examples/arduino: release 0.2.0, and record both board cores

[psiddh]

ETModel.h is new public API that sketches will be written against, so this is a minor rather than a patch.

The memory notes now carry measurements for both board cores instead of one. 0.55.2 shipped a 28 KB arena and that was correct; 0.90.0 fails it by 180 bytes, which is how a working example became a broken one without anything in the repository changing. Keeping both rows is the point. The 40 KB row under 0.55.2 is gone because neither of us measured it, and 64 KB has not been retried on 0.90.0, which the text now says rather than implies.

Sizes are re-measured on 0.90.0 with the wrapper in place: keyword spotting now uses 45,044 bytes of RAM rather than 47,084.

[psiddh]

ETModel.h is new public API that sketches will be written against, so this is a minor rather than a patch.

The memory notes now carry measurements for both board cores instead of one. 0.55.2 shipped a 28 KB arena and that was correct; 0.90.0 fails it by 180 bytes, which is how a working example became a broken one without anything in the repository changing. Keeping both rows is the point. The 40 KB row under 0.55.2 is gone because neither of us measured it, and 64 KB has not been retried on 0.90.0, which the text now says rather than implies.

Sizes are re-measured on 0.90.0 with the wrapper in place: keyword spotting now uses 45,044 bytes of RAM rather than 47,084.

[psiddh]

ETModel.h is new public API that sketches will be written against, so this is a minor rather than a patch.

The memory notes now carry measurements for both board cores instead of one. 0.55.2 shipped a 28 KB arena and that was correct; 0.90.0 fails it by 180 bytes, which is how a working example became a broken one without anything in the repository changing. Keeping both rows is the point. The 40 KB row under 0.55.2 is gone because neither of us measured it, and 64 KB has not been retried on 0.90.0, which the text now says rather than implies.

Sizes are re-measured on 0.90.0 with the wrapper in place: keyword spotting now uses 45,044 bytes of RAM rather than 47,084.

[psiddh]

ETModel.h is new public API that sketches will be written against, so this is a minor rather than a patch.

The memory notes now carry measurements for both board cores instead of one. 0.55.2 shipped a 28 KB arena and that was correct; 0.90.0 fails it by 180 bytes, which is how a working example became a broken one without anything in the repository changing. Keeping both rows is the point. The 40 KB row under 0.55.2 is gone because neither of us measured it, and 64 KB has not been retried on 0.90.0, which the text now says rather than implies.

Sizes are re-measured on 0.90.0 with the wrapper in place: keyword spotting now uses 45,044 bytes of RAM rather than 47,084.

[psiddh]

ETModel.h is new public API that sketches will be written against, so this is a minor rather than a patch.

The memory notes now carry measurements for both board cores instead of one. 0.55.2 shipped a 28 KB arena and that was correct; 0.90.0 fails it by 180 bytes, which is how a working example became a broken one without anything in the repository changing. Keeping both rows is the point. The 40 KB row under 0.55.2 is gone because neither of us measured it, and 64 KB has not been retried on 0.90.0, which the text now says rather than implies.

Sizes are re-measured on 0.90.0 with the wrapper in place: keyword spotting now uses 45,044 bytes of RAM rather than 47,084.

[psiddh]

ETModel.h is new public API that sketches will be written against, so this is a minor rather than a patch.

The memory notes now carry measurements for both board cores instead of one. 0.55.2 shipped a 28 KB arena and that was correct; 0.90.0 fails it by 180 bytes, which is how a working example became a broken one without anything in the repository changing. Keeping both rows is the point. The 40 KB row under 0.55.2 is gone because neither of us measured it, and 64 KB has not been retried on 0.90.0, which the text now says rather than implies.

Sizes are re-measured on 0.90.0 with the wrapper in place: keyword spotting now uses 45,044 bytes of RAM rather than 47,084.

[psiddh]

ETModel.h is new public API that sketches will be written against, so this is a minor rather than a patch.

The memory notes now carry measurements for both board cores instead of one. 0.55.2 shipped a 28 KB arena and that was correct; 0.90.0 fails it by 180 bytes, which is how a working example became a broken one without anything in the repository changing. Keeping both rows is the point. The 40 KB row under 0.55.2 is gone because neither of us measured it, and 64 KB has not been retried on 0.90.0, which the text now says rather than implies.

Sizes are re-measured on 0.90.0 with the wrapper in place: keyword spotting now uses 45,044 bytes of RAM rather than 47,084.

[psiddh]

ETModel.h is new public API that sketches will be written against, so this is a minor rather than a patch.

The memory notes now carry measurements for both board cores instead of one. 0.55.2 shipped a 28 KB arena and that was correct; 0.90.0 fails it by 180 bytes, which is how a working example became a broken one without anything in the repository changing. Keeping both rows is the point. The 40 KB row under 0.55.2 is gone because neither of us measured it, and 64 KB has not been retried on 0.90.0, which the text now says rather than implies.

Sizes are re-measured on 0.90.0 with the wrapper in place: keyword spotting now uses 45,044 bytes of RAM rather than 47,084.