Trying to solve some issues at work

[drcad2003237304]

We use sketchup. And already you guys know where I'm headed with this. What takes me 2 weeks to do in Revit our designers can do in sketchup in a day. Obviously, I can give them plans, a section schedules ect ect ect. However in the design phase I don't need all that yet. So I tried Vasari thinking it would be nice to be able to model something extremely fast. Yet, obviously since its based on masses, I can't place windows or doors.
In the end, the issue is we end up having these beautiful sketchup models and they go into the trash when we start CDs. Then I have to model an entire building in Revit. Client wants to know why it's taking so long ect. Project managers say elevations take 2 or 3 days in AutoCAD. The logistics seem really messed up and I was curious how everyone deals with this. Should I be using masking regions all over and fub it so it goes faster? I mean, we don't render off of these models, and we use Revit for documentation purposes only with our unified keynotes, legends and schedules.

[thechunkygentleman]

You know what? There isnt really an easy answer. The reality is that Revit isnt yet capable of doing what Sketchup does. You cant design as quickly and as flexibly in Revit as you can in Sketchup. Equally you can fudge up a set of CAD elevations quicker than you can get them out of Revit (mostly because you dont need to work everything out properly in CAD).

I think there are three realistic ways of dealing with this. 1. You accept the need to use other software early on and you only Revit-up at a later stage (often Stage C) 2. You manipulate your programme so you have more time up-front or 3. You run Revit and Sketchup/CAD in parallel early on (with consequent need for more resources on the job).

I have been through all three and having more time is the most satisfying but you cant always achieve that.

[damon.sidel]

From the very short description you present, it sounds like your firm has fundamental design/workflow/process issues that aren't necessarily about software. And not to minimize your frustration, but Revit has a pretty serious track record now with a lot of resources out there as to how to implement it. If you want to be the agent of change, I'd highly recommend lobbying for getting some professional help: a consultancy to come in and implement Revit or at least give a presentation, answer questions, and dispel myths.

In a few years after you've been through the transition, perhaps a few times at different firms, make sure to get a DIFFERENT kind of professional help: some therapy. You'll probably need it. ;)

That's my 2 cents.

[drcad2003237304]

Alright let me ask this. Does anyone's firms out there use Sketchup and Revit together? Are you running into these issues? Are people actually designing in Revit? Our designers say that is madness and they wont try it. I almost agree with them, Revit might not be flexible enough for creative ideas.

Right now we have our designers using Autocad to generate preliminary floor plans. Then those go to other designers to make an exterior sketchup model. Together, these go to the city for approval. Then I get a sketchup model and those autocad floor plans. It takes me 2 or 3 weeks to remake the revit model to incorporate the autocad plans and make the exterior look exactly like the Sketchup model. If it's even remotely different from the Sketch up model the project managers get annoyed. Making the plans in Revit is easy just make dimising walls, make some floors, exterior walls then link in the units. Issues I'm having is making the exterior look like the sketchup model in less than a week for the elevations. I find it time consuming to get all the walls to join properly and get the wall sweeps right.

Are these timelines appropriate in your guys' firms? Do you guys have the same workflows?

[cdatechguy]

Yeah, I'm using a very nicely produced sketchup model too and am trying to create it in ACA. It's taking me way longer in ACA vs if I did it in Revit. So why is this? Both are modeling programs right? Well, I have 8 years experience working with Revit and only 2 weeks experience working with ACA. I know ACA is capable of doing what I want, its just getting those steps down that are different than Revit's. I could have the building all done and working on the interiors if I was working in Revit!! Sound familiar? You get the same comments from Autocad folk moving to Revit.

You get good at a software, that is all. I switched from Autocad to ArchiCAD so Revit came easily to me after that.

[Paul Munford]

I'm having the same issue moving us from AutoCAD to inventor. The designers can model stuff so quickly in sketch up, it makes inventor look like a dog.

I'll be following this thread with interest...

[MikeJarosz]

I'm unrolling my yoga mat and sitting cross-legged. Close your eyes. Here is my mantra:

"Revit front loads the project"

You are doing twenty times more work in the beginning of a project than the Sketchuppers are. The payback comes when you move into DD and beyond. You will produce everything faster and much more accurately. At the end of the project the sum total of all the work on a project will be significantly less than the Acad way.

You are most likely dealing with people who know only the Acad workflow and don't want that to change. Revit has a completely different workflow from Acad. To use Revit effectively, you simply have to abandon the Acad workflow. Revit is upending the industry the way CAD upended manual drafting. That revolution ended a lot of careers. The BIM revolution will too.

Revit is taking over the industry. McGraw Hill, in an extensive survey of the design/build industry, has found that 71% of projects are starting in Revit. Clients, especially institutional clients are demanding it. Contractor acceptance is growing.

Start reading employment ads. You will probably not find one that does not mention Revit as a requirement. Imagine what happens to a resume that only has Acad 2009 on it.

[damon.sidel]

Alright let me ask this. Does anyone's firms out there use Sketchup and Revit together? Are you running into these issues? Are people actually designing in Revit? Our designers say that is madness and they wont try it. I almost agree with them, Revit might not be flexible enough for creative ideas.

Right now we have our designers using Autocad to generate preliminary floor plans. Then those go to other designers to make an exterior sketchup model. Together, these go to the city for approval. Then I get a sketchup model and those autocad floor plans. It takes me 2 or 3 weeks to remake the revit model to incorporate the autocad plans and make the exterior look exactly like the Sketchup model. If it's even remotely different from the Sketch up model the project managers get annoyed. Making the plans in Revit is easy just make dimising walls, make some floors, exterior walls then link in the units. Issues I'm having is making the exterior look like the sketchup model in less than a week for the elevations. I find it time consuming to get all the walls to join properly and get the wall sweeps right.

Are these timelines appropriate in your guys' firms? Do you guys have the same workflows?

Not to get into the philosophical realm, since many others have already commented on that, I'll make a few specific suggestions/comments.

The first step in the process you outline is "we have our designers using Autocad to generate preliminary floor plans." As you note, doing the floor plans in Revit is easy! Why not have the designers do this step in Revit? You'd get a huge benefit right from the beginning with GFA calculations, room tabulations, etc. These are all things you should be doing in early concept and schematic phases, right? You can still export to DWG and import to SketchUp. Then you have a Revit model to START and based on your timeline, it will take you only a week to get the exterior to match the SketchUp.

We definitely don't use Revit during concept design and we often don't use it for some or all of SD. It mostly depends on the familiarity with Revit for the specific team. We don't abandon other modelling softwares (Rhino in our case) even after we go into Revit. On some projects we continue to use Rhino for certain studies depending on the nature of the design (is it curvy?) or the composition of skills on the team (who knows what?). cdatech guy rightly points out that in most cases, it is skill level that determines how quickly people can model/draw, not something inherent to the software.

Ultimately, MikeJarosz is correct: we are moving into a BIM (and usually that means Revit) world. Personally, I embrace as many new technologies as I can and find the best use for them. I think we should strive to use Revit, SketchUp, Rhino, Autocad, etc. all together, all at once for whatever they are best at doing. The best tool for the job. (Sorry, got into the philosophical.)

[MikeJarosz]

It is unfair to pit a sketchup whiz who has been using it for years against a new Revit user.

[dkoch]

It is unfair to pit a sketchup whiz who has been using it for years against a new Revit user.

It is also unfair to compare someone generating a "close enough" model used only to generate some 3D visuals against someone accurately modeling the intended construction to generate contract documents capable of conveying the design intent as well being used for various building analyses, as a starting point for further refinement/receiving additional data during construction and ultimately serving as a database for the building owner or operator during occupancy.

It is frustrating to consider all of the effort that goes into various stages of a project, only to have that effort duplicated when a different tool is used for later stages. We have tried to encourage using Revit earlier in the design cycle, but still get a lot of pushback about it being too difficult to model certain geometries in Revit. Someday....

[Revitaoist]

......What takes me 2 weeks to do in Revit our designers can do in sketchup in a day..

If it takes you two weeks to produce a schematic design, then you are a novice with Revit. I can usually have any project massed in 8 hours. That includes all windows, doors, and materials. I've done projects with several buildings mass of exterior and site done in 8 hours. When I am done, I have a 3D rendering, working floor plans, elevations, and sections laid out for CD's - as a byproduct of doing it in Revit. The pro trick is to have your template set up with a simple building, so when you model your building, the sheets are already set up, you just have to print. If you know what you are doing with Revit, you'll have the sketchup and CAD guys trembling with fear of their outdated skill set.

[rtbrodie108]

The pro trick is to have your template set up with a simple building, so when you model your building, the sheets are already set up, you just have to print.

I am fascinated by your statement in quotes above... so let me make sure i understand what you are actually stating here (since myself being a Newbie, I DEFINITELY want to learn how to save time)...

Are you recommending to create a base template with, say, a ridiculously easy and fake building (i.e. say something like 10' x 10' and maybe even 3 or 4 floors too, just to establish potential multiple LEVELS in the file? and then when a real project happens, one just uses that template and quickly alters each 10'x10' level plan into the real project?

In other words, I am very interested your process, if you don't mind sharing your expert "secrets".

Thanks! :) :) :)

[Steve_Stafford]

In a sense yes. You could assume a two or three story building of a typical size. All the typical views could already be on sheets and schedules created ready for stuff to get modeled. The only thing missing is the actual design.

A good template should take in account all of these things, see attached. It's based on a document that dates back to the early years of Revit, even before Autodesk bought it. You might also take the time to read through THIS BLOG (http://malleristicrevitation.blogspot.com/2011/03/creating-revit-template.html) post about templates.

[Revitaoist]

I am very interested your process, if you don't mind sharing your expert "secrets".

Thanks! :) :) :)

Yes, I have a simple rectangle building with 4 walls, a floor and a roof. 3 levels: 1st floor, 2nd floor and T.O. Roof. Structural grid is set with a continuous dimension string EQ'd and you can just select a dimension and change the layout. If you've set it up right, the walls should move with the grids, and the roof should be attached to the wall. All the plan views are already set up on the sheets, foundation, roof, and structural framing. Elevations and sections are set up on sheets with the appropriate view settings. Save your template somewhere, then pushpin it to your dropdown. When you find yourself repeating a process, open your template and add to it. I have a bunch of typical details (2D), so I can do callouts and get the set looking good quickly, then fine tune it later. As Steve said, have your schedules all dialed in, and you can get a material list started based on what gets modeled. When I am asked just for a rendering of the exterior, I'll also get quantities like exterior wall stud count, siding and sheathing, roofing, cubic yardage of concrete, and whatever else gets modeled. Get away from the old CAD method of just going into old projects and copy/pasting, it takes time to load a file, especially if it upgrades, and you'll end up bringing in a bunch of garbage that will bloat your file.

[drcad2003237304]

If it takes you two weeks to produce a schematic design, then you are a novice with Revit. I can usually have any project massed in 8 hours. That includes all windows, doors, and materials. I've done projects with several buildings mass of exterior and site done in 8 hours. When I am done, I have a 3D rendering, working floor plans, elevations, and sections laid out for CD's - as a byproduct of doing it in Revit. The pro trick is to have your template set up with a simple building, so when you model your building, the sheets are already set up, you just have to print. If you know what you are doing with Revit, you'll have the sketchup and CAD guys trembling with fear of their outdated skill set.

Middle of the road I would say. You do sound pretty talented though. You say mass, but are you actually using masses? I assume since you're talking windows and doors you are talking actual Revit walls. I'm trying to find a use for Vasari at the moment for early design phases, but without the ability to attach windows and other items to masses I might just as well do these early phases in Revit.

[irneb]

I'm in 2 minds about this. Yes Revit gives you much more in the end, but the clients usually don't consider this and only see that your first design takes longer than the Sku/Dwg guys. It's the mindset that needs to change: Pretty pictures from Sku or schematic designs in Dwg needs to nearly be ignored by the client before they start accepting that BIM actually gives them a more efficient design process.

But all that said: I've come to realise that the "changes due to technology" is not all beneficial. I started in the late '80s on a drawing board (as a draughtsman mostly doing CD), moved to CAD in 1989 and BIM in early 2000's. Have done nearly every type of building in all 3 scenarios (from house additions all the way through to office towers / hotels / warehouses / prisons / hospitals / etc.) What I've seen is (in my mind) the deteriorating quality of the designers themselves - not necessarily their design, but the process they go through to get there. I.e. their capability to not just draw a 3d concept, but even just to "think" in 3d space.

In the '80s I got a decent sketch from the designer (even hand-drawn perspective), all aspects were thought out beforehand and I had little to improvise on the drawings themselves. The client was happy since the design was done interactively (even in some cases a situation of the designer sketching the concept right there in the meeting - 90% of the changes happened in that RFP meeting itself). As CAD came about the designers tended to rely more on the technicians (draughts-people) to do those "intent" drawings for them. At first you started doing isometrics, then actual 3d models in CAD just as a "what-if" analysis. This meant the client meeting became non-interactive, the designer would take notes, make a few proposed generalization sketches and ask the tech to draw it up and "prettify" it for proposal to client. Then have a 2nd meeting to propose the alternatives and get feedback from the client, repeating the process 3 to 4 times.

With BIM I see the same trend (only much much more extensively). Only recently it was hammered into me again: A curved office tower with huge basement which has been in design phase for the last year is now in DD, supposed to go to CD at the start of next year. And we're doing it in Revit. The SD was done through Sku and then (due to the extreme amoebic shapes even Sku had difficulty with) it was moved direct into 3ds - actually modelling it there. But after all that we are still fiddling with stuff like "how many cores", "what size floor plates", "does this area go to that floor or this one". And for each of these "what if we do this" questions from the client(single tenant) / developer we end up modelling the thing to show them the consequences of such change. And at each change now ("Because you're doing it using BIM") we have to provide them with stuff like GFA / GLA / Usable / Glazed areas / Landscaped areas / public spaces / etc.... at each of those feedback meetings. We're in DD, but for the last 3 months it was constant changes like these (i.e. not concerning stuff like how do the curtain walls tie together at the various junctions, but rather what is the shape & size of the different portions of the building, is the flow throughout optimized and not too long a walk to the cores, and other concept questions like these) at least once a week (more often than not twice). I've redrawn the entire building at least 5 times already - each time they expect a set of perspectives, elevations, plans, sections and near-tender-like documents (schedules and the like) for the QS to cost the change.

You end up doing shortcuts just so as you can give them something. E.g. on last Thursday afternoon (just after the last meeting) they wanted the entire GLA to be reduced (from 88k sqm back to 85k but still keeping the previous extra retail areas - which wasn't in the original design). This to be completed for the next meeting this coming Tuesday - plans, sections, elevs, perspectives, area schedules, etc. IMO we give them more info from Revit, which only means there's more possibilities for them to change.

So due to the technology you get a situation of unreasonable expectations from the client. But more worryingly, designers who can't advise the client at the meeting about the consequences of some decisions - needing to go back, have the model changed and re-present the change to find that it needs to be reverted and adjusted to accommodate the clashes they should have foreseen as and when the client even voiced the idea.

[rtbrodie108]

I'm in 2 minds about this. Yes Revit gives you much more in the end, but the clients usually don't consider this and only see that your first design takes longer than the Sku/Dwg guys. It's the mindset that needs to change: Pretty pictures from Sku or schematic designs in Dwg needs to nearly be ignored by the client before they start accepting that BIM actually gives them a more efficient design process.

But all that said: I've come to realise that the "changes due to technology" is not all beneficial. I started in the late '80s on a drawing board (as a draughtsman mostly doing CD), moved to CAD in 1989 and BIM in early 2000's. Have done nearly every type of building in all 3 scenarios (from house additions all the way through to office towers / hotels / warehouses / prisons / hospitals / etc.) What I've seen is (in my mind) the deteriorating quality of the designers themselves - not necessarily their design, but the process they go through to get there. I.e. their capability to not just draw a 3d concept, but even just to "think" in 3d space.

In the '80s I got a decent sketch from the designer (even hand-drawn perspective), all aspects were thought out beforehand and I had little to improvise on the drawings themselves. The client was happy since the design was done interactively (even in some cases a situation of the designer sketching the concept right there in the meeting - 90% of the changes happened in that RFP meeting itself). As CAD came about the designers tended to rely more on the technicians (draughts-people) to do those "intent" drawings for them. At first you started doing isometrics, then actual 3d models in CAD just as a "what-if" analysis. This meant the client meeting became non-interactive, the designer would take notes, make a few proposed generalization sketches and ask the tech to draw it up and "prettify" it for proposal to client. Then have a 2nd meeting to propose the alternatives and get feedback from the client, repeating the process 3 to 4 times.

With BIM I see the same trend (only much much more extensively). Only recently it was hammered into me again: A curved office tower with huge basement which has been in design phase for the last year is now in DD, supposed to go to CD at the start of next year. And we're doing it in Revit. The SD was done through Sku and then (due to the extreme amoebic shapes even Sku had difficulty with) it was moved direct into 3ds - actually modelling it there. But after all that we are still fiddling with stuff like "how many cores", "what size floor plates", "does this area go to that floor or this one". And for each of these "what if we do this" questions from the client(single tenant) / developer we end up modelling the thing to show them the consequences of such change. And at each change now ("Because you're doing it using BIM") we have to provide them with stuff like GFA / GLA / Usable / Glazed areas / Landscaped areas / public spaces / etc.... at each of those feedback meetings. We're in DD, but for the last 3 months it was constant changes like these (i.e. not concerning stuff like how do the curtain walls tie together at the various junctions, but rather what is the shape & size of the different portions of the building, is the flow throughout optimized and not too long a walk to the cores, and other concept questions like these) at least once a week (more often than not twice). I've redrawn the entire building at least 5 times already - each time they expect a set of perspectives, elevations, plans, sections and near-tender-like documents (schedules and the like) for the QS to cost the change.

You end up doing shortcuts just so as you can give them something. E.g. on last Thursday afternoon (just after the last meeting) they wanted the entire GLA to be reduced (from 88k sqm back to 85k but still keeping the previous extra retail areas - which wasn't in the original design). This to be completed for the next meeting this coming Tuesday - plans, sections, elevs, perspectives, area schedules, etc. IMO we give them more info from Revit, which only means there's more possibilities for them to change.

So due to the technology you get a situation of unreasonable expectations from the client. But more worryingly, designers who can't advise the client at the meeting about the consequences of some decisions - needing to go back, have the model changed and re-present the change to find that it needs to be reverted and adjusted to accommodate the clashes they should have foreseen as and when the client even voiced the idea.

I read somewhere that 8% of all bankrobbers in prison has had an architectural education. Nuff said, lol.

[rtbrodie108]

Yes, I have a simple rectangle building with 4 walls, a floor and a roof. 3 levels: 1st floor, 2nd floor and T.O. Roof. Structural grid is set with a continuous dimension string EQ'd and you can just select a dimension and change the layout. If you've set it up right, the walls should move with the grids, and the roof should be attached to the wall. All the plan views are already set up on the sheets, foundation, roof, and structural framing. Elevations and sections are set up on sheets with the appropriate view settings. Save your template somewhere, then pushpin it to your dropdown. When you find yourself repeating a process, open your template and add to it. I have a bunch of typical details (2D), so I can do callouts and get the set looking good quickly, then fine tune it later. As Steve said, have your schedules all dialed in, and you can get a material list started based on what gets modeled. When I am asked just for a rendering of the exterior, I'll also get quantities like exterior wall stud count, siding and sheathing, roofing, cubic yardage of concrete, and whatever else gets modeled. Get away from the old CAD method of just going into old projects and copy/pasting, it takes time to load a file, especially if it upgrades, and you'll end up bringing in a bunch of garbage that will bloat your file.

Thanks to you and Steve Stafford, this is VERY HELPFUL INFORMATION!

[rtbrodie108]

In a sense yes. You could assume a two or three story building of a typical size. All the typical views could already be on sheets and schedules created ready for stuff to get modeled. The only thing missing is the actual design.

A good template should take in account all of these things, see attached. It's based on a document that dates back to the early years of Revit, even before Autodesk bought it. You might also take the time to read through THIS BLOG (http://malleristicrevitation.blogspot.com/2011/03/creating-revit-template.html) post about templates.

THANK YOU!!!!!!!! :)