The 2D To 3D Bottleneck
At one of the many discussions that I was a part of at COFES 2008, someone started lamenting on how many years ago everyone said that 3D was the next big thing and it would trample 2D out of existence. And yet that has not happenned. Someone else suggested that it would never happen. It is my view that 2D will always have a rightful place in the CAD world. Not as the beginning of the design process, but rather as one of the end products of design. Things like workshop ready drawings, contour maps, etc. will always be in 2D.
But this begs the question. If 3D is "the" way of doing it, then why is so much design still being done in 2D? There are as many answers to this question as there are people trying to find an answer. And here is my own.
First let me ask you a question. What is the first thing you do when you start to model a new part in a 3D parametric solid modeling system like SolidWorks, Inventor or Solid Edge? Correct, you create a sketch - a parametric sketch. You then extrude it and then maybe create another parametric sketch on a face and so on.
So if parametric solid modeling relies on parametric sketches, which are essentially 2D drawings, then doesn't it make sense to add parametric sketching to 2D CAD systems to help people make the move from 2D to 3D. After all, the most popular reason given for not moving from 2D to 3D is the "you-need-to-change-the-way-you-think" excuse, which I believe, is a perfectly valid one.
A perfect example is AutoCAD and Inventor. Clearly Autodesk wants its AutoCAD customers to move to Inventor, but it has not added parametric sketching to AutoCAD in all these years. Why? Because if they do, they will essentially be opening a door for their cutomers to get out. But on the other side, their customers will have a choice of many doors to enter and only one of them will be called Inventor. A pretty nasty Catch 22 situation for them. So it is in Autodesk's best interest to keep AutoCAD users happy with wonderful things like non-rectangular viewports, fancy layer editing tools and pretty interfaces. If they choose to give their customers a helping hand across the 2D to 3D chasm, they run the big risk of losing them altogether.
Autodesk can easily add parametric sketching to AutoCAD. They have had it in Inventor since the start. They change their file format every three years. So they can accomodate just about anything into a AutoCAD DWG file, just like how they encode parametric sketching data in an Inventor IPT file. If there is any company that can help 2D users move to 3D it has to be Autodesk. After all, almost all of 2D is AutoCAD.
So am I blaming Autodesk for the 2D to 3D bottleneck? Hell no! Autodesk is not here to do social service. They are not going to do anything to reduce their customer base. And why should they? It is up to their rivals to find ways of relieving Autodesk of their customers. And it it up to 2D users to decide for themselves if and when they should make the switch.
I understand that the issue is not as simple as adding parametric sketching to AutoCAD. I used it only as an example, as one of many things that will help 2D users make the move to 3D. My point is that 3D CAD vendors need to make it extremely easy for 2D users to move to 3D. And last weekend at COFES 2008, I saw something precisely along those lines.
Not surprisingly it came from SpaceClaim. Roman Walsh from SpaceClaim took the time and showed me SpaceClaim 2008 in the lobby of the Scottsdale Plaza. As I watched him model a differential I was amazed to see that he was literrally "drawing" in 2D just like how you would in AutoCAD and 3D parts were simultaneously being "modeled" in another window. After he was done I asked him to rotate the 2D view, and just as I suspected, the 2D views he was "drawing" into were actually full blown 3D section views of the same model that was being constructed in the 3D window.
I turned to Mike Payne, the CEO of SpaceClaim, sitting across the table and said, "Mike, you guys are going directly after AutoCAD customers with this". He replied, "Naw... now why would we do that?" but could not hide a wicked smile.
I believe the approach of convincing 2D users to switch to parametric solid modeling is not going to work as well as the approach that SpaceClaim has taken, and now Siemens PLM Solutions is taking. As far as I can forsee, my kids are going to learn about parametric solid modeling, but only in their CAD history class. It will be a chapter titled, "The Great Skip" and it will describe how 2D users skipped parametric solid modeling and used direct modeling as their express elevator to the 3D world.
However, the 2 billion dollar question is: What role is Autodesk going to do play in that chapter?
But this begs the question. If 3D is "the" way of doing it, then why is so much design still being done in 2D? There are as many answers to this question as there are people trying to find an answer. And here is my own.
First let me ask you a question. What is the first thing you do when you start to model a new part in a 3D parametric solid modeling system like SolidWorks, Inventor or Solid Edge? Correct, you create a sketch - a parametric sketch. You then extrude it and then maybe create another parametric sketch on a face and so on.
So if parametric solid modeling relies on parametric sketches, which are essentially 2D drawings, then doesn't it make sense to add parametric sketching to 2D CAD systems to help people make the move from 2D to 3D. After all, the most popular reason given for not moving from 2D to 3D is the "you-need-to-change-the-way-you-think" excuse, which I believe, is a perfectly valid one.
A perfect example is AutoCAD and Inventor. Clearly Autodesk wants its AutoCAD customers to move to Inventor, but it has not added parametric sketching to AutoCAD in all these years. Why? Because if they do, they will essentially be opening a door for their cutomers to get out. But on the other side, their customers will have a choice of many doors to enter and only one of them will be called Inventor. A pretty nasty Catch 22 situation for them. So it is in Autodesk's best interest to keep AutoCAD users happy with wonderful things like non-rectangular viewports, fancy layer editing tools and pretty interfaces. If they choose to give their customers a helping hand across the 2D to 3D chasm, they run the big risk of losing them altogether.
Autodesk can easily add parametric sketching to AutoCAD. They have had it in Inventor since the start. They change their file format every three years. So they can accomodate just about anything into a AutoCAD DWG file, just like how they encode parametric sketching data in an Inventor IPT file. If there is any company that can help 2D users move to 3D it has to be Autodesk. After all, almost all of 2D is AutoCAD.
So am I blaming Autodesk for the 2D to 3D bottleneck? Hell no! Autodesk is not here to do social service. They are not going to do anything to reduce their customer base. And why should they? It is up to their rivals to find ways of relieving Autodesk of their customers. And it it up to 2D users to decide for themselves if and when they should make the switch.
I understand that the issue is not as simple as adding parametric sketching to AutoCAD. I used it only as an example, as one of many things that will help 2D users make the move to 3D. My point is that 3D CAD vendors need to make it extremely easy for 2D users to move to 3D. And last weekend at COFES 2008, I saw something precisely along those lines.
Not surprisingly it came from SpaceClaim. Roman Walsh from SpaceClaim took the time and showed me SpaceClaim 2008 in the lobby of the Scottsdale Plaza. As I watched him model a differential I was amazed to see that he was literrally "drawing" in 2D just like how you would in AutoCAD and 3D parts were simultaneously being "modeled" in another window. After he was done I asked him to rotate the 2D view, and just as I suspected, the 2D views he was "drawing" into were actually full blown 3D section views of the same model that was being constructed in the 3D window.
I turned to Mike Payne, the CEO of SpaceClaim, sitting across the table and said, "Mike, you guys are going directly after AutoCAD customers with this". He replied, "Naw... now why would we do that?" but could not hide a wicked smile.
I believe the approach of convincing 2D users to switch to parametric solid modeling is not going to work as well as the approach that SpaceClaim has taken, and now Siemens PLM Solutions is taking. As far as I can forsee, my kids are going to learn about parametric solid modeling, but only in their CAD history class. It will be a chapter titled, "The Great Skip" and it will describe how 2D users skipped parametric solid modeling and used direct modeling as their express elevator to the 3D world.
However, the 2 billion dollar question is: What role is Autodesk going to do play in that chapter?
posted by Deelip Menezes at
8:25 PM
6 Comments:
At the Autodesk Media Day, they told me they plan to enhance 3D in AutoCAD over the next three releases. That would fit in with AutoCAD 2010 scheduled to have a new DWG file format.
Also, the current AutoCAD has two undocumneted system variables that have something to do constraints.
So the tech is there, but not yet activated.
By
Anonymous, At
12:17 AM, April 23, 2008
In my opinion, making AutoCAD do SpaceClaim-like modeling will be akin to making an old dog to do new tricks. It is possible, but I think they will be better off fiddling around with Inventor instead. However going by their Digital Prototyping strategy, they may end up buying a company instead of developing from scratch.
By
Deelip Menezes, At
1:26 AM, April 23, 2008
I've used mostly "ottocr@p" (acad), but to seg users to any product requires seguing in a "proprietary" manner. $co's want to do that, but opensourcers don't need to.
also, your post title suggests an essay about 3d ui. :-)
that's the true bottleneck in all 3d apps.
By
4la, At
2:13 AM, April 23, 2008
Stop lamenting, stop wondering, simple stop talking about and pushing users into 3D; it is not profitable for all and once the one eyed 3D protagonist understand that it is profitability that determines the use of a business tool then maybe we will see some commonsense discussion.
Before anyone chooses to jump down my throat over this issue, I use 3D a lot and have done so for a long time and that is why I can say DON'T push 3D.
Autodesk, and other vendors, are pushing only for reasons of control and their profitability not for the profitability of their customers; this is evidenced by Autodesk total inability to answer any of the questions I have asked them relating to the profitability of using their 3D products, they simple don't know how too.
Your SpaceClaim 2008 example is exactly what Brian Hunter was achieving with his Autoplan/AutoCAD offering in the mid 80's; draw a floor plan, in plan mode, with his Autoplan(AutoCAD) software creating 3D and elevations simultaneously. Just one example of many, the process is hardly new; it was a great achievement at the time, but was only profitable for some and for others they simply used, and still do, his tools to speed up their 2D work.
2D will remain, on this we agree Deelip, but it will remain in all stage of the design and draughting process not just at the end if profitibilty/time to market is the driver.
Parametric or no parametrics is like the 2D-3D argument, a waste of time, they are complementary processes. AutoCAD has long had a vast amount of both non-parametric and parametric functions, many of the later have simply gone un-noticed, a result in part of Autodesk's effort to hide them with marketing bumpf. In some cases parametric functions are being discarded, disadvantaging those who have adopted their use.
It is the customer/users who can best decided when and if they should stay in 2D, move to 3D or use a combination of both and should be left to do just that; a RELEVANT CAD vendor is one that provides tools to allow customers to make these choices whilst using a single design/draughting product. It is possible, and they do exist.....That's is what the discussion should be about!
PaulW
By
Anonymous, At
5:48 AM, April 23, 2008
ACAD has had parametric abilities since 1992. First as the Designer add-on package for R12, and continuing as the Mechanical Desktop vertical packages which also include Vanilla AutoCAD. 16 years of old news.
By
Anonymous, At
10:34 PM, April 23, 2008
I disagree with your prophesy of the "great skip". If this technology is vastly superior to the parametric approach, you will see existing parametric users flocking to it because these are typically the forward thinkers.
Those who haven't made any effort to switch to parametric 3D because of laziness or fear of the unknown are also the types who will not appreciate this kind of technology. You can't teach an old dog new tricks, but when the trick happens by itself automatically when the old dog was just going about its business, the old dog won't appreciate it much.
By
djohnson1976, At
4:01 AM, November 15, 2008
Post a Comment
Subscribe to Post Comments [Atom]
<< Home