Interesting. Went back and did another test.

Source image from NASA's APotD: 
https://s3.amazonaws.com/fg_stuff/STS134VAB_cooper.jpg

ImageXXX() thumbs:
https://s3.amazonaws.com/fg_stuff/thumb_STS134VAB_cooper.jpg
https://s3.amazonaws.com/fg_stuff/thumbM_STS134VAB_cooper.jpg

<cfimage> thumbs:
https://s3.amazonaws.com/fg_stuff/thumb_STS134VAB_cooper2.jpg
https://s3.amazonaws.com/fg_stuff/thumbM_STS134VAB_cooper2.jpg

Looking closer, the <cfimage> image result is one pixel taller than
the ImageXXX() code. I use the auto-calculated dimension feature for
the ImageXXX() function:

ImageResize(imgData, gcPhotoThumbWm)

(gcPhotoThumbWm = 400, set in a defs file)

<cfimage> requires both parameters, so I calculate that out:

  <cfset intH = round(gcPhotoThumbWm/ cfimage.width * cfimage.height)>
  <cfimage action = "EDIT"
      srcfile     = "#argSource#"
      destfile    = "#argTarget#"
      width       = "#gcPhotoThumbW#"
      height      = "#intH#"
      type        = "JPEG"
  >

Hmm. I just did another test using the same calculated dimension
value, and while closer, the CFIMAGE tag still results in a file that
is smaller and has less "noise"; the ImageXXX() edges look sharper.

https://s3.amazonaws.com/fg_stuff/cfimage_STS134VAB_cooper.jpg
https://s3.amazonaws.com/fg_stuff/imagexxx_STS134VAB_cooper.jpg

Mostly aesthetics here, but I'd have to start digging through the
source's implementation to see why they differ, and I don't think it
really matters in the long run :D

-- 
official tag/function reference: http://openbd.org/manual/
 mailing list - http://groups.google.com/group/openbd?hl=en

Reply via email to