Kyria - new adventures in keyboards

Somehow, even though you think you got to the proper end of a journey, there is always another step to take.

Some steps sneak up on you too, like a valley you could not see from earlier in the path, the kind of which you intended to avoid but might as well experience now that it is in front of you, even if it is something of a delay.

What are you going on about?


description description

This is my new Kyria keyboard, bought as a kit from the excellent The hidden valley was the fact that I bought the kit including assembly, but decided to assemble it myself when I discovered the assembly service was unavailable for the forseeable future. The main thing I was after when I placed the order was to get another split keyboard, and to experience another key layout while at it. Learning how to solder was nowhere in my plan, but I decided to take the pieces and see what I could do rather than get my money back and wait for some possible future day to come.

Everything I had ordered was delivered in one box full of clearly labeled plastic bags, most of which also listed their exact contents and a link to the build instructions. I got a feeling similar to opening a Lego set: here is a box full of parts you will have a great time assembling, you can do this!

I took my time getting started though. After reading through the instructions a couple of times, my soldering skills felt like the big question mark. I did some reading, looked at some introductory soldering videos and listened to encouraging comments in Kodsnack's keyboard channel before I felt confident enough to get started.

Soldering a keyboard

The main reason I managed to finish this project successfully was the excellent set of guides Thomas of Splitkb provides. Every single step is clearly documented in text and images, striking that magical balance where there is no unnecessary fluff and every single question is somehow answered if you just read attentively.

To try and avoid the story lasting forever: everything was pretty straightforward. The two big soldering steps were soldering a diode for each key, and soldering the key switches themselves. Two hundred solder points in total, of which I ended up improving 25 or so. It was extremely helpful when I posted a photo of my first round of diode soldering to Slack and immediately got a comment about which points looked to be in need of improvement and why.

(The fact that I chose not to have any LED:s or OLED screens saved me quite a bit of work.)

The tricky part about diodes was getting them to sit as close to the board surface as possible, to ensure they would not get in the way of any other components. I did a good enough job to not have to re-do anything, but I did end up bending some diodes to the sides, so I pretty much used the margins I had. Perhaps I should have taped the diodes to keep them tight to the board as I worked? The amount of clearance needed is completely decided by the case and other components, so I understand why the guide could not provide a firm number for how low to keep the diodes.

As for the key switches, the main soldering uncertainty was soldering bridges - accidentally connecting areas which should not be connected. The guide had clear warnings about this, and it looked uncomfortably close in some places, but in the end I think it happened in at most two places. I had enough trouble with two keys that I ended up removing them completely and re-soldering after cleaning up as much as I could, so I am not comepletely sure if a bridge or just plain bad soldering was the cause.

Layout and use

I did of course realize the Kyria layout is very different to other keyboards I have used, but not until I actually sat down to type did it dawn on me just how much this could and should influence my keymap.

I think I was somewhat blinded by realizing the Kyria actually has more keys than my other 40% keyboards. While completely true, those extra keys all fall under the thumbs, laid out as if the bottom row of a standard keyboard had drifted toward and bunched up in the middle.

My current layout has put ctrl, alt, and command closest to their standard positions on the very left of the bottom row, but they are now nowhere near the finger positions they used to be at. The one advantage is that I find it easy to recall where they are, but I am sure they should move once I figure out the right destinations. It will take some thinking to find good places to put keys which are often combined. Each thumb cluster should only have one combo key each, the rest need to be combined only with keys outside of the cluster.

The "extra" keys allowed me to bring base-layer (that is, keys I do not need to hold down any other key to use) arrow keys back, but I am pretty sure that will not last.

(On my other layout, I have them on WASD when I hold down the tab key. Which means a lot of fingers get involved if I try to hold down shift and command while pressing the down arrow key.)

As I write this, I have an inverted t-shape in the right thumb cluster mapped to the arrows. While it is nice to have the arrow keys sitting right there, easily combined with modifier keys, it made me realize just how good an idea it is to move everything possible to the home row of keys. If I want to use these arrow keys properly, my right hand has to move away from the home row position completely, and that is no way to live.

As of typing this, my idea is to add a key which toggles a layer with arrow and modifier keys, so I can arrow around and use modifiers without needing to also hold down a layer switch key.

A final funny thought: if I do succeed in bringing more functionality to the home row by way of layers, I could find myself with even less use for my bonus keys.

Wrapping up, for now

I enjoy my Kyria a great deal, but its way too early to tell if it might become my new favourite. I still make too many mistakes, and I think it will take longer to figure out and learn a truly efficient layout. But if I do find a good one, I might find it very hard to go back to a more traditionally shaped keyboard.

Time will tell.