Anyone else?? Dimensioning Non-Linear Scaled Blocks in paper space

[curtisclark]

Issue

I dimension in paper space (not here to debate that, thanks), and rarely have any issues with it. I've recently started working more with non-linearly scaled blocks, which are amazingly useful. However, dimensioning these blocks in paper space produces outrageous and inconsistent values. (To clarify, blocks are in model space, dims in paper space, and DIMASSOC = 2)

Example 1:

For simplicity, here is a screenshot describing my issue with dimensioning non-linear scaled blocks in paper space. Note how the dimensions are created non-associative (yellow warning symbol with Annotation Monitor turned on):

 

Example 2:

In case my screenshot example isn't clear, I describe my exact situation (inconvenient to fit in one screenshot) below:

In my situation, I have a block of a bridge section in model space, where one copy of it is 'stretched' (scaled by 1.15470054 in the X direction), to account for how the bridge looks when the section is taken along the skewed plane.

For a point of reference, the span of this bridge:

= 2598mm measured perpendicularly to the abutments (the non-scaled block)

= 3000mm along the 30 degree skew / road centreline (the stretched block)

In model space, or through a viewport, dimensioning these two objects correctly reads the above values.

But in paper space, on top of the same 1:20 viewport, when I dimension both of the same distances:

= 2598mm measured perpendicularly (CORRECT on the non-scaled block)

= 150000mm along the 30 degree skew (50x the correct value (3000) on the block stretched only 1.154x?!?!)

The math doesn't make sense to me with the factors 1:20 and 1.154... involved. 

And what is even weirder is that the same thing happens when I dimension something vertically! (I get a readout 50x the correct value on the stretched block, even though the block is not stretched in the y-direction at all!)

Conclusion

I can't find anyone else experiencing the same thing, but it's a hard thing to search for. In order to stay organized I've been dimensioning through the viewport into model space, and then using the CHSPACE command to bring the correct dimensions back up into paper space, which seems to hold their value (but they are no longer associative this way of course).

Anyone have any ideas how to get around, or better yet, fix this?

Is Autodesk even aware of this anomaly?

Thanks!

AutoCAD Civil3D 2019

Windows 10 Pro, 64-bit OS

[cadffm]

I doesn't read the lonh test, just topic and pictures (sorry if i miss the topic now)

Objects in Modelspace?

Dimensions in Paperspace?

Your good dimension are (fully) associated with the modelspace objects,

in this case acad can calculate the "right" length (scaled by the viewport scalefactor).

The bad dimension is NOT fully associated and acad can not calculate what you want.

http://help.autodesk.com/view/ACD/2020/ENU/?guid=GUID-EEF0796B-9C18-476A-A0E8-876BC5346C97

[cadffm · Accepted solution]

Sorry, i was to fast with my finger on [enter]

-

And NO, you can not fix it, it's a documented limitation.

https://knowledge.autodesk.com/support/autocad/learn-explore/caas/sfdcarticles/sfdcarticles/Associat...

[curtisclark]

Yes, blocks are in model space.

Yes, dimensions are in paper space.

Sorry, I thought I included DIMASSOC=2 in my original post. Thanks for pointing that out. 

I've editted my original post to clarify the above.

And yeah, I can see they are not created associatively - though for some reason I never actually checked if those values were actually correct in paper space - and they are. *facepalm*

So, to clarify - my actual question is why aren't my dimensions attaching associatively to a block when it is non-linearly scaled?

[curtisclark]

Oh wow, I see it now in that list "Mirrored or non-uniformly scaled block references"

Thank you!

I guess Autodesk does know about it.

Should I mark your last post as a solution? (even though it doesn't solve the problem, it is the answer to my question).

[curtisclark]

I definitely meant "Non-UNIFORMLY" scaled blocks... not non-linear.

I ran out of time to edit the post, while editting the post.

If a moderator is reading this, could you please change all "linear"s in the original post and post title to "uniform" for the sake of clarity/searches?

Thanks,

[cadffm]

>"Should I mark your last post as a solution?"

Yes, please mark my answer with the link, thx.

[curtisclark]

Done.

Wish I'd found that link myself first - sorry to make you go looking. At least now others may have an easier time coming to the same conclusion without spending the time I did.