|  |  | From what I understood of the code, it looks like media method 3 uses a
fixed AA threshold of 0.2 and completely ignores the value of the
aa_threhold parameter. What did I overlook?
	I'm referring to lines 732 and 768 in media.c in function
sample_media_rec which I believe should be replaced by something like:
DBL aa_threshold=1000; /* or any very big value */
IMEDIA IMedia=*Media_List;
while (IMedia) {
  if (IMedia->AA_Threshold < aa_threshold)
    aa_threshold=IMedia->AA_Threshold;
  IMedia=IMedia->Next_Media;
}
if ( Colour_Distance(C1, C2) > aa_threshold ) /* this is line 732 */
...
	Or maybe, to get faster result, precalculate the threshold and make it
a paramter of sample_media_rec...
		Jerome
-- 
*******************************
* they'll tell you what can't * mailto:ber### [at] iname com
* be done and why...          * http://www.enst.fr/~jberger
* Then do it.                 *
*******************************Post a reply to this message
 |  | 
|  |  | You are correct!  This will be fixed in the next release.  Thank you.
-Nathan
Jerome M. BERGER <jbe### [at] enst fr> wrote...
> From what I understood of the code, it looks like media method 3 uses a
> fixed AA threshold of 0.2 and completely ignores the value of the
> aa_threhold parameter. What did I overlook?
>
> I'm referring to lines 732 and 768 in media.c in function
> sample_media_rec which I believe should be replaced by something like:
>
> DBL aa_threshold=1000; /* or any very big value */
> IMEDIA IMedia=*Media_List;
> while (IMedia) {
>   if (IMedia->AA_Threshold < aa_threshold)
>     aa_threshold=IMedia->AA_Threshold;
>   IMedia=IMedia->Next_Media;
> }
> if ( Colour_Distance(C1, C2) > aa_threshold ) /* this is line 732 */
> ...
>
> Or maybe, to get faster result, precalculate the threshold and make it
> a paramter of sample_media_rec...
>
> Jerome
> --
> *******************************
> * they'll tell you what can't * mailto:ber### [at] iname  com
> * be done and why...          * http://www.enst.fr/~jberger
> * Then do it.                 *
> ******************************* Post a reply to this message
 |  |