Solidedit and crossing mode

Falk Schacht

Hello,

I would like to use the "_solidedit _face _move" command (or any other solidedit command) like the stretch command. I know that in AutoCAD this is possible, which makes it really easy to "stretch" solids. The command itself works like it expected, with the exception that I cannot use crossing mode to pick more than one surface. Is there a reason for this or do I miss something?

Joe Dunfee

 And I would like it possible to stretch multiple solids!

-Joe

Roy Klein Gebbinck

See:
https://forum.bricsys.com/discussion/29338

Joe Dunfee

 Thanks for the reminder Roy.  I had forgotten about the special technique necessary to make it work.  Still, wish the standard stretch command would do the trick.

-Joe

Tom Foster

Roy, interesting you've brought up https://forum.bricsys.com/discussion/29338 again.

It covers, better than anywhere I've seen, issues of 3D UI consolidation and intuitive simplification around familiar Acad 2D-style commands, rationalising the duplication of functions and nomenclatures that have come from bolting-on excellent new capabilities in Brics. I'm really disappointed that v17 hasn't  tackled this before the beta-like confusion gets too permanent.

There is scope here for a quantum-leap in ease-of-use, leveraging the muscle-memory of countless Acad users - a leap that Adesk itself failed to take.

Not saying it would be easy - an Apple-like UX design challenge, which may not have been possible until AI now comes ready to guess users' intentions.
Until now such consolidation into fewer, unified 2D/3D commands may have necessitated multiple Microstation-like settings-boxes, to cover all the variant/ambiguities that machine-logic may need to know before completing the user's intuitive desire.

Tom Foster

Roy, interesting you've brought up https://forum.bricsys.com/discussion/29338 again.

It covers, better than anywhere I've seen, issues of 3D UI consolidation and intuitive simplification around familiar Acad 2D-style commands, rationalising the duplication of functions and nomenclatures that have come from bolting-on excellent new capabilities in Brics. I'm really disappointed that v17 hasn't  tackled this before the beta-like confusion gets too permanent.

There is scope here for a quantum-leap in ease-of-use, leveraging the muscle-memory of countless Acad users - a leap that Adesk itself failed to take.

Not saying it would be easy - an Apple-like UX design challenge, which may not have been possible until AI now comes ready to guess users' intentions.
Until now such consolidation into fewer, unified 2D/3D commands may have necessitated multiple Microstation-like settings-boxes, to cover all the variant/ambiguities that machine-logic may need to know before completing the user's intuitive desire.

Joe Dunfee

@Tom Foster:

Roy, interesting you've brought up https://forum.bricsys.com/discussion/29338 again.
 AI now comes ready to guess users' intentions.

The smarter a program gets, the harder it is to know what it is thinking.  And therefore the more frustrating it is to use a program.  Though,I too, would love to see the stretch command work with solids.  The fact that I had forgotten how to do it, and even forgot that it was possible, it a testament to the reason it would be helpful.

There may be some reasons, that are not apparent on first glance.  Sometimes keeping features separate is a good way to simplify it.

-joe

Tom Foster

@Joe Dunfee:

The smarter a program gets, the harder it is to know what it is thinking

There must always be a way to correct its AI assumptions - actually that's how AI learns, for each user.

Tom Foster

Something funny there, trying again:

@Joe Dunfee:

The smarter a program gets, the harder it is to know what it is thinking

There must always be a way to correct its AI assumptions - actually that's how AI learns, for each user.

Tom Foster

Still wrong. This is my reply to Joe:

There must always be a way to correct its AI assumptions - actually that's how AI learns, for each user.

Joe Dunfee

@Tom Foster:

There must always be a way to correct its AI assumptions - actually that's how AI learns, for each user.

The challenge to this is that the user must somehow be made aware of what the AI had assumed.  And and AI system must know to ask the question of the user, and not ask too many questions.  Otherwise people will just "click through" without bothering to read the question.

A lot of the process of doing CAD work, is not about reading and responding to a prompt.  Rather, it becomes a series of steps I take automatically.  An example is when I create a dimension, I automatically go through a series of keystrokes and clicks, without thinking about what is happening.  If a dimension becomes no longer associative in the process, what if the AI assumes that I meant for that dimension to be associative and then fix it?  What if I knew that a certain kind of action breaks associativity, and was later surprised when my dimension changed without my wanting it to do so. 

For a real-world example, I know that creating a paper space dimension on a polyline creates an associative dimension that may have its end points re-positioned if I edit the polyline. But,  what if I do something in model space, such as insert a break in a polyline that represents a wall, so that I can put a door there. Any dimensions already there in paperspace that show the length of the entire intact wall, will no longer show that dimension.

Knowing this, I may copy the polyline, and then deleted the original. This breaks the associativity, so that I can edit the polyline confidently, and not have to scrutinize every dimension on a multi-page documents for any errant dimensions.

-Joe