wifi_no_sleep = true for ESP32C3 still needed in T 14.5.0 ?

[GEAG]

In file tasmota/tasmota_support/support_wifi.ino there is still the following line enabled only for ESP32C3 targets:

wifi_no_sleep = true; // Temporary patch for IDF4.4, wifi sleeping may cause wifi drops

Since T has been ported to Arduino Core3/IDF5 for all versions >= T 14.0.0, I wonder if this still has relevance or if it is obsolete by now?

Our own tests and measurements have shown, that the energy consumption may be significantly reduced (up to -48%), if this line is eliminated from the code.

[therobveiller]

I’m very surprised you have no response to this, and I didn’t know about this issue and “work around” until your post.
Having recently bought a batch of C3 devices, I noted how incredibly hot they ran with Tasmota compared to my other ESP modules. For a test I pointed my IR heat camera at the C3 board running the version where wifi sleep is disabled (current “bug fix” of wifi_no_sleep = true) and the voltage regulator and ESP32C3 run in the 47-50 Celsius range. Using the same custom software build I use, on the same device of course, with no other changes whatsoever between builds except the code you are referring to commented out so that wifi sleep config follows the normal code options, the chips run at around 25-27 Celsius – a massive difference, and far more conducive to the longevity of the devices.
I have a dynamic sleep value of 50 set in both tests, but obviously that is ignored for C3 in a default build with wifi_no_sleep = true (as I read it).
I have not had any issues at all with wifi not reconnecting or dropping with this code commented out, just like you. And this is with a relatively poor RSSI of 42% on the device as it’s a long way away from the access point.

[Jason2866]

PR #23117