i've started hacking a bit on the MNT Pocket Reform keyboard firmware to decouple USB from input/display/sysctl handling using timer interrupts, i'll finish and clean that up next week. this should fix any "usb disconnect" issues.
Conversation
Notices
-
minute (mntmn@mastodon.social)'s status on Thursday, 02-Jan-2025 07:44:32 JST minute -
minute (mntmn@mastodon.social)'s status on Thursday, 02-Jan-2025 08:26:44 JST minute WIP for the adventurous (yes this needs more cleanup/refactoring, but works) https://source.mnt.re/reform/pocket-reform/-/merge_requests/23
-
minute (mntmn@mastodon.social)'s status on Thursday, 02-Jan-2025 08:31:00 JST minute pretty funny: at some point i had OLED updates and trackball motion polling in parallel and was wondering why the OLED was glitchy/only partially updating then. then i remembered that they're on the same i2c bus argh ^^
-
minute (mntmn@mastodon.social)'s status on Thursday, 02-Jan-2025 08:31:24 JST minute but then i also noticed that we can easily bump i2c from 100 to 400 khz and everything goes brrrr
-
Charlie Balogh (chainq@mastodon.social)'s status on Thursday, 02-Jan-2025 10:19:56 JST Charlie Balogh @mntmn ... does that mean it's still glitching, you just don't see it any more because it's glitching so fast? 🙃 (That's still a 100% valid fix btw, not criticizing. 😅)
-
minute (mntmn@mastodon.social)'s status on Thursday, 02-Jan-2025 10:23:05 JST minute @chainq haha no i have to still do that serially but when there's an oled refresh now (happens only if you leave open the battery status page) it's not taking so long as to noticeably disturb the trackball
-