Say for example you have an Easy Chart whose plotting area is 1000 pixels wide. You then ask for a 30 days trend of an input which has been logged every second - potentially over 2.5 million points.
Does the Chart request and get all 2.5 million points? If not how does it filter them? How then are the points displayed? Does the Chart try and plot all returned points on the graph?
I’ve seem some quite clever historical trenders that look at the width of the plotting area (in this example 1000 pixels), then get a maximum and a minimum for 1000 equally sized groups. They then draw a vertical line between the maximum and minimum values for each pixel. The result is a graph where you can see all perturbations no matter how far out you are zoomed, without worrying about biasing things (and losing brief transients) by carrying out averaging. The amount of data coming from the SQL database is also hugely reduced.
I’d be interested to hear what currently happens, and whether this kind of scheme could be an option for the future.