Interior Elevations vs. 3D Perspectives

JQL

I don't exactly agree that we think in 2d projection. We do that on plan views but we don't do that on elevations.

In my opinion people who read plans read them as schematics and it's fairly easy for them to map the plan views in their heads as they relate to them abstractly. They are completely different from reality.

Elevations, in other hand, relate to how they actually perceive space and sometimes the do lack information.

Horizontal distances are key for construction but they are usually represented in plan view. Heights and vertical relationships are easily perceived in perspective so an elevation could eventually be replaced by them. What perspectives lack is an accurate way of comparing dimensions as objects far away are smaller than closer objects... But as that is covered in plans, maybe we could use skecthup for perspective views which are more informative, and forget the elevations most of the time.

This could work on smaller scales like 1:500 to 1:50, while on scales that you need to use to build accuratelly, like 1:50 to 1:10, traditional ortho views with more details that could be actually measured on paper, should be required.

srx

@jql said:

I don't exactly agree that we think in 2d projection. We do that on plan views but we don't do that on elevations.

I did not say that architects think only in 2D projections. IMO they use, beside floor plans, sections as one of the most important analytical tools to perceive (and present) complex 3d structures and relationships, among the others. To solve complex problems, one has to analyze it first, break it into smaller peaces easier to swallow, and with more expressive capacity to show the IDEA. Sections are just that.

pbacot

I use both, but I don't take the time to model as much as I do draw 2d. So the model is generally less accurate and only to clarify a 2d drawing. I agree with SRX. Interior elevations tend to be important but how they are used in my projects varies greatly. Interior elevations can be as important for layout as the floor plan, so while one could do the same in SU as in CAD in this respect, I never have for interiors. It's often pretty exacting in determination what can fit in an area. And it can be more schematic in terms of visual realness. The lines are more important than the surfaces if you understand what I mean, and often I never get to the final styles, just the outlines. At any point I can pass it off to a cabinet maker or ID, and detailed realism is pointless. I find interior models more helpful in explaining than designing, mostly for the client but sometimes for the builder (then they do something different anyway) :roll

kyeric

Most of this is from my own practices and experiences, but I have found that many of the architects I have been either taught by or apprenticed under had an infatuation with the re-presentaion of architecture. The product was not the actual construction, but the delivery of the idea...the drawing. Line weight, scale, complexity distilled into a perfect diagram took an enormous amount of time to create. But, when this was complete, you were left with a sense that you understood the project more fully. I am absolutely infatuated with the representation of an idea as built form and have spent the better half of my career seeking out how to make it better, more legible and, well, beautiful. I love a great set of CDs!

The problem with this line of thinking is leaving out the other players in the project, namely the client and the contractor. In both of these cases (again from my own experience) they have utilized less of the time-consuming "beautiful" drawings and instructions that I spent a certain amount of time to create and opted for more of a schematic approach of 3D representation. Why? Well, SketchUp has become somewhat of a Rosetta Stone for design professionals, clients and stakeholders and the guys actually swinging the hammers. You can get an immediate response from the friendly face of a SketchUp model. I have never experienced this form of openness in any other software, and believe me, I have tried them all.

I find that the diagrammatic portion of architecture (plans, sections and details) are still absolutely essential not only for quick design iteration, but documenting and describing space to a client and construction to a contractor. The more pictorial parts of a project are getting less formal for me (exterior elevations, interior elevations, assemblies) and I find that more can be garnered from the textured 3D perspective with dimensions and annotations than anything that I have completed previously with 2D CAD.

Again, I am not trying to be Yoda and burn architecture and it's sacred methodologies of representation to the ground, but thoughtfully considering what is really used and not confused during a design presentation, on a review board and on the jobsite.

I love a good dialogue. Thanks for the critical responses, and YES, I loved The Last Jedi.

Eric

pbacot

Good point. I decided to look at Nick Sonder's take, and in his book he shows interior elevations made from SU model. Not to say he hasn't provided the 3d to builders and clients. When I look at them, with the important construction information, references to location, dimensions, I don't see providing as much information as clearly in a construction document as represented in the 2d version. I think you need both.

I like to do Exteriors from the SU model, but in 2d projections I feel the relationships can be more clearly seen for the builder and designer, and will absolutely be required for the planning official. In the end the layout in the field is orthographic, unless you are some sort of scupltor.

JQL

There's a recurrent story here about a studio that delivered the renders of a preliminary version and the full CD of the final version. They built the rendered output and disregarded the CD's.

kyeric

@jql said:

There's a recurrent story here about a studio that delivered the renders of a preliminary version and the full CD of the final version. They built the rendered output and disregarded the CD's.

And therein lies the problem. The CDs are contract, but the renders are infused design descriptions. Unless both are referenced or linked to one another dynamically then architects open themselves up to errors and omissions. This is where BIM makes sense. The difficult part, especially for me, is navigating the strict nature of BIM implementation versus the flexibility and speed of SketchUp.

And yes, I have followed so many approaches to utilizing SketchUp and LayOut as another method of crossing this divide, but I continue to find that most of the time, the client and contractors constantly refer back to a 3D view.

The hope is to have LayOut become more of integral part of SketchUp and not so much what I think it is now, which in my opinion is a bit too slow and precarious as compared to its older sibling.

I would really like to see that thread, JQL. Do you mind linking it in this conversation?

JQL

Sorry, it's not a thread, when I said "here" I was talking about a story I heard here in the country I work.

It was at the dawn of rendering too, not with the photorealist rendering we have now.

I think you should look at Layout again and take give a chance to Sketchup+Layout+Thea render as I do.

With these 3 I have a workflow where I achieve that sync between project iteration, CD production and 3D communication, still or interative, including photorealistic representations. (Of course I can only do this because I don't leave my hand sketchbook.)

The next step would be realtime rendering but that isn't convenient to create or the quality is not on par with the light studies I can achieve with Thea.

I've been moving away from photomontages or collages and physical models, because, as you said, I tend to feel that I'm working for the representation, not actually studying the building itself.

What I like is to keep iteracting with the building not it's representation. The more I pursue the building's representation the less I'm exploring architecture itself.

kyeric

I would love to see or even start another conversation with other architects on how they use SketchUp and LayOut in their personal workflows. The closest that I have seen the promised land was from Nick Sonder, Matt Donely and Mike Brightman. The approaches are somewhat similar, although Mike’s team developed plugins specifically built for documentation in LayOut.

I found both books, plugins and investigations fulfilling and worthwhile, but I have gravitated back to more of a hybrid Nick Sonder approach, as I still find it easier to iterate and share plan work in CAD but utilize LayOut more and more as it is becoming a better end-to-end solution with SketchUp.

Maybe we can get a small band of Sketchucationers together to share a Google Drive folder of projects, templates and workflow examples? I would be willing to share with a select few the things that I know work and the things I struggle with that could use some enlightenment. Or, just fly me to Europe and I will be an architectural intern for you, JQL. You wouldn’t mind a 40+ year old student, right?

JQL

@kyeric said:

Or, just fly me to Europe and I will be an architectural intern for you, JQL. You wouldn’t mind a 40+ year old student, right?

I wish I could, but it wouldn't justify the trouble. I have little to teach and most probably couldn't afford your fees... I have a very small office with very small projects.

kyeric

I was only joking about the Intern part, JQL! Although I am not a regular contributor on this forum nearly as much as I used to be, I still read and learn from it almost daily. Your posts and insight, in particular, are more valuable than you give yourself credit for. I will think about the best approach on the possibility of resource and workflow sharing as my tiny office is taking on more work and I need to think about how I will want to get even better at what I do.