Autocad limitations

[JoDCad2]

I 'd like to pose an open question to all forum members.

I am conducting a survey of Cad software in terms of a university project and I would to ask the forum members who care to reply, in their expertise what deficiencies or limitations would they think Cad software suffers from. I come from a construction industry background.

It could maybe be features that are considered beyond the scope of Cad, or that may not have been implemented well. It could also be some requested features that have not gained enough traction to appear in Cad or some that are only available by user add ons.

Some areas of issue might be the interoperability of files and user interface, or the input mechanisms of the software. For example with multi touch gaining ground, how is the traditional keyboard limiting or complicating the user experience. I would think the user interface would be a major gripe for some people. Here are some points I have come across in my research:

- How suitable is Cad software in the early design stages as opposed to the information production stage?
- How well established is a model based approach and how far has it come in integrating with cost and performance analysis software?
- How well does Cad actually improve the actual design process and opposed to offering merely a set of tools.

If anyone feels they can spare the time, I would really like to hear from you.

[StykFacE]

- How well does Cad actually improve the actual design process and opposed to offering merely a set of tools.

This question is relative. If you compare the drawing boards to a CAD application, then the improvement is a thousand fold. But for other things, such as a 3D BIM job it really is just a "set of tools" and doesn't offer much help when it comes to the actual design process because it cannot possibly know where a duct should route to, or if a wall should go to deck, or where exactly a plumbing fixture will go on a wall.... Hope this helps a tiny bit, since I only answered one small question.

[JoDCad2]

This helps a lot! Thanks for taking the time to reply. Excuse my ignorance, are you referring to some specific piece of BIM software that integrates with CAD or a stand alone BIM application. As far as I have understood BIM does offer some algorithmic design guidelines or maybe I haven't really grasped it.

[JD Mather]

Find the Keynote presentations in the On-line classes here http://au.autodesk.com they will start with the prefix KN
of course Autodesk is only one CAD company, there are others.

You might also check out some of the user presentations.

[Pablo Ferral]

JoCad,

CAD is only a tool. You can create a drawing of anything with a CAD package. From this point of view, there is no feature that is beyond CAD - and no limits.

Without a bit of industry knowledge about how you should describe the item you are drawing, it's possible that your intention will be misinterpreted. Without a bit of industry knowledge about how your item should be manufactured, it's possible that your drawing will be no good to anyone.

Most CAD software OOTB is as generic as a drawing board. It's designed to be this way so that the software producer can sell as many seats as possible. Some software producers (and 3rd party's) sell 'vertical' products that tailor CAD software to a particular industry's needs. Some companies can even afford to employ someone to custom program CAD software to work just the way they want it.

Is this a deficiency, because the CAD software doesn't do what the company wants it to do - or an opportunity to make use of someone else's hard work, rather than have to develop an entire CAD system in house?

Model based design does give us some new tools for examining and coordinating design elements. Ultimately though (particularly in the construction industry) there is no substitute for face to face meetings. The phone, Fax, and E-Mail allow people to hide from confrontation. This is when mistakes are allowed to slip by. Whether its a site meeting to look at and compare some drawings, or a web conference to examine clashes in a BIM model, it is important that people come together to communicate, coordinate and agree - the tools will never do this for us. A BIM model might show a potential clash, but it won't decide what to do about it, and (some would argue - more importantly) it won't decide who's cost it should be, or when the work needs to be done.

On a user basis, the software companies will always struggle. There will never be a consensus on the 'best' form of user interface, or the best input device. Again, is this a limit - or an opportunity for someone to develop niche input devices for specific markets?

I personally don't believe that we will ever have one piece of software that does it all. It would have to be too complicated for it appeal to so many different markets. However Like most people, I would like to see more file compatibility. We are seeing this 'in house' at Autodesk as they work toward making DWG accessible to all it's software. By default many software producers are making their software DWG compatible, in order to sell to companies who need to offer compatible files with the market leader. To go back to my previous point though, just because the information in two files is compatible, it doesn't mean that it is correct - or coordinated.

CAD is only a tool. For the time being at least, the person who is responsible for the output is more important than the tool. It's good to have the right tool for the job, but different tools suit different people and different situations require different tools.

I enjoyed that rant. I don't think I really answered your questions, but maybe your questions are a little generic. Maybe you would get the return you are looking for if you narrowed the scope a little ;-)

- How suitable is Cad software in the early design stages as opposed to the information production stage?

It depends on who is using the CAD software, what there approach is to design and what their expectations are.

- How well established is a model based approach and how far has it come in integrating with cost and performance analysis software?

This will depend on Country, industry, Company and individuals.

- How well does Cad actually improve the actual design process and opposed to offering merely a set of tools.

By what criteria can the design process be improved? Are you talking about making the design process more fun for the designer, more cost effective for the Client, or more useful to the manufacturer?

[MikeScott]

Very subjective questions.. It will depend on who you ask, what applications they use, and what the application is being used for.

I would to ask the forum members who care to reply, in their expertise what deficiencies or limitations would they think Cad software suffers from. I come from a construction industry background.

None except price and compatibility with other software. Everything else, is what it essentially "has to be" to get things done in a manner that is properly controlled by a user. There are a number of products on the market that permit users different control set-ups, or data entry systems. I could 3d scan a product, put it into CAD, and then use a 3d printer to create another one. The sky is the limit.

It could maybe be features that are considered beyond the scope of Cad, or that may not have been implemented well. It could also be some requested features that have not gained enough traction to appear in Cad or some that are only available by user add ons.

There's an awful lot of CAD programs out there to make any kind of generalizations in that manner. If the feature exists, and can be added by the community, that's enough for me. CAD users have a wide range of needs. If someone could fill them all, the program would be unwieldy and cost so much more.

Some areas of issue might be the interoperability of files and user interface, or the input mechanisms of the software. For example with multi touch gaining ground, how is the traditional keyboard limiting or complicating the user experience. I would think the user interface would be a major gripe for some people.

As I noted, the interface options are huge.. you can run multi-touch systems, or you can opt not to. If the keyboard doesn't do it for someone, they can input stuff another way, like through tablets, for example, or through custom LISP files activated by single keystrokes.

Here are some points I have come across in my research:

- How suitable is Cad software in the early design stages as opposed to the information production stage?

Suitable? How would it be unsuitable? unless the object being designed can't be realized through CAD.. like a song or something.

- How well established is a model based approach and how far has it come in integrating with cost and performance analysis software?

If you set-up your drawings to do it, you can get a complete materials listings from it. Performance analysis software would be pretty specific.. and not something I'd have a use for. How established is it? Depends what model-based approach you're referring to, not all CAD systems rely on that.

- How well does Cad actually improve the actual design process and opposed to offering merely a set of tools.

"Improving the design process" is an interesting statement. You can only establish an answer in a comparitive sense, ie- drawing by hand versus drawing by computer. I'm no artist, so there is an improvement in that regard, However, there's an increased usefulness in terms of drawing the item you wish to build, and then going through all of the fabrication steps using the same drawings.

CAD systems are a set of tools, they don't design for you, so the question is kind of moot, when you consider that pens and pencils are also merely a set of tools. Is it more effective than a pen or pencil? absolutely.. is it more effective than a 3d mock-up model sitting on a presentor's desk? Maybe not as much, though you can get a lot farther down the design process without costly prototype costs, when you render in 3d-versus make a physical model, especially if the client wants to see dramatic sweeping changes made.

Perhaps a better idea of how you want to use this information would make it more intuitve to answer. As it stands, it's far too broad and subjective to be able to answer it clearly.

Ie - you want to come up with the next latest great way of doing things? or you want to add to the CAD systems we already use? or you want to evaluate whether CAD is a useful investment for your company, etc..

[JoDCad2]

Thanks guys for the excellent replies, I am reading them very carefully while contemplating my next post.

[JoDCad2]

Very subjective questions.. It will depend on who you ask, what applications they use, and what the application is being used for.

I am focusing on the building construction industry.

None except price and compatibility with other software. Everything else, is what it essentially "has to be" to get things done in a manner that is properly controlled by a user. There are a number of products on the market that permit users different control set-ups, or data entry systems. I could 3d scan a product, put it into CAD, and then use a 3d printer to create another one. The sky is the limit.

That's a very confident statement, could it be perhaps a bit over-confident? Surely there must be some areas of annoyance, some blind spots that have users requiring a better approach.

There's an awful lot of CAD programs out there to make any kind of generalizations in that manner. If the feature exists, and can be added by the community, that's enough for me. CAD users have a wide range of needs. If someone could fill them all, the program would be unwieldy and cost so much more.

I wasn't aware that CAD was so modular, thanks for that info, is the closed source code an issue, or and sdk does away with it I was not aware that even established programs like autocad had that kind of modular community functionality.

Suitable? How would it be unsuitable? unless the object being designed can't be realized through CAD.. like a song or something.

Well, the opinion that it's not very suitable for the draft stages I something a have heard often, that's why I am reporting it. I don't know what the crux of the issue might be, I can make some guesses but that's all.

If you set-up your drawings to do it, you can get a complete materials listings from it. Performance analysis software would be pretty specific.. and not something I'd have a use for. How established is it? Depends what model-based approach you're referring to, not all CAD systems rely on that.

Well I was refering to the construction industry where performance analysis is of course a requirement. I don't know how established they are or if they integrate at all with Cad. Hey do you think you can elaborate a bit more on what CAD systems rely on a model based approach and if this has given them any market leverage?

"Improving the design process" (...) made.

I am not quoting it all for brevity, these are some very good points you are making. I would think 3D Cad modelling would be more effective in jumpstarting a design idea, or inspiring changes to the design.

Perhaps a better idea of how you want to use this information would make it more intuitve to answer. As it stands, it's far too broad and subjective to be able to answer it clearly.
Ie - you want to come up with the next latest great way of doing things? or you want to add to the CAD systems we already use? or you want to evaluate whether CAD is a useful investment for your company, etc..

No I am not looking to come up with the latest and greatest, more of an overview of what required additions might CAD afford, or the road ahead for it. In other subjects I am well versed at I could make a few informed assumptions on where things are heading and how they can improve. Love the Waterboys too btw!

[MikeScott]

That's a very confident statement, could it be perhaps a bit over-confident? Surely there must be some areas of annoyance, some blind spots that have users requiring a better approach.

Not when you talk about CAD systems. Now, if you were referring to a specific CAD application, like JUST AutoCAD, or JUST inventor, or JUST Solidworks, there are issues for each one, individually. However, as a whole, I make that statement using the understanding that any one of those systems merely scratch the surface of what other CAD systems can provide.

How about a laser scanner being used to replicate a 13th century fountain in pieces, throw the details into CAD, and produce molds that were put into a CAM environment and permitted the entire recreation of the fountain. We've already done that.

How about the marriage of CADCAM operations? They already do that too, normally through 3rd party products. As of release 2011, you can even print out your stuff to a 3d printer. (you could already do that, but required a separate program to run the printer).

Well, the opinion that it's not very suitable for the draft stages I something a have heard often, that's why I am reporting it. I don't know what the crux of the issue might be, I can make some guesses but that's all.

Opinions are based on the formation of subjective facts. Some people don't like the fact that you have to learn AutoCAD before you can use it, and it frustrates them. However, if you ask them how else they'd like to get the information into the computer, they have no answer for it. Much like,.. "I wish I could run it with my mind." As long are computers are involved, you're required to move your ideas from your mind, into your computer.

Well I was refering to the construction industry where performance analysis is of course a requirement. I don't know how established they are or if they integrate at all with Cad. Hey do you think you can elaborate a bit more on what CAD systems rely on a model based approach and if this has given them any market leverage?

I can't really elaborate along the lines of what you're asking, except to say I wasn't sure what you were asking.

By model based, I didn't mean as in Model-based design, I meant as in you create 3d items by giving lengths to each side as you make it. Later versions of AutoCAD stuck with that system, but made it progressively easier to adjust something you've already made. In the beginning it was hard to adjust those items, and often easier to rebuild it than it was to make hole position adjustments.

In parametric-based modelling, you have a shape that you can throw dimensions onto and merely adjust the dimensions to adjust the shape, or locations of features. AutoCAD 2010 and 2011, use this system as well as their original system. Solidworks uses this system, as do many CAM systems.

So I was speaking more in terms of the act of drawing/image creation, rather than design testing.

BIM moldeling (Building Information Modelling) (used by Revit) is true model-based design.. closer to the type of thing you're speaking of, if I understand you correctly now. http://usa.autodesk.com/company/buil...experience-bim

No I am not looking to come up with the latest and greatest, more of an overview of what required additions might CAD afford, or the road ahead for it. In other subjects I am well versed at I could make a few informed assumptions on where things are heading and how they can improve.

Well, this makes all the difference in the world, I think, with regards to answering your questions. In my opinion, the future of CAD isn't about making it easier to use at operation level, it's about discovering new ways to make the end client feel as though they truly understand what a company is creating/designing in a quick intuitative manner. I think it'll end up being photorealistic renderings with 3-dimensional walkthrus, and 3d glasses with 3d gloves for "avatar" navigation through the spaces.

Not because it saves paper, but because it's the ultimate expression of an idea, (though I'm sure that "thinking green" will be in the ad campaign for it.)

It's a market-driven entity just like anything else. My father's company got AutoCAD (back in 1985) because his company couldn't compete with their hand drawn lines anymore. CAD raised the bar, and presented inked lines with the precision beyond even the best drafters in the world. As the competition began using it, he was losing jobs to them as a result. So, he did what any other business owner would do.. He went for a system that would do it clearly and effectively, utilizing reasonably-priced availible hardware.

The other benefits of CAD? Most weren't even thought of at the time.. it just drew really well, and replicated easily, meaning you could draw intricate details one time, and apply them wherever needed, even in other projects, rather than knifing and taping vellum sheets together to be overlay copied, assuming you could find that detail you might have made for the Collins building or whatever.

The evolution of going from drawing a single line at a time, to drawing a fully featured wall at a time, based on predetermined specifications is well underway. These become specialty approaches to drawing, and will end-up in "architectural editions" of the primary system. It's nice for that kind of functionality to act as a time saver, but if you're not drawing walls, there' not much point in having it.

These days, we're on the edge of 3d printing, giving people a 3d model with which to "sell" their ideas, or communicate their understanding of someone elses idea. Much like folks have been doing by hand for decades.. Even Darren from "Bewitched," and Mike from "The Brady Bunch" had scale models they built and carried around to meetings with clients. We're already doing it, and have been doing it for years now, but it's becoming a little closer to mainstream now.

Eventually, I'm sure we'll also see things like automated Code checkers that verify a building's drawing is up to applicable codes in the area. This will also be a hot marketable point, and probably not too far into the future.. It's just that there are steps required to get there which won't make much sense until we arrive. Such as room identification according to a set list of occupiable spaces, and item identification.

ie- this is a handrail, it's in the bathroom, the bathroom is within a restaurant, it's made of tubular steel. Then the program will ascertain what the height requirements are and length, if any, and alert you if it finds a code discrepancy, alerting you to file for a code deviance, or correct the item to meet code. It might also enable Codes to be more specific, in terms of assigning load strength requirements and suchlike that could all be checked on-the-fly as the items are drawn and identified.

Heck, at that point the layout of restaurants and suchlike can become automated, and you end-up with intelligent space planner programming, so that if a client wants more tables spaces, it can mathmatically determine some layout options while maintaining the required distances, and using your preferences as a guide, etc.

The neat thing is that with things like parametric modelling, we are well on our way to doing that sort of thing. "Sure.. lemme' change this number right here.. and bingo! it's done! The ceilings are now 10 foot tall, rather than 8.. Now, how does it look to you?" with the convenience of being in the client's office with a wall projector plugged into a laptop that's rendering away.

Love the Waterboys too btw!

I "respect" the waterboys, but no, this is my given name. Oddly, most folks these days thinks I'm "Michael Scott" from "the office", or the pitcher for the Houston Astros, or even the sci-fi author.. I haven't been reminded of the Waterboys in awhile. The secret horrible truth is that I was named after two of the boys in Peter Pan, one as my first name, and another as my middle name. It's excrutiatingly common though.

[nukecad]

- How suitable is Cad software in the early design stages as opposed to the information production stage?

Early design stage - I use my brain and sketch a few things on paper; then when I have tried a few ideas I will use CAD to advance them and present them.

It doesn't matter which CAD, if you are designing something you have to design it - a programme can't do it for you.