OpenMP

It seems that in the latest ImageMagick a few algorithms have been 
updated to support OpenMP, for parallizing an algorithm across CPU 
cores.  The -blur algorithm is one which was updated.  I tried a 
comparison of the -blur operation between latest ImageMagick and 
GraphicsMagick and am confused by what I am seeing.  I can see that 
more than one core is being used (resulting in a report of 135% cpu), 
but ImageMagick is still much slower than GraphicsMagick:

imconvert EADT6030.tif -blur 0x1.0 -depth 8 crap-im.tiff
   46.34s user 2.55s system 135% cpu 35.969 total

gm convert EADT6030.tif -blur 0x1.0 crap-gm.tiff
   15.93s user 0.94s system 98% cpu 17.196 total

I had to add the '-depth 8' to the ImageMagick command since 
ImageMagick wanted to convert my 8 bit TIFF into 16 bits, which would 
be slower.  The CPU meter on my system indicated that the second CPU 
was used for about 30% of the total run-time.

Is the experience of others different than my own?  What open source 
packages have managed to retrofit OpenMP in order to obtain real-world 
performance improvements (and not just higher CPU utilization)?

Bob
======================================
Bob Friesenhahn
[email protected], http://www.simplesystems.org/users/bfriesen/
GraphicsMagick Maintainer,    http://www.GraphicsMagick.org/


-------------------------------------------------------------------------
This SF.net email is sponsored by: Splunk Inc.
Still grepping through log files to find problems?  Stop.
Now Search log events and configuration files using AJAX and a browser.
Download your FREE copy of Splunk now >>  http://get.splunk.com/