A whine about default device settings [message #58641] |
Sat, 09 February 2008 13:42  |
Kenneth P. Bowman
Messages: 585 Registered: May 2000
|
Senior Member |
|
|
The new 24-bit Z-buffer device is nice for generating graphics on a
device without a display, as is common with servers, for example. It is
easier to configure and use than Xvfb, for example.
Unfortunately, some of the default device settings are different between
the Z-buffer device and the X-Windows device. (I haven't checked the
WIN device because I don't have a Windows machine).
IDL> window, xsize=1024, ysize=768
IDL> help, !d, /str
** Structure !DEVICE, 17 tags, length=84, data length=84:
NAME STRING 'X'
X_CH_SIZE LONG 6
Y_CH_SIZE LONG 10
IDL> set_plot, 'z'
IDL> device, set_resolution = [1024, 768], set_pixel_depth = 24
IDL> help, !d, /str
** Structure !DEVICE, 17 tags, length=84, data length=84:
NAME STRING 'Z'
X_CH_SIZE LONG 8
Y_CH_SIZE LONG 12
Because IDL scales margins and plot sizes based on the character size,
switching devices means that the resulting graphs are not identical,
despite identical window/device sizes.
This can be rectified by setting the character size of the Z-buffer
device with
DEVICE, SET_CHARACTER_SIZE = [6, 10]
or doing the reverse for the X-Windows device, but this seems to me to
be a pointless and unnecessary device dependence.
I am annoyed enough to post about this because I just spend longer than
I should have figuring out why the graphics output from the two devices
was not identical.
Ken Bowman
|
|
|
Re: A whine about default device settings [message #58708 is a reply to message #58641] |
Tue, 12 February 2008 07:24  |
Mike[2]
Messages: 99 Registered: December 2005
|
Member |
|
|
On Feb 11, 1:34 pm, "Kenneth P. Bowman" <k-bow...@null.edu> wrote:
> Oh good, the X, Z, and WIN devices all have different default character
> sizes and pixels per centimeter.
You can't get that sort of consistency without years of development
and growth! On the positive side, knowing about this helps immensely
- I thought I was just misunderstanding something basic. Which I was,
but at least I'm not alone. Reminds me of the rounding errors that
change the slopes of plotted lines.
Mike
|
|
|