AC15 / Modelling

[Anonymous]

So now we know ArchiCAD 15 and 16 are to do with modelling, similar to the way 13 and 14 were centred around teamwork.

This will be interesting to watch, as the implementation will affect whether solo users see this as “yet another update for the big end of town”, or whether it has a general benefit to the usability of the program, and therefore is beneficial to all users.

I’ve long held the view that the 3D engine in ArchiCAD is archaic and in need of urgent overhaul. Not only is it slow, but the ability to create complex objects is limited to GDL scripting, and then there is the whole disconnect between the 3d model and 2d view. Specific complaints might include:
• Single threaded processing (AC needs to maximise use of mainstream multicore CPUs)
• Main accessibility via GDL scripting irrelevant to majority of users (AC needs a good GUI for modelling - urgently).
• Buggy SEOs that magically affect a 3D view without modifying the 2D view. (AC needs a robust modelling toolbox including nurb type shapes, with reliable transformations, that are graphically apparent in the final object).
• Complex model elements do not relate to the ArchiCAD toolset. (Creating complex walls etc still need to behave as walls and be able to receive doors and windows etc - think complex profiles on steroids. One of significant flaws in GS last attempt, the Maxonform add-on, was the way “intelligent’ objects lost their properties and became part of a dumb amorphous mass – and so could not be scheduled, dimensioned etc.).
• Inability to set up relationships between objects

Observations of the pace of development and resources available at GS suggest that they are struggling to keep up with the competition. Two years of AC updates were consumed implementing and fixing Teamwork, and we understand at least 3-4 years development preceded that. Other parts of the program have slipped into irrelevancy (e.g. Lightworks– though fortunately one can get around that with a plethora of capable renderers).

Nemescheck as a group has committed to IFCs for interoperability and so we see both GS and Vectorworks (VW) committing to that standard. Returning to the topic of modelling, VW have embraced Siemens Parametric technology going forward. In an ideal world I would hope GS could also embrace such – not only offering a terrific boost to AC, but also strengthening the Nemescheck group through a far more powerful interoperability than IFCs offer. We shall wait and see what unfolds next year….
Interested to hear what other users have to say on the topic....

[Anonymous]

Steve wrote:
Do you think some of these add-on companies have made deals with Graphisoft not to put them out of business ?

I doubt it. Software companies generally don't rely on third parties to decide development strategy. It is much more common for the best ideas to get incorporated into the software. It may be though that the presence of good third party products may reduce the urgency of including those features.

[Anonymous]

A huge leap forward in AC15 would be the ability to import or convert RVT objects. Hundreds of companies have already created their entire product lines using Revit. I have no desire to purchase Revit. However I would love to have access to this huge online database of objects created by all these companies that did not make the same product in GDL or as GSM. And think about it... they have already allowed for interoperability with a lot of structural and mechanical software (including those by Autodesk), so this seems the next logical step in improving the software.

[Anonymous]

SmileyMan wrote:
A huge leap forward in AC15 would be the ability to import or convert RVT objects. Hundreds of companies have already created their entire product lines using Revit. I have no desire to purchase Revit. However I would love to have access to this huge online database of objects created by all these companies that did not make the same product in GDL or as GSM. And think about it... they have already allowed for interoperability with a lot of structural and mechanical software (including those by Autodesk), so this seems the next logical step in improving the software.

This would be nice but I suspect that it would be somewhere from very difficult to impossible. Revit has a very different way of defining the parametrics and even if that weren't the case it would be difficult for any automatic function to figure out the intent behind all the variables. On the positive side Revit families are simpler in principle than GDL parts so it might be possible. I'm sure going the other way would not.

[Anonymous]

A bit frustrating.... They do not have to be functional 'with parametrics.' Since the Windows version of AC allows for SketchUp objects to be imported (which are not parametric), I see that as an unnecessary requirement, and nowhere did I request it.

IFC conversion is great and your point is well taken regarding the MEP parts. .... However that presumes you want to pay for Revit so that you can get the objects, convert them to IFC, then import them to AC. But at that point, are we making the argument to switch from AC?!

We need to realize that there is currently no incentive for any of these companies to make a GDL version of all their parts when they have completely created their libraries in RVT. So instead of trying to convince the thousands of companies to go thru their entire object-creation process again (which they will not), it seems most plausible for one company (Graphisoft) to make the necessary adjustment and capitalize on all the work and parts already created in the other format.

All that aside, the absolute smartest way to go would be for someone to make an IFC software product that companies can create their objects in FIRST so that their is no need for conversions between AC, Revit, Microstation, VectorWorks, etc etc etc.... just like HTML pages can be viewed in IE, FireFox, Safari, etc.

(and yes, I saw your other post)

[Anonymous]

SmileyMan wrote:
A bit frustrating.... They do not have to be functional 'with parametrics'.

That depends on the parts. For plumbing fixtures all you need are materials parameters (which is actually a bit tricky but doable between AC and RVT) but doors and windows tend to be highly parametric and quite complex.

From what I've seen of the Kolbe parts they would be a real challenge to convert without a lot of human intervention. They also have a SketchUp library but even there they are doing some pretty sophisticated stuff and I'm not sure it would be too easy to import them either.

I agree that the ideal would be to have a common standard through IFC (or something) but I doubt that Autodesk would be too eager to support it.

[Anonymous]

rwallis,

Can you please tell us if you only heard about conceptual improvements to modelling or did you hear about any specific details to modelling as well?

If it appears from what is being said here that AC15-16 are going more into free forming, that would need its own set of tools, so how would this possibly affect the normal (box) forming tools?

Can you spill the beans a bit more specifically here?

Thanks

[Anonymous]

Matthew wrote:
That depends on the parts. For plumbing fixtures all you need are materials parameters (which is actually a bit tricky but doable between AC and RVT) but doors and windows tend to be highly parametric and quite complex.

I disagree. The only division I see parameters 100% necessary is door/windows. But other than that, all other CSI divisions could be dumb objects. Personally I would prefer to have the faucet first instead of waiting for the "perfect parametric" faucet, GE appliances, etc.

Personally I do not see the need for an object maker at this time. There are several products out there that help with object creation [ArchiForma, OBJECTiVE, Extrusions, etc]. Graphisoft has already started the monumental effort of interoperability in AC14. The next logical step would be to continue this to objects/library parts.

Another preference would be that they expand on the power of the Curtain Wall Tool. The tools ability to include several components as part of its creation has allowed me to create floor framing, unique tilted acoustical ceilings, unistrut systems, etc. I think it would be better to spend time expanding the power of some of the other tools. If stairs used the CW tool ideas, then balustrades are more customizable, caps are Profiles, and maybe so are nosings. Think of walls that have the framing already apart of the tool instead of a secondary step with an AddOn. Floors and Roofs that can be assembled with their parts (like metal decking, tapered insulation, cants where they meet parapet walls, etc) instead of just line separations in the current components. Window and Door tools that allow you to select Profile shapes for customized door/window moldings. All of this technology and programming is currently available in AC14!! But it just has only been "tool specific." These are just some of the possibilities of each Tool is expanded to include what others already allow.

Either of these two ideas (reading of RFA/RVT object files or expanded internal programming) would be a monumental leap forward. I admit the first is a challenge, but it is necessary. The second is a lot easier since it is already part of the current programming.

[Anonymous]

I'm hoping that with the advent of IFD the issue of content on various platforms is eliminated. The manufactures just provide a text file that describes their product and it is up to the software vendors to generate the object/family based off of the text file. Sounds like it puts less work on the manufactures to keep up with the bim formats.

[Anonymous]

IFD?

[Anonymous]

http://www.ifd-library.org/index.php?title=Main_Page