Inventor Fusion and SolidWorks Confusion
The Lonely Engineer blog is dead. I have to admit, I never really understood what it was all about. The subtitle of the video blog read "One engineer's adventures with direct and parametric CAD", although I don't remember Roger ever mentioning anything about direct and parametric CAD. Too bad I cannot cross check because the all posts, except the last one, have been deleted. The last post, dated 19th May, is basically a suicide note (in terms of blogging) and a pointer to a new video of Inventor Fusion.
At COFES 2009, Buzz Kross, Senior Vice President of the Manufacturing Solutions Division at Autodesk, let me know that they would be releasing Inventor Fusion on Autodesk Labs in May ("Inventor Fusion is Coming"). Maybe the death of the Lonely Engineer blog is timed to match the release of Inventor Fusion. Lets see. But for now, I suggest that you see the video.
I wonder if Inventor Fusion will be the final straw for SolidWorks. Probably something that will finally make big brother Dassault let them use their V6 direct editing technology. It is an open secret that Dassault does not appreciate SolidWorks eating into it's market. That probably explains why direct editing technology, of the kind that we see in V6, has not yet found its way into SolidWorks, after all this time. Heck, Dassault still makes SolidWorks use the Parasolid modeling kernel by Siemens to do its basic modeling, thereby lining the pockets of competitor Siemens.
For its part, Siemens has made it quite clear that Synchronous Technology is for their use only and will not be licensed to competitors. So obviously, if SolidWorks is to move beyond Instant3D, it would have to either borrow/license direct editing technology from Dassault or someone else or develop its own.
So if I have understood this correctly, Dassault prefers that SolidWorks (a company that it owns) pays hefty royalties to Siemens (its competitor) for every license and subscription of SolidWorks that it sells, as opposed to earning royalties from SolidWorks (a company that it owns) for letting SolidWorks use its V6 technology. I am sure there is a very good reason for this kind of thinking, but it is certainly well beyond the limits of my comprehension.
Something else strikes me. Alibre has been constantly comparing itself with SolidWorks, more in terms of functionality and less in terms of price. I can now see the similarity between the two on another level. In the part of the mid range CAD market that I track, Alibre and SolidWorks seem to be the only couple that I consider to be still stuck in the past.
At COFES 2009, Buzz Kross, Senior Vice President of the Manufacturing Solutions Division at Autodesk, let me know that they would be releasing Inventor Fusion on Autodesk Labs in May ("Inventor Fusion is Coming"). Maybe the death of the Lonely Engineer blog is timed to match the release of Inventor Fusion. Lets see. But for now, I suggest that you see the video.
I wonder if Inventor Fusion will be the final straw for SolidWorks. Probably something that will finally make big brother Dassault let them use their V6 direct editing technology. It is an open secret that Dassault does not appreciate SolidWorks eating into it's market. That probably explains why direct editing technology, of the kind that we see in V6, has not yet found its way into SolidWorks, after all this time. Heck, Dassault still makes SolidWorks use the Parasolid modeling kernel by Siemens to do its basic modeling, thereby lining the pockets of competitor Siemens.
For its part, Siemens has made it quite clear that Synchronous Technology is for their use only and will not be licensed to competitors. So obviously, if SolidWorks is to move beyond Instant3D, it would have to either borrow/license direct editing technology from Dassault or someone else or develop its own.
So if I have understood this correctly, Dassault prefers that SolidWorks (a company that it owns) pays hefty royalties to Siemens (its competitor) for every license and subscription of SolidWorks that it sells, as opposed to earning royalties from SolidWorks (a company that it owns) for letting SolidWorks use its V6 technology. I am sure there is a very good reason for this kind of thinking, but it is certainly well beyond the limits of my comprehension.
Something else strikes me. Alibre has been constantly comparing itself with SolidWorks, more in terms of functionality and less in terms of price. I can now see the similarity between the two on another level. In the part of the mid range CAD market that I track, Alibre and SolidWorks seem to be the only couple that I consider to be still stuck in the past.
posted by Deelip Menezes at
8:17 PM
21 Comments:
Deelip,
If by 'stuck in the past' you mean still using the history based modeling scenario instead of direct modeling, I thought that discussion was done. Synchronous Technology turned out to be overblown. Co Create has had decades to make some headway, and have yet to look intimidating.
There has been room enough in this market for both history and direct modeling scenarios for a long time, and I expect there will be for a long time to come. CAD manufacturers have shown that they are willing to make the leap, but the other half of the market (customers) are not showing that propensity in significant numbers yet.
People who do real modeling work are staying with history based modeling. I can see where the fringes of the market that really don't need the power of history might instead opt for the simplicity of direct modeling, but if you've looked closely at it, the direct editing thing is not in truth all that simple. The rules for making smart selections are complex, and you've got to understand b-rep modeling to make it work.
I've used a couple packages, and remain unconvinced. There are magical spots, but there are also many ugly spots.
By
Matt Lombard, At
9:58 PM, May 22, 2009
I kind of liked the "Lonely Engineer" videos. I don't know if these new videos that are 2/3's entertainment and 1/3 marketing are good or bad but I do enjoy them. I even posted about them in my personal blog last weekend.
http://virtualvector.com/archives/329
By
Mark Burhop, At
10:01 PM, May 22, 2009
Direct modeling is absolutely not a replacement for history-based modeling. For example, I think SolidWorks has done a great job bringing the interaction of direct modeling to history-based modeling with Instant3D. Direct modeling simply lets more people be successful with 3D.
By
Blake Courter, At
12:20 AM, May 23, 2009
I think some CAD company should come with something completely new and innovative. The computer technology and interface technology aren't the same since the developed pro/e autocad, solidworks. There even completely different. But those programs relay very much on there basis, that was developed in those times. I am waiting for a CAD company that is willing to throw everything out and start new and fresh. It should be open it should be customizable, it should be working with the Internet. Everyone should be able to use it everywhere and with everyone. It really should be free, yes free. Product company's will pay to have adds in the library with direct links to there 3D models. Milling company's will pay for the adds on the CNC add-on. In assembly mode bolts and nuts suppliers will have adds in the library.
Hello CAD company's the internet generation is here! , viva la revolution :P
By
Marijn, At
12:54 AM, May 23, 2009
I think direct modeling is a nice tool to have, but I do not think that history-based modeling is going anywhere soon.
Also, SolidWorks showed some enhancements to the Move Face tool using Instant3D to enable more "direct editing"-esque capability at SolidWorks World 2009 in Orlando. It does add a feature to the tree; but personally, that's the way I prefer it.
Btw, I still see all of the Lonely Engineer posts in my feed reader...
By
Brian McElyea, At
1:01 AM, May 23, 2009
I’m with Matt as far as saying ‘that discussion was done’; but may differ in saying it was never needed. Sooner or later there is going to be a realization ALL the tools – 2D & 3D - we have seen developed, and used by many of us, over the past 38+ years have a place in the daily lives of designers and draughtspersons.
The problem, we users have created - and wrongly assisted developers with - is to align ourselves with one method or another based on our own requirements or what we ‘believe/perceive’ is the ‘correct/best’ way to create and or document a design.
Arguing one method of modeling/draughting over another is counterproductive and at odds with what we do! We design/re-design and document new and existing products and processes: to do so in the most productive manner we NEED to have, and be able to use an entire suite of skills/tools, whenever appropriate. We NEED a tool, or a seamless set of tools, that allow sketching, 2D draughting and 3D modeling, both ‘history and non-history’ etc, used in any order, NOT ONE OR THE OTHER.
Autodesk were always - and still are - the closest to this mark; but even they cannot see what is right in front of their faces; they actively pit their products against one another instead of having the commonsense to fully integrate so as to allow their customers a more sensible, and cost effective, path of selection.
Marijn1 says, “ I am waiting for a CAD company that is willing to throw everything out and start new and fresh. “ . Maybe that’s a solution, maybe not necessary if only we customers can put aside our ‘personal requirements’ and push the existing developers for tools that are more appropriate for the ‘industry/skill/craft’ of design and draughting instead of 2D or 3D and History or Non-History etc.
I see more of a problem here with ‘customers’. By polarizing our arguments around particular products and methods, based on ‘our’ requirements not the requirements of our wider industry; we allow developers marketeers to determine the path of CAD software development.
If we continue to argue ‘one method over another’ we continue to let the developers ‘off the hook’, with the result we will never have good, fundamental/base, CAD tools that can be widely applied, across the many disciplines that require these facilities. We will continue to waste huge amounts of time and money juggling new designs and documents through many disparate tools.
By
R. Paul Waddington, At
8:14 AM, May 23, 2009
I enjoyed your post. I foind it interesting that you referenced V6 a few times. To me, it hasn't earned any street credit. I work with CATIA users all the time. Not one is considering V6??
Also, I agree with others, I don't see direct modeling to replace history based. I think a hybrid of the two will win over users.
I think Fusion looks really promising. Inventor is coming along greatly. But the fact is, SW is still in the lead on workflow, ease of use.
If you want my opinion on who has the pieces given the correct direction, keep your eye on NX.
Derrek Cooper
By
Derrek Cooper, At
8:44 AM, May 23, 2009
Matt Lombard: "People who do real modeling work are staying with history based modeling."
Matt, that's quite a statement and makes me wonder if you realize its implications. Basically, you are saying that everyone who ever used a software like IronCAD never did anything serious enough to be considered "real" modeling. This images at http://www.ironcad.com/gallery/ look like pretty serious stuff to me. I believe some, if not all, those models ended up being "real" physical objects.
I am curious to know what you mean by "real" modeling.
By
Deelip Menezes, At
10:57 AM, May 23, 2009
I have to completely agree with Matt on this one. Inventor Infusion...anybody notice the ugly history that was created in the video? I did. A 5 feature part that took Infusion over 20 features to create. Have fun trying to find the features when the "Poor Direct Editing" doesn't work.
Its all a smoke screen when someone talks about the "Engine" of the CAD tool. Who cares if SolidWorks pays royalties. For the majority of the end users it is not about the engine. Ask around...it is about the interface or what some of us like to call the "Body". SolidWorks has in my opinion the Lamborghini body when it comes to the interface.
I get a good chuckle when folks start to show the love they have for Autodesk and how powerful they are. The ones who use 2D can stay 2D as far as I am concerned. It leaves more opportunity for those of us who appreciate 3D and have made the leap. I feel I need to mention something about Inventor. I cant tell you how many times I have helped companies that have purchased Inventor get SolidWorks up and running a few weeks later when they realized they purchase a "Yugo" and would like to drive around in a Lamborghini instead.
I also get a good chuckle when the SolidWorks bashing starts. Ever notice that most of the 3D tools on the market are starting to look like SolidWorks. I am ready for SolidWorks to start taking action on the patent they have for the Feature Manager...ever wonder what would happen to the "look alikes" if they did? Sorry to burst anyone's bubble but yes…it was SolidWorks who came up with the easy way to put features in a manager for ease of editing. If you are a SolidWorks user you know what I am talking about. We have been able to edit features easily in a manager for almost 15 years now.
When all is said and done and I go home for the end of the day…I am glad that I use SolidWorks. They found a way for me to easily work with history and take advantage of some "Direct Editing". I thought it was great to hear that Inventor Infusion has just now announced that they have this capability. Old news to me since I have used it for quite some time in SolidWorks.
By
About Me, At
7:16 PM, May 23, 2009
Deelip, you make an interesting point about v6 technology. The problem is that CATIA itself licenses from D-Cubed, which itself is part of Siemens. Also, the SolidWorks code base is what, 15 years old now? Well overdue for a rewrite from the ground up. As another said, CATIA v6 is not exactly flying off the shelves. Am I the only one who thinks it is all a bit gimmicky? Do I really want a video game like interface to my day in day out CAD system?
It would be a huge undertaking to move from Parasolid to CATIA. Not something that could be done easily. How do you handle legacy files? Would the development costs of moving to CATIA kernel outweigh the license costs to Siemens?
If there is one thing I have learned from SolidWorks is that they always have something on the backburner. Nobody can really say what any of the CAD companies have up their sleeve. As users we just hope we have chosen the right system to run with.
The only benefit I can see for Dassault to force SolidWorks to switch kernels would be to use SolidWorks as a way to encourage the CATIA user base to make the move to v6. Perhaps if that did happen CATIA would be restructured with SolidWorks becoming the core modeller module for CATIA? Who knows? Not me.
By
Kevin Quigley, At
8:13 PM, May 23, 2009
Kevin brings up a good point:"Also, the SolidWorks code base is what, 15 years old now? Well overdue for a rewrite from the ground up."
Bloated. SolidWorks has to add and remove functions in Service Packs now, for example, Display States Linked to Configurations, because, in my opinion, bloated and conflicting existing code.
This I believe is the elephant in the room: The cost to rewrite the SolidWorks code. It must be a huge cost. I believe they're looking very hard at this issue.
Devon
By
Anonymous, At
10:57 PM, May 23, 2009
> "Dassault prefers that SolidWorks... pays hefty royalties to Siemens.... I am sure there is a very good reason for this kind of thinking, but it is certainly well beyond the limits of my comprehension."
There are very good reasons. I don't think they're beyond your comprehension -- but they may be beyond your experience.
Talk to someone who has done kernel-level interoperability work, and ask them how easy it is to replicate the weirdnessess of one kernel in another.
The issues of surface parameterization and modeling tolerance alone are enough to make most developers run away screaming.
Dassault owns both the ACIS and CATIA modeling kernels. Either one could be "bolted in." Neither one would provide support for legacy SolidWorks models.
My sense is that the SolidWorks people are pretty smart -- but this is a pretty "core" issue.
By
Evan Yares, At
8:26 AM, May 24, 2009
I just spoke with the engineers that toiled for years in 2D CADAM and the like to produce their designs. They told me that they feel much sympathy for today's engineers who still don't have 'direct' modeling technology and are forced to use a history based parametric solid modeling tool.
By
Jeff Cope, At
8:59 AM, May 24, 2009
I feel the reason the history based tools are not going anywhere is because the engineering community is too lazy to learn. Refering to previous comments of "viva la revolution" I agree and the revolution is Inventor Fusion. Ditch the history base invest in some training and make the move. Don't pretend we know it all just because we have been in the field for 12 years. I learn everyday. And my best move ever was diving head first into Inventor software with some training and having them come to our office and do some consulting. I don't know who your VAR's are but I know mine was a huge help.
By
cbrink, At
2:44 AM, May 27, 2009
Deelip,
You have been over the top on Direct Editing since SpaceClaim hit the market. I recall your prediction that SpaceClaim would get snapped up by someone for Billions! What happened to that?
Greg Milliken responded to your original hyperbole about synchronous technology in your May 1, 2008 blog (http://www.deelip.com/2008/05/alibre-reacts-to-synchonous-technology.html), in which he stated that Alibre already has direct editing and the non-history based editing is simply a feature, not the basis for a product. Time has proven Greg's comments to be correct.
Why should Alibre, or SolidWorks for that matter, waste development resources on features that their customers are not asking for? Hopefully, we are not expected to write code solely based on what our competitors do and the opinions of industry pundits.
Paul Grayson
Founder and CEO
Alibre, Inc.
By
Unknown, At
4:36 AM, May 27, 2009
Paul,
I guess SpaceClaim did not (and probably will never) get bought because companies decided to "make" instead of "buy". Maybe the "billions" were the problem. I may have been wrong about other CAD vendors buying them or their technology. But I was not wrong about them adopting the direct editing technology approach in their products. Siemens, and now Autodesk, is proof of that.
As regards Alibre already having direct editing capability, I am pretty sure both of us know that this is not the kind of capability that we are talking of here. Adding a move/offset face feature to the feature tree for every direct operation, is not what is being discussed.
I completely agree that a company should do exactly what their customers are asking them to do and what they feel is right. And if your customers are not asking for direct editing of the kind that we are talking here, you should definitely not go down that road.
My opinions are based on my sense what the future may look like, which seems to converge with the majority of CAD software vendors. Only time will tell who is right. And when I say future, I do not necessarily mean one or two years.
By
Deelip Menezes, At
8:28 AM, May 27, 2009
Direct modeling will never replace history modeling. It will complement it. In history there is information and direct editing deliberately decides to ignore it.
I am one of the "someones" that Evan talked about. I worked in kernels for a lot of years and I can witness that kernel interoperability is a mess. It is
already difficult to have interoperability with previous versions of the same kernel. The consequence of this is that sometimes old models do not rebuild in new reelease of the same CAD system.
Roberto Ciarloni
CTO think3
By
Roberto Ciarloni, At
1:37 PM, May 27, 2009
Roberto, one of the selling points of Think3 back in 2000/2001 was that interoperability was not an issue as you can use GSM to modify the shape. The Think3 IGES translator at the time was the best in the business so once you had the geometry in you were able to manipulate it as desired using GSM tools. In those respects Think3 was ahead of the game - after all that is what Siemens etc are claiming now!
I would be interested to know, as a user, how much developement resource ina major point release goes into ensuring that history rebuilds features correctly between different versions?
Maybe this is the rabbit Siemens have in the hat? By throwing out history they can focus on deliverying more stable software as so much development resource is not taken up with tweaking the kernel?
The big question should be for all licensees of Parasolid - how much effort will Siemens put into development of the kernel to ensure version to version compatibility for history based modellers?
By
Kevin Quigley, At
9:45 PM, May 27, 2009
I think that a successful system will be one that can do history based modeling AND direct editing. I have been using Move Face and Delete Face in SolidWorks for years. (Long before all the Direct Editing hype hit the airwaves.) It has helped me in many situations and the best thing about it is...I have history in place in the event I want to "go back" prior to the direct edits.
I think far too many people are getting hung up on direct editing WITH history. So what? To me it still leaves some valuable options for the user. Could those tools be improved in SolidWorks...sure they can. Which is EXACTLY what they showed us at SolidWorks World 2009.
The key is to balance ease of use yet keep editable and UNDERSTANDABLE parameters in place for down stream modification. The history tree does just that. I can turn Instant3D on and model a little quicker vs. traditional methods yet still keep my normal feature tree. If I have an imported part that needs modification, THEN I move to the direct editing tools. Being able to rollback and show the part in its original imported state has been a valuable tool on the current project I am working on which saw a massive influx of parts brought in from ProE.
I can only speak for myself here, but I like the way SolidWorks is approaching this. Give the guys in Concord a little credit....they have been introducing time saving innovations to the industry for years. All of the other "mid-range" products are STILL trying to play catch up to many features SolidWorks has had for YEARS. I still believe they keep the user's best interests in mind much more so than any of the other CAD vendors.
Ricky Jordan
By
Ricky Jordan, At
8:15 AM, June 03, 2009
Hi Ricky,
Yes a system that does both is necessary (By they way, you should check out the upcoming events from IRONCAD in July -- hint hint). IRONCAD has had both for a while but are making large improvements in the next version.
Anyway, one other comment. The one issue that is a major drag in history systems is the strong relation dependency. Yes the history is powerful and helpful if used well or if major changes are not required (that may not be handled in the original design intent). What often ends up happening in this case, is the rebuild errors and endless time trying to repair the model. That is where the direct editing thrives. But as far as I know, IRONCAD is the only one that actually can support a history and direct edits (true direct edits that do not depend on the history) in a single part structure. What this means is that they have a history tree of features that can be parametrically related. Within this tree, they can make direct edits and it will only affect the history shapes that it needs too. In the end you have history features and brep features in the same tree. The brep features are not dependent on anything and therefore element rebuild errors unlike the SolidWorks direct feature. The SolidWorks approach (which can also be done in IRONCAD's next version), just adds the direct edit as a dependent feature which can open it up to the rebuild errors when changes are made. SolidEdge does not offer what IRONCAD is doing either. They have two independent part types (not mixed).
Anyways, you can tell I like IRONCAD. Both IRONCAD and SolidWorks have had unique items in the industry since 1995. However, I think IRONCAD's next version may be one that truly offers the best of both methods which allows the users to truly pick which is best for any given design (not forced down one particular path by the system.
By
Anonymous, At
12:19 AM, June 05, 2009
My Solidworks assemblies always look like they are FLOATING. How do i make it look like they are on the ground?
Solidworks
By
intercad, At
4:04 PM, July 10, 2013
Post a Comment
Subscribe to Post Comments [Atom]
<< Home