Author Topic: Transformation matrix vs Acads matrix  (Read 15375 times)

0 Members and 1 Guest are viewing this topic.

gile

  • Gator
  • Posts: 2507
  • Marseille, France
Re: Transformation matrix vs Acads matrix
« Reply #30 on: February 09, 2007, 08:40:25 AM »
To go a little further.

In ObjMatrixV2, Joe Burke gives a 'TransPt' function to transform a point coordinates from (or to) the coordinates system of the reference (block or xref, whatever its nested level) to (or from) the WCS (or UCS).
I made something similar here.

(transpt pt (last (nentselp)) 2 0)
returns the same as
(RCS2WCS pt (caddr (nentselp)))

and

(transpt pt (last (nentselp)) 0 2)
the same as
(WCS2RCS pt (caddr (nentselp)))

But the way to get these results is quite different. Perhaps, comparing the two methods should help to understand the way to work with transformation matrices.
Speaking English as a French Frog

Joe Burke

  • Guest
Re: Transformation matrix vs Acads matrix
« Reply #31 on: February 09, 2007, 10:56:08 AM »
Thanks Joe.

I still haven't got it working the way I want it to.  I now have a better understand of the 'normal' and I can tell, when looking at the numbers, where it is being looked at from, but I can't seem to code it correctly.  I have spent a couple days, off and on, on it, and still am lost.  This is driving me crazy.  :ugly:

Hi Tim,

Forgive me, but I can't find much sympathy for, "I have spent a couple days... and still am lost." :-)

I have no idea how many hours I spent on the ObjMatrix functions, other than I revised them many times over a period of about three years.

Was it worth it? Yes, no question.

And I'm not forgetting you helped me with that puzzle John Uhden posed. The origin of a block definition not at 0,0,0. Which prompted another revsion of the code.

I applaud your effort to understand the nuts and bolts.

I tried to design my code so someone could use it without understanding how it works. But in the end and like anything else, somone who understands the ideas behind it will put it to better use.

Regards