VectorScript objects and IFC

[Petri]

Can we expect that objects created with VS can be IFC-compatible?

A straight answer would be preferred.

[Gerard Jonker]

Hi Petri,

Why wouldn't they be? As long as you assign IFC classes, they are. After all, what gets exported is the drawn object and not the code.

Missing the plank entirely?

Gerard

[Petri]

Missing the plank entirely?

I believe so. The only available IFC-compatible object, Space, has a separate IFC section (behind a button) - but that object seems to have been created with C++. In the leading application, ArchiCAD, there is an IFC tab for all tools/objects.

Yes, one can "Create IFC data", but that is not the same.

----

So far, when one exports as IFC from VW, one cannot import it back: "Bad IFC". Nice!

[Gerard Jonker]

Hi Petri,

What about BIM? Do you expect any solution from that xth attempt to retrieve information from the CAD "models"?

Just trying a new plank,

Gerard

[Petri]

Yes, what about BIM? Do you know what it means?

I don't "expect" anything. IFC is here and, by and large, works. At least in the civilized world.

It may be that you are toying with "models", but we work with models. Why? Well, primarily because the construction industry wants it, as do big clients.

[Petri]

I note that in 24 hours or so, NNA has not been able to provide any information. I gather this means "no".

[Gerard Jonker]

I happen to know what it means, I also think we will be forced by the bigger clients to work BIM compliant.

That, when we hand them the key to their new building with the CD (DVD?) containing the BIM, they will shelf the CD and use a simple PC with and excel spreadsheet to manage the maintenance and use of the building, will be not a topic of discussion at the moment we get the contract.

On the architects side lies a more serious problem.

The problem in the average architects firm is still the devision of responsibilities. The drafts person, responsible for finishing a certain sheet of paper on a certain date will care less about the actual content of the file as his deadline comes closer. Then when the piece of paper is on its way to client or contractor he might be persuaded to check his model. After that the person in charge of pricing etc. gets the file, makes a few attempts to retrieve data electronically, gives up and starts counting by hand. Not quite the way it should be, but unfortunately still practice.

We still have a few years to prepare ourselves, before BIM takes over.

What you describe as "toying" with (3D-) models may be very important to the individual client. The 2D/3D models you are "working" with contain a lot of information, which is a lot of work to set up and, more importantly, to keep accurate over the year(s) the average project lasts. A lot of work we are only prepared to do when that data is actually going to be used. The data is only going to be used if there is a guarantee about its accuracy. People are still waving those responsibilities, and that is problem we (here in Holland anyway) are facing in the years to come.

So BIM or no BIM, the answer is in the attitude and not in a certain format.

15 years ago, my partners in the firm back then, had set up a beautiful system that created complete tables with everything to order for a building, the system was accurate and fast and allowed for very sharp pricing of buildings. It was done in MiniCAD 4.x!!!!

Regards,

Gerard

[Petri]

I happen to know what it means

Excellent!

That, when we hand them the key to their new building with the CD (DVD?) containing the BIM, they will shelf the CD and use a simple PC with and excel spreadsheet to manage the maintenance and use of the building, will be not a topic of discussion at the moment we get the contract.

Dear me. You don't know after all.

Then when the piece of paper is on its way to client or contractor he might be persuaded to check his model.

More evidence on your profound ignorance.

So BIM or no BIM, the answer is in the attitude and not in a certain format.

It is? It isn't? Prey tell me how one exchanges BIM data with clients and consultants without a format capable of that? If I send my idea of columns, beams and slabs to the structural engineer, the engineer needs to (i) know what they are, (ii) change them with whatever program they use and (iii) send the modified model back to me in a format that my program again can use. (Better still, put the modified model into a shared model database.)

I could not care less for irrelevant comments coming from some technological backwater like the Netherlands if it weren't necessary to rectify the disinformation you spread.

(Is it necessary? No, not really: from this on I'll just ignore you, being sure that the members of this forum are intelligent and knowledgeable to do the same without my commentary.)

I'm only interested in finding out whether one's own VS-based plug-in objects can efficiently be exported as IFC objects. Obviously they can't: the silence is deafening.

[Gerard Jonker]

I really wonder how much more information you might need to add to a column or slab (in whatever format) so your engineer might finally understand what you mean? Give the (wo)man some credit.

The decisions are made on the top. (do I get that contract or not.)

Whether or not the format gets to be used to its optimal capabilities is determined further down the line, because people are people.

What rests is, at the very end, the agreed CD with the complete model the client demanded.

What I fear is that the whole BIM idea will turn out to be nothing more than an improved Drawing Exchange Format, a lot of extra bells and whistles, none of which making the exact sound you wish, and a lot of extra work on the plate of the architect.

Ignorantly yours, (or should I write "Ignored-ly")

Gerard

P.S.

This eardrum ripping silence may, of course, also be caused by something other than the (possible lack of) IFC capabilities of VS.

[gScott]

gerard

don't rise to it,

when i saw petri was back i was delighted,

but by his last post to this thread i see

he's back to his usual , and now,

i miss his absence...

me, me, pick me, ignore me too, please!

[Petri]

me, me, pick me, ignore me too, please!

Dear me. Logged in from someone else's machine to verify that NNA has nothing to say and found this.

Fear not, Mr. Scott: you have been on my ignore-list since I established it. I have no interest in reading your stupid comments - you are not even funny in your ignorance as Gerard is.

[Petri]

I think we can now conclude that objects created with VectorScript cannot be IFC-compatible in any meaningful way. If in any way: I still can't import an IFC-file created with VW's IFC-export.

[Gerard Jonker]

Hi Gideon,

Thank you for the warning.

Anybody care to continue the discussion on a more gentle tone and see if we can get to some dept before start calling each other names?

The trap Petri seems to be falling into (in the humblest of my opinions, of course) is that of many computer "literates". (Please don't think I am not aware of his skills and insights.)

They focus too much on creating a model (2D/3D, an information model) of reality in their computer, the need to describe everything. From that point on they want to share that information with other parties. Is it not too much a need for control or too much fear things will go wrong if we delegate some responsibilities?

But the computer never holds everything, no matter how you try. The computer is not the source of the information, the idea ripens and gets shaped inside your brain. Sometimes you need visual clues, sketches, you are at that point communicating with your self.

Only after the point the idea has matured you start to look for media to communicate that idea.

Up to now VectorWorks has proven quite a good tool for communicating, even at sketch level, but we couldn't do all the communication with Vectorworks, sometimes we had to fall back to pencil and paper, fax, spreadsheet- and database programs and don't forget the telephone.

Now BIM is around the corner, it is not the first attempt to:

a: increase the communication abilities of a CAD program

b: channel more information through the CAD program.

I can only be very enthusiast about a, but b worries me a bit:

Who is going to feed that stuff into the computer?

Is it necessary to put it there, i.o.w: will it come out again?

Is a CAD program the best way to communicate that data?

The idea is that when you click on an object, some text pops up, displaying all nicessities (the first i is not a typo) about that object.

Now, if we return to the example of the engineer.

What would be the minimal information we need to send him/her?

What if we send the engineer a 3D model of the whole building.

That would probably enough, unfortunately the engineer has to start looking to what information is his/hers.

If we assign some classes (a standard would be nice) the engineer can extract the construction from the rest of the building, while that the rest of the model immediately gives some idea about use and therefor about loads. (If you are working on an exceptionally small building with an exceptionally heavy floor load you probably already talked about that when had the engineer on the phone.)

Now I can add information to the constructional elements: "This is a concrete column, or this is a steel column" But that is something the engineer can see from the shape of the object.

The next phase is the engineer has made the necessary calculations and sends the model back, the concrete column is still a concrete column, a suggestion is made about different size or materialization, either drawn or not, depending on your contact with the engineer.

When i click on the column now I can see how much steel goes into the column, this is great.

I pick up my MacBook and walk to the site. There I see the column, concrete ready to be poured.

-Hey, wait up! There is not enough steel there, my model say 10 rods, and you are putting in only 8!

-Yes, we ran clean out of ? 10, but we have plenty of ? 12.

-Ai, Ai, 8*12 is not the same as 10*10!

-Agreed, but is falls right in the margins. We phoned the engineer about this, this morning and he faxed a signed approval, I think by tomorrow your model will be updated by his staff, they are changing their models as we speak.

Yes, I can see BIM would work, Yes, I can see it would be great if the information was there.

But the weakest is link is often forgotten. As long as blueprints are used on site, those drawings have to be made, the drafts persons are the weakest link. They have to put in all data, also data that doesn't immediately appear on drawing, and that is where they will cut corners to get the drawing finished in time.

That is how it is how things are going right now and I see no reason why a BIM format would prevent that.

People in Holland hardly used classes in MiniCAD. They didn't need it for their own drawings, relying on an object oriented structure of groups and symbols. Only after Classes could also assign colours and hatches they started to use them.

I have yet to see a similar bate, to lure people in preparing good BIM models.

Just my 2 rather lengthy cts,

I'll try to be briefer next time,

Gerard

[Robert Anderson]

Petri and others, first let me apologize for not answering this sooner. (I don't actively moderate this particular topic in the forum, and for some reason, no one brought this to my attention.)

The short answer is, -any- object, PIO or symbol or whatever, can be made into an IFC object using the VW interface. Select an example of your object, choose 'Create IFC Entity' and decide what IFC object-type and which property sets and data you want your object to have.

But this is using the VW interface. As of right now, there is not a way to do this from within VectorScript. This is an issue I am working on right now. I'll report back when I know more.

[Dieter @ DWorks]

I do not really know what IFC is and stands for, I would like to know more after reading this, so can someone give me a link to a site with more info on IFC?

[Christiaan]

http://www.nemetschek.net/architect/bim.php

[Dieter @ DWorks]

So if I understand it correctly, IFC is actually a way to standardisize the way we draw for better exchange with other programs. We just need to say what kind of object it is and what values the parameters have. Then another program will know that too and can create calculations on this. So if you create your drawing in IFC then you are actually creating it like BIM but with kind of standards. So IFC is a standardisized BIM.

Actually you can compare this to the use of classes (in VW) or the use of layers (in AC). you can use your classes the way you want (BIM) but you can also use them in a standardisized (IFC) 'format' so that everyone knows what's in it and that everyone use the same names.

So everything is just evolving, first the use of classes and now BIM, that's great, I love it.

But there is a small problem with it: 90% of the architects, drawers etc. I know of don't even use classes. so how will they use BIM/IFC?

This standardisation thing is a really great thing, but I think it is first time to learn people how to use the basics (classes and BIM). Most people (not me!) do not see the advantage of using Classe/BIM, so they just not using it.

There is a lot of work to do.

[Petri]

This is an issue I am working on right now. I'll report back when I know more.

Thanks, Robert, you bring good tidings. For me it would be easiest if IFC data would be written with normal database procedures.

Maybe the "basic" IFC assignment could even be a user-definable property (as "default class" is). Not sure of the ramifications, though. The point, if any, is that there may well be situations where a user systematically uses an object for a non-intended purpose.

In fact, this issue relates to many standard tools as well: it should be made easy to map, say, walls as IfcBeams. At least en masse editing of IFC-attributes should be made easy!

Reports should also be able to be generated from IFC data & using that in criteria - but surely you already are working on that, too.

----

Speaking of beams:

http://www.steptools.com/support/stdev_docs/express/ifc/html/t_ifcbeam.html

describes the IfcBeam entity. I believe that even a cursory study of the possible attributes and references will be enlightening to those, who believe that IFC-based BIM is pointless and that everything useful and relevant can be easily processed by engineers etc eyeballing "drawings".

[wezelboy]

I believe that even a cursory study of the possible attributes and references will be enlightening to those, who believe that IFC-based BIM is pointless and that everything useful and relevant can be easily processed by engineers etc eyeballing "drawings".

It helps if they actually look at the drawings.

[Petri]

It helps if they actually look at the drawings.

Well, thank you very much for your irrelevant contribution. Obviously Gerard is not the only one who totally misses the point, so I guess I owe an apology to him.

Now, tell me how the engineer's computer program would "look" at the drawings? Am I missing something? There are computers and programs with full human intellegence and perceptive/interpretative capacity? Dear me - HAL 9000-series computers are here!

[Christiaan]

Yes, you're missing something (not an uncommon occurrence). He's talking about the engineers, not the engineer's computer program.

[Petri]

Oh yes, you are correct! I just didn't think that we would need a highly trained and paid Engineer to interpret drawings in order to distinguish what is a beam and what a column, since IFC can handle these mundane tasks.

My apologies for being so many decades ahead of London UK, Napier NZ, the Netherlands and an unknown country.

Edited July 31, 2007 by Petri

[Christiaan]

Apology accepted.

[wezelboy]

You obviously do not work in building design or your sarcasm gene is heterozygous, as opposed to non-existent- as evidenced by your feeble attempt at it.

Everyone knows that no one actually looks at the drawings!

[VectorGeek]

Petri,

I will remind readers that on May 12th, 2007, you said good-bye (for the 87th time, but who's counting) to the VW world:

------Quote-------

But never mind. I'll just take my playthings to another sandbox. I have other IP-addresses and could circumvent the Gestapo, but I couldn't be bothered. VW will never become useful outside the U.S. of A: the Powers That Be are ignorant and myopic.

So long and have fun with VectorToyWorks.

------Unquote-----

I am left wondering, what keeps bringing you back? Although there are many who would probably wish you would stay away for good, I have seen you come and go since the CompuServe days. Each time you say good-bye, you unleash some wicked hyperbole, stomp out the door and swear that you are done. Within weeks, you slink back in and begin to throw your barbs at other forum members. Many take it personally, I just laugh because like the Line Tool in VW, you will always be there in some form.

Keep up the good work buddy.