List of Dependent Features

[mavigogun]

Is there a mechanism to list Dependencies of elements in the Browser or Timeline?

[lichtzeichenanlage]

Nope

[TrippyLighting]

No, very much to my dismay there is not.

I've been trying to promote the idea of a dependency graph as a more complete  version of the timeline, but it has not caught on. 

The lack of visibility of dependencies and he inability for manipulate dependencies is a major limitation in managing complexity in Fusion 360.  

[lichtzeichenanlage]

Can you provide a link to your entry? I would vote for it.

[mavigogun]

---

@lichtzeichenanlage wrote:

Can you provide a link to your entry? I would vote for it.

---

Almost all the Idea Station submissions relating to dependency have been "Archived" to the waste basket; the only related idea is more rant than coherent.   If we want this to happen, we'll have to work out the details, pitch it to the community, then stuff the ballot box- else it will remain another basic  feature that is weighed by non-user programmers to be unimportant.

[TrippyLighting]

@mavigogun Spot on!

The Fusion 360 team feels likely that this would be too complex for most users.

And while that might be correct, this is a very, very slippery slope!

If you aim an application exclusively at that sort of user then you won't attract those users that can help you make your application competitive with other applications.

[mavigogun]

---

@TrippyLighting wrote:

The Fusion 360 team feels likely that this would be too complex for most users.

And while that might be correct, this is a very, very slippery slope!

If you aim an application exclusively at that sort of user then you won't attract those users that can help you make your application competitive with other applications.

---

Or, you know, they can just capture that whole "I wanna clean my STL before sending it to the printer" crowd.   Lucrative market, I'm sure.

[lichtzeichenanlage]

But Fusion 360 isn't good in STLs at all. Even the export is broken like ...

---

@mavigogun wrote:

---

@TrippyLighting wrote:

The Fusion 360 team feels likely that this would be too complex for most users.

And while that might be correct, this is a very, very slippery slope!

If you aim an application exclusively at that sort of user then you won't attract those users that can help you make your application competitive with other applications.

---

Or, you know, they can just capture that whole "I wanna clean my STL before sending it to the printer" crowd.   Lucrative market, I'm sure.

---

[jeff_strater]

yes, a display of feature dependencies would be useful, I agree.  But, as @TrippyLighting says, this is a pretty advanced tool for a lot of users, and is not yet very high on the priority list.

However... if someone is truly motivated, I think that there is enough info returned by API methods that one could conceivably write a script to return this information, at least at a basic level.  At one point, I was planning to start to write such a script in my spare time, but then realized that I had no spare time, so I never got around to it.  I have lots of things I never get around to...

One challenge is how to display these dependencies.  How to convey the rat's nest that will probably result in a way that could be understood is something I know I would struggle with.

[TrippyLighting]

---

@jeff_strater wrote:

I have lots of things I never get around to...

---

Yes, life is too short 😉

I don't think it would be a good "project" for an individual however motivated that individual would be.

This is something that should be taken on by the Sketch/Modeling teams as there are certain learning effects that will also certainly inform how features in Fusion 360 are going to be developed.

How to graph these things is not something that can be determined purely intellectually. One would actually have to look at a variety of different projects and see what dependencies are in these projects. Patterns will emerge that then inform what and how something can be graphed.

Node editors such as in Blender, or Houdini come to mind.

[PhilProcarioJr]

@TrippyLighting

"Node editors such as in Blender, or Houdini come to mind."

Node editors are insanely powerful and once learned a much better implementation of history/parametric environment. The best part of a node system is it is completely visual and you can clearly see all the information you need at once. Time lines and feature trees are clunky in comparison and only acceptable because CAD people have gotten used to them...

When ever a CAD development company creates a feature rich CAD system that doesn't use destructive editing they will corner the market.

I don't see why making a dependency graph is such a hard thing to do for the dev team. Just keep it basic...any thing is better then nothing which is what we have now....

For the life of me I can't understand why this is not a priority for the dev team....

[mavigogun]

---

@jeff_strater wrote:

yes, a display of feature dependencies would be useful, I agree.  But, as @TrippyLighting says, this is a pretty advanced tool for a lot of users, and is not yet very high on the priority list.

---

Wow.   This is such an important thread for the Autodesk, Fusion team, independent of this specific consideration.   I am not, by any stretch, an 'Advanced User', having many remaining opportunities to develop technique and tool familiarity.   That said, despite still laboring through the awkward beginner-intermediate transition, Fusion itself has impressed on me the utility and importance of such a tool- JUST AS IS HAS EACH OF YOU.   My conception of what a basic, fundamental tool is has evolved with my capacity and intimacy with Fusion.   If the needs of the neophyte dictate priorities, it is not risk but certainty that the shape of the tool will motivate advancing users to explore other options.  

Do not mistake me- focus on barriers to adoption in the resource prioritization and development process is necessary.   Make it easy to get the vehicle in motion, projects off the ground.   If success is measured in many short flights of fancy, such a goal is being met.   For those who need to go further than a short hop, reliable tools meeting the natural demands imposed are essential.   From the heights Fusion has afforded me- from that perspective -my 'flight plan' is limited by the tools at hand and anticipated tool failure.   A lack of feature dependency tracking is part of that estimation.

Initially, after learning here there was no undiscovered tool for exploring dependencies, I was conflicted, strongly believing reliability of existing tools should be priority one, the introduction of new tools detracting from that; however, dependency tracking isn't really a new feature- it's a missing cog in the existing parametric machine.    You've got a new guy here asking where the 4th leg of the table is, 2 long time users lamenting the absence, and a guy working at the table factory saying he's dreamed of making one.    How many Idea Station votes is that worth?

[PhilProcarioJr]

@mavigogun

A lot of us EE have been asking for the core functionality of Fusion to be the focus rather then all the fancy tools they have been focused on as of late, but there is an issue with all of this. The people that can't afford simulation, sheet metal and generative design see Fusion as a cheap way to get these advanced tools so the majority of the voicing population drowns out the real need for Fusions core development. I for one definitely want to see simulation, sheet metal and generative design, but not before the dev team can deliver a stable Fusion that has a rock solid core set of tools. Fusion's stability has gotten worse over the last few updates and yet we still have to use work arounds for rudimentary tasks that should have working tools to achieve. I would have thought that having ways to fix all timeline errors would be a priority but here we are still with no way to fix all errors in the timeline. Some will say if you plan everything out and work correctly it won't happen.... This is what happens when the sales teams drive development. Now I understand the concept that they need to make money to continue development of Fusion, but when all these users start hitting all the roadblocks in Fusion and all of it's limitations they are in for a lot of frustration. Their choice I guess. At the end of the day this stuff has been beaten like a dead horse for years and we are told that they disagree with the development direction headed down the wrong path. So as much as I hate to say it is what it is.

[mavigogun]

@PhilProcarioJr, much thanks for your testament; lacking an actionable lever to pull, knowing I ain't alone is appreciated- especially when so well expressed.   I've come to understand the Expert Elite label to be a signifier of grim pragmatism and endurance, resigned so to a certain level of frustration.  

[b.jacquesF9G7Y]

Almost 3 years later now, I'm adding myself to the list of those yearning for better handling of dependencies or constraints. I've started using Fusion360 only this January, but before that I've worked with Catia for years, so I do have some background, and I have some intermediate-complexity Fusion projects going on now.

To illustrate my current struggle a bit: I have a sketch with many symmetries, many of which use the major axes (X, Y, Z). This results in lots of symmetry icons along those axes (projected auxiliary lines of those axes, to be more precise). Now if I e.g. need to delete one specific symmetry, it's a needle in a haystack situation, having to zoom in for selecting individual symmetry icons, but then not being able to see the sketch elements they belong to, as they are out-of-window; and even if they weren't, the "highlighting" of the related elements is so subtle and only activated by hovering that it's almost useless (why not make the highlighting more obvious? Why not keep the highlighting when clicking on the constraint? And why not also do it the other way round, highlight constraints when selecting, or hovering, on one of the dependent elements?). Clicking a constraint only tells me which type it is, not which elements it belongs to, because those elements don't have unique identifiers.

All that fishing for constraint relations is often only necessary because they don't seem to work properly in the first place: Often, some sketch lines are shown blue, and visually I too have the impression that they are underconstrained. Yet, they are unmovable, and adding constraints is not possible. That's a first major roadblock, where analysis tools or a list of dependencies would help. E.g., hover over, or right-click on the offending sketch line and see which constraints are affecting it. Sounds simple and logical to me.

In that last situation, I can't fully prove whether it's an actual bug, or if there really is some very hidden constraint explaining that strange behaviour. But what I also observe regularly: Changing a dimension e.g. from 12 to 13 mm results in an error; but making 10 increments of 0.1mm works. Makes me question Fusion's solver once more. And then I'm not yet talking about the increasing slugishness of the solver when working with a bit more complex sketches - it's great that from one sketch I can derive many bodies and manipulations, but what use is that if the software gets jammed with the simple task of a 2D sketch? Yes, I've deactivated all performance-hungry options, and I've got a Zbook i7 32GB 6GB VRAM workstation, half of which is only used, and after restarting Fusion it often gets better, so I don't even get where the jamming comes from. The only thing that somewhat supports my trust until now is the absence of Fusion crashes (only Win10 automatic restart managed to make me lose work in Fusion; can be deactivated in the group policies of Win10 Pro if you're lucky enough to be the admin).

It's this kind of strange or buggy behaviour in the elementary functions which really makes me doubt whether I can trust Fusion enough for building my small business on; one would think that those are teething problems, but a 7-year-old should be able to walk properly if he's not sick. Same goes for the constant, Win10-style, cloud-based updating of potentially every aspect of Fusion. Of course, this is fine for a hobbyist where reliability is not important and fancy new features are cool. But in any professional situation (and I'm not even talking about NASA or nuclear power plants, this goes just as well for small businesses which have to rely on their IT systems), this is a simple no-go. They need something they completely own and control and which remains available and compatible for many years, in worst case as an isolated legacy system. I know the big pro-CAD tools for large companies, but the advanced hobbyist and the small business just fall in between those and Fusion. Quite a shame that Fusion can't seem to cover that gap, by showing just a bit more professionalism and reliability. Instead they go the Google way, where tools and features are introduced, changed, and cancelled at will, call it "perpetual beta", bananaware, or worse. If this is a sales tactic for Inventor, it might be working, but it might just as well lead me to other programs.

Sorry for this rant, but there hopefully is some relevance to this thread's topic, and if I can vote or contribute to these issues, please let me know, thanks.

[david.fosters9]

The only technique I have found is to suppress the feature and see what else is supressed. If anything, it shows more "dependent" features than fewer, but at least I can work through and make my own judgeemnt whether it truly is or not. Then decide what to do. It's a real fiddle, but....

[FrancoGabrielPerez]

It's been 4 years. This is hopeless. It seems that every time I have a problem o look for a way to do something I end up finding a thread from years ago from someone with the same problem with no solution in sight.
If the program knows that another feature is referencing the one I try to delete, can't it at least show me the name of that feature? even if it's not the most helpful it's better than nothing.

[kandennti]

Sounds like an interesting theme.

I can't guarantee that I will reach the goal, but I will give it a try.

An add-in that displays a tree of documents related to the active document, rather than a timeline, is available here.

https://github.com/kantoku-code/Fusion360_GOKOTAI/tree/main/GOKOTAI/commands/DocumentTree 

You should be able to find top-level documents easily.
However, all are in Japanese.

[laughingcreek]

@kandennti -a useful add on, but not what the thread is about.  the poster was asking to find feature dependencies. 

sorry @kandennti just re-read what you said.

for anyone from AD reading this thread, I think a re-read of what @PhilProcarioJr said in post 13 is in order.

[TrippyLighting]

---

@laughingcreek wrote:

...

for anyone from AD reading this thread, I think a re-read of what @PhilProcarioJr said in post 13 is in order.

---

Yep, and we keep saying that and we keep getting the same answer.