Okay, cool. Just be aware i changed the wiring, so old fw will not work out of the box…(rows are not shared between halves anymore, i hope this will fix some ghosting)
Ill try to update the fw in the weekend at the latest…
Hey, when you say you ordered 10 v2, Im getting a bit nervous… Im still working on v2, its not released yet? In fact i just received some prototypes, but havent tested them yet…
I have a v2 branch, but there Are no releases yet…
Hi. Im pretty sure you cant use pa11 or pa12, as they Are used for usb…
Check this: https://www.reddit.com/r/olkb/s/CqDGqABzg3
Looking good!
The function of the encoder is custom code, you need to Edit the encoder.c file to tweak it.
P.s. You know you can tweak the led? Turn it off/change color and brightness, all using the qmk rgb keys. (Just wondered if the thing on top is to dim the light:)
Looking good! I ordered 2 ch55 boards when i saw your fak announcement. I like the idea of minimalism taken all the way!
Nice work! I found myself ordering cables and connectors on Ali express before I had finished reading…
Just thought of one thing. Are you sure your network cable is not crossed? Could you test another just to be sure
Not sure what went wrong, but did you try the same uf2 file i did? Seems like official miryoku to me…
Ferris sweep uses choc switches which Are quite expensive. The Corne is not reversible i think, making the pcb more expensive. Also, both uses two pro micros which Are more expensive than the rp2040 of the cheapino, which also uses just one.
Not very big difference, but its cheaper…
That is awesome! Can confirm that putting rp2040-zero in dfu mode and just copying the uf2 file from your workflow flashes miryoku in the cheapino.
Looking good! Now you just need to set the led to solid pink to match the switches.
I think i need some see through keycaps myself. I have been looking at the fully transparent ones, seems fun to see everything beneath…
Hey, if you start with the «tompi» keymap, it should be pretty close to miryoku.
Hey. Cool! Send some pics!
Since there is no official qmk support for cheapino, you need to use my fork: https://github.com/tompi/qmk_firmware/tree/cheapino
I should maybe try to get my fork merged into official qmk…
For your question about miryoku, there isnt a super easy way. If you are confident with git and c, you can try merging my cheapino fork into miryoku.
I based my own keymap on miryoku, so if you start with «make cheapino:tompi:flash» you should be pretty close.
Im not familiar with the redox, but i see you got two unpopulated resistor footprints around the trrs connector.
Are you sure the resistors are optional?
Of course, that would be awesome. I am going to Make a v2, with hotswap and 5 pin support and fix some other small stuff.
The coolest thing would be if you made the reversible RJ45 available as a footprint, then anybody can use it without making the same mistake i did ;)
Hey, thats a very nice build!
Im only using Mac, so I mapped the encoder to undo/redo, which on mac is LGUI+KC_Y/Z, but unfortunately not the same on windows. On windows gui(win)+y/z just equals "y/z", so thats why you get y and z when you turn the encoder.
Like @[email protected] says, you need to edit the encoder.c file to tweak it to what you want. If you cant figure it out, let me know how you would like it and Ill send you some code.
I checked the default keymap, and there is indeed double space there, I guess nobody has ever used it...
Anyway, what you want to do is go into keyboards/cheapino/keymaps and copy "tompi" to e.g. "chracoon", and then upload the keymap.json file to the configurator. When you are finished tweaking, download to same place and flash with "make cheapino:chracoon:flash"
Let me know if anything doesnt work.
And like @[email protected] says, I did indeed make a little bobo when routing the reversed RJ45, but fortunately it was pretty easy to fix in software, the matrix has to be custom anyway... There are some ghosting issues which I have not found the root cause for, but havent really hade the time to dig into it(just fixed the ones I found in software...) I suspect these are related to my choice of doubling the columns of the duplex matrix, instead of the usual row-doubling...