| Re: Frustrated by 2 Data Plotting problems [message #76275 is a reply to message #76274] |
Fri, 27 May 2011 11:28   |
David Fanning
Messages: 11724
Registered: August 2001
|
Senior Member |
|
|
Paul van Delst writes:
> Dunno about the OP, but plotting lots and lots of points (i.e. scatter plot) can tell you a lot about the relationships
> in, and between, datasets. Especially if datasets derived using different algorithms/input-data/whatever are
> scatter-plotted with different colours. (a meaningless scatterplot scenario: red points show a linear dependency with a
> negative bias, the blue quadratic with a positive bias, and the green linear/+ve bias for low wind speeds, but inverted
> quadratic for higher windspeeds)
>
> I could also see plotting individual points using a color gradient to include, say, time information in said
> scatter-y-type plot.
>
> It wouldn't be the only way I would look at a dataset, but it is still a useful visualisation of the data.
I don't have any problem with scatterplots. I'm
just saying that you can't realistically "see"
4 million points on a line plot unless your
monitor is the size of, say, the Vietnam
Memorial wall!
I wonder how your visualization would differ
if you randomly selected one percent of those
points and plotted those. I would guess the
plot would not look materially different,
although the rendering speed might improve
dramatically. :-)
Cheers,
David
--
David Fanning, Ph.D.
Fanning Software Consulting, Inc.
Coyote's Guide to IDL Programming: http://www.idlcoyote.com/
Sepore ma de ni thui. ("Perhaps thou speakest truth.")
|
|
|
|
comp.lang.idl-pvwave archive
Members
Search
Help
Login
Home
