New option to control striding on 1D plots

[karlmsmith]

Reported by @AnsleyManke on 10 Nov 2010 00:14 UTC Putting this in the v7.2 Release Sprint milestone, but I am also willing to kick it out. Doing this has a low possibility of causing other bugs.

An installer writes that he commented out "GO LAS_set_strides.jnl" in the LAS_initialize_variable.jnl script because he didn't like the striding in time series plots. On line plots, striding can result in missing the highs and lows.

We should offer an option to turn striding off, and control how much is done.

Option:

* decimation strategy
      * no decimation
      * light decimation
      * normal decimation (default)
      * extra decimation

And possibly in the future, offer other means than simple decimation?

• ... more options
  • (Ramer-Douglas-Peucker filter)
  • (Kalman filter at 1 sigma) -- (line plot would need to somehow shade the filtered zone)
  • (Kalman filter at 2 sigma)

Migrated-From: http://dunkel.pmel.noaa.gov/trac/las/ticket/974

[karlmsmith]

Comment by @kevin-obrien on 10 Nov 2010 00:21 UTC

This is a new feature that has both Ferret and UI implications, and in that spirit might disagree on the low possibility of causing other bugs. On our short timescale, I would lean towards not including it in the v7.2 release sprint. Or maybe only if "all other things are done perfectly" but would be interested in what others think....

[karlmsmith]

Comment by @AnsleyManke on 18 Nov 2010 23:44 UTC Moving this out of the sprint

[karlmsmith]

Modified by @AnsleyManke on 5 Jan 2011 23:46 UTC