Sorry :)

["Arne Chr.Jorgensen" <instel-at-online.no>]

Hi,

I am dead tired - and should shut up :)  I read briefly over some
of the stuff in my mail,  you know it is also sent back to
yourself.  Sorry,  I should have read it before I mailed it.
Spelling errors,  missed something some places - and I have
swapped some words around - so they are not in logic order.  :)

But I may also have scared some of you from commenting on the
subject, by giving the impression that - here should some
authority in this field,  say something.  Well,  I certainly made
a fool of myself in many ways.  Why not take the risk ?
Any input is welcome.

I may not have made any sense with the stuff I wrote, - first I
say that we should try to go in one and the same direction, then I
favor two different systems,  tcl  and varkon.  Well, there is not
anything wrong by having two.  Tcl would be a light, fast to
modify, system that most would be able to join in on.  The other
is a bit different - it is larger, may take some more time, and it
is a bit above the level we have discussed.  But it would
certainly be nice to look into the future a bit on what can be
done.  Get me ?

So I don't see any problem in bringing up two different systems
here.  What do you think ?

I have also looked briefly at qcad as Urs Schütz suggested.  This
is a 2d system, that still needs a lot of work, and it lacks
something we discuss here.  Things more in the direction of
machining, gcodes, an more.

I just will sum up the two different systems that I can see
possible right now:

1.  Tcl/Tk
-------

Get a binding to Vogle for added "canvas" routines in tcl/tk.
Use what we could from TDCAD - like the DXF file reader and
display it.
Make this package output gcode.
Add some edit functions of the graphic screen.
Add some edit functions for the gcode.
Add some  nice Tk buttons and other stuff.
Maybe add some look ahead marker and position marker from the
machine.
Add some subroutine calls to run several small gcode programs in a
batch mode.
Maybe add Jon Elson "pocket" programs if he don't mind.
If Dave Anderson or someone else would add some visual machine
motions, then great.
Maybe have something for a tool changer to ?


2.  Varkon/ MRSEV/ NML

This is just to gather some info on what a future program
environment could look like.  This would be
a larger and more CAD/CAM - that is something that would be a
better  "total" design packet.

( With  Tcl/Tk,  I think a lot would use the CAD system they
currently use for the drawings, then export it and
read it as a DXF file. It is just to hard to implement all the
stuff you would like in a tcl package.  Just to
have ortho osnap - for example - takes just to much computing time
and update time of the canvas. In TDCAD
there is a cross hair, - and that alone took just too much update
time. With Vogle, you may improve a lot on
many graphic routines - but you would very far from a real CAD
system with a geometric data base, and stuff )

I will quote Dan here:

"In the last month or two, it appears that CAD companies are
releasing code faster than we can digest it."

Yes, and we have a problem to understand what we should be looking
for too.  I heard that  LinuxCad claim to be
Acad for Linux, - but this is far from the truth. I have heard
that it lack so many things, it is useless. The thing is
- how long time will we use, before we finally find this out ?

That is why I would like to hear from someone who have some
knowledge about Cad/Cam and give us shed some
light on this.  The fact that we would very much like to add
functionality to EMC and the Linux environment, brings the
thought of some input from the experts here.  From NIST.

And if any have some thoughts along these lines, please add in
here, whatever you think would be nice.


I hope this made things a bit clearer.


In closing, - we should should write a few notes of personal
opinion about packages we know a bit about, and
put it into the fact or the cookbook.  Then it would be easier to
"rank" these, and see what is of great interest,
what is of less.  Then we may not have to spend to much time on
those that don't fill "our" wishes.

//ARNE