Re: Two questions about plotting with Octave

[marco atzeri]

On 4/2/2012 3:00 PM, Zumsel wrote:
> I'm still a newbie with octave, so may questions may be pretty basic:
>
> 1. Say I have to matrices data1 and data1, with two columns and n and m rows
> respectively. Assume n<  m. The first column is a time-coordinate, so
> data2(1:n,1) is identical to data1(:,1). With
>
> plot( data1(:,1) , [data1(:,2) ; data2(1:n,2) ])
>
> I can plot the first column of data1 versus the second column of data1 and
> the first n entries of the second column of data2. However I can't plot
>
> plot( data2(:,1) , [data1(:,2) ; data2(:,2) ])
>
> because data1(:,2) is too short. Is it somehow possible to have data1(:,2)
> and data2(:,2) in the same plot, even though the first one is shorter?

you need 2 couples of data, of the same lentghs

plot( data1(:,1) , data1(:,2) , data2(:,1), data2(:,2) )

see "help plot"

> 2. With
>
> loglog( data1(:,1) , data1(:,2) )
>
> both axes are logarithmic. Is it possible to have only the second axis
> logarithmic? Of course
>
> plot( data1(:,1) , log(data1(:,2)) )
>
> would give the correct plot, though with incorrect labeling of the second
> axis.

semilogx, semilogy