PF4 : ICP3 !

[trimarco232]

there is a thing that is interesting for who needs a second capture input, that does not exist on atmega328p
typically if one want to decode manchester like signals , or uart , for witch it is probably better using ICP3 , because of the digital filter
ICP3 is on pin PF4 ; because PF4 is pin 3 of the 32 pins chips , witch is connected to gnd on the mini or nano boards , only the 48 pins board has it available
really ? according to the datasheet, port PF4 is connected to pin 3 of the 20 pins chip , together wit PD3 and PD4
maybe it worth the effort to mention it (or complete the drawings)

[dbuezas]

Indeed! @wollewald made an update today, but I think this is not included.

It's not shown In the QFP32 chip pinout, but it is there in the table

[dbuezas]

Oh wrong direction. Anyway still not connected. I guess one can still solder a wire to the pad, but that's beyond your fantastic pinout diagrams

[trimarco232]

Hi,
PF4 is ICP3 and OC1B
unfortunately , none of the pin#3 (PE4+PF4) nor the pin#6 (PE5+PF5) of the 32pin chip is connected to any pin of the nano or mini boards
this is because pin#3 and pin#6 of the atmega328p are connected to gnd
(may I suggest our Chinese friends to remake these boards and connect these pins instead of the two rst pins , witch have no utility)

[dbuezas]

Which would be particularly useful b/c f4 and f5 can be high current outputs

[wollewald]

I added ICP3. I also "found" PF4 and PF7 and added them. A pull request is coming. Feels like an asymptotic approximation to the truth.

[dbuezas]

Asymptotic indeed 😂

[trimarco232]

Hi, PD2 is analog AC0O (comparator 0 output)
this allows the complete access to the comparator , in combinaison with the pins AC0P and ACXN
(still few space between the 2 curves 😂)

[wollewald]

Why am I so blind? It seems I have worked too long with these pinout diagrams! But thanks a lot. At least a bit of the space between the curves will be filled. I will send the next PR in a few minutes.