Re: How to display NG created in a buffer? [message #77492] |
Fri, 02 September 2011 16:22  |
Paul Van Delst[1]
Messages: 1157 Registered: April 2002
|
Senior Member |
|
|
David Fanning wrote:
> Paul van Delst writes:
>
>> It's a bummer that a bunch of graphics objects I
>> created are so tantalisingly close (I can almost
>> taste them!) but are unavailable to me.
>
> To me, it seems as if some very bright ideas have
> gone into *making* these graphics objects, but
> no one has sat down yet to *use* them. Because
> when you start to use them, these deficiencies
> appear regularly. I hear this story over and
> over again. :-(
And the speed... don't forget the speed. The reason I'm plotting these graphics to the buffer is because if I produce
them the Direct Graphics Way, i.e. just plotting them on screen and saving them as I go, it'll take a looong time.
So, plotting them into the buffer and stuffing their references into a hash (where the hash key is the filename root) is
the easiest way I've found to generating the output quickly, e.g. for the "gref" hash I produce,
IDL> foreach graphic, gref, name do graphic.save, name+'.eps'
This produces my raft of EPS output for inclusion in a LaTeX document.
All of this would've taken much less time if I could display the buffer-contents for a particular plot, tweak it (e.g.
adjusting legends, changing symbols, whatever), and then output it.
I guess if the speed of NG output was the same as DG, I wouldn't be grousing about any of this because I could just
re-do everything instantaneously (exactly like I would if I was using DG).
Life is too short, and productivity demands too high, to put up with NG output being slower than a snail in a
straitjacket. Heaven forbid if I wanted to plot more than 100000 or so points.... I mean it's 7:30pm on a Friday and I'm
here at work making plots! Crikey... I'm going home.
cheers,
paulv
|
|
|