Page 1 of 1

still need first minor release for -fx hypot(x,y) function

Posted: 2008-04-08T14:26:41-07:00
by fmw42
Some of my scripts are breaking on older versions of IM due to an addition of hypot(x,y) in -fx to replace having to do sqrt(x^2+y^2). I would like to put a trap on the version number where this change occurred, if someone could tell me which minor version of IM when hypot first appeared. I would like to use hypot(x,y) on newer releases, if possible, as it is somewhat faster than doing thesqrt(x^2+y^2)

I have a note from Anthony Thyssen saying:

"
Date: Mon, 24 Sep 2007 13:34:21 +1000
From: Anthony Thyssen <anthony@griffith.edu.au>
X-Brightmail-Tracker: AAAAAA==

I was just informed by Cristy, that hypot() has been added
to make -fx handling of polar type images easier...
"

Perhaps this date will help someone identify the release. From the changelogs, there is no mention, but the date puts it at about the end of 6.3.5 or the beginning of 6.3.6.



Thanks

Fred

P.S.

if possible I would also appreciate knowing when the -distort perspective control points format change - what minor IM release was the first to use the new format.

Re: -fx hypot(x,y) function

Posted: 2008-04-08T15:55:29-07:00
by rmagick
According to my archives, hypot first appeared in 6.3.8. I have 6.3.8-11 installed, but I suppose it could've appeared in any of the earlier sub-releases of 6.3.8.

Re: -fx hypot(x,y) function

Posted: 2008-04-08T16:16:06-07:00
by fmw42
Thanks, but I do need to know the exact subrelease if I am going to trap accurately. I appreciate the reply.

Re: still need first minor release for -fx hypot(x,y) function

Posted: 2008-04-09T22:55:53-07:00
by fmw42
if possible I would also appreciate knowing when the -distort perspective control points format change - what minor IM release was the first to use the new format.

Thanks.

Fred

Re: still need first minor release for -fx hypot(x,y) function

Posted: 2008-06-05T18:02:19-07:00
by anthony
IM v6.3.6-0 was when the distort functions changed to the new format in preparation for the addition of 'least squares fit' to multiple control points.

However any -distort usage before this can be classed as 'experimental' and not for general use.