Re: logarithmic scale starting with 0
- From: matt271829-news@xxxxxxxxxxx
- Date: 20 Jan 2007 06:23:43 -0800
different wrote:
I'm trying to interpolate between to numbers using the logarithm
function.
I divided the inteval between the numbers in 20 parts and use the
equation in:
http://www.mpip-mainz.mpg.de/~deserno/science_notes/log_interpol/log_...
to compute the value at each step.
My problem is that my interval begins with 0.
I've already been suggested to shift the graph by 1. In this way the
equation becomes:
x = (x2 + 1)^f * (x1+1)^(1-f) - 1
But it doesn't work. If I use log10 on the interval [0;100] and f, the
position of the point of which I need the interpolated value, is 1/2 x
should be 10, but the equation gives another number.
Any suggestions?
I don't really understand what you are trying to do (and, btw, the link
seems to be broken). Given some number x, it seems that you want to
find some other number, y, that is the logarithmically interpolated
value, or translated value, or whatever you want to call it. To
construct a logarithmic mapping you need to fix exactly two pairs of
values. In other words, you need to stipulate that when x = x1, y = y1,
and when x = x2, y = y2, for some stated x1, y1, x2, y2. What are your
x1, y1, x2, y2? (It's also necessary to know which "way round" the
interpolation is supposed to work - that is, whether it is logarithmic
or exponential - but hopfeully that will be apparent from the values
you supply.)
.
- Follow-Ups:
- Re: logarithmic scale starting with 0
- From: different
- Re: logarithmic scale starting with 0
- References:
- logarithmic scale starting with 0
- From: different
- logarithmic scale starting with 0
- Prev by Date: Re: A help with a formal proof for an "obvious" fact, please
- Next by Date: Re: Spectral Radius question
- Previous by thread: Re: logarithmic scale starting with 0
- Next by thread: Re: logarithmic scale starting with 0
- Index(es):
Relevant Pages
|
