todo.txt

Processing todo:
* Normalize data over time not over samples
* Normalize data spatially not just over time
* Identify bottom types
* Identify changes in temperature
* Identify significant changes in depth
* Identify what look like fish in the log
* Identify what looks like logs, or garbage or other things
* Create some kind of summary map/data that marks key places for fishing


Reverse engineering sl2 protocol TODO:
This is secondary to me as the current data is sufficient for things I want to do at the moment (unless I can get an accurate time from the GPS unit and access to IMU sensors would also be nice but dont look like they are available)

It seems to me that the current known fields looks ok, but the flags are incorrect.
Additionally there is data available in unknown fields we could probably work out
Also I am seeing differences in the data I collect from my sonar and from the small.sl2 file downloaded from elsewhere. Would be useful to know how that file was generated
Really want to collect data from different sonar units, also try diff settings on mine and tes

* Log min/mean/max for each item per channel for those items that seem to be fixed values (maybe in hex also)
* Need to define way to compare between different sl2 files
* Add support for plotting sl2 in data sequence not separate channels only
* Add support for plotting sl2 in time (or sl2_index) sequence not separate channels only
* Start documenting relationships we see between unknowns
* Create code that logs warnings when assumptions are not met:
	For example unknown6 looks like a % 0 - 100, but is a 32 bit uint seems weird except maybe it is more complex field with 8 bit uint in bottom?
* Document flags that appear the to match, add code to compare similarity %
* In particular these have values non-zero:
	unknown3
	unknown6
	unknown8
	unknown9
	unknown10
	unknown11
	unknown14
	unknown5
* Flag correlation with data changes (trial each combination of settings we consider)