| Circle drawing request. [message #7981] |
Tue, 28 January 1997 00:00 |
D.Kennedy
Messages: 26
Registered: January 1997
|
Junior Member |
|
|
This must be a FAQ - anyone got a good circle drawing sub-routine?
Input - x, y, radius
Output - [x,y] vectors for POLYFILL etc.
I have been using one (shown below) but I find it produces fine lines
shooting off to small y values from the left and right edges of each
circle. And my fiddling with it has made it worse I'm afraid.
FUNCTION CIRCLE, xcenter, ycenter, radius
; Emailed by Dfanning 27th Jan
step = (radius/24.0)
x = FLTARR(25)
y = FLTARR(25)
; Construct a circle
FOR j=0,24 DO BEGIN
x(j) = j*step
y(j) = SQRT(radius^2 - x(j)^2)
ENDFOR
x = [x, Reverse(x)]
y = [y, -Reverse(y)]
x = [-Reverse(x), x]
y = [y,y]
; Center the circle at the specified coordinates.
x = x + xcenter
y = y + ycenter
points = FLTARR(2, 100)
points(0,*) = x
points(1,*) = y
RETURN, points
END
--
David Kennedy, Dept. of Pure & Applied Physics, Queen's University of Belfast
Email: D.Kennedy@Queens-Belfast.ac.uk | URL: http://star.pst.qub.ac.uk/~dcjk/
Hi! I'm a .signature virus! Copy me into yours and join the fun!
|
|
|
|
| Re: Circle drawing request. [message #7985 is a reply to message #7981] |
Tue, 28 January 1997 00:00  |
Wayne Landsman
Messages: 117
Registered: January 1997
|
Senior Member |
|
|
David Kennedy wrote:
>
> This must be a FAQ - anyone got a good circle drawing sub-routine?
> Input - x, y, radius
> Output - [x,y] vectors for POLYFILL etc.
>
> I have been using one (shown below) but I find it produces fine lines
> shooting off to small y values from the left and right edges of each
> circle. And my fiddling with it has made it worse I'm afraid.
>
David,
You might want to look at the procedure TVCIRCLE available from
http://idlastro.gsfc.nasa.gov/ftp/pro/tv/tvcircle.pro, which includes a
/FILL option for using POLYFILL. It uses a more sophisticated circle
drawing algorithm that was shown to me a few years back by Allyn Saroyn.
A possible problem with the simple-minded algorithm where the X,Y
coordinates of the circle are calculated at a preset number of values
(25*4 = 100 for David Fanning's sample code) is that the circle might
sometimes be smooth, and sometimes spiky, depending on the circle size
and the graphics device. TVCIRCLE gets around this problem by always
calculating the positions internally in device coordinates, and
calculating all positions where the X,Y device coordinates differ by an
integral amount. Thus, the circle is guarenteed (in principle) to come
out smooth on all graphics devices.
--Wayne Landsman landsman@mpb.gsfc.nasa.gov
|
|
|
|
comp.lang.idl-pvwave archive
Members
Search
Help
Login
Home
