Re: Displaying 3-D vector fields [message #32911 is a reply to message #32726] |
Thu, 14 November 2002 15:47   |
Mark Hadfield
Messages: 783 Registered: May 1995
|
Senior Member |
|
|
"Rick Towler" <rtowler@u.washington.edu> wrote in message
news:ar15bu$uku$1@nntp6.u.washington.edu...
> Now you speak up!
My apologies. I did submit a message to this thread earlier but
somehow it never showed up.
> I can easily create 20k individual vector objects so I don't know
> what you have run into in the past. This approach will cost him an
> extra 8MB or so with the IDLgrModel overhead but what the heck,
> memory is cheap!
OK, it's not memory that kills you when you have a large number of
objects, it's redraw speed. (Another apology, this time for
looseness.) I based this assertion on experience with scatter plots,
discussed on the group several times in the past: when you're dealing
with > 1000 geometric entities, combining them into one object gives
much faster redraws than having a separate object for each one. I
thought I should check that this principle out for the current case,
so I just compared a barb plot with 1000 line segments in one
IDlgrPolyline vs a barb plot with a 1000 IDLgrPolylines. Sure enough,
the single-object version is redrawn very snappily, the multi-object
version takes 0.5 s or so per redraw. So, on my machine at least, 18 K
objects would be pretty slow.
> True, but dealing with the vector head using this approach is an,
> um, headache.
Oh right, I hadn't thought about this much. I have given up on trying
to draw heads on my barbs (can't be bothered with all the geometry)
but my MGHgrBarbPlot object can draw a symbol at the base of each
barb. I find I prefer the look of this anyway.
>> Unfortunately my library is off the air at the moment. (I really
>> must do something about this.) > > Please do!
Yes, I really really will. I promise.
--
Mark Hadfield "Ka puwaha te tai nei, Hoea tatou"
m.hadfield@niwa.co.nz
National Institute for Water and Atmospheric Research (NIWA)
|
|
|