topic Re: Negative dimensions now exist in 2023, but not implemented properly in Inventor Forum

Negative dimensions now exist in 2023, but not implemented properly

msonderegger — Mon, 23 Jan 2023 14:14:52 GMT

Like many of the users of this forum, I too have wished for the ability to use negative dimensions in sketch values. Naturally the goal being that objects will actually flip from the positive side of the dimension reference back to the negative side.

Somehow 2023 has addressed this...or at least, they've addressed a very small part of it.

So, now in 2023, you CAN write an equation in the EDIT DIMENSION dialogue which results in a negative value. We migrated from 2020, and I'm pretty certain Inventor didn't let you do that in 2020. Before you rush off to test this out, its not what you're expecting...Not even close.

The sketched entity doesn't actually move to the "negative" side of the object it was dimensioned to! It's as if Inventor "displays" a negative value, but then takes the absolute value of that number anyway. Is this really the intent? Seems awfully dangerous to accept user inputs that the tool is not going to interpret properly. "Uh, yeah, you CAN use the number 5 in Inventor, but we're gonna interpret the number 5 as a 7 instead, and those negative numbers you wanted, yeah, you can use 'em, but we ignore those too"?!?!?!

What happens when an offset sketch plane goes from positive to negative? It moves in the opposite direction, as expected!!!

What happens when an assembly constraint value goes from positive to negative? - Again...yes, as expected!!!

But yet in sketcher negative numbers still have no effect. I don't think there is any other CAD tool that handles negative numbers this way.

Re: Negative dimensions now exist in 2023, but not implemented properly

kacper.suchomski — Mon, 23 Jan 2023 15:40:53 GMT

Hi, I think a better way would be to post this suggestion on Idea Station.
The forum is more for solving ad hoc problems.

Re: Negative dimensions now exist in 2023, but not implemented properly

CGBenner — Mon, 23 Jan 2023 17:46:45 GMT

@kacper.suchomski

I think it is ok to post this here in case this was unintended behavior. Perhaps someone from Autodesk will see it and have some thoughts. @johnsonshiue?

Re: Negative dimensions now exist in 2023, but not implemented properly

johnsonshiue — Tue, 24 Jan 2023 01:19:06 GMT

Hi! I think I know the issue here. Though negative values are allowed in the dimension input box and parameter expression, it does not mean the plus or minus value carries the sense of direction. It is still being solved as the absolute value.

I think the user expects by changing the sign, the sketch will move in an opposite direction. But, I don't think that is supported in Inventor 2D/3D Sketch yet as in Assembly Constraints.

Many thanks!

Re: Negative dimensions now exist in 2023, but not implemented properly

gregory_nickol — Tue, 24 Jan 2023 10:58:48 GMT

This behavior sounds concerning and poorly implemented to me, if intentional. First of all, it is an unnecessary addition. If someone intended for an absolute value to be used, the option is already included in Inventor to simply write: abs(parameter). Meanwhile, now values that would formerly have thrown an error for being out of the intended bounds, may now show up generating incorrect parts. One of the common themes I see with implementing changes to Inventor is minimizing how they may adversely affect existing workflows. It is highly possible that existing workflows rely on this error message to prevent issues of someone using values that are outside of the intended input range.

On top of being an unnecessary feature that saves marginal bits of time for the few people that want an absolute value but to not have to spend the extra 2 seconds to type abs(), it directly contradicts with the ability to introduce this idea board idea:

https://forums.autodesk.com/t5/inventor-ideas/negative-dimension-in-sketch-to-flip-dimension/idi-p/9252271

When ideas fail to gather support in a timely manner, they get archived, so you would expect the opposite to be true. If you sort by top voted ideas, this idea is on page two, so it should be hard for the dev team to miss, and it's one of the top by number of votes per amount of time that it has been on there. If we accept that we are willing to break potential existing workflows by now allowing negative numbers (in a way that I argue most users would NOT expect to happens and is illogical), then the only reason not to implement the above idea would be difficulty in programming it versus a change to workflows.