basic demo configuration error + cleanup + qq (#4717) #13
Comments
Yep, the esphome people like to make gratuitously-incompatible changes.
Right. This is similar to what I did last NY, then I wasted a day tracking down the task_wdt crash (esphome/issues#5317). The proper way to fix this is get the mentioned fix into ESPHome, then add the board definition to platformio platformio/platform-espressif32#1269
This component is very simple, so re-drawing the screen blocks the main loop. ESPHome doesn't like it and spams the above warning.
Where did you see that? We do epd_poweroff() |
They call it breaking changes on release notes 🤓. The updated code still doesn't look right. If we were to apply
That's exactly my point. There is no need to chase self-inflicted bugs and re-invent already existing platformio boards. Strip down your code and both problems will disappear by magic (please, refer to the compilation and run logs above). For some strange reason, the internet is full of confusion when it comes to There is an opinion that stating
👍🏻 Thanks a lot for confirming this! It's been very surprising to discover that the original thread #1109 has become polluted by s3-related discussions which were meant to be confined within #1960. And yes, there is a mention of a new variant V2.4. It seems to be compatible with your display component, thank you! EDIT: corrected link to Lilygo T5 ePaper 4.7 ESP32-S3 Development Driver Board variant V2.4. |
You're right, but it works and I've already spent an infinitely more time on this than I would've hoped for...
Just to be clear, I'm not in any way affiliated with Lilygo. I'm not getting paid by them or anyone really for this. I just bought their board to play around with, and I'm not even using it anymore. I might agree with your suggestions if you were talking to lilygo. For example I relayed your notice about the mismatch between the board definition and the README to an issue that sits ignored on their repo. Myself I'm not interested in testing if their board still works with newer PIO/arduino versions and parameters they have not tested it with! You should feel free to fork this repository and use the simpler code, and I'll happily accept PRs mentioning your approach. I'm willing to switch the base
No, I think this is an illustration of how you managed to pick a more appropriate base board and that things improve over time.
Both threads have unfortunately turned into support Q&A. Oh well.
Not sure I follow, wrong link maybe? |
|
This is a test
|
|
I was about to give up on ESPHome in favour of a plain reference code. The way the reference code does partial refresh seems to be far from perfect though. For some strange reason, the whole screen is visibly darker than the partial refresh's window area when it comes to white. In the other words, the whole screen isn't that white when using the reference code. I came back to this brilliant implementation which produces a much better contrast clock image. The white is much whiter. Unlike the reference code, this brilliant implementation doesn't do Reality check: when I see Oh! And a typo has been spotted! 🤓 L12 shall read esphome run t547plus.yaml |
|
You keep mentioning epd_powerdown reportedly being used upstream, but I just don’t see it: https://github.com/search?q=repo%3AXinyuan-LilyGO%2FLilyGo-EPD47%20epd_powerdown&type=code |
|
Mine bad! It's |
|
I don't run this board on battery, as it seems that you're trying to. But I have vague recollection of seeing this effect. Others noticed similar issues with the older revision esphome/feature-requests#1109 (comment) |
and others
@nickolay, I was trying to follow the steps on README, and it didn't work.
ota syntax error
I'd guess, with the ESPHome version update, basic.yaml requires
otakey to be aligned with the new reality and get aplatformsubkey?With reference to your observation, there seem to be a cleaner solution which doesn't require much pinning. I have to confess I dislike platrormio, because it solves Microsoft's problem, not mine. The whole
esphome.platformio_optionsbranch can be safely removed (yay!) in exchange for a seemingly compatibleesp32.board:esp32s3box.Here is the patch proposed:
basic.yaml.patch
And here is a test log:
esphome run basic.yaml
I haven't seen much logs with the warning I am getting (see below). It seems to be present with the original
esphome.platformio_optionsas well. Am I in trouble? Or is it just an inexpensive 'slow display'?The vendor recommends to turn off power from time to time (
epd_powerdown()), and not keep power on all the time to avoid permanent display damage. Does your custom platform code take care of this vendor's recommendation? Or do users have to make some independent arrangements?The text was updated successfully, but these errors were encountered: