Area boundaries are no longer functional

[jamesr]

The area boundary function continually cuts off small areas that we need to include in our area calculations.

Per our vendor:
"In 9.0 room and area boundary detection was switched to more robust algorithm with the goal of no more missed detections of bounded rooms or areas. As side affect code extended use of gap patching to narrow channels which would not fit reasonable human (up to 6-12")."

This destroys our abilty to modify and/or generate BOMA calculations for our clients. Buildings of complex geometry cannot be evaluated correctly for area and we are forced to implement, by policy, sloppy drafting. Ironically, this appears to be a set of programming intended to make up for someones sloppy drafting. Worse is that this was not an issue in 8.1.

This is an issue of such substantial importance to every architect using Revit that waiting for the next release is not acceptable and a patch should be deployed.

[patricks]

I never trust Revit's automatic boundary placement anyway. I always just use the pick tool, uncheck Use Rules and place them according to BOMA standards.

[jond]

The abstract "average human" rule prevents one from being able to do this. Once the lines get close together revit will close/cut off the chain and give you the resulting area that it made - even if you made the lines first.

[David Conant]

The gap closing behavior has been present in Revit for many releases. What has changed is that more elements can bound a room. Prior to r9, columns were not room bounding. Thus, Revit did not see a narrow gap. To restore pre 9 behavior, select the offending columns, and uncheck their Room Bounding property. Revit will ignore them when calculating rooms and areas.

On a side note, I recall from my practice days that BOMA allowed a tolerance of +- 2% in area calculations. The 1sf difference shown in the example would exceed the tolerance only in a room of 50sf or less. Allowing a tolerance recognizes both the difficulty of making truly precise area calcs and variations in construction that can lead to differences between area as drawn and area as built. A .5" error in placing a 24' wall loses 1 sf of floor area. Many things like small bumps, niches, and pilasters were ignored simply because it was far too laborious to measure them. While CAD can deliver what seems to be a higher level of precision, it may be no more accurate in the end.

[jamesr]

So following that line of thinking, we have a highly precise tool that no longer allows architects to have control over accuracy. Let me explain:
BOMA provides 2% leeway from what was drawn versus what is built. This would assume that architects are using a precision tool and using professional judgement to verify the accuracy of their design from which field conditions and revisions inherent to construction produce changes. BOMA recognizes this and provides tolerance to limit liability. Revit 9 area boundaries now takes way the ability to graphically verify the accuracy of the design at its start point thereby removing the starting point of argument in a litigation setting. This is a serious problem in my view that has the ability to cost me substantially more money in the end than having my team deal with the little bumps, niches etc as they do the area calculations. Having been in litigation due a discrepancy between the calculated areas and the built areas, I know that the only leg I have to stand on is an accurate area boundary from the start.

I'll remind everyone that not all area calculation are base upon BOMA. The project I'm currently working on is a developer driven condo where they need every square foot to be counted in order to make the proforma viable. Assuming my area calculations are off by 2%, of which I must assume they are low, I am professionally obligated to tell my client that of the 600,000 sf of sell-able area, they are loosing 12,000 sf due to the program I'm using for design. With a sells price of $650/sf, that equals $7,800,000. Their proforma no longer works and I've lost a project. This is applicable in BOMA calculated buildings as well. This is obviously worst case, but it illustrates the need for a precision tool to work like a precision tool and stop trying to be smarter than the person using it.

We've tried your suggestion in removing the room bounding property, but our clients building area rules are complex enough that we elected to use area boundary lines. When a space is small enough - about 12" or less, such as a column next to curtain wall, the area between the column and the curtain wall is removed from the calculated area. This is area that is sell-able, and as such is very important to our client.

My company does not feel short cutting area calculations is an acceptable practice, and having it forced upon us is both frustrating and disheartening. If we wanted to do area calculations in the manner being suggested, that should be our option. It should not be a decision made for us.

[funkman]

Wow, I don't know if am more astounded by the extent of the area innacuracies and (albeit consequential) litigatation issues or by Autodesk's response (it's close enough).

I would hope that this is rectified by the factory extremely quickly, as James has so eloquently clarified the issue.

[greg.mcdowell]

I think there are a couple issues here... one being on what we should and should not be responsible for and how BIM helps and/or hinders that... the other being what level of accuracy we should be expecting.

As to the first issue... explicity quantifing anything in our profession is potentially problematic as we have very little to no control over the actual field dimensions. Ours specs allow for a certain degree of "fluff" in construction (called tolerance) and I don't think it's appropriate to hold us to something we can't control. What we can be responsible for is "design intent" and that, I think, begins to lead to the second issue...

The level of accuracy we need and/or expect from our software is, to some extent, customizable. We can choose to set the units to greater or lesser degrees of accuracy and this will DRASTICALLY affect the outcome. Even setting it to the greatest possible accuracy there can still be errors (you can move an object in AutoCAD, for example, 1/512" but you can't measure it) though they're small enough to be neglible... but that's the real trick, at what level does the rounding error affect the "design intent"?

You could argue that we need be only as accurate as the accuracy in the building program... and if that's the case is it a problem if the calculated and building areas differ by a couple units?

The BOMA issue of allowing aras to be off by 2% is an extreme. I would guess that the accuracy we're talking about in Revit is much, much smaller (to the point that quantifing it as a percentage would be impractical)... although the larger the area being measured the more potential there is for rounding to make a difference... and in cases like this perhaps simply increasing the accuracy could be enough to bring it back to within acceptable limits.

[greg.mcdowell]

Yeah... I know what you're saying... and we definitely need to be able to have some trust in our software (I mean if a calculator added 2 and 2 and got almost 4 that would NOT be good).

Part of what I'm trying to get a handle on (and where my post was coming from) is how the ability to quantify everything is going to affect Architecture. It's sometimes hard to get across to the interns I've worked with but we are not in the business of producing shop drawings...

According to the AIA contracts we're also not responsible for providing quantities and the like to the contractor... that's their job. And if we start to assume the responsibilities of our contractors then we also start to take on some the liability... I've run into this myself (as has a poster above it sounds like) even before BIM (in actuality we've been using BIM, as a concept at least, in AutoCAD for years... the drawings have information that we're always querying) and I am constantly reminding myself of this fact.

So what do we do about it? We can point to the likes of Frank Gehry all day long and his relationship with fabricators and how they pull shop drawings directly from the computer model. But what we don't know is the sort of contractual relationship he might have with these fabricators. I'd be willing to bet that they've got something written up that would relieve him of at least some of the liability. So maybe part of the answer is for Architects to begin forming relationships with fabricators... that and have a really good lawyer! <grin>

I'm just rambling a bit here... I do agree, as I said before, that we need to be able to rely on our software and if there is a "problem" with how Revit calculates areas we need to let the developers know (though this isn't the correct place for that) so they can isolate and correct the issue.

[iru69]

So following that line of thinking, we have a highly precise tool that no longer allows architects to have control over accuracy.

It seems like the issue kind of leapt off the logic rail from how much tolerance Revit allows for a "gap" when room bounding to Revit not being accurate.

As Patrick pointed out, if every inch counts that much that millions of dollars are at stake, then can't you manually place the boundary lines yourself?

[Wes Macaulay]

I think I'm okay with Revit approximating room areas, but when it comes to area plans, the gap feature has to go... and preferably by the next build!

Imagine a 10" wide area -- say a mechanical shaft. I can't include that in my areas any more!

This situation constitutes itself as an emergency. In Vancouver, buildings are designed to maximize the built area, and how are we supposed to determine what these little areas contribute to the project? Over 60 floors, this could put us over our maximum floor area for the site. David, Lilli, Matt... I hope you can get someone to fix this real soon!

Edit: I have submitted a support request to have this changed. I sent along the file that I have also attached to this message.