting... never thought about it like that but yes, when flat i do see it. instead of coming straight back down the column, if your hand is more relaxed/normal positioning (palm arching towards center) you can shift almost a full column from row 3 to row 1 (if spacebar is row 0)
Keeb.io Viterbi, Apple m0110, Apple m0120, Apple m0110a, Apple 658-4081, Apple M1242, Apple AEK II, MK96, GH60/Pure, Cherry g84-4100, Adesso AKP-220B, Magicforce 68

Offline Gorbon

  • Thread Starter
  • Posts: 17
  • Location: Greece
Re: A symmetric stagger, ergo design
« Reply #19 on: Sat, 27 February 2021, 15:45:38 »
The layouts is impactful on FLAT keyboards because of how deeply the wrist has to rotate inwards.
With Proper Tenting, 45* Degrees and up,  you don't need these complicated arcs, it makes no difference.
Yes, you are right and we have been in agreement about tenting since earlier in the thread.

A tented, column and height staggered, split keyboard with concave keywells in a non-qwerty layout, is considered (although not universally) the holy grail. If e.g. you are an author typing millions of characters, you should definitely consider one of those. Although a stenotype would be significantly more efficient, not to mention speech-to-text has come a long way. And to think that there was a time, not that long ago, when people actually used pens to write books…

My goal here is not that holy grail (there are other keyboards aiming at that), but to find out how far can you push (ergonomically) a flat, one-piece keyboard, without requiring specialty keycaps, while trying to keep shape and dimensions contained. Apart from the Willow64, I haven't seen any other keyboards taking into account the -tented palm on a flat keyboard- aspect of typing.

This is the same principal which applies to the discussions on the ergodox's thumb section, where people complain it's too far from the core.
It's not too far,  it's only too far, if your wrist has to rotate inwards to hit it.  TENTED, it feels at the perfect distance.

That's a good observation and also putting the two halves further apart (shoulder-width apart) might also help, since that would require less/no tenting.

For this reason, I believe that tented and flat keyboards shouldn't have the same layout and if you design a keyboard that works well tented, you should offer it in a tented format, like the Esrille for example, and not as an optional extra.

Offline tp4tissue

  • * Destiny Supporter
  • Posts: 13172
  • Location: Official Geekhack Public Defender..
  • OmniExpert of: Rice, Top-Ramen, Ergodox, n Females
Re: A symmetric stagger, ergo design
« Reply #20 on: Sat, 27 February 2021, 17:11:18 »
the amount of tenting required is determined mainly by the elbow's height relative to the desk that the board rests on.

How far the arms are apart does not change the tenting angle by much.   

To bring your hands closer together usually means lifting your elbow, that's why you may feel like the tenting angle needs to go down,  but that's not due to the hand's distance from each other, it's due to the Elbow change