New 24-Dec beta available now
 1-20  21-35

Previous
Next
 From:  Schbeurd
263.21 In reply to 263.20 
Hi Frenchy,

Welcome back and Happy New Year ! ;-)

Hey, you come back with a pretty cool suggestion ! Array with a scale option would be really fun (and useful) I think !
So, Michael, what do you think of this ???
  Reply Reply More Options
Post Options
Reply as PM Reply as PM
Print Print
Mark as unread Mark as unread
Relationship Relationship
IP Logged

Previous
Next
 From:  Michael Gibson
263.22 In reply to 263.20 
Hi Pilou, welcome back and Happy New Year!

re: Array scale function

Are you talking about for grid-type array, circular-type array, or curve-type array?

Could you describe a little bit more about how it would work?

- Michael
  Reply Reply More Options
Post Options
Reply as PM Reply as PM
Print Print
Mark as unread Mark as unread
Relationship Relationship
IP Logged

Previous
Next
 From:  Frenchy Pilou (PILOU)
263.23 In reply to 263.22 

> Are you talking about for grid-type array, circular-type array, or curve-type array?

Yes these three one :)
That works very simply : at any time an objet is redraw during the array function, its "scale" is increased :)
For example if "scale" is 1.2
You will obtain 1, 1.2, 1.4, 1.6 etc
if "scale" is 0.5
You will obtain 1, 0.5, 0.25, 0.125 etc
And all that for the X Y Z
So you can have an homothetic scaling if X =Y =Z or not if X <> Y <> Z :)
Is it more clear?
Of course user must make a good choice of scale :)

Very practical for natural shells for example :)
(a sphere on an arc will be a very cool snail :)

A supreme refinement will be the "line action" of the redraw of object (gravity center, corner, side, ...)

EDITED: 7 Jan 2007 by PILOU

  Reply Reply More Options
Post Options
Reply as PM Reply as PM
Print Print
Mark as unread Mark as unread
Relationship Relationship
IP Logged

Previous
Next
 From:  Michael Gibson
263.24 In reply to 263.23 
Hi Pilou - for grid scale this isn't very clear, since grid scale doesn't just copy an object along one single path, it can go in the x, y, and z directions.

I'm afraid this one may have to wait for a while... I agree it would be cool and good for seashells, but not quite so practical in general.

I get a little worried about adding too many parameters on things. Adding parameters can add to the power of a tool, but there can get to be a certain point when there are so many parameters that the ones that need to be used most frequently become obscured in the big list...

Maybe this would be good to have as a sort of "super-array" special plugin later on at some point.

- Michael
  Reply Reply More Options
Post Options
Reply as PM Reply as PM
Print Print
Mark as unread Mark as unread
Relationship Relationship
IP Logged

Previous
Next
 From:  Frenchy Pilou (PILOU)
263.25 In reply to 263.24 

< for grid scale this isn't very clear, since grid scale doesn't just copy an object along one single path, it can go in the x, y, and z directions
It's sure than some objects will be intersected :D

So only on the "Curve array" will be terrific
With only that : it's just one parameter to add :)

  Reply Reply More Options
Post Options
Reply as PM Reply as PM
Print Print
Mark as unread Mark as unread
Relationship Relationship
IP Logged

Previous
Next
 From:  Michael Gibson
263.26 In reply to 263.25 
But wouldn't the circular array be the one for making shells?

There's kind of another problem though too - scaling doesn't happen just to an object by itself, scaling happens around an origin point. Where would you expect the origin for the scaling to be?

I guess probably the most expected scale origin would be around some kind of pivot of the object, but as you know right now MoI does not have the concept of a separate pivot axis associated with each object.

- Michael
  Reply Reply More Options
Post Options
Reply as PM Reply as PM
Print Print
Mark as unread Mark as unread
Relationship Relationship
IP Logged

Previous
Next
 From:  Frenchy Pilou (PILOU)
263.27 In reply to 263.26 
< circular array
...is just a particular aspect of the "Curve Array" :)
An artificial Pivot of the first object can't be X,Y,Z max/min of the object selected ?
So the scale will be possible to calculate
  Reply Reply More Options
Post Options
Reply as PM Reply as PM
Print Print
Mark as unread Mark as unread
Relationship Relationship
IP Logged

Previous
Next
 From:  Michael Gibson
263.28 In reply to 263.27 
> An artificial Pivot of the first object can't be X,Y,Z
> max/min of the object selected ?
> So the scale will be possible to calculate

That's certainly one possibility - but it would have some side-effects. For example if you had a line, as your line scaled down it would shrink around its own midpoint which would in effect move it away from touching the curve, even if it started out touching the curve. I'm not really sure that would always be the expected result - the problem is that you may need to control the origin of the scaling to get what you want.

This is one area where having a pivot point available on each object can be helpful.

- Michael
  Reply Reply More Options
Post Options
Reply as PM Reply as PM
Print Print
Mark as unread Mark as unread
Relationship Relationship
IP Logged

Previous
Next
 From:  Frenchy Pilou (PILOU)
263.29 In reply to 263.28 
< This is one area where having a pivot point available on each object can be helpful
Of course yes :)

I don't know what is exactly your actual "point P" of an object when you make an array curve function
(but you have something else the function is effective :)
It's not possible to make that ? (Image follow)
Replace the previous by the scaled at the same Point of curve tangence

EDITED: 23 Jan 2007 by PILOU


  Reply Reply More Options
Post Options
Reply as PM Reply as PM
Print Print
Mark as unread Mark as unread
Relationship Relationship
IP Logged

Previous
Next
 From:  Michael Gibson
263.30 In reply to 263.29 
Yeah, for array curve (not grid or circular), there is that point on the curve available as an origin point. But using that would mean that an object that was to the side of the curve would get progressively moved closer to the curve on every scaling, instead of it scaling "in place".

It kind of feels odd to me to put in scaling as an option just only for array curve - I mean right now array curve does not have other "per step" options for additional movements (like Array Circular)... Why scaling and not movement and/or rotation options per-step? Is uniform scale enough or does it need separate scales in x/y/z ...

It just doesn't quite feel right.

- Michael
  Reply Reply More Options
Post Options
Reply as PM Reply as PM
Print Print
Mark as unread Mark as unread
Relationship Relationship
IP Logged

Previous
Next
 From:  Frenchy Pilou (PILOU)
263.31 In reply to 263.30 

< Is uniform scale enough or does it need separate scales in x/y/z ...
Seems Uniform is suffisant, Separate will be a refinement :)
but we have also a proverb "Qui peut le plus, peu le moins"
Who can more, can less :)
And you have that in the normal Scaling function :)

< would get progressively moved closer to the curve on every scaling, instead of it scaling "in place"
Object "snaped on a curve" is the more easy to understand for this sort of manipulations

All that was a suggestion wait if some people are also interested of that :)

  Reply Reply More Options
Post Options
Reply as PM Reply as PM
Print Print
Mark as unread Mark as unread
Relationship Relationship
IP Logged

Previous
Next
 From:  Michael Gibson
263.32 In reply to 263.31 
I'll see if I can write a macro to do the scaling after the array is finished. If this is possible then it would also be possible to adapt the script macro to do other effects beyond scaling as well.

- Michael
  Reply Reply More Options
Post Options
Reply as PM Reply as PM
Print Print
Mark as unread Mark as unread
Relationship Relationship
IP Logged

Previous
Next
 From:  Frenchy Pilou (PILOU)
263.33 In reply to 263.32 
Cool ! Make as you want before after or during just compt the result :)
  Reply Reply More Options
Post Options
Reply as PM Reply as PM
Print Print
Mark as unread Mark as unread
Relationship Relationship
IP Logged

Previous
Next
 From:  Michael Gibson
263.34 In reply to 263.33 
I think I need to add one scripting method to make it work, so that it is possible to get the point coordinates of a point object from a script macro. What I'm thinking of is that you could do an array, and include a point object along with your object that is being copied. The point object would act as the local pivot for the object. Then the script macro would go to every result object from the array and scale it around that point object, and then delete all the point objects.

One nice thing about this method is that you could use any type of array command to generate the object+point copies.

But I think I'd better wait until I finish the image background stuff before trying this though.

- Michael
  Reply Reply More Options
Post Options
Reply as PM Reply as PM
Print Print
Mark as unread Mark as unread
Relationship Relationship
IP Logged

Previous
 From:  Frenchy Pilou (PILOU)
263.35 In reply to 263.34 

Take your time :)
Seems a cool method
Relax : we have some ideas for helix for the moment in another thread :D

  Reply Reply More Options
Post Options
Reply as PM Reply as PM
Print Print
Mark as unread Mark as unread
Relationship Relationship
IP Logged
 

Reply to All Reply to All

 

 
 
Show messages:  1-20  21-35