[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re: Slowness in function 'open'
From: |
Mark B. |
Subject: |
Re: Slowness in function 'open' |
Date: |
Thu, 21 Jun 2007 14:31:55 -0700 (PDT) |
Thanks again Dmitri for taking the time to trying to help me :-)
Dmitri A. Sergatskov wrote:
>
> Since you have not mentioned what is your hardware, it is hard to make a
> meaningful comparison.
>
That's why I included the Matlab times, so that you can get a meaningful
ratio because the actual time in seconds is not really very meaningful. For
example, "is it 90 seconds too slow for a 30 Mb file?" Nobody can with 100%
certainty say yes or no. Maybe it's the best that can be done with any
hardware (it's only an example).
Dmitri A. Sergatskov wrote:
>
> But in any case on my CoreDuo II (2GHz / 2GB)
> Fedora 7 octave 2.9.9 reading 5 rows x 512000 lines file xxx.dat (25 MB):
>
> octave:3> save -binary "a.bin" a
> octave:4> tic; load("a.bin"); toc
> Elapsed time is 0.037651 seconds.
>
Yep, I will have to use binary files if nobody else can provide a better
solution with text files.
Dmitri A. Sergatskov wrote:
>
> It is a known issue that reading a formatted ASCII takes longer than
> reading
> a binary file.
>
Surely, but it does not look as a very difficult or strange task to load a
text file. For example (please this is only an example) Matlab is ~3.5 times
faster using 'load' and up to ~7 times using fscanf. It could be nice to
improve this. However, I understand that this is an open source program and
nobody can make any guarantee.
Dmitri A. Sergatskov wrote:
>
> Hope that helps.
>
It surely does.
Thanks
--
View this message in context:
http://www.nabble.com/Slowness-in-function-%27open%27-tf3960902.html#a11242113
Sent from the Octave - General mailing list archive at Nabble.com.
Re: Slowness in function 'open', Dmitri A. Sergatskov, 2007/06/21