Hi Sean, Concerning the tree structure, I meant the object hierarchy. Archer already has a good outline of it, I only miss the possibility to edit the objects parameters directly in their tables. But I understand that it is still in the alpha version.
As for OpenSCAD, it would be great if its input data style (i.e. fully and simultaneously editable input parameters of all objects in the completely visible code-like structure, updateable by a single click, with pretty good STL export ) could replace mged CLI. Regards, Mike -----Original Message----- From: Christopher Sean Morrison [mailto:brl...@mac.com] Sent: Thursday, November 17, 2016 5:09 PM To: BRL-CAD Developer Mailing List <brlcad-devel@lists.sourceforge.net> Subject: Re: [brlcad-devel] Newbie to BRL-CAD Hi Mike, > You are absolutely right that it would be the best solution to completely > avoid export to STL for 3D printing and to make a direct export to g-code > instead. But I am afraid that it would take very very long time and BRL-CAD > might meanwhile miss a great opportunity for attracting new users and new > developers from explosively expanding 3D community, which could in turn > rapidly auto-accelerate the BRL-CAD development. 3D printers are becoming so > cheap and their use so simple that they will soon be considered as normal > household appliance like PC or laserjet today. It will create millions of new > CAD users, which, like me, will soon want to move from the "kindergarten > CAD" like SketchUp, 123D or Tinkercad to something more powerful but still > free. Completely agree. The only caveat is that it shouldn’t take as long if we collaborate with guys that already speak gcode (e.g., linuxcnc). BRL-CAD already has highly optimized logic for understanding where solid objects exist, much more than the non-solid “kindergarten CAD” as you put it. > 1. User friendly GUI. BRL-CAD has a notorious image in the CAD internet > community as very mature and powerful extraordinary solid modeler but with a > terrible user interface "like from the 1980s". If the first-time users do not > find the easy graphical editing, visualization and manipulation features they > are used to from "the kindergarten CADs " they will quickly move to try > alternative CADs like FreeCAD, OnShape or Fusion. Agreed. And our new MGED/Archer GUI brings us into the 90’s. A necessary stepping stone infrastructure-wise to get us positioned with a fully modern GUI while preserving all of BRL-CAD’s best capability. Usability has been priority #1 for a very long long time for BRL-CAD. Just takes a long time to get there when there are other funded and community initiatives that have different priorities. > 2. Good STL import/export. The 3D printing users have available millions of > existing models in STL from the internet repos or their own past work which > they want just slightly modify in the CAD and export in STL to a slicing > software they are used to (Sli3er, Simplify3D, Cura). The same is true for > new models they make directly in the CAD, they do not have a need for > avoiding the slicing software they know. BRL-CAD’s strength (solid representation) is a weakness when it comes to polygonal export. A solid modeling system is traditionally supposed to reject exporting such notions and we’ve had to retool (and are still retooling) to support more robust export that might not necessarily be solid (and more robust solid export). Import isn’t a problem. STL files are trivial. > If I can daydream a perfect CAD for 3D printing and more, I would love to see > this combination: > 1. Mature mathematical and geometrical background and raytracing of BRL-CAD. *nod* > 2. Easiness of SketchUp GUI for graphical manipulation, editing and > visualization. Best if written in Qt as a mainstream platform known to a > large amount of developers. *nod* (Qt is the plan for our next gen GUI, some progress in a branch) > 3. Project tree structure and direct object parametric editing of FreeCAD (or > SolidWorks or Inventor). This is surprising! That is, the tree structure aspect, not direct parametric editing. Both MGED and Archer support some direct editing, but MGED actively hides it and Archer isn’t even beta. Are you referring to the object hierarchy or filesystem tree structure or? Please elaborate. What aspect do you like/want from the structure? > 4. Editable from-beginning-to-end complex code-based parametric modeling of > OpenSCAD, replacing current "command-enter" CLI of BRL-CAD, for more advanced > work. We’re nearly compatible with OpenSCAD’s object definitions. Some of our work of recent has been to bridge us closer so we can directly speak their syntax and use them for parametric evaluation. It needs someone to champion the project. > I hope that I do not sound too negative or naïve, I really started to love > mged more and more with every tutorial I already followed :) I am just > little bit pity that such nice system like BRL-CAD would become undiscovered > by 3D printing community only for missing the user friendliness they are used > to. I completely understand and totally appreciate the feedback. Keep it coming. ;) Cheers! Sean ------------------------------------------------------------------------------ _______________________________________________ BRL-CAD Developer mailing list brlcad-devel@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/brlcad-devel ------------------------------------------------------------------------------ _______________________________________________ BRL-CAD Developer mailing list brlcad-devel@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/brlcad-devel
