handlejitterremoval option

[MakotoMiyakoshi]

Hello. I just wanted to point out that one of the options in load_xdf() 'handlejitterremoval' is 'on' by default. This could be a dangerous assumption as I showed in the following URL link. I suggest that before applying linear regression on the time stamps, it should evaluate the amount of jitter etc so that users can decide whether linear regression is an appropriate solution or not. Also, the default for this option should be 'off'.
https://sccn.ucsd.edu/mediawiki/images/d/d8/How_to_make_irregular_and_missing_sampling_points_uniform_in_LSL_data.pdf

[cll008]

AFAIK VideoStream was the only app to violate the 'more or less regular sampling interval,' and handleJitterRemoval should be fine for all other apps. @dmedine did most of the work to correct the app itself 2 years ago. The old VideoStream app would not sample at a regular intervals because 1) the read and write threads weren't being locked properly and some frames were lost but the header info was still assigned and 2) some cameras have variable frame rates depending on the load of the computer. Both of these make the reported 'effective sampling rate' wrong and cause 'handleJitterRemoval' to fail.

I wrapped up his quick fix before he left SCCN but never got to supporting multiple cameras, different camera types, and different codecs. I know David had also started on a new an improved version, on his own time (not Brain Products').