Error on displaying a 20k x 20k image on DS9
nomo17k wrote on May 10, 2010
I'm trying to display 20k x 20k image on DS9 but having a problem.
I have copied $IRAFPATH/iraf/lib/imtoolrc and $IRAFPATH/iraf/lib/graphcap to my home directory as .imtoolrc and .graphcap, and added an entry:
64 1 20000 20000 # imt64|imt20k
to .imtoolrc and:
imt64|imt20k|20K frame buffer:cn#64:xr#20000:yr#20000:\
:LC:BS@:z0#1:zr#200:DD=node!imtool,,20000,20000:tc=iism70:
to .graphcap.
On IRAF, I do:
cl> reset graphcap = /home/USERNAME/.graphcap
cl> gflush
cl> reset stdimage = imt20k
and then issue DISPLAY or IMEXAM while DS9 is running in the background.
This always ends in the following error (this session in on PyRAF but I get the same thing on IRAF cl):
--> display data/image.fits 1
z1=-2.064185 z2=2.07611
Killing IRAF task `display'
PANIC in `/usr/local/iraf/iraf/bin.redhat/x_tv.e': fatal error during process shutdown
Traceback (innermost last):
File "<CL script CL1>", line 1, in <module>
iraf.display('data/image.fits', '1', _save=1)
IrafError: Error running IRAF task display
IRAF task terminated abnormally
ERROR (723, "AWRITE transfer not aligned on a device block boundary (node!imtool,,20000,20000)")
I wonder if there is any way to fix this problem! I believe I followed a standard procedure to add an entry to imtoolrc and graphcap as above, but I'm not sure now what I'm missing. By the way I've had the same problem both on Linux and OS X (Snow Leopard).
Any input would be appreciated!
Thank you for your time,
Taro
I have copied $IRAFPATH/iraf/lib/imtoolrc and $IRAFPATH/iraf/lib/graphcap to my home directory as .imtoolrc and .graphcap, and added an entry:
64 1 20000 20000 # imt64|imt20k
to .imtoolrc and:
imt64|imt20k|20K frame buffer:cn#64:xr#20000:yr#20000:\
:LC:BS@:z0#1:zr#200:DD=node!imtool,,20000,20000:tc=iism70:
to .graphcap.
On IRAF, I do:
cl> reset graphcap = /home/USERNAME/.graphcap
cl> gflush
cl> reset stdimage = imt20k
and then issue DISPLAY or IMEXAM while DS9 is running in the background.
This always ends in the following error (this session in on PyRAF but I get the same thing on IRAF cl):
--> display data/image.fits 1
z1=-2.064185 z2=2.07611
Killing IRAF task `display'
PANIC in `/usr/local/iraf/iraf/bin.redhat/x_tv.e': fatal error during process shutdown
Traceback (innermost last):
File "<CL script CL1>", line 1, in <module>
iraf.display('data/image.fits', '1', _save=1)
IrafError: Error running IRAF task display
IRAF task terminated abnormally
ERROR (723, "AWRITE transfer not aligned on a device block boundary (node!imtool,,20000,20000)")
I wonder if there is any way to fix this problem! I believe I followed a standard procedure to add an entry to imtoolrc and graphcap as above, but I'm not sure now what I'm missing. By the way I've had the same problem both on Linux and OS X (Snow Leopard).
Any input would be appreciated!
Thank you for your time,
Taro
Mike Fitzpatrick wrote on May 10, 2010
The IIS protocol uses a header of signed 16-bit values meaning that under the best of circumstances the max frame buffer size allowed is 32768 columns. However, due to the way this was implemented long ago the actual value is 16384 (I think somewhere in the code is tries to write at least two full lines of data), the error simply means you're trying to write a non-integral number of lines to the frame buffer.
I'd have to dig through the code again to see if I can remove that factor of two, until then you'll need to keep in mind the max FB size, use DS9 to load the image directly, or perhaps use something like a 10K frame buffer so there is an even blocking when you display with the 'fill' parameter enabled to scale the image.
-Mike
I'd have to dig through the code again to see if I can remove that factor of two, until then you'll need to keep in mind the max FB size, use DS9 to load the image directly, or perhaps use something like a 10K frame buffer so there is an even blocking when you display with the 'fill' parameter enabled to scale the image.
-Mike
Last post on May 10, 2010