We know what you DON’T want, so what DO you want?

It gets tiring always being a downer. Always being “against”. I really do feel that SolidWorks hasn’t given me as a design professional much to cheer about in the last several years. There are obviously some other people out there that feel the same way. But this downer stuff isn’t the message I would choose. If Dassault wants to make SolidWorks into something entirely different, I can’t do anything about that, but I don’t have to participate in it either. The ship is headed in a direction that doesn’t benefit me, and I can’t do anything about the direction, so it’s time to find a new ship.

If I were to speak with my own voice rather than trying to speak for a corner of a CAD community, I would ask for a midrange software that is focused on product design. I would take elements of Rhino, Solid Thinking, VX, Solid Works and Solid Edge and make something that was intended from the ground up to be a shape modeler. I would love it if there were a new CAD developer with some funding who was looking to do this kind of thing, or an existing CAD developer who wanted to augment their business into the consumer product or swoopy shape areas.

I’ve talked a lot about Solid Thinking, but I discovered that their software can’t do simple (and necessary) things such as shell and rib. So it turns out to be NURBS software that only makes “looks-like” models. It’s a very niche use case for ID development, and that’s about it. You can’t really use it for completing the part engineering. Seems like a waste. It could have been so much more. NURBS is about precision. If they only want “looks-like” models, they should just be using polygonal modeling, like Modo.

Anyway, you know what I wouldn’t do. Now, what would I do, if given the opportunity? The goal here is shape development and product engineering. So it has to have both the artistic shape development tools – tug-and-pull as well as polygonal modeling – and the analytic engineering tools to get manufacturing details correct – shell, rib, draft, fillets.

  1. both NURBS and polygonal modeling. You need NURBS for manufacturing and precision, you need polygonal for shape and reverse engineering (laser scan)
  2. both history and direct edit methods available, somewhere between the implementations of Solid Edge and Solid Thinking
  3. tug and pull capabilities to create shapes, with analytic control for surface curvature range, draft, inflections, etc
  4. c2 edge constraints available on everything
  5. skin over point cloud selections
  6. n-sided patch with mesh direction control
  7. non-history based sketching in a single 3D sketch environment

Maybe the system I’m describing already exists in the form of NX. But it needs to be in the $5k range, rather than the $25k range. Trim NX down to the tools needed for product design, and you could sell it to a whole new market. Or add a few features to Solid Edge.

We have several CAD packages kind of focused on machine design, and some aimed at ID type visualization. We need something that merges the two. I think the industry really needs a shape modeler that can be precise and also do the part engineering. Something for which complex shapes are as native and natural as straight lines and extrudes are in SolidWorks.

32 Replies to “We know what you DON’T want, so what DO you want?”

  1. @Matt… if SW and Modo were able to ‘talk’ to each other 100%, would that be a fair compromise? In other words, Sub-d modeling with a history based features. I think the Tspline options, that will be native in SW sounds like it could be the much needed answer.

    When it comes to SolidThinking and it’s functionality, in many respects it blends both Rhino and Solidworks into one program. So one might say “Hey that’s just what I need” But what it causes, at times, is for the program to have an identity crisis. It just doesn’t know what it wants to be. Also no true assembly modeling environment, which is not necessarily a bad thing as you can model much like Rhino or SW in one Part file with multi-bodies. Listing some pros/cons:

    Great sketching tools.
    Has more rendering options than you can shake a stick at.
    Great surfacing features
    VERY poor history management. (imagine windows explorer with 20 folders each as a subfolder of the one above it) It’s not an truly icon driven tree which would help it makes sense as to what feature is what.
    Also, unlike Solidedge with ST, once you switch a model to the polygonal state, there’s no going back to using the history features.
    Inspired, while amazing in many ways, is one of those hit or miss things. Not sure when the last time any of you were inspired by ‘bone growth’ around your model as a source for shape creation.

  2. The result is a 64 bit value. The expression is lost. The value is not shown as I typed it, so I cannot change one part. PI cannot be perfectly represented by 64 bits. 6.022E+23 is too big for SW. Otherwise they do the job perfectly.

  3. A feature that would be excellent for any relational CAD is dimensions as entered. These may be formulas or expressions. Such as: pi/3, sqrt(2)*4in, 5280 mi, 3.57mm, 14″, 7/16″, 5/8″, 30mm + 0.025″ 3.579525E-10. Previously defined constants could be used: L + 0.25″ etc. This type of representation is exact, no rounding, no limited range, no limited resolution, and a much more compact representation.

  4. SolidThinking update: price, as of 2009 starts at about $5500: http://www.tuaw.com/2009/01/07/macworld-2009-solidthinking-offers-speed-and-simplicity-for/

    Yes, ST does have a shell feature. I read somewhere on the ST site that they recommended making the shape in ST and taking it elsewhere for shell and rib. Here’s the link that mentions the ST shell feature: http://solidthinking.com/Navigation.aspx?top_nav_name=Product%20Information&item_name=aboutst

  5. Sorry for the double post all. Tried the first time and after a few hours figured it did not work and tried again. So now there is two of them, and a testament to the clouds reliability to boot!

  6. I’d like to see an upgrade from PV360 2011 (and 2012) to PV360 2010. One of the most critical features for getting work done quickly, or applying variations in color/material, is the ability to cut/copy/paste said materials on surfaces, bodies, and parts. This was removed in the “revolutionary” disruption of interfaces between 2010 and 2010. Really. Such basic function simply disappeared.

    That’s not to say there aren’t means of getting around this, as Rob and others (to my gratitude) illustrated on the thread I started ( https://forum.solidworks.com/message/252598 ), but those means are simply much less efficient (as work-arounds always are). Copy and Paste functionality was literally lost between one version and the next! Can you imagine working in a text application without such a primary tool? With 720 views of this thread (since October), it appears to be a hot topic and a question others are asking.

    Please bring back this functionality, SolidWorks. It shouldn’t be too difficult. You’ve crippled your software, and now we have two full releases (unless it’s solved in SP1 of v2012?) with no fix. Incredible.

  7. @Donovan
    He seems focused on open source, which isn’t really part of my goal. I’m not against open source CAD, I’m more focused on the “product design” side of things. Meaning complex shapes with analytical tools. If open source is the way to get that, then I’m ok with that.

  8. @smartin
    Well, I think you’re right, the information the tool is designed to present, actually DOES have the lion’s share of the area: it’s a marketing piece. The calendar is just an excuse, and from the looks of it, a bad one.

    @R.Paul Waddington

    We’re back to that? I know I can’t write any code worth distributing. I just wish someone would make this happen. Product design and engineering is an unserved market. Not as big as the machine design market, but one that no one owns. With the demise of Think3, and with Vx going to the Chinese, there’s no one really focused on mid-range product design. Some folks have recently pointed out that NX does have a $6k bundle, but that doesn’t include advanced surfacing. $6k NX WITH surfacing would be a big deal to me.

  9. The duplication of dates is a crazy error, but the non-marking of Thanksgiving is not a big deal to me. What caught me immediately is that the day markings have one of the smallest chunks of area on the calendar – the information the tool is designed to present should have the lion’s share of area, not a small corner. This isn’t a calendar, it’s a cutesy picture with a small calendar tacked on.

  10. Neil,
    Without Dassault coming right out and saying so you are close to what I believe the truth to be. Since all of this stuff is driven by creating more cash for the pockets of Dassault, lets look at it in this light. I begin to believe the number crunchers have done an analysis of where Dassault can generate the best bang for the buck and future for growth. It is not the cad market as this is relatively mature in developed countries and most third world countries work off of cracked versions so big BIG growth there is questionable.
    Look at the absurd valuations assigned to junk like Google and Facebook. The potential for growth chasing this social stuff may be far greater for a company like Dassault. I believe they may be choosing not to be primarily a cad company and are no longer very worried about geometry tools. Assuming of course that you are not to concerned about preserving the cad base in your cad programs and you think you can make a lot more money by gamification and the rest of it’s partners being inflicted upon Dassaults cad users as you chase bigger markets why not do so. I have to assume people like Bernard are not total idiots but rather driven by greener pastures and the SW cad users are not it.

  11. One possibly is that DS is becoming so big that they are warping space time….
    Or perhaps they have taken asset ownership to the next level – buy 365 during November and get another 3 for free!
    Hail the new masters of the universe. Where there is a will there is nothing that can’t be recoded in C++.

  12. So I have this calendar in my office and it has some cool geeky CAD pictures…guess what’s wrong with it !

    Click on the picture and stare at it….don’t get carried away by the cat or the ipod. Look carefully at what a calendar is meant to show – the numbers and the days..

    No, my company didn’t make this calendar..It was sent to me by a French company 😉 It’s not surprising that we don’t buy anything from them.
    [img]http://www.dezignstuff.com/blog/wp-content/uploads/2011/11/DS_Catia_Calendar_November.jpg[/img]

  13. @cloudman
    Well if you are listening, which you arent, people dont want something revolutionary they want continuing evolution of what was working well.
    The innovation people want DS to focus on is confined to things that directly benefit the technical tasks they undertake not companion selling or reinventing and incorporating ancillary junk.
    I am glad you identified that cloud proponents are crazy rather than visionary though.
    History is littered with crazies who tried to impose big plans and failed or were overthrown after considerable inconvenience to the populace.
    Having used a slide rule in the past I can tell you the great thing about one is ‘it just works’. No bugs, no batteries, no makeovers required. It is entirely functional and quite elegant in its simplicity. No one tried to fit an empty bean can and a length of string to the end of it to increase sales. You got only the tool you needed. If you ever had to buy a new one you didnt have to sign up to a carrier pigeon service or an air tube delivery system because it was all the rage at the time.
    If DS are going to pull this expedition to Shangri La back from a major fail they really need to get out of the tent and talk to the footsoldiers about practical detail instead of having their soothsayers retell tales of fabled lands or treasure just beyond the next hill. Without being able to identify tangible benefits they aren’t going any further with the present leadership. About the only thing left to do to save face and arse at this point is to consult the stars or magic tea leaves and claim divine redirection.

  14. Neil,
    Without Dassault coming right out and saying so you are lose to what I believe the truth to be. Since all of this stuff is driven by creating more cash for the pockets of Dassault, lets look at it in this light. I begin to believe the number crunchers have done an analysis of where Dassault can generate the best bang for the buck and future for growth. It is not the cad market as this is relatively mature in developed countries and most third world countries work off of cracked versions so big BIG growth there is questionable.
    Look at the absurd valuations assigned to junk like Google and Facebook. The potential for growth chasing this social stuff has far greater potential for a company like Dassault who may be choosing to not be primarily a cad company worried about geometry tools. Assuming of course that you are not to concerned about preserving the cad base in your cad programs and you think you can make a lot more money by gamification and the rest of it’s partners being inflicted upon Dassaults cad users why not do so. I have to assume people like Bernard are not total idiots but rather driven by greener pastures and the SW users are not it.

  15. @matt
    Any article is going to be superficial given your specific requirements. PowerSHAPE started as a way to prepare models for PowerMILL.

    I couldn’t quickly find a spec sheet listing all its capabilities, but I did find out a bit more. This PDF ftp://arrow.delcam.com/pdf/powershape/en/PowerSHAPE_Variants.pdf shows the versions; only the Pro version supports meshes (but it’s good to see Drafting listed as supported).

    Deelip’s series is at http://www.deelip.com/?p=886

  16. @Tony
    Tony, no, I’ve never really looked at it in detail. That article is a bit superficial, but I’ll have a look at the software. Thanks for the suggestion. I always thought of Delcam as CNC software.

  17. Dan, SolidThinking is about £3k for the standard version, £4k for Inspired here in the UK last time I looked – used to be a lot cheaper (as in around £1000) but then Altair bought them, so of course it probably costs $3000 and $4000 in the USA 🙁

    Matt, I don’t agree with you about a “looks like” model. The only reason I use SolidWorks for production CAD is becuase I get quality drawings out of it. In terms of modelling – specifically shape modelling – I could use another system quite happily, but having a linked model/drawing/history tree is what I want. SolidThinking, along with most other apps of its sort offers great shape modelling but falls short on the drawing side. If all I had to supply was models, I would probably use a more flexible system.

    SolidThinking (like others) does actually offer shelling, and all the other solid features you associate with SolidWorks or SolidEdge. I accept that these apps don’t necessarily have the shortcut tools like ribs, bosses, lips etc but you can usually model these easily.

    That aside, to do everything you want NX would still fall short – and it would cost you over $25k and then some on subs. Same with CATIA. The only app I have seen that can genuinely cover all the shape bases is Powershape (as Tony mentions).

    Powershape grew from the original DUCT surface modeller and has always been a niche product in the mould making business. What Delcam do, they do very well, and you can mix and match point clouds, meshes, solids and surfaces as needed. You can create your swoopy surfaces then add intricate surface detail to it – as geometry.

    For the full system you are getting north of the £10k price but all the toolmakers and customers I have who use it swear by it (rather than at it).

    Getting down to more affordable levels there is a great package called Shark FX that sells itself as a precision concept modeller. This is built on the general 3D interface of the original Ashlar 3D modelling systems and (in my opinion) that interface is still one of the best for shape creation in wireframe.

  18. @Dan Staples
    Dan,

    I’ve never used it for a project, but the scheme is to build a history-based model, push/pull to create shape, update the history-based feature and the push/pull direct edits are also applied to the update. The problem to me is the lack of features for real manufacturing modeling.

    It also does polygonal modeling and has an integration to some popular reverse engineering hardware.

    The pricing I’ve heard to be between $5k and $8k.

    I think the “high-end” look is because it’s all that money for something that is not very broad-featured. It’s very ID niche. It’s priced better than say Alias. I think it’s a little process-intense for most IDers, and not analytical enough for most plastic part engineers. It’s a great idea left in a pretty weak position. They need a bit of what you’ve got ;o)

    Plus, the parent Altair outfit seems to specialize in higher end simulation.

  19. Matt,
    Have you ever looked at PowerSHAPE? It handles solids (via Parasolid), surfaces, and meshes, and they’ve added some sort of direct modeling. Pricing isn’t great: starts at £5,000, although you can try PowerSHAPE-e for free (not crippled, except exporting costs ££). Develop3D has a overview here http://develop3d.com/reviews/delcam-powershape-2010 and Deelip has covered it a bit.

  20. I seem to want the same product as Matt is describing. I want smooth, precise, controllable complex curved surfaces. The sketch spline through points should result in the smoothest spline that contains all of the points. Smooth should be continuous in rate of change of curvature. Sometimes a crease is needed, we can do that by mutual trim of two surfaces or have an influence factor across the crease guide curve. Splines need good control handles, the Adobe style spline control handle works better than the 3way Solidworks handles. I need to add notes to my models; sketches need simple text.

    CAD systems need to flush all of this non-geometry creation stuff. The only thing that all CAD companies need is a price list. Maybe Frank can figure out a nice package and price for us.

  21. Matt/Others…

    Solid Thinking looks really cool on the website. Lots of stuff that is very impressive. Have you actually used it in a production oriented way? Is it as good as it appears? What about pricepoint for SolidThinking — it seems to have a lot of “high end” customers. Is the price high end too? (what is it?)

  22. @Cloud Man
    Rain Man,

    Yeah, we’ve heard that Ford quote to justify a lot of stupid things. Ford also made something called the Edsel, which I guess dinosaurs just didn’t know that they really wanted. Really like the incognito stuff, it increases your credibility, dude.

    I’m glad you clarify this stuff, because, you know, we “dinosaurs” distinctly remember a time when we all ran software off of a common central server. This wasn’t in the future, it was in the past. We’re glad to be rid of that innovation.

    Revolutions aren’t always good. Change for the sake of change is not progress. Castro and Khadafi came to power in revolutions.

  23. @Neil

    When we are at the junction of technological revolutions it’s not useful to ask users what they want. I heard a quote of Henry Ford – “if I had asked my customers what they want then they would have replied – a faster horse”.

    Same goes with apple products – please read Steve Jobs bio.

    If you want something revolutionary then you have to trust the crazy ones.

    Otherwise please go back to using a slide rule and drawing board – dinosaur.

  24. “Maybe the system I’m describing already exists in the form of NX. But it needs to be in the $5k range, rather than the $25k range. Trim NX down to the tools needed for product design, and you could sell it to a whole new market. Or add a few features to Solid Edge.”

    NX comes in different tastes; the bundles don’t all cost $25k
    Ask for the NX Mach Advantage bundle and maybe you will be hapilly surprised

  25. It depends how you frame up ‘against’.
    If you are opposed to things that make life difficult or are wrong headed and say so then that’s a positive and worthwhile thing in my book. You can be for peace by being against violence. It depends which way you look at it.
    You aren’t about pulling something down for the sake of it, you are about warding off stuff that you anticipate or identify is detrimental or unwanted.
    Why be a silent victum? If you are aware the program isn’t delivering what you need its definitely in your and everyone elses interest to speak up about it. Once you commit to a CAD vendor you want them to keep on keeping on. You bought one particular program for a specific reason not because it would morph into something else on the makers whim.
    Over the time I have used SW I have endeavoured to do my bit for truth, justice and the light exactly because I found the decision makers stray from the plot quite easily especially being too readily influenced by marketing and diverted by flakey reasoning.
    Users definitely have a vested interest in saying what enhancements and new functionality goes into the tool they rely on. It is a technical tool not a fashion statement or xmas present wrapping contest.
    SW certainly haven’t wanted for ideas and opinion from the community about what to do better if they cared to listen.
    To my knowledge the feedback has been all about improving things to the benefit of customers and the company.
    Unfortunately despite fairly robust feedback from the user community SW have veered away from what was wanted and no longer give the impression they care much about serving mechanical engineers and designers as much as they want to build empires via social media and virtual experiences.
    I am sure many people regard this self styled adventure as a rather monsterous violation of their trust and a serious inconvenience to their business and livelihood. Honestly I don’t think DS appreciate the ill will they are causing, either that or that or they just don’t care because they are so cock sure of their path to guaranteed profit.
    I think though you need to distinquish between the way many people would like SW to continue to be, both users and coders, and what has been imposed from the top down.
    There is what is now an old song by Paul Simon along the lines of Ever since the world began Everybody got the runs for glory Nobody stop and scrutinize the plan. While SW users may have scrutinised the plan and found it wanting DS just don’t want to know. DS doesn’t speak the same language as its customers any more.
    In fact they dont speak at all or if they do they are pretty much only talking to themselves about themselves. People aren’t impressed by a CEO with idea diarrhea or going about dropping statistical BS.
    Given this its not surprising people would stop trying to be ‘against’ wilful stupidity or ‘for’ common sense and apply their time, brains and energy to something that will benefit them.
    Who would blame even the most dedicated and loyal SW fan from recognising the futility of the situation and moving on….
    Bring on something new that does what people need.
    It would be great to be inspired and enthused about CAD software again as we were in the early days of SW. I wonder what Jon H. is up to….

  26. In my opinion, think3 also have a good features both for surface and mechanical. But we dont know how those company will going because of internal conflict.

  27. Hi Matt,

    Do you think that SolidWorks not enough for your needs to cover both surface modeling and mechanical design ? In which area that you think SolidWorks still poor ?

Leave a Reply

%d bloggers like this: