@flyingjeepman on/off button on? is M5Burner detecting com port when usb is connected?

Hello @alex9 it looks like yes, GPIO0 is TX pin for RS485. See pinmap here. The platformio.ini example can be found here. Running a quick search I did not find an RS485 example, sorry. Edit: Maybe try this RS485 library. Thanks Felix

@LETSOC Thank you for the confirmation LETSOC. I attempted to update m5stack libraries and device files for arduino IDE this week and compiled the code again, but HID functionality is still not working on Tab5. I was excited to try using the Tab5 as an advanced pocket macropad, but without HID this is not possible yet. I'm glad to at least have the Cardputer working.

@teastain hello just wondering is the screen going to be off the entire while doing this? and im wondering how do i reset the firmware with it bricked and the screen off? ive consitently tried the button thing and plugging in at same time, maybe not getting it right idk. im on linux (chromebook) and i got the burner unzipped loaded up and i couldnt get the cardputer to go onto the os files system i could only find it through port and in the linux usb sharing. so i instead tried to download the firmware onto the local machine and then to the device in thw port and then it turned off when i did it. did i add one firmware o to the other making it brick or crash or just sent to cardputer bad?

this should go as reply to https://community.m5stack.com/topic/1838/ncir-unit-weird-values accidentally posted as new topic.

@galehill The "5mm" holes, actually 4.85mm, are for Lego pegs to attach the Atomic base to Lego Technic parts like beams. Most M5stack things have Lego-compatible mounting points. If you look under the Atom Lite, there's a small yellow M2 hole marked on the label which you can use to fix the atom to the atomic base from below. There's a M2-sized hole in the atomic base. Just screw through the atomic base from below, into the atom label and there's a thread inside. M5stack used to supply a M2 hex-headed bolt of the correct length and a little Allen key to mount the atom on the base. But usually just mounting the atom on the atomic base pins is enough for most uses.

@andyzot so you can't find location where downloaded firmwares are stored on MAC? [image: 1770374964770-m5burner_firmware0.jpg] see Win10 example

Check the main menu or dashboard after logging in, usually the actual working page is under Home or Dashboard, not Tutorials. If you tell us what you’re trying to access, someone can point you directly.

@PitDuran This discussion highlights questions about adding AI features—like voice commands, image recognition, or anomaly detection—to embedded/IoT projects and choosing between on-device and cloud AI. It reflects the real challenges developers face when balancing performance, power, and usability in smart devices. For teams tackling such embedded AI integration, ai application development services https://artjoker.net/services/ai-app-development/ can offer expertise in designing custom workflows, tools, and models that fit both edge and cloud use cases, helping turn complex ideas into reliable solutions.

M5Paper quirks can definitely slow you down. Curious what steps you’ve tried so far and if any improvements helped.

Connectivity and login problems tend to pop up with firmware tools. Curious if anyone found a reliable workaround.

@stackman This has better sensor for dashcam - https://docs.m5stack.com/en/accessory/CamModule_SC850SL Some inspiration - https://www.linuxjournal.com/content/building-dashcam-raspberry-pi-zero-w

@brian111 Both ax8850 and ax650n are not the identical SoC, but shares the same software stack. Make sure you use the ax650n documentation for development.

Never mind, this is a model problem, I downloaded another model from hugging's face and it worked. Most of the models I tried from there do not work, I only got two of them to process. It was qwen2.5-7B and the Deepseek one. Both were a little slow, so I compared them using the CPU and the AX8850 and the CPU was slower, but not by much, This is on a pi 5 16GB board. I also have a Kinara ara-2 to work on with 16GB memory and probably would be faster, but I am having problems converting models to its format. I got the pi 5 to recognize it via driver, I just need a converted model to try it. I also have a Hailo-10H coming, can't wait to test it for a project, its supposed to work better with Home Assistant.

Hello @astro_m5 have you tried to flash UIFlow2.0 or CoreS3 UserDemo using M5Burner? Does that recover the backlight? Thanks Felix