Sloppy Modeling that isn’t Making Me Sad
One of the reasons why I love working in ARCHICAD is that there is no one true way to use the software. There are specific tools for various building elements, but you don’t HAVE to use them. You could build a model without using the Wall Tool. You could make all your doors with the Window Tool. You could never use the Slab Tool, instead only using the Roof Tool—or the Shell Tool for every flat or sloped surface. None of those concepts are necessarily great ideas, or for the faint of heart. But the merits of those solutions don’t matter to me, not today. What does is that the program is flexible enough for us to work the way we want to.
There are of course better ways to work. Smarter ways to work. Faster ways to work. None of this is new. And sometimes I will unequivocally say “if you aren’t doing X in ARCHICAD, then you are doing it wrong”. But that is typically about bad methodologies and decisions. Breaking the rules of appropriate ARCHICAD usage isn’t wrong if you are doing it wisely. That was the point of my recent MEP article. There’s a better example of this though. My favorite ARCHICAD blogger recently wrote an awesome post, Sloppy Modeling Is Making Me Sad, that laid out a rigid right and wrong way to work in ARCHICAD. In the article, which you should definitely read, James talks about imprecise data entry and modeling—and all the ways it will ruin your model. And he’s absolutely right. If you are lazy about placing elements, you will get horrible tiny fractional distances that might vanish when dimensioned. If you aren’t careful you can make a wall that is 10′-0 1/64″ long and will dimension as 10′-0″. Put enough errors like that together and you’ll get wonderful math like this: 10′-0″ + 10′-0″ + 10′-0″ + 10′-0″ = 40′-0 1/8″ That is wrong. James lays out the proper ways to place elements and avoid all those rounding errors (errors that can exist in both Imperial and Metric units). You should read and heed his advice.
James is 100% right that sloppy modeling is not easier. His advice on how to avoid this trap is spot on. So now let me show you where sloppy modeling is still sometimes the right answer.
I was in a client meeting recently. We were talking about the layout for a new bathroom and closet. I didn’t bother printing things out because I had my laptop and prefer to show things digitally. I especially like showing early schematic designs directly in ARCHICAD. It takes less prep work and also makes developing ideas in the meeting easier. I am not an architect who brings tracing paper and markers to a meeting. I sketch within ARCHICAD.
I showed the client four variations, none of which were quite right. I knew that going into the meeting. Slamming a piece of paper down on the table and screaming “genius!” was not my intention. Instead I wanted to present some ideas, then together develop a solution we both liked. Below are the four original schemes. Yes they were all in the same file (if the design brief were more complicated, I would probably have used multiple files—if that helped things move faster). The variations used Renovation Status, but I mostly just deleted elements to be demolished (I found that tracking demo elements at this stage of the project got in the way). Not everything was noted and dimensioned because I only wanted the annotation that would further the conversation (though what was noted was done like this). I forged ahead, silencing my perfectionist demon who wanted me to make each scheme complete and ready for construction documents. These four models weren’t going to last forever, so to listen to that perfectionism would have been a mistake.
Once I presented the original four ideas, we began iterating. And here’s where things got loose and fast. These schemes led to discussions, and a few more schemes. I’m not sure that we looked at much in 3D. At this stage, it wasn’t necessary. Everything I was doing was powered by ARCHICAD’s 3D capabilities; the tools allowed me to move fast and smart. We could jump to 3D at any point, but we chose to keep the discussion focused on the plans. In fact it’s probably better to think of the four plans above, and the grid of variations below, as detailed diagrams.
You’ll notice that in the array of floor plans above one is crossed out (we quickly agreed that it was a terrible idea, albeit one worth seeing) and two are circled. The two circled plans are the ones that survived to round two. One of the plans was done in the comfort of my home office, so it’s nice and precise (and Onland approved). Let’s ignore that one. The other one was done quick and messy in the meeting. Here it is:
Let’s play find all the bad modeling. Walls don’t align; walls don’t connect; flooring doesn’t extend all the way to the walls; the shower Label text could be placed in a better location; I’m sure things are floating in 3D; the doors aren’t in precise locations; one of the sliding doors isn’t centered in the wall; there are definitely fractional inches that would make James sad…I could go on and on before even listing the things that are missing. But none of that matters. When we were building this plan in the meeting, I moved elements around until they felt right. Worrying about the details would have been a distraction. As a result, the above plan is a loose sketch. It was intended to be orphaned. Once we reached this point, I went home, fiddled some more with the design, then rebuilt it from the original, pure as-builts. At that point I returned to working as James describes. I was exact and perfect and methodical. I didn’t try to fix the plan above. I used it as reference and began again from something that I knew was done right.
Hand sketches get thrown out or redrawn. When we look at a piece of paper, scribbles are scribbles and hardline drawings are fact. No one needs to write blog posts about that. We are cognizant of when to be loose or rigid when working with traditional tools. We need to have the same understanding when using ARCHICAD. With BIM, we focus on how the preliminary wall becomes the final wall. And I love when that happens. We can draw a wall on day one and have that wall and that file continue uninterrupted through construction, both evolving. However, that is not an immutable truth. We don’t have to live with everything we make. Sometimes our workflows (and thought processes) aren’t as linear. In that case, our journey from day one to permit will be more convoluted. Having both options is a strength of BIM and digital design. We get the benefits of linear and non-linear workflows. We can go from Point A to B to C or from Point A to Points B through F simultaneously and then grab parts from all those solutions to arrive at Point G. If we model well, we can take that data directly. If we model loose, we rebuild it when we need it. Done wisely, both workflows—orderly or chaotic—can lead to great results. We just need to remember when order and chaos need to be paused. In my example above, I was exact through existing conditions and my first four schemes. Then I paused and was loose, letting the design evolve unfettered. Once a direction had been settled on, I returned to the perfect and unpaused.
- Sometimes we use ARCHICAD for design, documentation, and analysis simultaneously. Sometimes we don’t.
- Sometimes speed is important to maintain your creative flow.
- Sometimes you have to ignore how messed up things will be.
- Sometimes keeping things in order does not help you reach the solution.
- Sometimes you have to treat Save As as a design tool.
- Sometimes you need to redo work because redoing is faster than being careful.
- Sometimes it’s not sloppy modeling, it’s smart modeling. And sometimes smart modeling means ignoring certain constraints that you’d never ignore if what you were doing wasn’t ephemeral.
Remember James Murray is Correct
This post was tough to write because I’m trying to show that two contradictory views are each key to true success with ARCHICAD and BIM. Just because computers can be 100% precise and accurate doesn’t mean we need to be precise 100% of the time. Designing in the computer with the elegance of analog tools when one tracks every sixteenth of an inch at all times is just as unproductive as letting sixty-fourths of an inch appear in your model when your work will remain in the project. We need to understand the difference between sketching and drafting in a BIM environment. We must understand when to stop being loose and start being rigid. And the reverse. When we sketch, we can be loose. When we draft, we need to be perfect.
The balance of precision and messiness is something we all need to struggle with. And hopefully we are, rather than just submitting to one extreme or the other.
Three Bonus Tips
- ARCHICAD does have the Nudge Elements command (which works in 2D and 3D). That might be a great way to balance the fluidity I’m describing while maintaining dimensions that don’t annoy James. I have never really given the Nudge Elements command a fair chance—probably because my keyboard shortcuts were all messed up until now. I think it’s time I try it. The trick will be finding the two spacing options that gave me the proper grain for making big and little moves during design. But of course, when working by hand we aren’t constrained by something artificial like that. So maybe it’s a dead end for my purposes (though still a useful tool for ARCHICAD users in general).
- Don’t forget about the power of Undo to move back to a clean model. If you have your Undo steps set to 99, you can make a whole lot of changes in an effort to find the right solution before undoing everything and making the correct moves properly.
- If you are working sloppy and fast in a meeting (or at your desk), make sure you are using a copy of your main file. This way you don’t need to worry that you are messing something up (because you will have the clean file sitting safely elsewhere).
Are you following Graphisoft North America on Twitter? Click Here to keep track of all the latest ArchiCAD News in North America (and beyond).