A feature in history system, please - tracking of a snap.

Next
 From:  Allan (ALLANKIIPLI)
3036.1 
Hello everybody!
I have used Moi program lately to design many fonts by constructing them in a viewplane in 2d.
Moi has outstanding tools for such task. But one suggestion to the developers, if i may:

Please do it as follows: When snap system is on: Can we have by default also on a so to say tracking of a snap: Once snapped to a location,
then when object is moved, the snapped point or tail of two points tracks its target. Is it possible?

It feels wery cool that blend between curves tracks the curves when history is used. But how about conic?
  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
3036.2 In reply to 3036.1 
Hi Allan - that kind of "snap history" is something that I had thought some about when originally working on MoI, but unfortunately it would take quite a lot of effort to make the history system work in that way.

Currently the history system is focused just on tracking objects that are selected by "object picking", like where objects become yellow when you click on them.

When you pick a point that happens to also be snapped on to an object, that is sort of a different category of thing, the system only records the 3D point locations and does not currently know how to also remember the objects that the points were snapped on to.

So that's the difference between Blend and Conic - with Blend you are picking 2 objects by "object picking", while in Conic you are picking points using "point picking" where only the final point locations are recorded for conic.

I would like to try and improve the history system in the future, but probably the main focus for improvements initially will be more in the area of trying to chain operations through operations like booleans.

There also tends to be a problem where too much history being turned on automatically can lead to some surprising results when someone does not expect other pieces to mutate when moving what seems like an independent part. A "point snapping" history could fall into that problem area pretty easily, so it would probably be something that you would have to turn on rather than it happening completely automatically.

So anyway, I've got some ideas on that but there probably won't be any action on that particular thing for some time to come.

- 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:  Michael Gibson
3036.3 In reply to 3036.1 
Hi Allan, also could you possibly show a little bit more about what you mean by having the "tail of 2 points" move - I guess you mean moving a curve's endpoint and the second one so that the tangent direction of the curve stays the same?

If you could show me an example image of how you wish that it worked, that would help me to make sure I fully understand what you are talking about.

- 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:  Allan (ALLANKIIPLI)
3036.4 In reply to 3036.3 
Hello Michael Gibson!
And thank you for reply!
Basicly my wish was to track yes tangency to the snapped curve.
One more suggestion in the history theme: Is it possible to do track also weight of this curve?
Maybe also a weight of blend.

Attachments:

  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:  Allan (ALLANKIIPLI)
3036.5 In reply to 3036.4 
More pictures:
It shows structure of blend curve.
It is very cool that history updates while i am draging around attached lines.
You see real time change in points and how they maintain continuity with lines.

Attachments:

  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:  Allan (ALLANKIIPLI)
3036.6 In reply to 3036.5 
Maybe a first solution to the problem is to add functionality where selecting a blend curve
and toggling on history in a panel.
Then it gives attributes as were available when constructing these curves: Bulge and Continuity.
Maybe it could also recognize when similar history type is handled over multiple selected blend curves.
  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
3036.7 In reply to 3036.4 
Hi Allan,

> One more suggestion in the history theme: Is it possible to
> do track also weight of this curve?

Unfortunately that's not easy to do, since the Conic command does not take any weight parameter as part of its input, it only takes 3 points.

So that means the only history that is recorded for it is 3 point x,y,z locations and not any other factors.


> Maybe also a weight of blend.

Yeah the "Bulge" factor for blend is tracked currently - it is reapplied when you move the input curves into blend around.

This one I do have some hope of making the bulge factor editable in an enhanced history edit function in the future.


- 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:  Michael Gibson
3036.8 In reply to 3036.6 
Hi Allan,

> Maybe a first solution to the problem is to add functionality
> where selecting a blend curve and toggling on history in a panel.

Yup, that's definitely something that I want to add in the future which would be a major enhancement for History.

The only problem is that it will take quite a lot of work to make that happen.

I do hope to make some progress on it in v3 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
 

Reply to All Reply to All