Deelip.com

Monday, April 21, 2008

WTF is Digital Prototyping?

That's the question I asked many people at COFES 2008 last weekend. Not exactly the same words, but pretty much something along those lines. I got several replies ranging from decent to colorful, depending upon the level of intoxication of the person replying. Replies like "it's nothing new" and "it's just a marketing thing" to "load of crap" and "complete bullshit". However, the average reply centered around "it's being done for many years now" and I tend to agree with that line of thought. But I know that Autodesk Marketing is the best there is and when they say something, I listen and wonder.

So WTF is Digital Prototyping? Just like PLM means different things to different people, I happen to have a view about Digital Prototyping as well. If you care to know, let's back track a little here. I had a long and animated discussion with Scott Harris, the co-founder of SolidWorks at COFES 2008. Among many things I asked him about the growing functionality being added to SolidWorks every year, and how this could eventually step on the toes of CATIA, if it hasn't already. Scott put his drink down, sat upright and said, "No way! We each cater to a different set of customers who have different needs. We operate completely independently of each other and are free to do what we want."

Later that evening I had another long conversation with Suchit Jain, Vice President of Strategy of SolidWorks, and asked him the same question. His response was more elaborate than's Scott's but not very different.

The next day I was having lunch with Stan Przybylinski, Manager, Market Intelligence of Dassault Systemes and I found myself asking him the same question. I was expecting a similar reply as the other two, but was pleasantly surprised to hear him say, "Yes, we are concerned about this, especially since we are interested in making our solutions more affordable to a larger cross section." So I was right. Dassault sees a conflict of interest here, and rightfully so. In these days when companies are looking to cut costs more than ever, Dassault runs the risk of losing the bottom tier of their customers to SolidWorks. No, not the Airbus type of customers, but the smaller companies which barely manage to afford CATIA and do not need all of its functionality.

So WTF does have to do with Digital Protoptyping? Patience, I am coming to that.

Now taking this argument to the Siemens household, we can see that Solid Edge runs into the same problem with NX. Whether they admit it or not, both SolidWorks and Solid Edge have to internally fight their way up. But there is no such glass ceiling for Inventor, the last of the three in the mid-range CAD market. It can push all it want's to. And push it will. Like I mentioned on WorldCAD Access, Robert "Buzz" Kross, Sr. Vice President, Manufacturing Solutions of Autodesk, the man in charge of Inventor, was very clear when he said that over the next ten years Autodesk intends to acquire companies that will aid in giving Inventor the capabilities of Digital Prototyping.

So again, WTF is Digital Prototyping? My definition of Digital Prototyping is simply "the difference between modeling and designing". Allow me to explain. Modeling is the act of fiddling around with the mouse and keyboard to arrive at a solid model, technically a set of trimmed NURBS surfaces, knitted into a watertight body. Designing, on the other hand, is something totally different. Designing is the act of testing and verifying whether a solid model will fulfill it's intended use, modifying it if it cannot, and if it can, optimizing it to make it do a better job. In designing, you need to apply real world conditions such as loads, heat, stress, etc. and determine whether the part or assembly will work satisfactorily and efficiently. Although the "D" in CAD stands for "Design", we have actually being doing modeling, that is, creating and altering geometry. When you "design" you essentially iterate though many versions of models of the same part or assembly and finalize on the best version which fulfill all your requirements. That, in essence, is "Digital Prototyping" according to me, and maybe according to Autodesk as well. I repeat, "Digital Prototyping is the difference between modeling and designing.

I believe Autodesk's strategy for Inventor actually involves giving Inventor the ability to "design" parts and assemblies, as opposed to just "model" them, which it can do already. And so can SolidWorks, SolidEdge and just about every other software capable of creating a watertight solid model. Modeling is a solved problem. It was solved long ago. The "enhancements" that we see in every new release of SolidWorks, Solid Edge and Inventor and just more elegant methods of doing the same thing - modeling. However, designing is a problem that is far from solved and only the high end CAD systems like CATIA, NX and Pro/Engineer offer solutions close to what is required to design. And for reasons mentioned above, SolidWorks and Solid Edge will always be restricted in offering design solutions.

So when Autodesk says things like "Digital Prototying", "ten years" and "aquire companies", I believe that over time they are looking at offering a solution, or rather a set of solutions, that spans from the mid-range to the high-end CAD market. If that indeed is the case (and it should be if they wish to continue making a lot of money), then Autodesk may feel the need to make some noise about their plans. And I believe that "Digital Prototyping" is that noise. But more than noise, to me it appears to be a firm foundation of a sound business plan to truly rule the CAD world - 2D and 3D.

Now we know that Autodesk has loads of money to acquire companies and make this happen. They are not taking the stairs here, it's the elevator. Buzz was very clear that they are not interested in developing these digitial prototyping solutions on their own. They intend to just go out and buy tried and tested solutions that work (or can be made to work) with Inventor. And when (not if) that happens, the rules of the mid-range CAD market game are going to change.

So while most of the world thinks of Autodesk's Digital Prototyping mantra as a "load of crap" or "complete bullshit", I would suggest that their rivals give them a little more respect and probably do something to about it.

12 Comments:

  • Interesting argument - I hadn't considered it that way. I'm just not sure about the definition of design though. The problem for all CAD is that if you cannot model it you cannot render it/analyse it/test it/create full drawings from it or export it to CAM. Modelling is, intrinsically a part of the design process. Shape design is the single most important issue as far as I'm concerned - and by shape I mean the interactions with assemblies and existing parts.

    In that respect we are a very very long way from having solved modelling. There are some fabulous technologies out there that certain go a long way to making shape design easier - T splines, Haptic feedback modelling system, global shape modelling, but there is no one single solution that wraps all this up neatly into an interface that any designer can access easily without being a full time CAD person.

    Without the ability model anything of any complexity easily no amount of rendering excellence or FEA technology or knowledge based systems are going to help.

    I see what you are saying about Autodesk and I think they will succeed if only by strength of numbers, but the real winner will be the company that gets the interface right that masks the power that can truly model anything easily, quickly and integrate with point cloud data sets, churn out quality drawings and link seamlessly to CAM.

    Maybe it will be Siemens - I see the webcast tomorrow is claiming "product development 100 times faster than ever before" - some claim!

    By Anonymous Anonymous, At 1:20 AM, April 22, 2008  

  • It is interesting to bring out the fundamental differences between Modeling and Designing.

    Designing is the earliest form of Art and Engineering, in which many creative thinkers have come out with different aesthetic and functional forms of solutions.

    For many engineering problems , the selection of proper form to enable the function is what matters.

    I think "modeling" makes emphasis to only "form" and not function. But Design as a whole makes emphasis to both Function and Form.

    There are number of fantastic tools available now to find the pleasing and aesthetic "form".

    Except the FEA for simulating the Functional feedback, we have to go a long way for actually mimicking the "Functional Simulation" i.e, Tools for Design for Assembly, Design for Disassembly, Design for Manufacturing, Design for Quality so on.,

    By Blogger gisku, At 10:07 AM, April 22, 2008  

  • Hi Deelip, thanks for taking the time to post your COFFS reports.

    Question: "Designing is the act of testing and verifying whether a solid model will fulfill it's intended use, modifying it if it cannot,"..etc.

    Is this why 2D is considered by many as 'dead'? All these years we have been slaving over drawing boards, and calcs, creating new goods, products , aircraft, ships, buses and cars all without actually designing; we've just been pushing graphite and ink all over paper and velum.
    If we were not designing then why do we need it now?
    Was the A380 designed and the R100 only drawn? ;-)
    PaulW

    By Anonymous Anonymous, At 11:53 AM, April 22, 2008  

  • PaulW: "If we were not designing then why do we need it now?"

    Because, if you could do FEA with pencil and paper, you probably would. And we could then smash all our computers to dust and live happily ever after in our 2D world. But unfortuntely that isn't the case, now is it?

    By Blogger Deelip Menezes, At 12:00 PM, April 22, 2008  

  • Deelip, what is the answer to the second question; "Was the A380 designed and the R100 only drawn? ;-)"
    PaulW

    By Anonymous Anonymous, At 12:44 PM, April 22, 2008  

  • PaulW: "Was the A380 designed and the R100 only drawn?"

    The R100 was drawn, prototyped, tested, redrawn, re-protytyped, re-tested... and so on till they ended up with a design that worked. They then redrew that working design to optimize it, re-prototyped and re-tested it. Then they optimized it a little more, re-prototyped and re-tested it... I think you get my point.

    Digital Prototyping aims at reducing (and in some cases eliminating) physical prototypes so that we use computers and science as tools so get products to market faster, cheaper and safer.

    I am not sure what mix of science and simulation was used in the A380, but I'm quite sure people would not use paper and pencil to design an airplane these days.

    By Blogger Deelip Menezes, At 1:04 PM, April 22, 2008  

  • I want to add something more to discussion.

    Digital prototyping for Form evaluation is what we have doing with the current available CAD/CAM/CAE tools.

    Using CAE, we evaluate the form with the objective to satisfy the Functional requirements such as Strength (Static and Dynamic)etc.,.

    Using CAM, we are finding the Manufacturing codes so that the form can be machined.

    With the available CAD products, the definition of Computer Aided Designing not yet achieved. The knowledge of designing new products have not been fully decoded into the software so far.

    Digital Prototyping is not a new
    paradigm. For Autodesk, it is just a marketing terminology. We should not dissect it too much.

    By Blogger gisku, At 3:12 PM, April 22, 2008  

  • "The R100 was drawn, prototyped, tested, redrawn, re-prototyped, re-tested....I think you get my point."
    Not so and therein layeth the myth of 'digital prototyping'. Its not that it does not exist in the minds of those that espouse it, it is just those designers of one offs know more about reality and how to achieve it than those who choose to talk about it.
    Only one R100; it was designed/drawn, successfully built using slide rules, pencil and ink drawn documents, tested and flown successfully: it was the prototype and the finished aircraft in one and one of many one off projects CAD vendor's spokespersons and those that want to talk up 'digital prototyping' should bone up on. It is just another way of saying what all designers have said and done thru' the ages, 'I intend to design and build an item. Use a 'drawing' to document it and then build one!'
    'Prototyping', digital or physical, prior to the creation of many large and small projects is neither new nor an important, or required, step for many just as having CAD and FEA are not necessary and do not make a person a designer!
    PaulW

    By Anonymous Anonymous, At 3:12 PM, April 22, 2008  

  • At the end of the day we are talking about the application of tools to enable human creative thinking processes. I do not believe I am a better designer because I use 3D CAD. I think I can have more control over what I design, but there is the "issue" - this control brings with it more responsibility, and until the legislative frameworks are revised to cope with this there will always be demarcation in design function and delivery.

    Ask the architects about this! With the great move to BIM they are having to rethink how they bill for jobs, and how to limit the liabilities when things go wrong.

    How long will it be then, in the Autodesk nirvana of digital design until some designer or company gets sued because something has failed for their lawyer to turn it on its head and say "well actually we are suing Autodesk because our client used an Autodesk "intelligent design system" and that said the product passed all the tests"?

    The simple fact of the matter is that for a lot of product design tasks it is simply FAR more efficient to stop the digital design at a certain stage and switch to physical testing, then revise.

    The huge worry I have with all this is that companies like Autodesk push these things so far and it gets picked up by mainstream media and thereby the customers who genuinely believe the hype and have no concept of the time and associated costs involved in keeping everything digital.

    For huge scale projects like a new airliner series or mass production car yes digital prototyping is essential - and indeed already a reality - but for a typical small scale product?

    The reality is that the designer or project leader needs to decide what tools are appropriate for the development - and this will depend on budget, market, timescale, materials, physical size etc. The more quality tools we have the better for sure, but as they say, a tool is only as good as the person using it!

    By Anonymous Anonymous, At 3:38 PM, April 22, 2008  

  • I think that Deelip and many of you have made some good points and I will be interested to see what ideas Autodesk (and certainly others) have in store for us in 'digital prototyping.' But, it occurs to me that while their is great conceptual distinction between modeling and designing, in practice their are few who only model and never design. The practical distinction lies more in the process and responsibilities, rather than in terms of the modeling tools that you use. In my mind design improvements, have more to do with the interactions people, information, and devices have with that tool and in what designated stages of the production process designing is allowed to occur.

    Many designers have been designing through modeling making iterations based on the compounding effects of changes and through near-instantaneous feedback and analysis. (I do it using BIM as a contractor, but only in the restricted framework limitations of 'means and methods' and not over-stepping into 'design intent'). We tend to take for granted the interactions we have while modeling: as we create something, change it, instantly see the results, and modify it again and again... in the modeling/designing phases. What IS changing today (and where we may be getting confused) is the interactivity with the modeling (e.g. intuitive GUI interfaces, alternative control devices, collaborative visualization and control tools, automated design feedback, etc.) is getting much better and much more interesting: allowing true design and modeling activities in a collaborative and efficient way.

    Autodesk is making leaps and bounds in this arena (among others), just spend some time on YouTube and "research". As Virtual Design & Construction Manager for a mid-sized construction company, I can assure you that these new interfaces along with the underlying backbone of BIM (or SolidWorks, Rhino, Catia, or 3DS for that matter), will result (and ARE resulting) in revolutionary changes to the construction industry. Let there be no doubt: design IS occuring through modeling in MUCH more efficient and effective ways than ever before: I deal with it every day.

    By Blogger Spinnacre, At 2:46 AM, April 23, 2008  

  • Impact of lesser usage of DP on the Indian industries and suggest steps to be taken
    to narrow this gap

    By Blogger Unknown, At 8:39 PM, November 26, 2009  

  • Hi Deelip. Can u tell me what are the debates on digital prototypes and drawbacks of it??

    By Blogger Amit Ahirrao, At 7:20 PM, December 04, 2009  

Post a Comment

Subscribe to Post Comments [Atom]



<< Home