2D or not 2D?

[Martin P]

I have been working on drawings that our other Revit user has been producing, and have found that he has been resorting to using a LOT of 2D linework to rpoduce his drawings. I must admit that Imyself suffer from the opposite, I tend to go overkill on wanting everything to be either 3D, or at least a detail component.

The thing that bothers me about it is, the 2D linework method that David is using seems to be allowing him to produce drawings at a faster rate than me, as I get bogged down constantly and try and do everything "properly" (with regards to revit).

I must admit that I am coming around to davids way of thinking, the whole point is that you do stuff faster isnt it? He comes from an archicad background BTW, which he teels is how it works.

I am going try and get away from trying to produce a "perfect" model, and if, for example the stairs dont work with modeling, or the handrails - I am not going to spend hours messing about with it, I will do it 2D and forget about it.

What are your experiences of this, I find it hard to think like this, but must admit I cant really see any problem with his method - after all it is DRAWINGS we are producing not 3D models.......

[bclarch]

If your intention is to produce construction documents then I think that the hybrid method (3D overall model & 2D detailing) is effective and faster. If your desires include renderings, walkthroughs and material take-off then you have to stick with true 3D as much as possible.

[hand471037]

On the jobs we have that are more information management than drawing production, I.E. Facilities or Leasing Management, almost everything is 2D. We tend to only make something 3D if it needs to be 3D, if we're not doing elevations or perspective views we don't even worry about 3D at all, and just use the level tools for information organization. Some of our leasing models are all 2D, one level per floor, all levels 10' apart, one DWG-linked plan per floor with leasing/area/whatever info done in Revit over the top. On the more design/production side of things, there are times where, if something only is shown in plan (like furniture) and no interior rendering is going to be done, that we will just use simple 2D furniture families in the plan.

We try to keep a top-down approach, where we only define things within Revit enough to serve to purpose of the job at hand; if the scope changes we can always come back and define them more and/or swap them for 3D versions. This is kinda like Hand Drafting, in that you only do as much as you need to get by. :)

[designer56644]

I am sooo glad to see this thread...

I came to the board today to discuss this very topic.

[Henry D]

Very timely thread... I've been redrawing an ADT project and 75% of my time has been spent creating families. I can draw an entire floor plan in Revit faster than I can create some families. Unfortunately, a lot of them have to be created because they appear in plan and elevation. But as to Baths, Kitchens, stairs for this particular project- no one will ever see these in 3D.

[PeterJ]

Since we discussed this on the phone earlier I've been thinking it through in a little more detail.

You have to know what stuff you will ever need a section through/elevation of. Interior fittings such as kitchens and bathrooms depending on whether commercial or residential and then on the value, we would be likely to produce internal elevations of so 3D representation is important, but it can be limited, so a 3D WC that is a filled region as a nested component in both plan and elevations is a good thing in as much as it will show right (masking tiles etc.) but won't ever be required as a subject in a render.

However, for all that, the initially much derided drafting tools in Revit do what is required of them and do it pretty well so anything which is hidden or will only show in one view can be drawn pretty effectively as a 2D image. When I get paid to render something then I'll take the pain of getting it all correctly modelled.

P

[J-G]

Interior objects are easy in 2d becasue often they are only represented one on the floor plans, but I wonder how much 2d line work are you using on your exterior elevations? Window trim and exterior details are really only going to be shown in elevations so you won't be gaining any time by doing it in 3d...but there is such a temptation to do it in 3d becasue Revit can give you a nice renderable model and help the owenr visualize the project.

In ADT I used the rule that if it was going to be represented in more then one view, i.e. sections and plans, or elevations and plans, then we would model it so we didn't have to draw it twice. With ADT we never used the elevations for anything more then a background that we could add detail to, but as I say now with Revit I am more then tempted to do it all in 3d. Even stairs...do you really think drawing them 2d would safe so much time? If you have to cut a section through a complex winding stair the 2d work can be time consuming also.

[aggockel50321]

I try to force myself to do everything 3d. Every time I take the 2d shortcut, it seems that someone comes in & says "can I see a 3d view with a section bpx cut??" All that 2d work up in smoke, and time to tie myself to the A__Kicking machine.

The other problem are the software interfaces. I keep trying to execute Revit commands in Acad, & get shot down. That's probably related to being over 50.

& then there's the layer thing. I'm now trying to work with a site plan developed by three different consultants, & it's a layer nightmare. Just trying to set up a pen plot stb file to plot it makes me want to upchuck.

[beegee]

I still follow the maxim - " if its going to show in more tha 1 view - model it ".
It should not take long to model the majority of elements that appear in most buildings. Its usually just a matter of tweaking an existing family or using one as the basis for the new type. Of course, if you do a lot of custom work, that could be different. Custom stairs and balustrades still take a bit of time, but IMHO, must be modelled anyway, as they will invariably be the subject of multiple plans, elevations, sections and details.
I quite agree with Peter J and Andrew G, that you can never really be sure where you may need sections etc in the future, so play it safe and model wherever possible and releatively easy to do so.

beegee

[Martin P]

It is interesting to know that some do a lot of linework in revit. I really am going to look hard at the way I do work with Revit, aside from anything I think trying to do a lot of 3D work often makes things very complicated, when they needent be - leading me to become very irate and stressed out with it. I also feel that I spend WAY too much time creating families, I can have a basic building done in half a day, then spend 1 or 2 days to get the doors,windows etc working. I think this may be slightly down to the Revit tutorials etc, they give you the impression that families should be parametric, or indeed done at all - there is no tutorial on how do just do a "get by" if you are in a hurry - and can I tag 2D linework as door or window or whatver?

BUT - if I am to start doing more 2D drafting with Revit, can I suggest they get rid of that "cannot keep lines joined" warning box - and more imortantly why the hell cant I export detail to a detail component family???? - and groups need to be sorted, they are great for 2D drafting - they allow you to isolate stuff to work on, move it as one object etc, but I am always wary they are going to make me crash.

Anyway, I'm going to give it a go - whether importing autocad stuff or doing it in Revit, I am about to make things easier on myself, and get the job done quicker. And at the end of the day... the drawings look better I think if you have done a bit of actual drawing yourself

[PeterJ]

Martin

There you were with your inline nested detail component families associated with jambs and heads and cills and your overlay views to get transparent filled regions.

Poised for Revit greatness and now you want to become 2D jockey again.

I think the answer may lie in the question that Brittany Fortin posed regarding templates. If one has a REALLY good template with the right stock details, views, view templates levels and so on preset the time taken getting your drawings to a point where you have the luxury of spending an extra couple of hours on a family will come more readily and the better the family the better the model. Not only that the better the model the better the schedule so overall your documentation time might be kept down.

What I think may emerge is a divide between buildings where everything can be lifted from a catalogue - and therefore in years to come may be available as a family for download from the manufacturer's website - which will respond well to a high percentage of modelling and those with a large amount of bespoke work which might be more easily created through drafting. Of course, the latter building types are the ones we will all find ourselves under greater pressure to model for rendering purposes.

Pete

[Martin P]

I know Pete :( but at the moment my employer is unfortunately looking for quantity - we have a whole heap of work to get through, and not many revit users!. It goes against the grain for me to do things like this, but at the moment I am under a lot of pressure to get stuff out the door, so long if it looks right on paper, he is not bothered - and our other Revit user seems to be churning stuff out faster than I do, and he has only used it 6 months - I find it hard to explain why I take a bit longer because I do it "properly" whereas our other user does it quickly, ie 2D, never bothers making new families etc. At the end of the day on paper it looks the same - sometimes on paper the heavy 2D method looks better, and was quicker!. Obviously long term I will be better off having made lots of families etc, but long term isnt an issue for the boss at the mo, just "get the drawings out" move on to the next job and forget about it.

I fear though that this is going to trip either david or myself up, and we are going to have errors between say a plan and section because of working like this - But then when the s**t hits the fan, we will be asked how we managed to make the mistakes!! I am kind of being forced down this route......

[aghis]

I currently try Revit, however I have a long experience in AutoCAD 2d and 3d and ADT.
My actual technique is to model the totality of the project for every phase and add details as the project develops.
At the beginning I need a good modelling of the site, the walls (the "enclosure" of the building) and levels (slabs). In that way I can have basic output sections, 3d etc. to work with other architects, landscape architects and engineers.

At this "first stage", doors are "basic doors" satisfying basic characteristics (width, height, and alignment to the wall) organised in "Styles" or "Families" according to their use or future types. Later on, I ad 2d information that is based on online catalogs of manufactures, this information is added as view depended 2d information ("custom components" in ADT, or "symbolic lines" or "detail components" in Revit) and the "Families" "Styles" and "Multi-view blocks" are redefined to reflect the new information. I do this since a lot of the online resources of the manufacturers are 2d DWG files, and I need to do the less possible work on them (clearing up the drawing, taking of the layers etc.).
This approach gives me the possibility to be fast and to be able to use different products for each project without having to create complex family elements from the beginning of the project.
To the final stage I model the ultimate choices in 3d to produce sections and renderings and of course I do the annotated details on 2d.

[christo4robin]

I am wondering specifically about wall base relative to this 2d/3d discussion. Seems to me that 2d in interior elevations, using a filled region, is very easy--but then there is the clients "oh, can I see it from that doorway". Given that I go 3d with wall base--it seems like a wall sweep isn't flexible enough--that leaves generic family.

Anybody have a better way than generic family?

[aaronrumple]

So why isn't the wall sweep tool working for you in this instance? I've done some pretty detailed trim with this and a couple of wall hosted families....

[bmadsen]

Yea, 3D/family and let Revit to do the coordination for you, or 2D and go back to cross-checking your sheets? As some said, it depends on the job requirements.

It also depends on the longer-range plan. While building families you are also building your templates for future work.

But can we be a little more specific here? What are the components you find you are having to build? What's missing from the basic set of Revit families?

[bclarch]

Martin,
It sounds like your problems are related to a great extent to the office politics of 2D vs. 3D rather than to the actual methodology. I would suggest a "clear the air" meeting with your boss. Lay it on the line and let him know that you can get things done faster by resorting to more 2D work but that it might increase the chance of errors creeping into the documents. Ask him for explicit instructions on how he would like you to proceed. Then, when the s**t hits the fan it will have been his call. (Of course, that doesn't guarantee that he won't still blame you but you can at least say "I told you so".) :) It depends a lot on what type of relationship you have with your boss but it sounds like you need to tackle the political issues in some fashion.

[christo4robin]

I have a stair that comes right up to the wall. At that stair/wall intersection, the trim just cruises right on past (where it would not actually be). Also, the wall in several instances (level changes) has different base constraints yet still the same floor level. Consequently, the sweep moves up and down the wall inappropriately. I could make several different wall styles, and that is where I began to ask myself if it would be easier with an inplace generic model.

I'm open to suggestions!

C

[christopher.zoog51272]

I have a stair that comes right up to the wall. At that stair/wall intersection, the trim just cruises right on past (where it would not actually be). Also, the wall in several instances (level changes) has different base constraints yet still the same floor level. Consequently, the sweep moves up and down the wall inappropriately. I could make several different wall styles, and that is where I began to ask myself if it would be easier with an inplace generic model.

I'm open to suggestions!

C

forgive me if this sounds dumb...
but did you know that you don't have to make a compund wall to add a sweep? Just use the host sweep tool, you can "pull back" and contrain the trim where it needs to be.

if that's not it, perhaps a image would help us understanding what the problem is?

[PeterJ]

Funny, Z. I read this query and just thought wow he must have the most fantastic skirtings, some kind of panelled or castellated affair. Perhaps it has just been too long a week.

P

[christopher.zoog51272]

....Perhaps it has just been too long a week.

P

well said..... almost the weekend here.....almost :beer:

[christo4robin]

Your imaginations probably paint a picture more beautiful than the reality. It is a house that has existing and new, and some odd transitions between. Tops of walls are not consistent, and neither are bottom of walls, so the sweep (just had a thought, I've been working with sweeps in the wall properties>edit>sweep area, not modelling>hosted>sweeps). So, the wall base sweep does not have a consistent datum from which to place itself.

Did I just muddy the waters more?