[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re: fscanf accuracy for floats
From: |
David Bateman |
Subject: |
Re: fscanf accuracy for floats |
Date: |
Thu, 03 Jan 2008 22:03:07 +0100 |
User-agent: |
Thunderbird 1.5.0.7 (X11/20060921) |
gOS wrote:
> I'm attempting to read in data from a file: (ex, -0.285121188808)
>
> There are multiple columns in the file etc. And I'm able to import all of
> the data into Octave. The problem is that -0.285121188808 becomes -0.28512
> which is not precise enough for the calculations that need done with this
> data. I can't seem to find anyway to make this more accurate. I've switched
> between f,e, and g. I've tried multiple permutations of the format, and it
> just isn't panning out. It's worth noting that in Matlab -0.28512 is
> displayed on screen when printed, but the rest of the data is still there
> (multiply by 10 for isntance and you get -2.285121). Octave yields -2.28512.
>
> Is there a better way to do this, or does this qualify as a bug?
>
Have you tried using the "format long" command to see if all of the data
is really there and just truncated for display?
D.