The Global Archicad Community

Stay informed. Get help. Share your knowledge.

Discussions closely related to Archicad. (Example: Do we need a Linux version of Archicad?)

Moderators: Karl Ottenstein, LaszloNagy, ejrolon, Barry Kelly, Gordana Radonic, nbalogh, mnguyen, gkmethy, mtron, Csaba Kézér

Moonlight wrote:As an old ArchiCAD user, I have found out that Graphisoft is not that type of company that will do what ever that is asked for to do, but will give you the answer you need, but lately I'm starting to have my doubts.

Me too. These days I am finding myself with mixed feelings: Something revolutionary good is on the corner; OR Graphisoft is being eaten alive by Nemetschek. :?

1. May be my last comment came late to you, but I meant that even most of those who have disposable time don't want to learn it, at least in my own direct experience.

2. Grasshopper is a graphical programming interface, yes it may lack a coherent 2D representation, but not per-se, I mean if you can program a 3D model, in Grasshopper, you may also use it to create it's 2D coherent representation, just in GDL, which can depend on automatic generation, of script a custom appearance.

3. As I said before, GDL have a very limited bi-directional data exchange compared to other platforms solutions, so creating a Graphical GDL under the same context is reinventing the wheel for no real benefit, besides, if you already have a solution that has already made the solution that is dynamic enough, almost limitless for your uses, and can solve bi-directional data exchange, then use it, and put this time and effort for solving other more important issues.

An example for what I meant of bi-directional data exchange:
Moonlight wrote: Tue Feb 11, 2020 12:49 pm @BRICKLYNE CLARENCE

I see that I wasn't clear in my previous explanations.
2. About Graphical GDL & "reinventing the wheel":
2.01. Many of you may not know that GDL is capped in what is refereed to bi-directional communication data exchange (i.e: the data that is sent/received between library parts and the program is limited), when right now, our practices are in the verge to conversion to adopt IT technologies and data, which is something that is somehow is already/almost solve/WIP solved with Grasshopper... so creating a Graphical GDL will not solve the problem of increasing ArchiCAD's user base........

As an old ArchiCAD user, I have found out that Graphisoft is not that type of company that will do what ever that is asked for to do, but will give you the answer you need, but lately I'm starting to have my doubts

Actually, I think that you also might have misunderstood a little bit why people are requesting such functionality (a graphical user-interface for custom GDL object creation) in ArchiCAD.
It's not necessarily for the purpose of (directly or expressly increasing the user base or number of users that would like to adopt the program.
Certainly not directly in that way.

But when you do this, you make the lives of the people who DO use the program easier, and in turn make them happier with the product they have purchased and justified in the money they have spent in doing so.
And they in turn actually DO play the part in helping increase your user base - indirectly - by either speaking well about it (and indeed about Graphisoft as a responsive company), and recommending it to their colleagues.
There's less of that that happens nowadays than use to happen when I began using this program years ago and most people I know now who know how to use the program only know it either because they were introduced to it in school as a free learning tool (and therefore had little to no choice but to learn it) or because it was what was being used in one of the first jobs they had (and therefore they had little to no choice but to learn it). Not because they heard great things about it from colleagues they respected and decided to try it out for themselves. At least not as much.

Furthermore, a lot of people often complain (justifiably so) about the dearth and scarcity of good third party supplier products and library parts from manufacturers they use in the projects, that are available to download or obtain online (as opposed to Revit library objects or the plethora of stuff on the (formerly) Google Warehouse).
If I'm a third party manufacturer and supplier who would like Architects and designers to use and spec my products in their projects, do you really think I'm going to invest the time to get a person to learn how to code in GDL to create my user library with smart intelligent and well-coded objects for the availability for Architects using ArchiCAD to download (when it is already hard enough for the architects themselves in the industry to find those same individuals to hire them in their firms - let alone learn coding GDL themselves)? Especially compared to how much easier it is for them to find a basic Revit user to create basic Revit family library objects where the learning curve for that (for the average Revit user) is far less steep than learning GDL scripting is for the average ArchiCAD users.

So it's no wonder why it's so much easier to find Revit library objects for products online than ArchiCAD objects.
Even more so for Sketchup objects.
And both of which are lousy when converted to ArchiCAD through the various translators.

And I always return to what seems like the failure to launch of Graphisoft's own BIMComponents online repository idea modeled after Google/ (now) Trimble's Warehouse idea.
Classice Cart before the Horse strategy thinking on Graphisoft's part there, in that they created the Warehouse for people to find the objects before they improved the means by which people can create the objects that they can then put in said Warehouse to share with other users and have that online repository (and save themselves long-term from having to create new (smart) library objects when some of that overhead can by carried over by users themselves. I mean, some of this long-term thinking just speaks for itself).
As it is, a large section of BIMComponent objects now just come with the descriptor "Imported as .SKP object" which basically means its versatility and usability in the ArchiCAD environment wlil be severely limited beyond just being a cosmetic placeholder.
And I don't think this was what Graphisoft had in mind when they created that online repository.

But that's what happens when you approach these things in such a backwards (or as we say here in North America, such an 'A$s backwards' ) way.

So improving the ability of users to create smart GDL library custom objects (with a graphical interface) isn't directly going to increase their user-base.
But indirectly, it will help increase awareness outside of teh direct user base thanks to happy users, as already mentioned (While simultaneously, hopefully preventing others from leaving the program out of frustration and feeling their needs are not being met and that they're not being heard - which also indirectly ties back to how people speak about the program to non-users).

And that can't possibly be a bad thing, can it?
Moonlight wrote: Tue Feb 11, 2020 12:49 pm @BRICKLYNE CLARENCE

Few years ago, in ArchiCAD España FB group, a member asked how to do something (I don't remember what was it exactly), so many proposed many different ways, and I proposed to use Grasshopper to do simple operation, since it was going to be quiet quick to preform, and I would stay online to give that member the indications to create the algorithm step by step in case he didn't knew how to use Grasshopper.
Then a member started to laugh and to joke to undermine the idea of using Grasshopper, long story short, he said, I don't want to program, nor to learn programming, I want something that does what I want when I want to it to be done how I want it to be done.

And as far I know and seen, most ArchiCAD users are like that whether it was for Grasshopper or GDL, except for a minority and those who are really busy and don't have some free time to dedicate it to learning.

Also, fair point from your story about the ArchiCAD user being reluctant to learn Grasshopper any more than they would be willing to learn GDL programming because,...."they don't want to learn programming but only want the program to do what they want it to do"

This is true and is the basic under-pinning of the argument we're making here.

But even given that, if you DID have to do some scripting or programming and you're coming from a graphical and visual design background that most here inevitably come from, would you much rather do it in a line-coding environment that we now have in GDL scripting or in the node-based component visual scripting type environment that Grasshopper now offers?

And not that I'm even saying that Graphisoft should use Grasshopper as a template for how to revamp the GDL scripting interface.
But rather the point being when Grasshopper was created, it was done so with the thinking and keeping in mind of whom the intended target user was going to be and how they are accustomed to thinking and in what way.

Line-coding and typing hundreds of lines of code is a far far cry from being able to visually see what you want your code to see by graphically connecting nodes and components and getting direct feedback without having to debug things or run code over and over again like you're in some 1980's computer hacking movie minus the cool 1980's music as a soundtrack.

This new graphical GDL scripting environment could just as easily end up being a (vast) extension of what they seem to have already started with the current Profile editor where you add customized controllable parameters to graphical properties of objects you've already created in the 3D window (or will create) with a wrapper or "interface" that allows one to add more define-able parameters and properties if needs be.
In other words you work backwards (so to speak) from the 3D window and creative environment where you CREATE to the environment where you add intelligence to the object or the created object based on what you can see.
There's a lot of chatter here about the programming as part of the design process. As someone who started on CAD some 35 years ago let me just make a few observations. I started on Autocad, and back then it was literally a digital drawing board so the advent of LISP programming provided huge productivity benefits to every user. Ten years later I moved on to using Archicad where we had GDL available. The advances in moving from 2D CAD to a 3D architectural package was huge and much of what had been done with Lisp was now just part of the regular 3D drawing process. However, GDL was available and I spent many an hour on various tools and working through DNC's Cookbook to address various shortfalls in the toolset. These days my programming interests have been abandoned as the current version of AC is sufficiently enabled that it is possible to do most of what I need without spending unpaid hours on programming. That is, from my perspective, a huge accolade to Graphisoft. Now I'm not saying that all the Grasshopper chatter is irrelevant. If you have the scale of project, client budget and technical resources then please feel free to push the boundaries, but for the average small practice where design is a visual process with the right tools available, programming is pretty much an unnecessary overhead. I will say all the talk of visual GDL has me amused because the only reason I see a need for it is because to date Graphisoft have failed to provide the ability to create blocks/modules that can be simply modified within the main drawing using standard tools. I do find it strange that a piece of advanced software in the 21st century is still missing what many competitors provided 20 years ago. Anyways if you want a bigger market share keep your customers happy and keep the workflow simple.
Bricklyne Clarence,

Let me talk you about frustrations:
1. Do you remember like 5-6 years ago, when the ability to clip a 3D view was introduced, well I have waited for that options like 2 years, and it was out of the box in Revit ever since I started to use it.

2. Do you remember how many years we had to wait so that we can make a schedule with some spreadsheet capability, like 5 years (at least).

3. Do you remember how much time I have waited for an option to automatically create topographic contour lines for any given terrain, well I have waited from version 6.00, until today ... and by the way, here is a manual I have created in its day to help a friend of mine. ... rhinogh-1/

4. Do you remember when ArchiCAD had an ODBC database connection out of the box ? Are you aware that it have taken out of ArchiCAD in version 18, and till today as much as I know (correct me if I´m wrong) it have not been substituted with another type of database connection.

5. What about graphical over-ride ?? How many years ever since users found that option in Revit ?

6. Or other line-type be implemented in 3D, like dashed lines

7. How many years have we waited so that Graphisoft change it's rendering engine ?

And like those issues we as user based we have a lot.

On the other hand, if you followed Rob Jackson's comments on Bond Bryan twitter account, he says that in their office they only use generic objects.

So that guy, who is an IFC & BIM expert, and an international reference on those issues says that, I guess it's better to leave Graphical GDL interface for the time being.
Bricklyne Clarence,

I´m not ranting, but right now, I see that we are in the era of programming/algorithmic design in our field, similar to when ArchiCAD was the dawn of BIM when everyone was on CAD.

The good news is that ArchiCAD have a good base and infrastructure for the coming era, and have took good decisions respect that issues, but it's not doing enough for the coming era, and its competitors are making huge strides on those specific issues, and it have not taken the advantage of selling ArchiCAD as the natural BIM of choice for algorithmic design done in Grasshopper.

So either Graphisoft mend those issues ? Or we may find ourselves forced to leave the usage of ArchiCAD.

To have a better idea of what I´m talking about, watch the Webinar I have posted few comments before, and you will understand.
DGSketcher wrote: I will say all the talk of visual GDL has me amused because the only reason I see a need for it is because to date Graphisoft have failed to provide the ability to create blocks/modules that can be simply modified within the main drawing using standard tools.
Exactly. We are missing the target focusing in a Graphical/Visual GDL interface. What architects want, is to create parametric "Things/Forms/Buildings" inside Archicad using all AC Tools (Walls, Slabs, Roofs, Objects, etc..) .

As Bricklyne Clarence well mentioned, a good start has been made by Graphisoft with the Offset Modifiers inside the Custom Profile Editor. Extending this capability to the 2d/3d environments should do the trick without having to create a new App. Lets see if it will evolve, because right now it is only scratching the surface.
@Mr. Moonlight. Topo contours!!!! THIS is one of my main gripes with archicad, how is it possible you cant generate terrains out of topo contours? Ill have to check your proposed solution! (that is, when i finally get to learn grasshopper, my new year resolution :lol: ) Anyway, manly tears were shed.

More digressions:

The Pareto principle states that 80% of the results come from 20% of the effort. Personally, I will have to agree and disagree with some of the previous statesments made here by fellow users, but lets see if we can first agree on the following:

-As per most accounts, Revit/autodesk has/is a big merchandize machine
-As per most accounts, Graphisoft does not seem to have a clear concept of what merchandizing is, at least not in a traditional way.
-Revit offers an unenjoyable building modeling experience but offers a capable/parametric object creation environment, which is known by the masses as "families and family creation environment".
-Archicad offers a VERY VERY enjoyable building modeling experience (as in, im smiling the whole time with a childish grin on my face enjoyable) but also offers an extremely obscure and inaccesible (for the mayority of users) object modeling environment, that is, if you want them to be parametric. Of course you can create any object with the modeling tool and save them as objects but other than specifications they lack any kind of parametricity althoug they can be somewhat stretchable.
-Revit users rave about parametric object modelling capabilities (i can do families!) when they should be modelling buildings.
-Archicad users dispair about lack of object modeling capabilities, when they should also be modeling buildings.

From the above, we can conclude that Revit has marketed their sole main ability that is better than archicad extremely well to the point that a lot of people are able to forgive all that software shortcomings and clunkiness just for the fact that they can do families. Kudos to them.
Meanwhile, Graphisoft has failed to promote its superior modeling environment and has let the lack of object modeling escalate to the point it is a concern for longtime loyal users.
(i dont mention the MEP modeling because lets face it, Archicad cannot compete there for the time being). We are talking about Market Share here. More outside the box thinking (which Graphisoft seems to like) needs to be done.

More later, i have got to go to kick a client´s rear...
Last edited by jl_lt on Wed Jun 24, 2020 5:53 pm, edited 3 times in total.
  • 1
  • 5
  • 6
  • 7
  • 8
  • 9
  • 25