Wishes
Post your wishes about Graphisoft products: Archicad, BIMx, BIMcloud, and DDScad.

New Door & Window Features

Anonymous
Not applicable
I just got what I think is a great idea for improving (dramatically) the usability of doors and windows, and it shouldn't even be very difficult to implement.

Allow the user to associate multiple library parts to a single opening. By treating the doors and windows as assemblies at the user level many current problems could be quickly solved.

For example, it would be possible to select all windows and assign the same exterior casings, sills, lintels, etc., and then select subsets separately (such as main floor, second floor, east wing, conference room, etc.) and assign common interior treatments to each subset. The sash types (casement, awning, double hung, etc.) could then be specified in common or independently without messing up (or having to redo) the other components.

Another big advantage is that the libraries can become mix and match. It would be possible to use any combination of sash styles and door types from various libraries with exterior and interior treatments from still other libraries and mix in home grown custom entry doors and stained glass windows.

The beauty of this is that it only adds capabilities. Current libraries would still work as before used as single parts. The reason I think it should be easy to implement (from Graphisoft's side) is that it only requires that additional tags be added to call the other parts and some adjustment to the parameters settings and custom interfaces in the settings dialog. It would of course be up to all of us making library parts to fully take advantage of the possibilities, but I suspect this would happen rather quickly.

It might be a bit messy if the multi part capability were made available to all the current libraries which have been elaborately developed to get around the current limitation. It would be good to review the global parameters and establish guidelines for new doors and windows to help ensure that they will play well with others. It would probably also be good to include a new flag to indicate whether a part has been developed according to the new standards.

The possibilities for doors and windows go far beyond the limited examples I have had the time to come up with here, and beyond doors and windows there are stairs, skylights/dormers, and all the objects that could benefit from the same capabilities.
6 REPLIES 6
Geoff Briggs
Mentor
Indeed this does seem like a viable approach. One that stands in stark contrast to the current state of the art as exemplified by Door/Window Builder.

DWB is an amazing achievement but if anything has become overly complex. While it attempts to be modular and offers opportunities to save presets it is quite laborious to work through all the dialogs, many of which are confusing due to how much is crammed on to each pane. And in the end many simple North American window profiles remain difficult or impossible to properly represent, especially in section.

So the component based approach not only has the advantage of flexibility but leaves the prospect of creating or customizing a needed sub-assembly as a real possibility. As is stands now one would have to dig pretty deep into GDL to add something as simple as an apron, only to repeat the same modifications to multiple window objects used in a particular project. That or hope to convince the developers of these complex objects to that your need justifies adding yet another custom parameter.

One thing that would really make this fly for me is a graphic development toolkit. Even if some GDL tweaking were required down the road, it would be great to build that sash from parts rather than code, in an environment where you could designate which dimensions became parameters.

Interesting subject.
Regards,
Geoff Briggs
I & I Design, Seattle, USA
AC7-27, M1 Mac, OS 14.x
Anonymous
Not applicable
Geoff,

Thanks for the response. It seems I can always count on you to help develop and expand on my ideas. I really appreciate your contribution and enjoy the collaboration.

What you said about DWB is very much on point. It is an extraordinary example of dealing with the problem of managing multiple complex assemblies with multiple complex relationships using the limited tools available. It is inevitable that the product becomes overly complex and difficult to manage. (Not to criticize the achievement, by all accounts it sounds like an exceptional product.) The fact that such a thoughtful product is ultimately difficult to use and limited in it's flexibility is a sign of fundamental problems in the foundation on which it's based.

The great strength (IMHO) of my proposal, from the programmer's point of view, is that it makes it truly possible to modularize the GDL code. It is a bit unusual in that the end user becomes an active participant in assembling the modules. I can't think of any software that does this, except perhaps the way Garage Band lets you assemble instruments into compositions. (Kind of a nice analogy come to think of it.)

GDL macros don't really work like truly modular, object oriented code because they don't have the "black box" quality of structured input & output. What I imagine is that the component macros behave according to some basic rules so that when you add a casing or sill to a window unit they will relate in a simple and predictable way. I don't think this requires some great forethought to create comprehensive rules and categories. Simple and reasonable guidelines should be sufficient to start with and with the proper feedback loop between the developers and users (i.e. ArchiCAD-Talk) these guidelines would be self correcting and evolve into a more complete and robust set of rules than anyone could possibly create in advance.

I think this change would dramatically improve the quality and quantity of libraries and parts that people produce. I'm sure library developers' lives would become much easier. I know from making my own custom doors and windows that most of the time is spent coordinating all the macro calls, parameter settings, and custom interface elements needed to manage all the complex interrelations between the various shared components.

Your comment about a graphic development toolkit is quite interesting in this context. Of course numerous attempts have been made with varied success, but they have all been limited by this one item one part restriction. What you describe take the concept one step further into sub-components. I am imagining building a sash from frame, muntin and glass parts.

This naturally brings up the next extension of the concept in the other direction; super-assemblies. The obvious example is ganged windows in a single opening. It would be great to be able to arrange, say, several double-hung windows with transoms, or some combination of casements, awnings and fixed lights, etc.; then merge and mull them into a single unit; and finally add single interior casing treatment and exterior masonry opening details to the whole thing.

All exciting ideas as far as I'm concerned.

Here's to hoping they come to fruition.
Petros Ioannou
Contributor
I 'm not sure I understood , but are you refering to something similar with the way Architectural Desktop 2004 is handling window-door styles?It has a clever way of asigning diferent door-windows to an infill of a curtain wall (also diferent profiles) and can easily create different assemblies. Too bad someone cannot use it (in my opinion) because ADT 2004 is miles away from being user friendly...
ArchiCAD 22 4023 UKI FULL,
Archicad 21 6013 UKI FULL, ArchiCAD 20 8005 UKI FULL
iMac Retina 5K, 27-inch, 2017
4.2 GHz Intel Core i7
32 GB 2400 MHz DDR4
Radeon Pro 580 8192 MB
Dave Jochum
Advocate
As I'm not a code warrior or even GDL literate, I don't follow the complete discussion here, but I would be very supportive of a system that would simplify the creation of doors and windows while at the same time increase the power and flexibility of the process. I use D&W Bldr. only when I have to build a unit that is not possible with the AC library, due to its over the top complexity, confusing icons and window layouts, and slow running scripts. But it is powerful. And of course the problems with the AC library parts are well documented.

This is an area that sorely needs help in AC. I hope GS takes this discussion to heart and posts a response. Incorporating a powerful, flexible, simple to use D&W system would be well received by the users and should easily pay for its development and then some.

Keep it up Matthew and Geoff!
Dave Jochum
J o c h u m A R C H I T E C T S http://www.jochumarchitects.com
MBP 16" (M1 Max) 64 GB•OS 13.5.2•AC 27 Silicon (latest build)
Aussie John
Newcomer
just to continue some comments on another thread, objects could be linkable also (eg stair and handrail, column and beam, series of columns -at particular centres or a number along a certain distance)
Cheers John
John Hyland : ARINA : www.arina.biz
User ver 4 to 12 - Jumped to v22 - so many options and settings!!!
OSX 10.15.6 [Catalina] : Archicad 22 : 15" MacBook Pro 2019
[/size]
Anonymous
Not applicable
...and to expand even further,

The stair type library part would include standard functions for passing rise/run, floor to floor, etc. to the stair rail part for accurate and automatic operation.