My gdl skills are equal to 0, i have tried to understand hiw it works, like almost of Archicad users, i failed,
I believe nothing can be done without a good guidance, and the gdl side isn't explored by almost of users,
There is no serious youtube videos that explain well how to create gdl objects that can be used for daily work, even the famous Archicad youtubers avoid this subject,
Graphisfot didn't introduce any good video series about this subject, is that a choice from the company or they just don't rely on gdl for the future and they will abandon it,
How a core future of Archicad is ignored line that ?
AFAIK, almost of Archicad commands turn around gdl, it should be introduced well to the community,
My gdl skill is also close to zero, despite I can comfortably code in C# and Python for Revit.
Some well structured resources I can find:
- Graphisoft, previously, had a GDL series on YouTube, but I don’t know why the series disappeared when I tried to find it last year. You could still find some of them on YouTube, which are reposted by someone else.
- Graphisoft GDL center https://gdl.Graphisoft.com
- the famous GDL cookbook. You can Google it.
The lack of enthusiasm of talking about making GDL objects across the forums also baffles me. How do companies create custom objects? The way my office does is to import sketchup object. But I don’t believe this is the most common or “proper” way.
Having said that, I’m more interested in the future of GDL though. The reason for me of not learning GDL is its use is limited to Archicad. I kind of read somewhere that the vision of creation of GDL was for describing geometry with information in very light weight codes so that it can be adopted outside Archicad. The logic goes that maybe 3D printer could accept GDL, online model viewer can accept GDL, or even GDL could be a way IFC stores parametric objects.
But all these never happen. GDL can’t even be directly translated into other geometry format without going through IFC.
As GDL has limited use and I can easily work around by importing sketchup objects without bogging down model performance (as Archicad is so good handling / managing large model), I would rather invest my time to learn something else. (E.g. BlenderBIM)
Anyone also feels the same as me?
what can be done in it that justifies that amount of time?
Just about every parametric object you have ever used will have been scripted with GDL.
So I guess they would justify the use of GDL.
Whether it can justify the time you spend to learn and create the objects is another question.
For a single operator it could be a useful but very big overhead and therefore just not practical.
But for someone like me that works for a largish company, and we rely on custom objects, this can be a more worthwhile effort.
It may still take me days to develop a single object, but as it is used by dozens of people, over and over again, it is probably worth it.
If you can find 3rd party objects that do what you want then all well and good.
Of course some are free but some are not (because of the effort to create them more often than not).
But if you truly want an object to do exactly what you want, then there is no option other than to script it yourself (or pay someone else to do it).
I taught myself with the GDL Cookbook by David Nicholson-Cole.
Best investment I ever made (now it is free although rather old and not up to date with current GDL- but the basics are still good).
There is still much that I don't know and I am certainly not an advanced coder, but I can usually get it to do what I want.
GDL is such a fundamental building block in AC I doubt it will disappear anytime soon. It was written as a simple language very like BASIC to enable simple "plain English" statements. The problem is grasping the workflow before you try to run with it. This is why the GDL Cookbook is still mentioned on this forum, it takes you through getting started and beyond. I assume DN-C has probably retired. It would have been good if Graphisoft had (still could) taken up the challenge to reconfigure or move that knowledge forward with the GDL updates. Certainly the working examples are fantastic for learning the code and how to put something together. Once you get the hang of GDL you will laugh at the capabilities of PARM-O.
I was really enthusiastic about param-o at the beginning. But... well. You said it.
As I always state, there are a few things that I always state that could make GDL a little less prohibitive/ scary to newcomers, like line numbering (Yes, I know about CMD+L to jump to, its not practical since you can't look at) and also ANY kind of color coding, like making comments red or green for example, would make things much more readable.
I do use LP2XML, but, I do think VSCode is an overkill sometimes for something like GDL (which is quite simple and doable without having to save, convert, reload libs to see that you made a simple mistake that you could see on the go through the GDL feedback window).
I just read through the gdl odd manual. Then started reverse engineering Archicad objects to get a better understanding.
I use gdl for my own cabinets, appliances, shelving, sidewalks and streets, doors, windows, framing, exterior coverings, eaves, soffit and fascia, trusses, hand framing, posts and beams (I don't use Archicad), I even wrote a stair and railing tool but stopped working on those as Archicad stair has more power. Knowing gdl turns Archicad into an even more powerful program.