Jump to content

Existing/Demo/Proposed File Organization


Recommended Posts

Early in my career I managed existing, demolition, and proposed work via Layers. The majority of those I work with do this however via Classes; especially with the use of Stories. I'm sure much of it is personal preference, but I'd welcome any input on the matter (maybe reference an earlier thread or video even). Thanks.

Link to comment
6 hours ago, CW2020 said:

Early in my career I managed existing, demolition, and proposed work via Layers. The majority of those I work with do this however via Classes; especially with the use of Stories. I'm sure much of it is personal preference, but I'd welcome any input on the matter (maybe reference an earlier thread or video even). Thanks.

my preference is to use classes. I know many people recommend the separate layer system, but the reason i don't recommend this way is that you have to physically change two layers when you make a change to the demo and proposed design. With the class method, i keep it all in the same layer, and I can ensure that it is all up to date. With the separate layers, you could forget to go back to the demo layer and change the design. 

 

Classes can be used to control the graphic style of the walls so you can use the same objects on the demo viewport and the existing viewport without having to duplicate them. 

 

The other point you mention is that if you use Stories, it is easier to use classes than layers. 

  • Like 3
Link to comment
6 hours ago, CW2020 said:

Early in my career I managed existing, demolition, and proposed work via Layers. The majority of those I work with do this however via Classes; especially with the use of Stories. I'm sure much of it is personal preference, but I'd welcome any input on the matter (maybe reference an earlier thread or video even). Thanks.

 

I use Classes in the first instance, where my primary concern is controlling visibility, but when I also want to control the appearance of existing/proposed/demolished objects in 2D + 3D I use Record Formats + Data Visualisation as well. I use the Data Manager to automatically attach a 'Construction Status' Record to all objects placed in my 'Existing'/'Proposed'/'Demolished' classes with the 'Construction Status' field pre-populated with the appropriate value based on the class (so I just need to make sure objects are placed in the right classes + the Record Format part of the process happens automatically). Data Vis gives me a lot more options for controlling the appearance of the objects than classes alone.

  • Like 2
Link to comment
1 minute ago, Jonathan Pickup said:

Tom W, I like the idea of the data visualisation, you would still need classes for visibility, but the data would handle the graphic style. 

 

You can use Data Vis to control visibility as well but classes are easier (in my opinion). The way I have it set up (using the Data Manager) it is no extra work using the two together (classes + Records).

  • Like 2
Link to comment

I use Layers.

 

I have an "as existing" layer and a proposed layer.

 

The "as existing" geometry largely stays as-is once I have draw things up from a survey. I find it useful to always have there in the background, to refer back to.

 

In theory it can be done in classes, including with classes that are for objects that are existing, but to be demolished. In practice though, I find this just adds an extra level of complication to everything that doesn't actually provide much benefit.

 

In particular wall objects can become very difficult to set up, where you have a run of wall that in the existing/proposed conditions is going to connect with different wall types at different places, may have various finishes added/subtracted, and openings removed or enlarged or added, and so on. You end up with a whole load of bits of walls that are then very troublesome to edit and this is more work than just copy-pasting the existing into the proposed layer and then making the modifications needed.

 

My demolition drawings tend to be the "as existing" with demolished portions highlighted manually in annotations.

  • Like 4
Link to comment

All very helpful. 'Line-weight' referencing wall objects is a big part of this consideration. Organizing walls/doors/windows to be demolished (for instance) in floor plan is straight forward enough. But as I almost exclusively work in 3D these days, there tends to be questions and dilemmas amongst studios as to the best approach. I'll draft a sample shortly in-case anyone is interested. Thanks for now!

Link to comment
37 minutes ago, CW2020 said:

Organizing walls/doors/windows to be demolished (for instance) in floor plan is straight forward enough.

 

Doors + Windows since Wall Closures actually got a lot harder because previously you could have a to-be-removed Window in a 'Demolished' class + a replacement Window in the same opening in a 'Proposed' class + just switch between them depending on whether you wanted to show the existing or proposed conditions. But now, if you're using Wall Closures, you have to include a section of Wall containing the Window in the arrangement i.e. have a section of Wall in 'Demolished' (containing the to-be-removed Window) + a duplicate section of Wall in 'Proposed' (containing the replacement Window). But I still think I'd prefer this to creating duplicate models on different sets of layers if that's the alternative...

  • Like 1
Link to comment
40 minutes ago, line-weight said:

I don't see the duplicate models as a problem, because you are not really trying to maintain and keep up-to-date two models. The "as existing" model, once you've built it, doesn't then change. By definition it doesn't change because it's frozen at a moment in time.

 

I hear you. I do this already with Site Models actually (have an existing Site Model + a proposed Site Model) because I haven't found it possible to do everything in a single model (although I keep meaning to check whether the Site Model Snapshot improvements now allow this). And now I think about it have actually done it with buildings as well on occasion: my ramshackle barn conversion project where I modelled the existing conditions in 3D Solids then used the BIM tools to create a separate proposed model.

 

I think the advantage for me (+ you already alluded to this) of doing the building in a single model is that you are positively identifying the removed architecture + can represent this geometry in VPs + report on it if you wanted. Plus I wouldn't find it ideal having two sets of retained Doors + Windows in the file. But pros + cons with all methods. Wall Closures would definitely be better using two models.

  • Like 1
Link to comment

We use a combination of Design Layers and Record Formats and Data VIs (and a bit of class setup) to manage existing, demolition, and proposed work.
Typical setup:

Design Layers:

EXISTING - This layer contains the as-built plan, which remains untouched and serves as a snapshot of the existing conditions.

EXISTING/DEMOLISHED - This is the design layer where we draw the existing conditions and indicate demolished elements during the design process.

PROPOSED - This layer is used for all proposed objects. During drafting and design, this layer is always overlaid with the EXISTING/DEMOLISHED layer to maintain context.


We also incorporate Record Formats with a 'Construction Phase' field that has three options: Existing, Demolished, and Proposed.
This allows us to use Data Visualization for both visibility and graphical overrides, similar to Tom W.'s setup.


We find that keeping the EXISTING layer untouched as a snapshot helps us refer back to the original conditions easily.

Edited by twk
  • Like 3
Link to comment

I have almost nothing to do with architecture, but I am curious. Do you really need to model the to be demolished walls to the level of detail of including the wall closures?  Wouldn't just an inserted window or door be close enough for something to be demolished? So would that mean you could just do the closures for the new and existing items?

 

  • Like 2
Link to comment
5 hours ago, Pat Stanford said:

I have almost nothing to do with architecture, but I am curious. Do you really need to model the to be demolished walls to the level of detail of including the wall closures?  Wouldn't just an inserted window or door be close enough for something to be demolished?

 

 

The case I was describing wasn't that the wall was being demolished, it was that an existing window was being removed + replaced with a new one in the same opening so the Wall Closures needed to be present in both cases. I can't remember exactly why I had to use a new section of Wall for the replacement Window: I think because the new window was deeper requiring the lining to the reveals to be cut back so the same Wall Closure couldn't be used in both cases.

 

5 hours ago, Pat Stanford said:

So would that mean you could just do the closures for the new and existing items?

 

This wouldn't work because if I used my proposed Window to apply Wall Closures to the Wall, those Closures would be present for the existing Window as well (unless two Walls were used). If you turn off the class the Window is on the Wall Closures associated with that Window remain in place.

 

  • Like 1
Link to comment
7 hours ago, twk said:

EXISTING - This layer contains the as-built plan, which remains untouched and serves as a snapshot of the existing conditions.

EXISTING/DEMOLISHED - This is the design layer where we draw the existing conditions and indicate demolished elements during the design process.

PROPOSED - This layer is used for all proposed objects. During drafting and design, this layer is always overlaid with the EXISTING/DEMOLISHED layer to maintain context.

 

When you say 'layer' singular do you actually use a single layer in each case or do you have more than one...? I am looking at a file of mine of a single storey building that uses eight (!) design layers for the building model. So if I was to follow your method I'd start off with eight 'Existing' layers then duplicate them to give me eight 'Existing/Demolished' layers, then I'd create additional 'Proposed' layers to contain the new geometry? To create drawings of the finished design I'd display the 'Existing/Demolished' layers + 'Proposed' layers at the same time?

 

Would I use less layers if I was using Stories?

  • Like 1
Link to comment

We use stories for new builds only.

I can't recall from memory where using a story in an alteration/extension job was necessary.

To your question about 8 existing plans. Yikes! Is this an apartment complex or multi story project, that youre doing alterations to each level?
Our rationale of using an existing snapshot design layer came about when there were too many instances in the past where we're doing design options, and we have nothing to revert to.

I would in your case of 8 existing plans, keep that in a separate file altogether. Then you'd have a 2 Design layers for each altered level (Existing/Demolished, Proposed Works)
And then yes, the sheet setup for Proposed Plans would have a viewport with those 2 layers visible and classed/data vis'd for each phase
For the sheet for the Existing/Demolished, we would have the viewport only show the existing/demolished design layer visible, again with appropriate data vis/class setups. With the new Viewport Styles, this has dramatically saved time with drawing management.

Edited by twk
  • Like 1
Link to comment
3 minutes ago, twk said:

Is this an apartment complex or multi story project, that youre doing alterations to each level?

 

No it's a single storey building! But I have layers for the structural floor slab, the floor finish, the walls, the roof structure, the ceilings, the roof finish + in this particular file the roof coves. I have high level walls on a separate layer as well so that gives me eight. I just find it easier to separate things out like this rather than having too much all on the same layer. But then everything else (existing, demolished + proposed geometry) is done using classes. I've always been able to get all the drawings I need very easily + get around the model very quickly using saved views.

 

  • Like 2
Link to comment

I always use a class based system for all the reasons mentioned above and:

You can put a new window inside an existing wall because the wall and the window are both on the same layer. You can't embed a window on layer 'new' inside a wall on layer 'existing'...

In our region (Benelux, France & Poland) we use prefixes for existing and demolition classes.

'O_classname' for objects that are existing and will remain.

'X_classname' for objects that will be demolished.

'classname' for new objects.

So we get 'X_walls', 'X_doors', 'X_sanitary', ...

Then we have a command to switch the visibilities of those classes based on their prefixes. The command will switch class visibilities for the drawing or for the selected viewport(s) depending on wether viewports are selected or not.

  • Like 1
Link to comment

I'd be interested to see a file set up to accomodate the following simple situation, using only one layer:

 

Existing: a solid brick wall, with two windows in it. Plaster finish on the inside of the wall, which wraps into the window reveals.

 

Proposed: one of the window openings is enlarged, the other reduced in size. The plaster finish has been removed, and replaced with insulated plasterboard. This also wraps into the window reveals.

  • Like 2
Link to comment
1 hour ago, Peter Vandewalle said:

I always use a class based system for all the reasons mentioned above and:

You can put a new window inside an existing wall because the wall and the window are both on the same layer. You can't embed a window on layer 'new' inside a wall on layer 'existing'...

In our region (Benelux, France & Poland) we use prefixes for existing and demolition classes.

'O_classname' for objects that are existing and will remain.

'X_classname' for objects that will be demolished.

'classname' for new objects.

So we get 'X_walls', 'X_doors', 'X_sanitary', ...

Then we have a command to switch the visibilities of those classes based on their prefixes. The command will switch class visibilities for the drawing or for the selected viewport(s) depending on wether viewports are selected or not.

 

Yes our class naming conventions are similar. ie prefixed (.DEMO-class_name, .EXTG-class_name, class_name <- for proposed/new).
With regards to new windows in existing walls, this is a common headache. Our workaround is to mark the extent of the existing wall being affected by the new window as demolished on the EXISTING/DEMO design layer, and copy and paste in place that portion onto the PROPOSED design layer, however this pasted wall is classed/tagged as existing. Then we place our new window joinery onto this 'existing' wall on the PROPOSED design layer. Sounds confusing to write out, but super easy in practice, with the correct classes and tags applied. The great thing about data vis as well is that it works in 3D, and if you tag a plugin object you dont have to worry about internal plugin class structures. (Which was the headache we had with using WinDoor).

 

 

29 minutes ago, line-weight said:

I'd be interested to see a file set up to accomodate the following simple situation, using only one layer:

 

Existing: a solid brick wall, with two windows in it. Plaster finish on the inside of the wall, which wraps into the window reveals.

 

Proposed: one of the window openings is enlarged, the other reduced in size. The plaster finish has been removed, and replaced with insulated plasterboard. This also wraps into the window reveals.

We would approach this with our workaround 🙂

  • Like 2
Link to comment
2 hours ago, line-weight said:

I'd be interested to see a file set up to accomodate the following simple situation, using only one layer:

 

Existing: a solid brick wall, with two windows in it. Plaster finish on the inside of the wall, which wraps into the window reveals.

 

Proposed: one of the window openings is enlarged, the other reduced in size. The plaster finish has been removed, and replaced with insulated plasterboard. This also wraps into the window reveals.

 

In that specific scenario I would have an existing Wall with brick + plaster components in a 'Wall-Existing' class + a proposed Wall with brick, adhesive, PIR, plasterboard + skim coat components in a 'Wall-Proposed' class + only have one visible at a time. So basically two models separated by class.

 

But if it was my project I wouldn't be stripping off the existing plaster, I'd be gluing the laminated plasterboard on top of it so would do it with two walls side by side: an existing Wall comprising brick + plaster then a second proposed wall lining. The former would always be visible + the latter only when showing proposed views. It does get a bit fiddly with the Windows + Wall Closures (as mentioned earlier) but doable. For example a recent project done this way:

 

As existing:

Screenshot2024-07-24at13_15_30.thumb.png.e0c2fe47c2c3ccffda203a251b12aa01.png

 

Proposed:

Screenshot2024-07-24at13_15_03.thumb.png.975d7edc66e6364216a07f6303e7dfd1.png

 

 

Not saying it's better than any other way but it's the way I mostly do it + it works for me.

  • Like 1
Link to comment
15 hours ago, Tom W. said:

 

No it's a single storey building! But I have layers for the structural floor slab, the floor finish, the walls, the roof structure, the ceilings, the roof finish + in this particular file the roof coves. I have high level walls on a separate layer as well so that gives me eight. I just find it easier to separate things out like this rather than having too much all on the same layer. But then everything else (existing, demolished + proposed geometry) is done using classes. I've always been able to get all the drawings I need very easily + get around the model very quickly using saved views.

 

This approach makes sense for this particular setup

Link to comment

This has been hugely educational. I don't have much to offer as you all seem to be much more experienced... I was unfamiliar with several terms/tools from the start. But I'll say that I'm going to focus on a structured approach based on classes as most of my current clients use stories and only add additional layers if absolutely necessary. I also think I overthink such problems in Vectorworks because I see them addressed so many different ways. Anyways. Thanks!

Edited by CW2020
Link to comment

Join the conversation

You can post now and register later. If you have an account, sign in now to post with your account.
Note: Your post will require moderator approval before it will be visible.

Guest
Reply to this topic...

×   Pasted as rich text.   Restore formatting

  Only 75 emoji are allowed.

×   Your link has been automatically embedded.   Display as a link instead

×   Your previous content has been restored.   Clear editor

×   You cannot paste images directly. Upload or insert images from URL.

×
×
  • Create New...