Very puzzled by how .sl2 files from my Lowrance are handled!
Attached is a spread sheet with some data relating to 20 .sl2 files I have collected this summer. I have tried importing them into Reefmaster with slg2txt.exe enabled and then with it not enabled. The start times are different in some cases and also different from the file time stamp. This is a slight problem (!) as the area is very tidal. Separately I have used slg2txt to convert the raw files to csv files. The sizes of the files are given in my table. Clearly some have a problem (frequency error 11) and hence the csv file is not complete. The ratio of the slg file size to the csv file size shows this - it should be about 18.
The interesting thing is the number of "elements" that Reefmaster gets from the files. In general it gets significantly more elements if slg2txt is enabled. The only exceptions are where this "error 11" occurs. In these cases the number of "elements" without slg2txt is significantly higher.
Can the gurus advise what the heck is happening?

Start time for Lowrance tracks is not recorded (unlike Humminbird logs), which is a significant drawback. Start time is calculated, as best it can, by subtracting the length of the track from the last write time of the file - which is usually recorded correctly. This *usually* gives the correct result, but in your case with the missing points from many of the RM processed files the calculated start times are also very different.

Usually when track-points are missing from the RM import it is because either RM was unable to recover from some corruption mid-file, or there is an issue with the track-points themselves that stopped them from being imported (e.g. the lat/lon were invalid). It is not common, generally, which makes your results very interesting. If you could share a few of the files that show the biggest difference in track-point(element) count that would be very useful - the easiest way to do this is to zip and upload to DropBox or similar and send the link through to This email address is being protected from spambots. You need JavaScript enabled to view it. .

Are you also seeing big differences when you view the tracks within RM?

I have a look into this and, although most files agree (approximately) w.r.t. the number of elements imported natively into RM or Slg2Txt, U have found a number that do not.
The image below shows a section of track which has been imported using both methods. The track on the left has 3545 points and the track on the right has just over 7000. Our logic for importing a new point is very simple: we look at each ping in the primary channel (typically 2D 200Khz) and record a new position if the location has changed (i.e. either lat/lon/both is different to the previous ping). I would need to investigate further to understand how slg2txt is generating the extra values - it could either be interpolating or taking extra values from other channels. In any case, I don't think this is an issue that will affect the quality of the data in any way. The RM imported points have a frequency of approximately 1 per metre and any increase in this value will not result in a more detailed map.

The time issue still needs looking at, although with my own test examples I don't get significantly different calculated start times (a few seconds at most).