I have tried building all three of the
I have tried building all three of the firmware options listed here (https://github.com/qmk/qmk_firmware/tree/master/keyboards/nullbitsco/tidbit/keymaps), none of which seem to make the rotary encoders work when rotating. The VIA build allows me to modify the key assignments, but only the click action works. When I download and use the precompiled snailmap or via build from here (https://github.com/nullbitsco/firmware/releases/tag/latest), the rotary encoders work, but they are not mappable in VIA.
20 Replies
Unknown User•9mo ago
Message Not Public
Sign In & Join Server To View
@*fin do you mind uploading the firmware you built so I can test on my end? You could also try downloading the precompiled nightly (not the official release)
Here you go. Tried the nightly (I assume you mean from here: https://github.com/nullbitsco/firmware/releases), but no snailmap (which I would like for the OLED). I also noticed there was an OLED enabled mapping, but I didn't know where to find the source file for that (I might just be misunderstanding how to access it).
@jay | nullbits The image is for reference on the layout and the build has been simplified for testing since I first posted, but the results are the same.
Thanks! Let me take a look. The tidbit stuff in the QMK repo should be 100% working, so if not, I’ll figure out what’s up with it and get a fix in.
Love your keycaps and layout btw 😍
TY! Love the nibble and tidbit! Was my first build from scratch.
I used the rotary encoder that came with the nibble on the tidbit as the second encoder. Could that be causing an issue?
That shouldn't pose a problem at all! I think there's just some software/configuration funk going on. I was able to confirm the encoders work with VIA on the latest QMK when I build it locally, so let's see if we can figure out where things are going astray for you. Btw, I did confirm that the last official release (which was almost a year ago) of the precompiled firmwares on GH is no longer compatible with VIA. So there's nothing you're doing wrong there, at least.
I also confirmed that your compiled hex is doing the same thing. So my guess is this is related to how you're setting up QMK on your end. Do you mind describing how you have QMK set up?
Sure.
First, I used this guide: https://docs.qmk.fm/#/newbs_getting_started
Installed QMK MSYS
Ran
qmk setup
Created defaults as noted with the tidbit as the default keeb by using qmk config user.keyboard/nullbitsco/tidbit
Created a new keymap defaulyt with qmk config user.keymap=new_map
Created a new keymap with qmk new-keymap
Opened keymap.c in VSCode
Edited, added a rules.mk with previously noted lines
Compiled using qmk compile -kb nullbitsco/tidbit -km new_map
Installed QMK Toolbox
Ran QMK Toolbox with tidbit in DFU (ATmega32U4 automatically selected) and flashed with new hex file
Attached is the file structure map
Awesome thank you. That helps a lot.
If instead of making new_map with the QMK shortcut, could you try making it by coping and pasting the via keymap folder? If using VIA, you can skip binding default keys in the keymap, since you can set with VIA.
Can you also run a command from the CLI to check which version of QMK you’re using for the build to make sure it’s up to date? From within the QMK directory, can you run
git log -n1 HEAD | tee and copy the output?@jay | nullbits
commit 654dc267db5c34d5b9beb49c21966f08ebe358c2
Author: Allie <[email protected]>
Date: Tue Mar 12 17:28:20 2024 +0000
Fix Magicforce MF17 RGB matrix (#23263)
I copied via to a new file called via_new and compiled the firmware. After flashing the firmware, the rotary encoders still do not work. Via does let me change them however. Something must be incorrect with my setup.
Cool, okay. Your QMK is up to date. That’s good news.
Why don’t I send you the file I compiled and tested on my end? That’ll help confirm if it’s a software thing (fairly certain it is).
Actually, this is kind of crazy but with the latest latest QMK I might be seeing the same thing? Would you mind doing two things for me?
1. Testing this known working hex file that I built on my end -- just to make sure your encoders are physically correct (I think they are)
2. checking out this commit with
git checkout 62e2cb92861820cf64440e754cd0fed155e431fd and then qmk clean before compiling?Okay, so try the file you sent and then run those two commands?
Just to confirm?
Encoders work in your build and are editable in via.
@jay | nullbits I'll check out the commit now. Do I run that in the QMK directory?
Yeah! Same as the previous git command.
I think I have it narrowed down to https://github.com/qmk/qmk_firmware/commit/3e1ac7a38fa4e6885053a762bc75f7c4e068eccb but am interested to see why no one else has reported encoder issues on non-split boards.
We have a winner. That build works with my keymap.
Awesome!!! I will investigate some more tonight and open an issue with QMK. My guess is there is something funky with the TIDBIT (perhaps the multiple encoders?) because I'd imagine that this would be impacting more people if it affected every single board with an encoder using encoder map.
Thanks for taking the time to troubleshoot and for being so patient during the whole thing 👍
Absolutely! If I can help with anything else, please let me know.
Unknown User•8mo ago
Message Not Public
Sign In & Join Server To View