Re: Testing gnuplot with libcerf

2018-11-06 Thread Joe Zeff
On 11/06/2018 12:17 AM, Samuel Sieb wrote: Most languages have a math library containing a pi constant that is as precise as the computer can store.  How is using a trig function going to be more accurate? This was from back in the early 1980s, when languages didn't have those constants built

Re: Testing gnuplot with libcerf

2018-11-05 Thread Samuel Sieb
On 11/5/18 9:07 PM, Joe Zeff wrote: On 11/05/2018 09:44 PM, Patrick Dupre wrote: pi = 4 arctan (1) if you need Excellent!  That was exactly what I was going to suggest.  Decades ago, the late Dan Alderson (The man who wrote JPL's main space probe navigation package.) told me that you should

Re: Testing gnuplot with libcerf

2018-11-05 Thread Joe Zeff
On 11/05/2018 09:44 PM, Patrick Dupre wrote: pi = 4 arctan (1) if you need Excellent! That was exactly what I was going to suggest. Decades ago, the late Dan Alderson (The man who wrote JPL's main space probe navigation package.) told me that you should always use that because it gave you

Re: Testing gnuplot with libcerf

2018-11-05 Thread Patrick Dupre
> > On 11/05/2018 07:49 PM, Patrick Dupre wrote: > > I am sorry for the confusion. > > I use gnuplot > > I don't, so I presume that pi is predefined. Yes, pi = 4 arctan (1) if you need > ___ > users mailing list -- users@lists.fedoraproject.org > To uns

Re: Testing gnuplot with libcerf

2018-11-05 Thread Joe Zeff
On 11/05/2018 07:49 PM, Patrick Dupre wrote: I am sorry for the confusion. I use gnuplot I don't, so I presume that pi is predefined. ___ users mailing list -- users@lists.fedoraproject.org To unsubscribe send an email to users-le...@lists.fedoraproje

Re: Testing gnuplot with libcerf

2018-11-05 Thread Patrick Dupre
> On 11/05/2018 06:31 PM, Patrick Dupre wrote: > >> > >> On 11/05/2018 05:32 PM, Patrick Dupre wrote: > >>> sqrt_pi = sqrt (pi) > >>> > >>> z (x, y) = x + I * y > >>> w (z) = faddeeva (z) / sqrt_pi # Normalized > >>> plot real (w (z (x, 1)) > >>> > >>> > >>> Good luck. > >> > >> How do you initial

Re: Testing gnuplot with libcerf

2018-11-05 Thread Joe Zeff
On 11/05/2018 06:31 PM, Patrick Dupre wrote: On 11/05/2018 05:32 PM, Patrick Dupre wrote: sqrt_pi = sqrt (pi) z (x, y) = x + I * y w (z) = faddeeva (z) / sqrt_pi # Normalized plot real (w (z (x, 1)) Good luck. How do you initialize pi? ___ I do

Re: Testing gnuplot with libcerf

2018-11-05 Thread Patrick Dupre
> > On 11/05/2018 05:32 PM, Patrick Dupre wrote: > > sqrt_pi = sqrt (pi) > > > > z (x, y) = x + I * y > > w (z) = faddeeva (z) / sqrt_pi # Normalized > > plot real (w (z (x, 1)) > > > > > > Good luck. > > How do you initialize pi? > ___ I do not have.

Re: Testing gnuplot with libcerf

2018-11-05 Thread Joe Zeff
On 11/05/2018 05:32 PM, Patrick Dupre wrote: sqrt_pi = sqrt (pi) z (x, y) = x + I * y w (z) = faddeeva (z) / sqrt_pi # Normalized plot real (w (z (x, 1)) Good luck. How do you initialize pi? ___ users mailing list -- users@lists.fedoraproject.org T

Re: Testing gnuplot with libcerf

2018-11-05 Thread Patrick Dupre
ue, France === > Sent: Monday, November 05, 2018 at 9:12 PM > From: "Ronaldo Mercado" > To: "Community support for Fedora users" > Subject: Testing gnuplot with libcerf > > Hi, > > I would like to tes

Re: Testing gnuplot with libcerf

2018-11-05 Thread George N. White III
On Mon, 5 Nov 2018 at 17:03, Ronaldo Mercado wrote: > Hi, > > I would like to test an update to gnuplot from bugzilla #1476616. > https://bugzilla.redhat.com/show_bug.cgi?id=1476616 > > I am using fedora 28 and I managed to install the package okay from copr > I don't know my way around gnuplot.

Testing gnuplot with libcerf

2018-11-05 Thread Ronaldo Mercado
Hi, I would like to test an update to gnuplot from bugzilla #1476616. https://bugzilla.redhat.com/show_bug.cgi?id=1476616 I am using fedora 28 and I managed to install the package okay from copr I don't know my way around gnuplot. I tried the simple plot from http://gnuplot.sourceforge.net/demo/s