Jump to content
  • 0
Sign in to follow this  
Jershaun

Clash Detections

Question

It would be nice if Vectorworks had a "clash detection" of 2 or more objects conflicting with each other. IOW they're occupying the same place in space or overlapping each other.

Share this post


Link to post

25 answers to this question

Recommended Posts

  • 0

Shaun, i know what you mean.

Being told by VW "there's a screw loose in your design! We don't know which screw it is, but check each of the 363 elements that make up this Symbol and edit it accordingly!" really doesn't help a great deal.

Any form of pointer towards the rogue element would be a nice touch. But then i've been asking for a little more info from the dialogues when this happens for some time.

Got to admit though, since the advent of 2009, the annoying little unhelpful comments of VW have been reduced to an almost acceptable minimum . . . .

"All Hail Parasolids!"

Just in case nobody out there noticed, this is well deserved praise to the "Naughts and Ones Boys" at NNA. Even if there are still a few kinks to sort out.

:)

Dexie, maybe a "Find Duplicates" would be more useful, then you decide weather to delete or not once you see what is highlighted.

Share this post


Link to post
  • 0

The problem is that everything is an object of one sort or another. And all those Handles can be composed into infinite groupings which then describe object Types.

If two vertices overlap is that sufficient criteria for deletion.. or two lines ... or nodes ?

Global deletion is a very dangerous procedure ... especially within hidden Groups, Layers & Classes.

Try writing the If>Then statement for this simple procedure :

"Computer find every duplicate and then 'delete' "

Share this post


Link to post
  • 0

If two vertices overlap is that sufficient criteria for deletion.. or two lines ... or nodes ?

Global deletion is a very dangerous procedure ... especially within hidden Groups, Layers & Classes.

Edward, i understand and agree with what you say, but not being a code wiz could never find out what writing it would do, but it does make sense. The "2 overlaping vertices" was often the cause of VW's refusal to accept certain commands to complete an operation . . . Adding or subtracting solids and it can take days to find which vertices are to blame. If code has been written to ensure the operation is not completed within these criteria then surely it must be possible to write code that can distinguish what elements are causing the discrepancy and report the "Law Breakers" to the user, or are we entering into another dimension here?

But, like i said, with VW2009 a lot of these problems have been smoothed out.

It still happens, but not as much.

The "Delete Copies/Duplicates/Anything" from a single command is definitely a real bag of worms. Probably a disaster waiting to happen and much too easy to trigger. A bit like the "Purge Unused Objects . . ." command!! Scares the pooh out of me. :)

Share this post


Link to post
  • 0

An example of what I mean is:

One has beams (concrete or steel) supporting a roof or ceiling. Air-conditioning ducts flowing horizontally and vertically to those beams. Now sometimes these beams and ducts will clash because of varing heights.

Another example (civil):

One has pipework (sometimes above and below ground) that might clash with each other.

What I'm asking for isn't for primary objects but for 3D objects that are sharing the same space in 3D, iow clashing with each other.

Share this post


Link to post
  • 0

Shaun, i still like the idea, i'm just not sure how long it might take to persuade "the powers that be" that it is an essential requirement . . . . .

no matter how many of us ask for a similar addition to our kit.

it does make sense though.

Edited by AndiACD

Share this post


Link to post
  • 0

This requirement can be dealt with eg. Solibri Model Checker, using IFC-data. See

http://www.solibri.com/

for details. But obviously there are alternatives, since they say that ?A lot of tools can do interference detection, but a lot of them are just comparing geometry to geometry.?

Share this post


Link to post
  • 0

From what I read in their website, collision detection in Archicad seems to work only for the MEP modeler extension which would require an additional purchase.

Share this post


Link to post
  • 0

i don't like the idea of a separate stand alone program like the MEP modeler as i feel this defeats the object of the request. 3D modeling animation Apps have this as a default part of the kit now, would it be so difficult to add this ability to VW?

What ever form the resulting solution takes, Plug-In or "Part and Parcel of", it's an essential part of a CAD tool.

Share this post


Link to post
  • 0
i don't like the idea of a separate stand alone program like the MEP modeler as i feel this defeats the object of the request.

That depends. In a normal architectural project there are several design disciplines involved, using a number of different programs. It is much easier, in my opinion, if all parties provide their model in a single format and the clash detection is carried out with a (truly capable) separate program.

The likelihood of (problem-causing) clashes in one's own design is, I think, relatively small, and can be further reduced by appropriate measures and use of intelligent objects (if available in your country.)

Share this post


Link to post
  • 0

Petri, the fact that we obviously work in areas with different disciplines would back up your needs in your field and normally i would prefer a solution that doesn't increase the strain on the VW core by working in a separate stand alone App, but clash detection on a rudimentary level within VW would, i suggest, in most cases solve some basic problems that are inherent in VW and have been so for a long time.

Mechanical Engineering also needs clash detection for various reasons that may differ from those in Architecture, but the results they are needed for, though different, are aimed at a similar end.

If NNA do put their minds to building a stand alone product for this purpose, i just hope they make separate specialised versions for Engineering and Architecture with dedicated tools.

Share this post


Link to post
  • 0

I do not disagree! The question in my mind is whether there is sufficient demand for this functionality inside VW, to pay for it. We do not want another half-baked module, do we?

Share this post


Link to post
  • 0

Too right! Half baked products often cause too many calamities! Modules certainly aren't a bad thing . . . . . if done correctly. Look at C4D.

Maybe a tool could be written as a "work in progress" a sort of "beta module under scrutiny" and if they don't make a complete b@@ls up of it, after a few years trial for us to see if NNA are any good at it, they could offer us the option of having it included in VW or an external module depending on how good it is. Or they could talk to some third party about the option of taking on the task.

Fat Chance! :)

i have to say, if they could create a tool like this for Machine Design that REALLY did it's job properly at a sensible price i'd love to have it in my arsenal.

A Stress Assessment Tool would be really nice too . . . . . . .

A home in the Bahamas . . . .

win the lottery . . . . .

a country retreat . . . . .

Share this post


Link to post
  • 0

Ray, that's what i've been getting at for ages. Years in fact.

VW, now more than ever needs a "Ground Up Rebuild" so that it CAN make use of technology that is now available.

Multi core processors, 64bit architecture, Cocoa savvy so that all of our machine's RAM can get used to it's full potential. i've got 4.5GB of RAM, but VW never uses more than 1.5 to 2GB tops with BIG files. Less than 0.5 goes to running the show and incidentals and the other 2Gig just sits around drinking coffee reading the Radio Times. til i open up another App or 10 . . . . . Oh yeah, there is the tiny little matter of 35Gig of VM being consumed.

Biplap, please get rid of the 28 year old code and write some new stuff so we can all benefit from the machines we now have.

Please keep the GUI as it is though. :)

Share this post


Link to post
  • 0

AndiACD, I agree with you. To add what you said, Hyperthreading came out in 2003 (correct me if I'm wrong) and since then CPU companies haven't looked back and forged ahead with multicores. The days of single core CPU's are over and Nemetschek has to stop hoping that one day it's going to come back.

In another thread someone asked why one would need 32 gig's of ram. I think this is an ignorant question. [sorry :( ] If one browse the net, one will see that the more ram one has on one's machine, the better. eg. when using Maxwell, one can model everthing like trees (with each leaf modelled), each strand of grass, humans with real hair, every bolt & nut so to speak, etc. the limitation is only your ram and operating system. Nemetschek needs to embrace 64bit and use this technology to make Vectorworks better. In 1995 average ram was 16MB. One would have asked then, why would we need 2gig's of ram but here we are today with more than 2gig's. In another 10 years we'll have 32gig's as the norm.

This is what frustrates me with Nemetschek and Vectorworks. They wait a long time to use the latest technology and when they're ready to use it, that technology is old. Look at Maxon, when 64bit cpu's and operating systems were announced, they were the first among 3D Animation packages to release their version of 64bit. Look at Graphisoft, they already have multicore support and according to Architosh: "Graphisoft multi-thread parts of ArchiCAD 12 with great performance results".

Some people here hate rendering in an external package (eg. C4D) and they say the best is to have Renderworks built into Vectorworks (not my opinion BTW), however some people (same or not) are now suggesting an external package for collision detection. I'm very confused about that. Vectorworks at present is already broken up into respective modules and I suggest not another module but to add features into the respective modules. eg. Collision Detection should be in the Mechanical module, if not, then put it into the full Design package only - that way I'll have an excuse to upgrade from Architect to the Full Design package ;) .

I appreciate all the hard work done by Nemetshek coders and I know it's extremely difficult therefore I hope the above provides some contructive suggestions and I hope Nemetshek make huge leaps forward.

Thanks,

Share this post


Link to post
  • 0

Shaun,

Comparing rendering and collision detection is, I think, misleading. The former can be ?good enough?, the latter can't. For the last week or so I have struggled (once again) with VW's terrain modeling and that is a much better point of reference: what use is a DTM or collision detection that is plain wrong?

It appears that, under the circumstances, NNA cannot produce professional quality add-on or additional functionality. I believe that they have the skills, but not the resources.

Share this post


Link to post
  • 0

Shaun and Petri, this is all down to the "Bucket of Bolts" Factor. NNA keep on throwing more "Odds and Sods' into VW and the bucket is getting much to heavy for VW to support it all.

i have to admit, i am a bit of an "If it ain't broke, don't fix it" type, but VW IS "Broke" and the rewelds and riveted patches aren't going to hold water for much longer. Parasolids technology is what VW needs to add a bit of Fiz to it, unfortunately all VW's wires are frayed so we are unlikely to ever see much of what it is capable of doing.

Rebuild! Rebuild! Rebuild!

Share this post


Link to post
  • 0

I don't believe it is necessary to rewrite everything again at one go. If I understand correctly, the advantage of object-orientated programming is that replacing objects with better ones is relatively easy (compared to old techniques.)

The problem, I think, is that many of these ?Odds and Sods? are more ?features? than useful functions. The reason for this is that the expectations of existing users and the criteria of prospective ones focus mainly, if not solely, on price.

Unfortunately many people think that if they need to spend even nearly as much money in VW than in competition, they'll be better off buying AutoCAD, Revit or whatever.

Share this post


Link to post
  • 0

CAD features are like soap bubbles ... they inflate until they break !

Thin event horizons exist between the product costs, learning curves, functionality, productivity, scaleability, sale-ability, and ultimate utility.

As long as Users are satisfied with the over all CAD experience ( their expectations are reinforced ), the software will succeed in the market.

Clash Detection is a worthy objective requiring implementation within a structured limited modular environment.

Share this post


Link to post
  • 0

Correct me if I'm wrong but didn't VW do a "ground up rebuild" around eight years ago with version 9 when it switched to a floating point database?

Ray, that's what i've been getting at for ages. Years in fact.

VW, now more than ever needs a "Ground Up Rebuild" so that it CAN make use of technology that is now available.

Multi core processors, 64bit architecture, Cocoa savvy so that all of our machine's RAM can get used to it's full potential. i've got 4.5GB of RAM, but VW never uses more than 1.5 to 2GB tops with BIG files. Less than 0.5 goes to running the show and incidentals and the other 2Gig just sits around drinking coffee reading the Radio Times. til i open up another App or 10 . . . . . Oh yeah, there is the tiny little matter of 35Gig of VM being consumed.

Biplap, please get rid of the 28 year old code and write some new stuff so we can all benefit from the machines we now have.

Please keep the GUI as it is though. :)

Share this post


Link to post
  • 0

Ariel, if you take a look at the link Ray posted above in the interview with Biplap of NNA whilst you are probably correct that VW9 was a a big hike for VW, in the interview he states that some of the code is almost 28 years old.

In another statement by another NNA staff member it is stipulated quite clearly that multi processor machines would not get much attention from the programmers because the "Other" processors only carries out menial tasks and so has no importance to NNA's planing. This is probably because their present configuration cannot take advantage of the "other" processors, not because the "multi" set up can't deal with the extra load.

Technology has changed incredibly since VW9 and VW has not yet caught up with it and as Shaun mentioned, Maxon makes a clean sweep of everything that needs it in C4D on a regular basis just to keep up with the machines it's designed to run on and it performs very complicated 3D solids computation in a relative blink of an eye because it squeezes every last drop of sweat out of the RAM, CPU's, VM, architecture and Grafix cards that it is able to get from them.

i'm afraid it is time they took another shot at a new rebuild using Parasolids tech as the nucleus around which to construct it . . . . Big Oaks + little acorns and all that stuff. :)

Share this post


Link to post
  • 0

Thank you for the clarification, AndiACD. I think you're making an unfair comparison since C4D isn't exactly a CAD application where accuracy is a high priority.

Considering that Minicad 1 was released in 1985, I believe 28 years is a bit off the mark though.

Share this post


Link to post
  • 0

Ariel, sorry, i think i may have lost the 28 reference when i was chasing the "New Tech" info accredited to Dr. B Sakar which i can no longer find(unless it's been edited) but i know it was there . . . . somewhere. i should have copied and pasted it when i found it.

'85 was it's release, but MC wasn't born over night. . . .

Please don't get me wrong, i love using VW and what i can do with it, but it often leaves a little to be desired. i do realise that number crunching puts a strain on any ones code skills, but that's what these new hard and soft technologies are there for, to improve results. Modifying code to suit may allow a stable, known, existing Apps to adapt to what's out there but it does cause it to trip over it's own boot laces eventually, 'sides, you can't put diesel in al petrol engine and expect it to go very far.

Share this post


Link to post

Join the conversation

You can post now and register later. If you have an account, sign in now to post with your account.
Note: Your post will require moderator approval before it will be visible.

Guest
Answer this question...

×   Pasted as rich text.   Restore formatting

  Only 75 emoji are allowed.

×   Your link has been automatically embedded.   Display as a link instead

×   Your previous content has been restored.   Clear editor

×   You cannot paste images directly. Upload or insert images from URL.

Sign in to follow this  

 

7150 Riverwood Drive, Columbia, Maryland 21046, USA   |   Contact Us:   410-290-5114

 

© 2018 Vectorworks, Inc. All Rights Reserved. Vectorworks, Inc. is part of the Nemetschek Group.

×
×
  • Create New...