LUSP Data Formatting Notes
In an effort to simplify handling and viewing LUSP messages, with the exception of each message's header (1st 5 bytes) and "end of SYSEX" (the last byte in the message) message, all data appears in a "nibblized" format. Ultimately, this makes interpreting the data extremely straight forward: unnibblize then look for strucutred fields.

By nibblized" we mean that each byte of data is split into two separate bytes, each containing one nibble of the original byte. The high nibble of the resulting bytes are filled with zeros as follows:

For data organized as 16 bit "words" (2 bytes) , are sent low nibble of the low byte followed by high nibble of the low byte, followed by the low nibble of the high byte and finally the hi nibble of the high byte. For example:

The implication of this is that some messages will be unnecessarily long. We recognize this but felt the trade off was worth it considering the complexity of the protocol, the box itself and the actual data contained in the messages (dumps, etc...).