The Global ARCHICAD Community

Stay informed. Get help. Share your knowledge.

GDL language and objects, API, Add-ons, Script…

Moderators: Karl Ottenstein, LaszloNagy, ejrolon, Barry Kelly, gkmethy

What is your opinion about this Wish?

Essential
53
90%
Important
3
5%
Average
No votes
0%
Not Important
1
2%
Not Needed
2
3%
User avatar
By Lingwisyer
#303947
Apparently there is work going on in regards to revamping the GDL IDE. To what extent, Marta did not know. It is not due soon though...
By Raoul Cenan
#304104
GRAPHISOFT PLEASE LISTEN!

We need a faster genuine ArchiCAD tool for doing genuine parametric objects! And this really quickly.

Otherwise, as already mentioned,

ArchiCAD IS ON HIGH SPEED ON THE HIGHWAY TO LOOSE THE BATTLE IN FAVOR OF OTHER SOFTWARES LIKE REVIT.

GDL is good but compared to other softwares this tool didn't evolve. The scripting interface's evolution is almost null in 15 years. There are new GDL commands with each new ArchiCAD release but using this programming language demands an effort of trial and error that nobody has the time to spend today!!! In the past when were no other choices this was good. But now we need a faster and more adaptive tool.
A graphical node based GDL tool would be possible.

Grasshopper as an external tool means paying twice the price for one tool and it's really cumbersome. Is it the way to go? I don't believe it. Not on a long term. I don't imagine an external tool working like an internal one. It never does.

Could we imagine the integration of Grasshopper in ArchiCAD ? This would be great!

If not, a new visual tool for parametric objects is REALLY CRUCIAL for the future competitiveness of ArchiCAD on the market!
By gavinNZz
#304125
Raoul Cenan wrote:
Sat Nov 23, 2019 1:42 pm
GRAPHISOFT PLEASE LISTEN!

We need a faster genuine ArchiCAD tool for doing genuine parametric objects! And this really quickly.

Otherwise, as already mentioned,

ArchiCAD IS ON HIGH SPEED ON THE HIGHWAY TO LOOSE THE BATTLE IN FAVOR OF OTHER SOFTWARES LIKE REVIT.

GDL is good but compared to other softwares this tool didn't evolve. The scripting interface's evolution is almost null in 15 years. There are new GDL commands with each new ArchiCAD release but using this programming language demands an effort of trial and error that nobody has the time to spend today!!! In the past when were no other choices this was good. But now we need a faster and more adaptive tool.
A graphical node based GDL tool would be possible.

Grasshopper as an external tool means paying twice the price for one tool and it's really cumbersome. Is it the way to go? I don't believe it. Not on a long term. I don't imagine an external tool working like an internal one. It never does.

Could we imagine the integration of Grasshopper in ArchiCAD ? This would be great!

If not, a new visual tool for parametric objects is REALLY CRUCIAL for the future competitiveness of ArchiCAD on the market!
Hi again Raoul,

Totally agree and NO Graphisoft won't listen as they have had their head in the sand for the last 15 years!

Best solution for GDL is to ask Epic Games to build a toolset within Unreal engine to allow us to build GDL scripts in there.

The only problem with this solution is that you are asking a racehorse to give birth to an elephant!

The elephant needs to be put out to pasture along with GDL. We are only still using it because no one has got the balls ($) to develop a new piece of BIM software with a modern code base.

It is hugely frustrating.
User avatar
By leceta
#304431
Visual GDL editor? no, GDL is already good as it is. Creating a new language is probably not a reallistic undertake.

Even is possible, graphic GDL wouldn´t have the potential that logical construct of plain scripting permits, let alone the maintainability of complex logics.

In my opinion, GDL should stay as it is, of course developing on features as recently added Dict data structure. Those are real advancements. Without doubt, the IDE could be vastly improved.

By the other hand, the decision of using Grasshopper as a side companion for Archicad, has still a lot of potential and this need a lot of time an effort to be achieved. This potential arrives even for using GRASSHOPPER GRAPHICAL PROGRAMMING ENVIROMENT FOR COMPLEMENTING GDL . IMO, the GH + Archicad combo is all about this: Complementing each other for a sinergetic product.

I already have some working object where half is made in Grasshopper and other half is made in GDL: I left my objects geometrical logic to be developed inside GH, where it shines, and let GDL create the Archicad friendly lightweight objects.
By joteroar@icloud.com
#304897
Whaaat? It is absolutely not ok as it is. Sorry, GDL is indeed ok as it is; the way we are forced to use it is absolutely not ok. Architects can and must draw, not programm. How many archicad users you think are using GDL? 0.1% maybe? My goodness! they even use two complex profiles, one vertical, one horizontal for doing objects in order to not having to programm! or Morphs everywhere. GDL is very powerful to let it just die or leave it like a freaks feature. It is also veeery slow for the everyday making of parametric objects. We NEED for sure a graphic interface for drawing parametric objects instead of programming them. There's no need for a new language, just some tools to conect drawing with scripting. That's basic. I don't think there's one single architect who wouldn't like to have that. I don't know how exactly but there are plenty of ways to implement that. Morph is a great tool. It would be amazing to have a way to turn morphs into objects keeping every dimension parametric, just like the new complex profiles and just drawing the 2D plan with the same parameters instead of writing scripts. Sorry, that's just my opinion but every single architect that I know says the same about that: "GDwhaaat?programming???are you crazy??I won't do that!"
By Braza
#304919
leceta wrote: I left my objects geometrical logic to be developed inside GH, where it shines, and let GDL create the Archicad friendly lightweight objects

The problem is that the current interface to create "geometric logic" GDL objects is not logic for an architect mindset.

You are talking about two different things: Creating GDL objects AND managing them as a simphony.

I agree GH is an amazing tool to conduct an orchestra of GDL objects.

But what really a Maestro can do with mediocre musicians?

The thing is 99.9% of architects can't even think of how to create a "geometric logic" GDL object.

My 2 cents
User avatar
By leceta
#304925
No, I think I know what are you referring to, and not, I´m not using as an "orchestration" tool, I´m actually transporting GDL geometrical operations (local, self-contained, not another agent dependant) to grasshopper. I format resulting grasshopper data to GDL friendly data structures, and those data structures are read by the GDL objects. It could be a long list of vertex coordinates for a mesh, a transformation matrix, whatever...

And yes, i´m agree that there is actually potential for grasshopper to orchestate GDL objects, but this is other question...

"GDL objects is not logic for an architect mindset". Then this mindest you talk is not prepared for grasshopper neither.
By Braza
#304929
So. You create GDL scripts from scratch and manipulate them with GH tools then save the result geometry as a single GDL object inside a pln file?
By Braza
#305000
Thanks leceta.

Very interesting workflow. Would you mind posting the file here?
leceta wrote:"GDL objects is not logic for an architect mindset". Then this mindest you talk is not prepared for grasshopper neither.
Sorry I didn't make myself clear... I was trying to say that: The current Archicad interface to create GDL objects is not suited for an architect mindset. Not the GDL object itself. That's why I am suporting a Visual GDL editor.

Anyway. GH/Archicad connection seems a very powerful tool to enhance GDL potencial. But IMHO its not an "Architect Friendly" interface. For me those wires, switches and gizmos are more for an engineer mindset. I like it myself (probably because I must be a frustrated engineer) but it's not for the average architect.

What do you think?
  • 1
  • 2
  • 3
  • 4
  • 5
  • 8