Question

I have a set of CSV data where it is a range of data from 2000 – 2000000 bytes, say, and I want to plot every single one of these points. I however don’t want it to say 2e6 for my data but rather just 2, so I can set my x-axis title to “megabytes”.

That is, how do automatically override the visible text on the x axis bar (via some sort of formula relating to the underlying data?) using gnuplot?

Answer 1

You can arithmetically manipulate data in the using directive. If you have a data file Data.csv:

3.7567e+06
7.7169e+06
9.8955e+06
6.0108e+06
2.1852e+06

you can plot this with this script:

set boxwidth 0.6 relative
set style fill solid
set ylabel "Size in MB"
plot "Data.csv" u 0:($1/1e6) with boxes

Note the ($1/1e6) using directive. The $1 references the first column of the data file. The brackets are mandatory!

You end up with a plot like this:

If you have data ranging though many orders of magnitude, maybe using a logarithmic scaling would be interesting to think about.