RE: Revit

From: Jonathan Chertok (
Date: Tue Apr 22 2003 - 23:17:10 EEST

Charles, Makai and List,

I have just finished a summary of a project that I have been working on and
would like to prod this discussion with the following:

My experience with the architectural field is that you will find a spectrum
of different attitudes towards the profession at a variety of firms.
However, I believe it is possible to break firms roughly into "design" and
"production". While this is a very general description, it is instructive in
my opinion and in many cases - especially with large projects - one will
find that a "design firm" has almost completely turned the project over to a
local firm that takes control of Contract Documents ("production") as well
as overseeing construction (Construction Administration).

"Building Management" (a natural follow-on to Construction Administration)
would be of interest to these "production" firms, who are often rather
corporate entities that are very focused on a kind of assembly line process
with the attendant concerns (a generality with lots of give and take at
these firms).

"Design" firms would be more interested in the process of creating the
project, how this process affects the creation and perception of a project
and who - in many cases - will have an element of research built into each
and every project.


Architectural Models and Architectural Software are a portion of this
"design" or "production" process. On the one hand architectural models could
be used in order to consider various abstract aspects (substitute any
"architectural concern here) of the project, while on the other hand one
could make a presentation model that will sit in a sales office to
facilitate condo or office space sales.

In my opinion it is a mistake to expect RP to serve to recreate in full a
presentation model. When I worked at Renzo Piano's architectural office in
Paris, we used a process called "coupe chemique" wherein one made an opaque
rendering in Autocad that was sent to essentially a metal screenprinting
shop that would etch the void in this rendering out of a thin piece of
metal. This final product was then bent etc and attached to a presentation
model, or competition model made out of a variety of other material. It was
a process used to represent a very specific aspect of the building and
really contributed to the impact of the model. I see RP serving a similar
process - taking into consideration the materiality - or immateriality - of
the RP process one is utilizing.

The other immediate aspect of RP in architecture would be with respect to
individual parts that make up a building. In some cases architects (or
sub-contractors) will design custom connection pieces for a project. In most
cases these are sketched and manufactured in a shop using traditional
techniques. In some cases they are drawn in 3D by a designer and then made
in a traditional shop. Eventually these pieces would be cast out of metal
via relatively traditional manufacturing methods (lost wax etc). I think
there is a lot of room for RP to enter the profession in this area with
respect to the design, distribution and creation of these custom pieces.

I think the production of an RP model should be considered as a separate
architectural activity to producing the information necessary to actually
build a project.


Finally, from a conceptual standpoint I could state that architecture has
traditionally gained it's "meaning" in reference to ideas from other
disciplines. Of late there may be a move to allow architecture to gain
meaning from things specifically architectural - like geometry. In this vein
of thinking one could return to Gaspar Monge's invention of Descriptive
Geometry and say that this was a tool that was leveraged by the engineers
who eventually used it to create dams, boats, airplanes and the like.

Then in the 1910's and 1920's architects - who were blown away by these new
shapes - began to imitate these shapes without having any real grounding in
the geometry used to create these shapes.

Currently architecture is undergoing a transformation similar to that
allowed by descriptive geometry. This is assisted by the computer and could
be assisted by various RP processes. Any parametric, topological, analytical
and other feedback from Architectural Software would simply help in
accomodating a more general change in architectural concerns. Most
interesting to me is the ability to control a project's 3D geometry through
the computer and to then pass this information along to vendors, who may be
able to use this Digital File to manufacture portions of the building.

There is an approach to arhcitecture that thinks that building materials are
just so much swiss cheese and it is just a question of poking holes in the
cheese in the right places. Architecture has to be much more deliberately
and finely layered than this, and as a consequence RP's contact with
Architecture should be similarly nuanced.

I suppose one could take another approach and extrude architecture like
pasta, or structural steel, but I think such an approach must be considered
as specifically not architectural.

Hope that was of interest to a couple of folks,

Jonathan Chertok

Universal Joint
Full Service Design and Construction

Jonathan Chertok. Principal
Austin Texas [512] 407 9628

Buildings + Digital Manufacturing

>From: "Charles Overy" <>
>Reply-To: <>
>To: "Makai Smith" <>, <>
>CC: <>
>Subject: RE: Revit
>Date: Tue, 22 Apr 2003 11:06:16 -0600
>I think that the generation of watertight output, ie "valid stl" should not
>be a primary criteria for Architectural CAD selection. The adoption of BIM
>and parametrics as well as the integration of tool sets that more
>model the actual architectural design flow are far more important.
>My view is that CAD and related computer technologies are predominately
>used in the architectural fields as a method of documentation not as tools
>to facilitate creativity and collaboration. (Part of my concern over
>Laiserin's drive to BIM is that it appears to me to prioritize
>documentation as a method of increasing the value of CAD in AEC).
>What I believe was significant in the adoption of computer technologies in
>other industries is that the automation of some of the design tasks
>definitively improved the aesthetic and functional quality of the designs.
>The frequent feedback I get from my customers it that this is far from a
>universally accepted truth in architecture.
>Clearly I am biased by my own efforts but I believe rapid modeling will be
>useful to the AEC profession. However, it will not be successful if it
>requires significant user intervention for the generation of "watertight"
>stl files as we, in this forum, currently understand that requirement.
>(Happy to expound upon why that is not possible if you would like!)
>OK, to cut a dissertation short, watch Revit for what it is, AutoDesk's
>generation parametric AEC modeler. As I understand BIM, Revit is it. I do
>not see any other real competition in this class although as I mention I am
>not very familiar with Bentley systems. Whether Revit will morph into
>AutoCAD AEC or the other way around is probably a function of AutoDesk's
>internal workings. Again, BIM-centric programs, and most of the big AEC
>modelers, to me, add very little, if anything to the creative process for
>the majority of current architects. For this lightweight, but inherently
>programs like Sketch up seem to have an edge.
>As for RP, I see two scenarios. One, the RP industry has to be
>significantly compelling for AEC CAD vendors to incorporate viable RP
>into their software. To truly create good RP output from a wide variety of
>AEC models this would require parametric substitution based on output model
>scale when the export file is generated. This is far more complex than
>generating valid STL export. The second road is the one we have taken, the
>initial R&D of "middle ware" that specifically addresses the AEC issue.
>Perhaps it is the road less traveled. I certainly hope it is not a dead
>-----Original Message-----
>From: []On
>Behalf Of Makai Smith
>Sent: Tuesday, April 22, 2003 5:48 AM
>To:; Bob Olsen;
>Subject: RE: Revit
>Very interesting. To bad Revit dropped the ball. I am looking forward to
>seeing your resurfacing tool. There seems to be a lesson for architects
>the allied arts in what you are doing: that we cannot wait idly and use
>tools vendors provide us. Out of necessity we must have an active role in
>creating them.
> >From your experience with architecturally oriented software, can you cite
>any examples of those with strengths in both parametric creation and
>watertight output (, for < $25k/seat?) IOW, that which might support
>both BIM and RP? We are currently an AutoCAD shop. While encouraging a
>move to next gen software its easy to suggest staying Autodesk, but I
>if there are better alternatives. Anyone?
>|\/| /\ |< /\ |
>NOTICE: This email message and all attachments transmitted with it may
>contain confidential information intended solely for the use of the
>addressee. If you are not the intended recipient, you are hereby notified
>that any reading, dissemination, distribution, copying, or other use of
>message or its attachments is strictly prohibited. If you have received
>this message in error, please notify the sender immediately by telephone or
>by return email, and delete this message and all copies and backups
>Our systems are checked for viruses but please note that we do not accept
>liability for any viruses which may be transmitted in or with this message.
>-----Original Message-----
>From: Charles Overy []
>Sent: Monday, April 21, 2003 9:12 PM
>To: Bob Olsen;
>Subject: RE: Revit
>In a word NO.
>More Detail:
>Revit, only has three "built in" export formats
>DWG - AutoCad
>DXF - generic CAD export
>DGN - Bently Systems
>As of last fall it was not exporting ACIS or other solids data to Autocad,
>only a tessellated mesh.
>Great you might say, a tessellated mesh, I can go to STL from a tessellated
>mesh via 3ds or the like.
>However, meshes that I exported from Revit were HIGHLY degenerate. They
>had holes, flipped faces, etc.
>I would be happy to send you some if you want to have a look. I have not
>worked with the DGN as I do not have Bentley. The DXF closely mirrors the
>DWG. They could be used as a snap grid to redraw the geometry but were
>useful in anything more that very basic cases as a starting point for mesh
>In addition, if you are considering building Architectural models from
>files you will face the problem that I call radical scale down. Basically,
>even if you own a 810, Titan or the like you are going to be running an
>model at say 1" = 8' or smaller. 1/96th the original size to maybe 1/400
>the real world size. Now Revit is great program because it is truly
>parametric. You can call up a window and place it into a wall. Revit also
>knows the thickness of the glass in the window. Even expensive .25" glass
>1/96 then is .0026" thick when you make the prototype. Now imagine a small
>office building with a "curtain wall" of glass representing 60 percent of
>the model and the model at 1" = 20' . You just built a Vat O' Dust.
>Because it is a parametric modeler, you could create window substitutions,
>but depending on the window library you cannot always just say you want 1'
>thick glass. Sometimes you have to modify each window.... Now what about a
>steel I beam with a .25" section... or a cable holding up a fabric
>Have I been there... yes.
>I have had meetings with Revit's old director of sales and their current
>CEO. At one point they were going to write STL export into the code for us
>(we were going to guarantee sales or pay) but that was before they got
>purchased by Autodesk. We ran several sample models for them which they
>liked. As of last fall there was no movement that I could ascertain to
>embrace RP.
>In the end we resorted to development of a more general solution to at
>least some of the problems that I describe above. The software solution
>solves some of the radical scale down issues and can work with existing RP
>mesh preprocessing software like Magics, Marcam, Stl Editor, to create a
>buildable mesh. The plan is to have operational stl file resurfacing
>available late next week.
>Please let me know if you would like further information on Revit, RP in
>architecture, or our solutions.
>Best wishes
>Charles Overy
>Director of Engineering, LGM
>-----Original Message-----
>From: []On Behalf
>Bob Olsen
>Sent: Monday, April 21, 2003 4:32 PM
>To: ''
>Subject: Revit
>Anyone know if AutoDesk's Revit software (an architectural package) can
>create an STL file, or how that is done? Thanks in advance!
>Regards, Bob
>Protogenic - Always on time,
> Always great Customer service,
> Always top quality.
>Bob Olsen sales manager
>1490 West 121st Avenue, Suite 101
>Westminster, CO 80234-3497
>* Direct: 303-453-3990
>* Main: 303-252-0212
>* Fax: 303-252-0223
>* web:

The new MSN 8: smart spam protection and 2 months FREE*

This archive was generated by hypermail 2.1.7 : Sat Jan 17 2004 - 15:17:23 EET