In that son of 'butter dish lid' file - I can delete all solids and surfaces (including those hidden), but they still exist in the history tree ... why would that be ?

Jol

I do not know the inner workings of the code to be sure. But are they Ghost remnants of surface work? Can you click on them in the tree, RMB and "Show" them to see if they are hidden, then delete them?

Ahh...didn't read "those hidden"...mea culpa.

Edited by user Monday, February 5, 2018 12:21:49 PM(UTC)  | Reason: Not specified

Hi Bill - what you say is true

However, my point is that whilst you CAN drive geometry directly from the split edge of a split surface ... you CANNOT drive associative geometry directly from the split edge of a split surface.

Why is this an exception to the rule ?

It's surely an omission !?

The only way to do it is to also do a 'surface intersect' - and drive your associative geometry from the intersect curve

If this is as intended - the 'split surface' help docs should sate that you cannot drive associative geometry from the resultant split edge

I'm only making a fuss about this cos I've been using this software since year 2000 and I didn't get this ( ... and I really needed to).

So, devs - is it a problem or not ? - will it be fixed, will it be ignored - I'd really like to understand

Jol

So you're advocating that a history tree should be no more than 2 levels deep ? OK

A different example ... Imagine slicing an 'infinite plane' through a surface tube - in order to cap it with a 'cover surface' at a certain height or a specific angle.

Well you can !

But if you later change your mind about what height or what angle that cap should be (ie you move the IP) .... you're out of luck !

(Example attached)

This is an associative modeller - each function is built on the results of the previous function.

That's how it's supposed to work ! ... or am I wrong ?

If some tools do not and cannot play the 'associative' game ... fine ... let's call them out - put it in the help notes, make them bright purple - something.

However, if this is simply an oversight on the dev side - let's get it fixed !

If I'm being stupid - also fine, but make me understand why

But for heaven's sake let's get rid of all these inconsistencies and misunderstandings - then perhaps we can spend more time modelling beautiful things and less time debating the tools !

This works as expected until you put the cover surface in place. Then the end of the cylinder freezes.

I take your point Murray - for example, you could never extrude a solid from a rectangle with a radius'ed corned - and later change the value of that radius. This was never an issue as you generally don't radius an object until - at least until it's a solid.

I think changing a radius value (as the driver of a solid) was possible when we had constraints - if I remember correctly

Either way - in an associative modeller, perhaps it would be smart to define objects that can't update associatively in the History Tree with an Asterix or something - as from this screenshot ... it looks to me as if what's below the radius can be updated if I change the value of the radius.

I know this is an old thread and I'm an old time user of Shark. The butter tray should have been started with a solid and should have been chopped up using your surfaces. That's how I would approach it. There's a hierarchy in designing with modern 3D CAD and solids should be the first in MHO. Surfaces should be the last approach. This intersection of surfaces just went over my head as far as determining what is the base control surface.

I do agree that the intersect tool does not update automatically if you move the intersecting surface and that should be looked up by Tim.

I agree with you Jol and ask the same question : why does a simple example like that fails ?

To link to my favorite topic "Model to sheet" : Have you tried to do a 2D drawing from your tube surface ?
--> It works, but ...
-----> it has no associativity. (modifying, moving, scaling the surface has no effect on the 2D views).
I know it's not common to make a drawing from a surface model, but a warning message should blink...

I agree with you Jol and ask the same question : why does a simple example like that fails ?

To link to my favorite topic "Model to sheet" : Have you tried to do a 2D drawing from your tube surface ?
--> It works, but ...
-----> it has no associativity. (modifying, moving, scaling the surface has no effect on the 2D views).
I know it's not common to make a drawing from a surface model, but a warning message should blink...

Triplicate yes !
I swear I posted only once...
In Germany this week, not the best place to have red wine.