Managing element based VG overrides in RA2008

[Calvn_Swing]

To start off, Kudos to the Revit team for turning out a great product yet again! Given, I didn't get a single one of my own wish list items, but most of the additions were sorely needed, and seem to work beautifully so far.

Now, I'm reveling in all the new visibility controls, but I have to admit I'm having nightmares as well. You see, I can now picture each user editing the VG overrides per element in every bloody view with no way for a project documentation manager to track the changes. Now, instead of forcing the users to be intelligent about how they model things (by category and subcategory) and how they modify the display of these things (by changing the settings for an entire category), they can build things wily nilly and cover their screw ups by fudging the element display alone. All of a sudden view templates become problematic as they may override these element settings. Also, I can no longer promise consistency across multiple views as each view has so many unique settings that can't be consistently applied...

Anyway, you're catching on to my apocalyptic tone I hope. Point being, though I love all this fine grained ability to change element display characteristics in RA2008, I can also see this power being grossly abused by the unknowing and inexperienced (or simply rushed by a deadline) Revit user.

Has anyone figured out some strategy to manage these kind of changes? Does anyone thing I'm crazy and blowing this out of proportion?

Chime in!

[twiceroadsfool]

I dont think youre crazy one bit. While i LOVE a lot of the changes and flexibility we were given in this release, a lot of it will be a CAD Managers nightmare.

My technique is: Tell people not to Overide things. Occasionally select every element in the view, Over Graphics in View > By Element > Reset. It puts it back (i believe) to its native appearence.

We havent rolled it out yet, but im especially concerned about maintaining the integrity of the model. For instance, anyone can now change the Surface pattern of a floor, in a particular view. Well, if youve modeled floors of different types, BECAUSE they have different patterns, this compromises your works integrity.

TO be clear, i am NOT in any way slamming the new tools. I think they are all great, and i think factory did an amazing job. My favorite release since i started in 8.1.

Its like Referencing Other View, it just leaves a door open for us to insert our own mistakes... But you cant take human error out of the model completely, someone has to drive...

[Calvn_Swing]

I know what you mean. Integrity integrity integrity. It's worse for us since we're an "integrated" firm. We use the architectural model for construction so it has to be "right."

I love it when the developers give us new features like this, I just wish they'd also give us the ability to track them, manage them, or reset them that is part of the new feature set. The reset option you mentioned is a nice "wipe the slate clean" kind of solution, but it doesn't help you figure out what was changed. If something legitimately needed the change, then you don't want to remove it.

Anyway, glad to know I'm not alone...

[muttlieb]

The reset option you mentioned is a nice "wipe the slate clean" kind of solution, but it doesn't help you figure out what was changed. If something legitimately needed the change, then you don't want to remove it.
In order to help manage graphic overrides, it would be nice to have a 'Reveal Graphic Overrides' button that would highlight all objects that have had their graphics overridden in the view, much like the new 'Reveal Hidden Elements' feature.

[Calvn_Swing]

Or, have an additional tab appear in the visibility graphics dialog called "element overrides" that has a list of all the element related overrides for the view. Or, a new element based dialog that could incorporate these element based VG overrides, the new hidden objects, etc...

Although I like the graphic interface for seeing hidden objects, it is time consuming to open that in multiple views, or to try to apply those settings across multiple views that already exist. If there was also a dialog to interact with in these cases, it would give us a text based interface that is much faster, and it could also have it's own View Template Settings.

While I'm on the subject, I think it would be nice to be able to turn off VG tabs in view templates. For instance, we like to manage our larger campus projects by linking building files. As such, we have view templates for the views of these files. Unfortunately, the VG settings for the host file are consistent, but each view usually has slightly different settings for the linked files. I'd like to make a view template that could set the host VG settings and leave the linked revit files settings alone. Or, change all the category based overrides but leave the element based overrides alone.

[aaronrumple]

. All of a sudden view templates become problematic as they may override these element settings. Also, I can no longer promise consistency across multiple views as each view has so many unique settings that can't be consistently applied...I think view templates become almost useless with the introduction of element graphics overrides.

As a designer - I'll abuse this feature more than anyone else. But for production drawings - it is going to be a real time waster. It looks like an AutoCAD UI designer was turned lose in Revit to me....

I've already seen people mixing up tempo ray hide and hide forever. So expect things to just "vanish" from your prints on a regular basis. Better up the omissions insurance...

I think they really missed the boat on this feature - big time. What people wanted was a way to create depth in drawings. Users repeatedly pointed to the ADT section view depth tools as a model that could work even better in Revit than ADT. In this instance ADT is more Revit-Like than Revit.

As for hiding annotation, it is a stop gap measure. You add a section mark. Then you run around to each view and hide it. Add another - repeat the process. Very time consuming. Many users have asked for the ability to hide callouts/elevations/section and detail markers by type. This would have integrated well into the visibility settings and view templates. It would have been very easy to manage and virtually automatic.

I give this feature an "F" and think it is the first feature introduced that works completely against trusting Revit as a total BIM system....

[Calvn_Swing]

"I give this feature an "F" and think it is the first feature introduced that works completely against trusting Revit as a total BIM system...."

Man, tell us how you really feel Aaron!

I love the tool as a designer as well. There are just some things where I want to be able to control it to heck with the rules!

I'm not quite as against it as you are (on principle). I think this tool is needed in rare cases, and I think all the potential pitfalls could be overcome by some simple management tools. The problem for me is that no such tool is provided. I can't manage element based settings at all. I can manage category/sub category based setting using view templates. I can manage project settings via object styles. I just need to be able to manage element based settings.

My understanding of Revit comes from a database background. Finite element control is just as important in databases as category-level controls. However, you are provided with tools to manage all levels of hierarchy, and in databases you can create new levels of hierarchy or add divisions within a level of hierarchy. What frustrates me about Revit is how inconsistent it is with it's approach at being a modeling database. I want to create my own bloody categories. I want to decide what categories can and cannot be part of a schedule. I want tools to manage things like levels in a project, or named reference planes, or how about a schedule of all the parameters in a project and what their settings are by category of element.

It drives me crazy that there is all this "hidden" magic that goes on behind the scene in Revit. To me, this is what makes the software difficult to learn and implement. It is hard to get a consistent understanding of the software because there is so much inconsistency.

It's easy to dimension and label families off of reference planes, but try explaining why you have to do it to someone just learning the program... "Because you have to" or "That's just the way it is" or even the eloquent "Because I said so." These are the educational tidbits that make Revit work? Please.

If I export to ODBC and import to Access, I can filter, sort, group, edit, etc... any bloody parameter I want. I can have infinite control of the properties of any modeling category, subcategory, or element. If there is something I can't do, I can add a parameter to do it. I can use text fields in formulas, I can write complex If/Then statements that Revit can't handle because it doesn't have a null function. (If A then B, else don't do jack). I could use a "family type" parameter to control any other parameters I like. Blah blah blah... I could go on for hours about what I can do in a true database app, and what I can't do in Revit.

Now given, some parameters need to be restricted since the program needs them to function. But currently Revit is WAY overprotective. If Revit gave me a little more freedom, I could put together a schedule that mapped every element with an override (because there is a parameter to handle this) and what view it was overriden in (there is a parameter for this too) and then build my own management tool.

Of course, Revit won't let me schedule either of those parameters. It won't even let me schedule one of them individually of the other. Arrrrrg!!!!

Anyway, I'll stop the Rant for now. This is a big deal to me, obviously, and I think it really inhibits what the software can do, and more importantly what the user can do with the software. It isn't like Revit is breaking new ground in this sense. There are tons of capable database applications out there that could be used as a model, and technology that could probably be licensed relatively inexpensively - heck, they probably already have it licensed. I just wish they'd trust us users a bit more with it. We'll probably screw up royally a few times, but that's no different from what we're already doing...

My two (thousand) cents... (Ok, it's only 708 cents (cents = words))

Sorry for the novel.

[Calvn_Swing]

Many users have asked for the ability to hide callouts/elevations/section and detail markers by type.


Of which I am one. This would be so much more useful. Especially since we can create types of our own. It gives us complete control over what shows and what doesn't. Sigh - database issues yet again... (we should be able to write a filter for this currently! No bloody excuse!)

[Mr Spot]

This was the first thing i thought about when i saw this new tool. The reveal hidden elements tool releaved some of my concern, but its still a bit of an issue.

It would be great to have an administrators setting that would enable a Cad Manager to disable this function on certain workstations. Then perhaps only one person "the cad manager?" has authority on whether something gets overridden by element.

That said, i've used it many a time already for CD and it is a rather handy tool to have and very important when you are using dependent views. In order to control your annotation for each part...

Perhaps it just needs a bit more thought with regards to managing this tool...

[iru69]

I can certainly appreciate the perspectives here, and I agree that additional "management" tools could be helpful. However, while there seems to be a lot of vague suggestions, I fail to see how anything has fundamentally changed for the worse.

The whole "CAD Manager" thing definitely has an important place, but it also needs to be kept in perspective. Revit is a means to an end, not an end in itself. Revit exists for the architects (and drafters), not for the CAD manager.

The whole "BIM" thing gets tossed around here endlessly, and I'm sure it carries great weight for those that are interested in it, but for the vast, vast, vast majority of users, Revit has for far too long been too heavily weighted towards appeasing such lofty concepts at the expense of an easier work-flow. None of the half dozen Revit users in our office knows what BIM is (and guess what? they don't care). What they know is that they spent two hours trying to get the freaking downlights on their electrical plan to not show up on the wrong floor.

I've incorporated so many work-arounds into my families in order to graphically achieve something that the new overrides and masking tool make possible in a few seconds, that to explain the domino-effect would take several pages of writing.

Revit had to give in somewhere. The refusal to incorporate anything resembling layers brought us here. Users were using worksets as layers, filters as layers, phases as layers... for all the ranting about how much layers suck, there's a reason just about every cad/graphic application on the planet has some form of them. Revit can't continue to impress by saying that it doesn't need layers, when we're simply calling layers by different names. It seems to me that the new tools have offered a smart compromise.

[Chad Smith]

My first thought when I saw this feature was definitely sloppy modeling and family creation, and still is.

With the addition of Filters back in 9.0 (I think), this solved all my override problems, while still conforming to strict processes. But with the new overrides this all goes to hell. And unless someone can make a differing argument, I think Filters was a short lived and now obsolete tool.

[Exar Kun]

I agree with all of the above. Coming up against a deadline shortly I've already found myself 'cheating' with some things and promising myself I'll go back and fix them later.

As others have mentioned - a full listing of hidden/overidden objects somewhere would be a good addition but I do agree with Chad that this has pretty much made filters obsolete.

[davidcobi]

Filters have Filter Rules based on parameters available in the families. Graphic Overrides require you to hand pick and choose what you want filtered. There very likely offices out there that have setup more strict standards for parameter usage in families and, for some tasks, rely on the more automated parameter driven visibility override controls that the Filters tool provides.

[Calvn_Swing]

The whole "BIM" thing gets tossed around here endlessly, and I'm sure it carries great weight for those that are interested in it, but for the vast, vast, vast majority of users, Revit has for far too long been too heavily weighted towards appeasing such lofty concepts at the expense of an easier work-flow. None of the half dozen Revit users in our office knows what BIM is (and guess what? they don't care). What they know is that they spent two hours trying to get the freaking downlights on their electrical plan to not show up on the wrong floor.

I admittedly come from a unique position in these discussions. My firm is focused on an "integrated process" which is marketing speak for "design - build." For us BIM is THE feature in Revit that matters. By using third party applications we can effectively use our model to estimate, schedule, and build a project. We are talking about modeling things like cranes and job site trailers, scaffolding, and temporary barriers. To us the "document" we are producing is the "model," not the drawing set. That being said, I don't care that much about these new features long term that I'm worried about short term.

Why? Because short term most of our consultants, subs, and suppliers are still using 2D drawings. By enabling these element level visibility settings I now have to worry about us sending incomplete information in 2D form to these groups. Now given, I had to worry about that before. But, before my worries were along the line of "have we modeled it?" If so, I had to only check one setting to make sure it was showing up, or not. I've found that the previous category and sub category based settings were sufficient for 90% of the visibility changes we wanted. I've found that filters calling from parameter values are not only sufficient for the other 10%, but that they are desirable. In addition to giving us an ability to graphically override an object, it allows us to then cost, schedule, and simulate those same objects differently than their peers. In short, we very rarely need to change an object's visibility simple for the sake of a pretty print. Most changes we need are because there is some fundamental difference between one element and another of the same type or family that is at a scale Revit does not currently model at effectively.

A great example of this is walls. All out interior walls have parameters to denote if they are rated or acoustic partitions. Revit doesn't give a fart about this difference graphically because it doesn't model down to the detail of cutting the drywall to fit the grooves of a metal deck and filling any openings or gaps with fire/acoustic caulk. However, this difference makes a big price difference for us. So, we have an instance parameter that when set activates a filter in our code plans to display rated walls differently. We have a graphical check that is automatic from all our template files. It took maybe 5 minutes to set it up once, and set it up right, and now all our projects do this automatically. I'd consider that a much better solution than having different wall types for rated/non rated or for changing only the graphic display of the wall without changing any of it's information. Then again, I'm going to be using the model to build a building in 6 months so I have a reason to care.

Short term it scares me because I know people will abuse this feature and use it to inconsistently change the way the model displays, and that this will likely be to the detriment of the drawing set we send out.

Long term, I don't care about this feature because long term I believe that we'll be sending models to all our subs, not plotted sheets of paper. I also believe that long term we'll all be forced into using the "I" part of BIM because it will give those of us that do a huge advantage in the marketplace. Those that don't will either catch up, or be marginalized to niche markets where BIM still doesn't matter. And, those markets will shrink to nill eventually, though it may take 50 years.

What's frustrating to me is that strict architecture firms are resisting the "I" part of BIM, and this is influencing how the construction industry thinks about it. The vast majority of Architects not using BIM to it's potential are the reason why the AGC's guide to BIM recommends to contractors using BIM to discard the Architectural model and start from scratch with their own. Am I the only one that thinks this is a waste of time and effort?

Sorry to rant yet again, it is just frustrating that I see so many professionals in architecture taking this great leap forward, just to take a few steps backwards (admittedly, a biased opinion). It frustrates me because this tendency is holding the entire industry back as well, and I want it to move forward now. Then again, I shouldn't complain. The more resistance it faces, the longer my company can hold on to a significant advantage in the market.

Go BIM!

Edit: dduarte, amen! Filters are hardly obsolete. They are parameter driven, unlike any of the other visibility controls. (Ok, sure the VG settings are "type" parameters of a "family" but since we can't access, schedule, or add to them, they're effectively useless for representing real "information") Seriously, Filters are the MOST important visibility tools for us. For me, the new element level settings are useful in extremely rare situations, or for presentations only. Once I have better control over how families display in Plan and RCP which I pray I'm given in the next release, I think I'll only use these element level settings in presentation views.

[Scott D Davis]

To reset all over-ridden elements in a View, window select all elements in a view (over-ridden or not) and then right click on an obejct, go to the override dialog box and click Reset. All elements are then reset to default display settings.

[aaronrumple]

To reset all over-ridden elements in a View, window select all elements in a view (over-ridden or not) and then right click on an obejct, go to the override dialog box and click Reset. All elements are then reset to default display settings.
However that has to be done view-by-view and then you also have to check hidden by element separatly. It is a multi-step process.

[Calvn_Swing]

To reset all over-ridden elements in a View, window select all elements in a view (over-ridden or not) and then right click on an obejct, go to the override dialog box and click Reset. All elements are then reset to default display settings.

This is not an efficient management process. Nor is it even effective. What about the objects you want to remain overriden? Not to mention that in some views "selecting all" can take a lot of time and resources - in some models we have it can take over a minute just to select all. Then to try and make a change....

I know this is a workaround. However, this is something that REQUIRES more than just a workaround...

[Teresa.Martin]

Hi. Good comments from everyone. I am actually having one of our in house programming gurus check into this and see if there is a way to run some sort of macro to reset everything per project or at least give a report (something like the XML report for interference checking would be nice!!) I will keep you posted if he discovers anything. I would be a useful little tool.

Best regards,

[sbrown]

Ok, lets calm down for a minute. The database isn't changing just the way we look at it. So these "overides" are being done for the printed page(something we hope to get away from in the future). So my suggestion is to have model views(you can create them anytime that have no overides. then you use overides when you need them for printing issues. Has anyone tested to see what happens when you export to dwg using these overides?

The linework tool was the same, this is just easier and more powerfull graphically which users have been complaining about for the life of Revit.

If a users isn't trained enough to know that you don't overided a stucco wall to look like brick, thats the trainers fault. These tools shouldn't be used to make one material look like another, but just to make your life easier graphically.

[arq.rfernandes]

I think, integrating SQL selection with Revit (like Quick Select in AutoBAD), could solve this kind of problem.

ie, we can "Select all the object with status=hide" (sql) ... and then right click on the specific object --> override dialog box --> click Reset.

All objects are then reset to default display settings, with options for one or more windows view.

[rjcrowther]

From memory (and it has been a while since I have used it) ADT has several levels of display control and the suggestion is to use the last level (equivalent of by view) very sparingly. The reason is to avoid the same issues that the new overrides will (can) create in Revit.

So, really, it gets to the users. And really, Revit is like most other things on the planet, it is only as good as the person behind it.

If it is going to take 2 hours to get some down lights to do the right thing then override them - unless of course you can justify to your client that 2 hours at ?? dollars (pound, rand, roubles, rupee, etc, etc) per hour is money well spent to remove a given number of down lights from a piece of paper.

If filters are a better way to go to help guard against sloppy modelling and family creation then use filters. Both tools are now available, use them both.

If the BIM concept is heavily involved in your perceived market advantage then use Revit to your advantage. If that means avoiding this new feature then avoid it.

If you are a CAD manager then I would say I don't envy you.

To me, based on a similar level of control in ADT, the message is to:

use the new feature but use it sparingly.
Maintain the integrity of the model where you can but don't get overzealous about it.
Use filters as rule because that is more likely to hold me accountable when modelling.