I actually have been forced to use matplotlib and it is sufficient but
much slower than gnuplot for the same amount of data processing for a
graphics image.
owner@localhost:~/FAF> time python plot-hour.py day175.hr24.tukey.log.dft
hour.png
real 7m33.299s
user 5m28.269s
sys 0m2.773s
owner@localhost:~/FAF> time gnuplot day192.hr17.tukey.log.plt
real 0m3.487s
user 0m1.902s
sys 0m0.099s
453.299 seconds versus 3.487 seconds.. gnuplot is MUCH faster.
~
~
On 8/11/23 17:01, bri...@pounceofcats.com wrote:
On Fri, 11 Aug 2023 16:34:30 -0700
American Citizen <website.read...@gmail.com> wrote:
Hello:
I am working with the NASPMON project (see
https://www.ig.cas.cz/en/naspmon-project-website/) on an original
research project pertaining to seismic detection of underground magma
movement prior to a fissure eruption (which Iceland has).
I have processed the seismic data, using fftw to run the discrete
fourier transforms on the data, in order to find the frequency
correlations. Where I am running into a problem is using splot to do the
3d surface map.
The data is huge, typically 3600 by 4000 data elements on just one plot
(of 425) of an hour's worth of data.
Has anyone used 3d surface plotting software (other than gnuplot splot
with pm3d) ??
python/matplotlib ?
it does surprising well even with a large amount of data.
