Podolsky wrote:
My opinion that PARAM-O is quite limited in term of modelling capabilities. PARAM-O is generating GDL script, that later ArchiCAD is reading.
Knowing GDL well - it's possible to achieve much faster better results. But only if you know GDL well.
But it's good tool to start learning GDL.
I wholeheartedly agree with Podolsky apart form one thing: You will be able to grasp some of the concepts (and limitations) behind param-o much faster if you know at least basics of gdl coding.
On that front, I was very dissapointed how much messy the code coming from param-o is.
Now to my personal experiences: I have opened it for a first time about a week ago. My goal is to eventually create a series of historical windows objects.
The idea being I will be able to put measurments into archicad as I am on the site. The logic of measuring and the measurment you get is very different from what you would imput into traditional archicad window object. And as we are leading practice in conservation projects it seems to us to develop a custom object for most common (historical) window types is worth the effort. Or at least it would if param-o was a little bit smarter.
Currently, as far as I am aware there are serve limitations to what can be achived and I personally got quite discuraged when i realized that very likely this tool is not (yet anyway) mature enough to achiveve what I want it to. Currently I have a frame for what Germans call "KASTENFENSTER" (not sure if such windows were used much outside europe)
The limitations I have encoutred so far are that it is very complicated to do anything that curves and bends and honestly I dont understand why primitives such as arc, arch and so on are not avalible inside gdl for years. Lack there of must make the work for the library developer, even inside graphisoft, very dificult indeed. Some of those limitations could, ofcourse, be resolved by having the ability to stetch profile along curve, but currently that is not a featute that param-o posseses. Or if it does, then it is not documented very well.
Which neatly brings me to the next, and biggest gripe with it: The documentation for the nodes is completely unusable and does not explain anything. in stark kontrast to the gdl guide (and archicad help) that is actually fairly usable.
The last gripe I have with param-o (and I hope this get resulted pretty quickly) is that if you want to do something more complex, it is greatly advatgeous to make few simple scripts and call it from the other. In my window example, I wanted to make separate param-o, that will be the window frame, and insert into it another script that will be the wing - opening part of the window - However currently only four parametes can be edited in the nested object and that is nowhere near enough to work for me (actually five could work in this case but I degess)
it would be cool if more paramtres could be added for a single library part.
Also there needs to be more of actuall programing possibilites - all those if, else and so on. Also it would be very good if graphisoft made some functions more accesible than the gdl - for example if you programed a editable hostsot in gdl you know the pain of putting in one point for the start then another for the end and then adding witchcraft to make it actually manipulate some value. It would be great if hotspots were already one box i just plug stuff in.
And one other small request in the end - it would be awesome to be able to select part of the resulting geometry in the preview window and had have the responsible node highlighted. This would do so much for more complex work, debugging and so on.
And here is the fruit of my param-o work so far. (please ignore the cube in the bottom. It is soon to be a bottom frame of the window)
From Archicad 8.5 all the way to the present.
Passionate Autocad hater as well.
Wish to save palets and menus individualy in work enviroments:
https://bit.ly/3pg3Bx7
