When using the Bridge library, where is the data stored? Using the AVR's 2.5KB ram, or using the Linux side 64MB ram?
Secondly, would Bridge storage be a good solution for storing strings to save memory on the AVR?
steinvb12:
When using the Bridge library, where is the data stored? Using the AVR's 2.5KB ram, or using the Linux side 64MB ram?
It's stored on the Linux side.
Secondly, would Bridge storage be a good solution for storing strings to save memory on the AVR?
I don't think so, there is a lot of overhead to access that data.
My personal opinion is that programming a Yun efficiently needs a rather different mindset than the other Arduino boards. You're used to doing everything in the sketch. My experience with the Yun is that it's most efficient (and easier to develop) to do as much of your work on the Linux side as possible. Do all of the heavy processing down there, especially any network communications and data storage. Send only the information necessary between the sketch and the Linux side.
I try to keep the sketch as a dumb I/P processor: it reads it's sensors and inputs and sends the raw data down to the Linux side, and it receives data from the Linux side and writes it to its outputs. The processing of the data, and any decisions to be made, are done down on the Linux side. I use Python, since it's easy, but there are other programming choices as well. I like to use the Process class in the sketch to start the Python code on the Linux side and run it asynchronously. Then, anything that the sketch writes to the Process object is read by the Python code exactly as if it were typed on the keyboard, and anything the Python code prints can be read from the Process object in the sketch.
Great info. I will definitely keep this in mind.