@honoluludesktop said:
It appears to me that SketchUp will forever be linked to its roots, that is a easy to use architectural building modeler.
I think it would be even easier for architectural modeling if curves are introduced (A sketchup newbie still thinks that the curves are real till he sees the 'input sides' option. By that time he has already spent 3-4 days into sketchup and gained some level of proficiency). Also, import and export to and from other softwares would be seamless and it would not split the 3D max models into ten thousand triangulations.
I often wonder why the people think that the absence of curves is what makes sketchup easier.
@unknownuser said:
I'm one of SketchUp's biggest supporters but I recognize there are some programs out there that do some things better and I think for those who need to do those things, they need to look at those other applications.
To be honest I do not think that sketchup is the best architectural modeling software. Older softwares like autocad microstation can handle the job better and with 100% precision. But still we have shifted to sketchup because of the amount of fun it offers, the cost and time equation, the shorter learning curve. Sketchup is still not THE software for any PARTICULAR purpose so why cant it just serve every purpose in the best way it can.......
@honoluludesktop said:
It would double (I think) the physical size of every new model in the Warehouse, and on Google Earth. Wounder if that could be a problem?
I think the size of the NEW files would actually be smaller. Suppose with the CURRENT sketchup if you have a curvy architectural model worth 10MB, then with the NEW sketchup the file size would hardly be 4MB.
@alan fraser said:
However it's difficult to see how the SU inferencing engine could be made to work with NURBS surfaces which effectively contain nothing to inference off. You have to remember that the inferencing engine itself is at the very core of SU's functionality.
Thats a great point worth pondering. But if you look closely sketchup's inferencing engine works best with flat surfaces only. I do not think it works accurately with curved faces as you do not know which is the BASE POINT of any curved surface. (for example- sketchup circle's do not have any quadrants and nor can any TANGENTS be made). So, as per me inferencing for curved surfaces is never 100% accurate.
When autocad was started back in 80's , I do not think that they had concepts of splines. When 3D MAX was launched , did it have nurbs modeling.... I wonder.
All these softwares have progressed to their current level NOT because they had to do it someday, but because it was the DEMAND of the industry and they HAD to meet it......