Why would one be better than the other? Relations vs. Constraints

[lococnc]

I am working with an assembly file that contains a skeleton. This skeleton fully defines the design intent of the parts that are based off it.
There are a number of features within the higher-level parts that refer back to geometry contained in that skeleton. However, I have found that the original author used relations (i.e. d877=d23:0) instead of simply using constraints such as use-edge or offset edge, etc.

My question is -- Is there a reason other than personal choice to do this? Is one way more robust than the other?

Good example, skeleton contains a curve with two line segments joined at the ends and at an angle to each other. They created a reference dimension for the angle (angle is defined by other references). They then used a relation to build geometry in a part that lies directly on top of this curve instead of aligning it to the line segments. Was there a point in Pro/E revisions that did not allow constraints in this manner and you were forced to use relations?
This thread is inactive and closed by the PTC Community Management Team. If you would like to provide a reply and re-open this thread, please notify the moderator and reference the thread. You may also use "Start a topic" button to ask a new question. Please be sure to include what version of the PTC product you are using so another community member knowledgeable about your version may be able to assist.

[HugoHermans]

As far as I know, the difference is in the author. For this case, he
end result is the same, as you know. Constraints are a graphical
approach, so you may expect in favor for most designers. Relations on
the other hand are more for the computer 'nerd', how loves to write and
edit.

Their might be a difference in performance, though. Relations are
evaluated differently then constraints.

Met vriendelijke groeten,
Kindest regards,

Hugo Hermans

-

NV Michel Van de Wiele
Michel Vandewielestraat 7
8510 Kortrijk (Marke)
Tel : +32 56 243 211
Fax: +32 56 243 540
BTW BE 0405 450 595
RPR Kortrijk

[dgschaefer]

Relations are 'buried' constraints, as far as I'm concerned.  They have
their place, but I find that constraints more obviously convey the
design intent.  You have to go digging for them to find out what's
controlling the geometry, but constraints are right there when you edit
the feature.

Doug Schaefer

[lwh]

There is a difference, though:

The relation (as stated in the example) depends on a certain parameter or dimension in component 0 in the assembly, without actually relying on a certain object. So You may be able to replace the skeleton, and as long as the parameter is again present the assembly will update accordingly.

A simple constraint (in the assembly from the component to the skeleton) will break, if the skeleton is replaced or the component is used in a different assembly (where it will still require the original assembly for updating).

Using a copy of published geometry as local reference for the component constraint is a solution for he latter, but still needs the original skeleton.