SourceForge has been redesigned. Learn more.

newbie, asking for basic brl-cad information

  • Marco Rigotti

    Marco Rigotti - 2008-03-20

    Hi all,

    I was looking for a software that can substitute Autocad and I came to know of brl-cad. I skim read through the documentation, but I couldn't figure out a few things, so I decided to post a message to have users feedback.

    As a start, in which way may I get information on the geometry of a model? I read about shotlining, but in my job I'd need to get 2d cross sections of a 3d model, or to have the ability to easily know the area or the volume of an object.

    Secondly, I often have 3d points data, like terrain elevation data, describing the geometry of an object (dams, houses, anything ...). Is it possible to import such kind of data in brl-cad and build a 3d model from the point data?

    I'm aware that brl-cad can do many things and it seems to me a pretty advanced software, probably it does much more than I need, but I simply need a tool by means of which I can abandon autocad, as I'd like to use open source tools only.

    I'd appreciate if any of you had comments or suggestions to make on the feasibility of this going from autocad to brl-cad.

    Thanks in advance,


    • Sean Morrison

      Sean Morrison - 2008-03-21

      BRL-CAD is a suite of a lot of different tools that all work together. 'nirt' is good for determining distances and thicknesses along a given shotline.  'rtarea' computes projected and exposed areas.  'rtweight' computes volume/mass/centroids.  'rtedge' will generate a 2D hidden line rendering.   There are a variety of other commands as well (more than 400 overall), but that's the core that relate to analyzing geometry.

      For terrain elevation data, we have a terrain primitive (the 'dsp') that will take a regular grid of elevation data (like contained in dem data).  For arbitrary 3D point data from a scanning system, that usually needs to either be meshed at the device or converted first (e.g. cy-g for cyberscan data) or be imported as an 'ars' primitive (which takes points along the waterlines).

      As for replacing AutoCAD, that's a much deeper issue as it's a CADD package, which is a rather different application domain focused on 2D and drafting approaches (for which BRL-CAD is ill-suited at present).  Our back-end engine can pretty much support most of the constructs employed but it's not exposed through a user interface; it would require new development effort.  Our more immediate focus is in the 3D solid modeling domain with strong engineering analysis and performance focus.

      We are probably by far the "closest" open source project to being a viable AutoCAD replacement some day, but there is much work to do still -- particularly in the user-interface and general usability.  For that we need development effort and getting more of the CAD community involved in development so we can accelerate development in needed focus areas (like AutoCAD's domain).

      Here's a diagram of where BRL-CAD presently is situated in the CAD industries:
      Expanding to the left and down are long term priorities after we get the underpinnings of a new interface worked out (which is active/current development).



Log in to post a comment.