How ultra low power is the Moteino Mega?

[WhiteHare]

Obviously a lot has been learned (and documented on this forum) about how to operate a Moteino R4 very efficiently.  How does the Moteino Mega compare in terms of current consumption?  I mean, yes, I can look at the datasheet, but that by itself doesn't really help enough: a lot can depend on non-obvious methods.  By that I mean: a lot about how to save power on the atmega328p seems obvious in retrospect, but wasn't obvious beforehand.  It took a lot of trailblazing, and "standing on the shoulders of giants," to get "real life" power consumption as dramatically low as it is today.

So, at this juncture in time, what are comparable current consumption benchmarks for the Moteino Mega, especially as compared to the Moteino R4?  Can you run it for 20 years on a coin cell while doing similar work to Moteino R4 that's been optimized to do the same thing?  I mean, what are the comparables, and what are the present limits?  Obviously, I'm asking this in reference to either battery operation or capacitor operation tied to energy harvesting.

[Felix]

I think most of the research done for Moteino is equivalent to MoteinoMEGA (non USB). The same peripherals are on the MEGA and the chips have the same low power technologies and expectations (according to the datasheets). The 1284p is just a bigger chip with lots more memory and IO, otherwise I would look at it as very similar. Of course that's just me saying this without empirical testing, and based on the fact that hardware wise everything is the same, except the MCU.