Borrar filtros
Borrar filtros

scatter3 in log-log-log

16 visualizaciones (últimos 30 días)
Nicolas
Nicolas el 28 de Jun. de 2011
Comentada: Walter Roberson el 10 de Jun. de 2020
Hi,
I was wondering it was possible to generate a scatter3 plot but with one, or two or the three axis in log?
Thanks
n.

Respuesta aceptada

Oleg Komarov
Oleg Komarov el 28 de Jun. de 2011
After calling scatter3 use this line of code:
set(gca,'Xscale','log','Zscale','log','Yscale','log')
  4 comentarios
MatteP
MatteP el 27 de Sept. de 2018
9.2.0.556344 (R2017a)
Walter Roberson
Walter Roberson el 27 de Sept. de 2018
Editada: Walter Roberson el 27 de Sept. de 2018
MatteP, would you be able to post a few lines that reproduce the error? I just tried in R2017a but was not able to get that message (though I think I might have just encountered a different bug.)
(Confirmed that I encountered a different bug, but that bug involves too many points being drawn in some circumstances, and no error message -- not likely to have any relationship to what you are encountering.)

Iniciar sesión para comentar.

Más respuestas (1)

Chen
Chen el 10 de Jun. de 2020
Editada: Walter Roberson el 10 de Jun. de 2020
hello
Can someone tell me the 'log' in
set(gca,'Xscale','log','Zscale','log','Yscale','log')
is in 'natural log'n or in log10?
I want to change to log10, after I generate a scatter3 linear figure.
How to do it?
Many thanks
Chen
  1 comentario
Walter Roberson
Walter Roberson el 10 de Jun. de 2020
It does not matter whether it is natural log or log10 or log2 or log98.7 -- any log scale will be a linear multiple of natural logs . So log10 would be 2.30258509299405 units per increase of 1 exponent, compared to natural log. But the linear scale of a plot is determined by the data units spanned, so data that was in some sense 2.30258509299405 times "more compact" would get magnified by a linear factor of 2.30258509299405 because there is a fixed vertical or horizontal space to hold all of the data units. The result is that no matter what base the log is, the linear scale turns out the same.
Where it would make a difference would be if you were placing a tick at each increase by 1 in the log -- are you placing ticks at exp(0), exp(1), exp(2), and so on, or are you placing ticks at 10^0, 10^1, 10^2, and so on. And that is a different question than the question of whether internally it is natural log or log10 . The answer is that no matter what internal base is being used for log scale, that the automatic tick choices are based around "nice" numbers in log10 .
So if you are relying upon the automatic tick choices, you would not be able to tell the difference between it being natural log or log10 internally.

Iniciar sesión para comentar.

Categorías

Más información sobre Log Plots en Help Center y File Exchange.

Community Treasure Hunt

Find the treasures in MATLAB Central and discover how the community can help you!

Start Hunting!

Translated by