Parasolid-based Mechanical CAD, for free. You read that right. Unfettered use of the entire Onshape system for up to 5GB and five private documents at a time is completely free, and a mere $1200/yr for unlimited documents. That’s right folks, the much ballyhooed Onshape just stepped into the ring swinging, and they are poised to cause a shake-up the likes of which haven’t been seen since Gangsta’s Paradise was a number one hit.

I’ve worked as a consultant for Onshape a few times. Given that the views expressed here are consistent with articles I wrote before I knew about Onshape, however, Josh and I thought it was fair to publish.

Onshape’s beginnings

Since stepping down as co-founder and CEO of SolidWorks in 2011, the venerable Jon Hirschtick (aka The J-Hirsch) has been oddly quiet. We always envisioned him living in a smoke-filled yurt two hundred miles outside Ulaanbataar milking goats and living off the land. We were way off. As it turns out, he’d been amassing an Dream Team of MCAD and cloud tech elite under the intriguingly nondescript “Belmont Technologies” moniker. Fast forward a couple of years and Belmont becomes Onshape, suddenly revealing that they’ve built a team of over forty engineers, an army of interns, and six of the eight top leadership slots filled with SolidWorks top brass from years gone by.

Their project: a collaborative, 100% browser-based, parametric MCAD system, built from the ground up to run on any platform or device. No installs, service packs or registration codes.

It’s hard to overstate what a monumental accomplishment this is. Parasolid was architected for workstation computing, not distributed cloud computing, so the mere act of adapting the kernel for efficient use across a server rack is an eye opener in itself. Add to that the fact that they’ve built a surprisingly responsive MCAD UI around it that’s cross-browser and mobile compatible from day one, and what they’ve achieved in such a comparatively short time is truly humbling. (Cue the pitter-patter of a thousand face-palms echoing across a Boston suburb.)

Cool technology, no doubt, but the real kicker? Brass tacks.


Onshape pricing

First of all, you can use every feature Onshape offers for free. Full Parasolid based parametric mechanical CAD. Unhindered import/export. Unlimited team size and unlimited sharing. Version control, branching, merging, 5GB of storage. Zero Dollars. Zilch. Free as in Beer.

Students and classrooms? Zero dollars, zero installation required, zero IT overhead. No license servers, no node locks, no dongle-berries. Works on any machine in any lab or classroom on any platform in any building on any campus. Oh, and it works on your iPhone, too.

Then, for $1200/yr, professional engineering teams can frolic unfettered in the flower-laden fields of Onshape. We’ve been assured by persons in the know that the $1200/yr price will continue be all-inclusive even as new functionality is added. That’s it.

Compare that with the new SOLIDWORKS 3DEXPERIENCE applications around $2800/yr per product, not to mention extra charges for showing your work to a client or colleague, and (a) we get kinda excited in ways we don’t discuss in public, and (b) we’re likely to receive some concerned emails with thick accents in the next few hours. (Duck and cover.)

Apples and Oranges

Granted, Onshape is a brand spankin’ new product still wet behind the ears, and that comes with limitations. Its feature offering is impressive for a first-gen MCAD platform, and it’s growing at a break-neck pace. That said, even casual CAD pushers will find its modeling feature set quite limiting at present. Give it a test drive, and you’ll find that it’s, well, very much a work-in-progress.

Given that Onshape is still in its infancy, it’s not really fair to compare its pricing with more established engineering platforms. Or is it?

Go back to the classroom example: why bother with the IT overhead of a traditional CAD system–not to mention spending a few hundred bucks per seat–when I could just point my students to and get going on day one? What about all those folks buying ShopBots, OtherMills, Carveys, Form 1’s, and those love-to-hate Makerbots? Every stinkin’ one of ’em just signed up for Onshape. What about the rapidly increasing number of engineering contractors, freelancers, and boutique design shops all over the world, Dads with too many power tools, kids who take things apart, or anyone currently using Sketchup?

Why wouldn’t you sign up for OnShape, even if only to use it occasionally?

Apples and Apples

“Now wait” you’re thinking (I’m talkin’ to you, @ScottMoyse), “what about Fusion 360?” Well said, Scott. Just where we were headed.

Fusion 360 has a free version too. It’s also got a $300/yr product that is more capable than Onshape at this stage, and a $1200/yr product (same exact price as Onshape… hmmm….) that is far, far more capable. One can only assume that Onshape’s pricing is a stake in the ground rather than a realistic valuation of its current capabilities. We won’t do a feature comparison chart just yet, but suffice it to say that Onshape wouldn’t stack up well against Fusion if we were to compare them today.

One might point out that F360 requires an installed app on your OS of choice, while Onshape requires only a browser. Most people don’t remember that Fusion 360 was originally, like Onshape, an entirely browser-based product. Users pushed back due largely to performance and graphics limitations inherent to WebGL, not to mention the near-constant stream of IP data required to keep an in-browser tool running means that any little blip in connectivity causes serious problems. The fact that F360 is an installed product today is not because Autodesk couldn’t build a browser version, but because users demanded a level of performance that the browser simply couldn’t provide.

So why, one wonders, is Onshape a bigger deal than F360? It’s not, really. Fusion 360 was exciting when it came out, and we’ve been watching it eagerly over the last couple of years. The difference is that now there are two players in the cloud CAD space. The addition of Onshape means that cloud CAD is no longer just a little experiment in Autodesk Labs. Fusion 360 was a huge step forward, but one little product does not an industry revolution make. With Onshape in the ring, things get interesting.

Fight The Power

I can’t stress enough what a big deal this is. Forget about features, workflow, and the normal bla bla bla about cloud, collaboration, multi-CAD compatibility, etc. It’s all about The Benjamins. Even if Onshape itself turns out to be a dud (it won’t), mark my words: the game has changed.

It’s no longer enough for The MCAD Syndicates to sell CAD by the “seat” with complex bundling restrictions, bizarre and arbitrary licensing schemes, and pricing that was dreamed up in a backroom sales negotiation at Boeing or GM. Even if Onshape is only moderately successful as a product in itself, it will force the industry to re-examine regressive revenue models in favor of simpler, more straightforward, customer-oriented value propositions.

Adobe was first with Creative Cloud, offering their entire product line for a single, simple price. It’s easy, affordable, and a no-brainer for customers like me. Autodesk was second with Fusion 360, offering an impressively feature rich CAD platform with a dead-simple two-tier pricing model, another no-brainer for small and medium businesses who need Mechanical CAD without the overhead of a full PLM system. Now it’s Onshape. It will take a good long while for bigger players to change course, but rest assured, they will.

And, I need to re-emphasize a crucial point: this isn’t about merely cheaper pricing for its own sake. My point is not egalitarian so much as good practical business sense: with the explosion of interest in extruder nozzles and mill bits in the public consciousness, the potential market for 3D creation tools just went bananas, b-a-n-a-n-a-s.

Is this a race to the bottom of the Feature Tree? Is advanced MCAD about to become a cheap commodity in a price-driven push market? Not a chance. There will always be a need for costly high-end systems that are only supported by an elite niche, and therefore demand a premium. But as the need for competent 3D design tools becomes mainstream, those tools will get cheaper. High end simulation and manufacturing tools will probably always the expensive prerogative of serious engineering businesses customers, but geometry creation and assembly design tools will not.

Sometimes industries have to be “dragged kicking and screaming to the money tree, and have it shaken for them.” Onshape is doing just that.

A nuanced debate

I’ve always said that if a CAD tool makes me twice as productive in a business context, that tool is worth my salary. If I can get $160k worth of work out of a $80k engineer by using XYZ CAD system, then I should be willing to pay XYZ CAD Systems Inc as much as $80k for it. Anything less than that, and they’re making me money. If they charge $40k, suddenly they just made me $40k. If they charge $5k for it, buying XYZ CAD just made me $75k. Hell yes.

That’s why it’s so important to understand that what I’m saying here is not about mere “affordability” or “democratization” or any such Utopian fluff. Professional MCAD systems are valuable. Good tools not only pay for themselves, they make me money. And I like making money. Dassault, PTC, or Siemens can promise to double my output, I’m willing to pay handsomely for that. But the market for 3D tools like these has fundamentally changed, and that’s the real point here. 3D design tools aren’t just for professional engineering firms anymore, so the entire value proposition is shifting.

If I told you I could sell a million seats of XYZ CAD over the next year, would you be willing to sell it for $100? Of course you would. And that’s what’s happening. It’s an exciting time to be in this industry.

Beware, Ye Cloud Deniers

On a slight tangent, I can’t end this article without a word to those of you–you know who you are–still holding steadfast to your little LAN-locked utopias, pining for the days when Men were Men and Computers were Big Beige Boxes.

CAD in the Cloud has arrived. It’s real, it’s here, and everybody’s doing it. Haters hate and nay-sayers say nay, but whether you like it or not, over the next decade desktop CAD users will become increasingly lonely, frustrated, passed over for slices of cake at office birthday parties, and bizarrely obsessed with Swingline staplers. As with any industry transition, there are problems-a-plenty with the Cloudy Thinking that’s pervaded the MCAD world in recent years, but the trajectory is clear: desktop CAD is sooooo Y2K. (Oh Mighty Interwebs, flex thy bulging quadriceps!)

What about the evil PLM Dragons hoarding your Precious Data at the heart of the mountain, sending you daily email reminders that “All Your Data Are Belong To Us?” It’s a big topic. Lets have that conversation over a glass of pure grain alcohol (“Our precious bodily fluids!”) and binge on moon landing conspiracy documentaries afterwards.

Oh, and what happens to your Cloud CAD when the internet is down? Hold that thought, I need to go refuel my backyard generator and restock the shelves in the bomb shelter downstairs.

It doesn’t matter whether you stand up, wave your fists around and scream about Cloud this or Rentalware that. If you want to live off MRE’s while working in a locked basement with nothing more than a telegraph, a stack of acid-free paper napkins, and your favorite archival India Ink rollerball, that won’t change the fact that the industry is changing. It will happen whether you’re along for the ride or not. And that’s a good thing.

What do you think?

I’ve never been accused of having too few opinions, or for holding them too loosely. That said, I don’t hazard public prognostications lightly. It’s true that I’ve done some consulting for OnShape, just as I have for lots of other companies over the last six or seven years. I’m a consultant. It’s what I do.

But what I’ve written here isn’t really about OnShape per se. This article, along with the last few I’ve written, is really about a changing industry. As loth as I am to give credence to grandiose claims about a so-called 3D printing “revolution,” (exasperated eye roll) one has to admit that the public exposure the 3D CAD industry has seen as a result has already been a force for change, and one that my daughter will benefit from.

By the time my daughter is in middle school, she’ll be using tools like OnShape to design things for STEM projects as a matter of course. And that, friends, really gets me excited.


Adam O'Hern is an industrial designer, designing products ranging from laptops to power tools, classroom toys to bathroom fixtures, and pro audio gear to guitar tuners. In 2008 he founded, and in 2010 co-founded EvD Media with Josh Mings of, and the two collaborate on the podcast.


  1. I have been motivated lately to get my hands into some of these cloudy CAD systems, partly because of the sentiment you relayed in the last sentence. My boys won’t be using the tools I use now, and I don’t want to be left behind… too quickly at least.

  2. Bearing in mind what is stated in the article about this being the early stages, feature wise; how would Onshape stack up to Autodesk Inventor (what I use everyday)?

    • If its behind Fusion 360 (and from the article it certainly sounds that way) then it’s behind Inventor

      I’m still waiting for a proper drawings capabilities to come to Fusion 360, I’m surprise at how incomplete their drawings are considering that they already have Inventor, I wonder why don’t they just port the drawing features found in inventor and improve on it rather then using ACAD as their drawings code base (from my understanding in one of their posts)

      Though in terms of drawings, I’ll say that among the mainstream CAD (NX and Catia exempted), SolidEdge is the gold standard for it.

    • It really depends on what you use it for. As I was comparing Solidworks, Inventor and Onshape, I found that Onshape would probably be fantastic for teams doing strictly mechanical designs. However as basically a one man shop, doing primarily aesthetic surfacing work, it didn’t have much for me. Inventor, for my needs, is a much better option, and even then, really only because of the T-Splines tools they just added. Before that, it wasn’t even up to what I needed.

  3. Adam,

    Are you and Josh coming to D3DLive This year? Man – you gotta get a ring side seat πŸ˜€

    Great post – keep up the good work!

  4. Great rundown. I mostly agree, though I think you laugh off the difficulties with always-connected software a little too quickly. I know at my workshop, the only broadband provider available, is down at least once a month, for a period of at least two hours, and that window has absolutely fallen in the middle of a late night rush to deadline before. I can’t even imagine the fury I would have felt, if I had to call up my client to tell him the design would be a day late because Time Warner sucks.

    That said, this is certainly the end of my days using Solidworks. I tried the Onshape preview, and it doesn’t come close to doing what I’m looking for right now, but between this and Fusion 360, it was enough to convince me to drop Solidworks, move over to an Autodesk subscription, and wait and see where the market goes, since I’m sure in the next couple of years, there are going to be enough changes that I would feel stupid sitting around paying maintenance for Dassault to keep trying to sell me on Catia Subscription Light.

      • Using is probably a strong word at the moment. I bought it, and I am learning it, but I am still leaning heavily on Modo and Solidworks to get things done in this transitionary period. I can see a workflow where I am entirely in Inventor, or maybe Inventor and Fusion 360 together, but I’m not there yet. Luckily, I already paid my Solidworks maintenance for the year, so I have until next year to decide for sure on what tool I’m going to make my primary.

        • scottmoyse Reply

          Have you had any transitional training? Going completely off topic from this post here. Sorry Adam.

          • Lee Lloyd

            No, I haven’t, but do keep in mind, I have been working in 3D since the late ’80s, and have used more software than I can even keep track of at this point. I have seen just about every possible way of doing any particular operation in 3D, it is just a matter of figuring out where the buttons are.

            That said, and keeping in mind that I only have a week or two of experience with Inventor at this point, conceptually it seems like Inventor with T-Splines is very similar to Solidworks with the Npower PowerSurfacing plugin. I am not yet seeing anything that gives me any worries, because interface wise, the two seem very much like copies of each other. There are some weird differences, and I’m sure as I dig in more, I will find some problems, since you always do, but it seems like a fairly straightforward transition at the moment.

            In fact, if this gives you any idea how similar they are, the biggest difference I’ve found so far, is that in Inventor cutting is just an option of the construction tools, instead of a separate tool of its own. If anything, the T-Splines make the transition easier, because I have previous experience with T-Splines for back when I was using Maya.

          • scottmoyse

            You will love the next release then πŸ˜‰

          • Lee Lloyd

            That has me intrigued! Hopefully you are hinting at more T-Spline functionality. Of course, what I would really love is to see all the Alias surface tools in there, but I’m sure Autodesk would never want to miss a chance to charge me extra for those.

    • Kevin De Smet Reply

      I get what you’re saying, I really do. But a day late isn’t such a big deal. Again, I get it. And you wanna get things done on time and that’s great. But lots of people are wasting a lot more time, all over the place, than being a mere single day late. Let’s be real here.

      • It’s going to depend entirely on the type of customer and projects you have. You are totally correct, it is not a big deal, unless it is. I of course try to avoid it, but there are times when people pay me quite well, to turn around a project on a very tight schedule. If you are working on something like a movie prop, or assets for a trade show presentation, you can find yourself in a position where you only have a few days to get the part out the door, or where there are changes being made to a design up until the very last possible minute. In situations like that, unexpectedly losing a few hours can be the difference between a happy customer, and never working with that company again.

        I am not trying to overblow it, and you are right, for many shops I am sure it is not a big deal to lose a few hours here, or a few hours there. It is going to be highly dependant on what industry and market you are talking about though. For me personally, I have worked on enough tight schedules that any possible point of failure is something I need to take into account when budgeting my time.

        • I’m a stickler for deadlines too, but I know I’ve often been delayed by persistent crashes, motherboard malfunctions, software glitches, accidentally deleting data (doh!), and, yes, even slow internet access. I agree that relying wholly on connectivity for access to my work is a bit scary, but it comes with advantages as well. If my computer goes down, I can just use another one. I’ll never lose data. Etc.

          For me the thing that changed my mind was a day a few years back when Google was down for a few hours. My world basically came to a halt. To be honest, if the internet is down, I can’t really do my work anyway. I think we’re already dependent on always-on connectivity more than we realize.

          • Lee Lloyd

            No doubt there is a certain amount of irrational fear, much like ‘range anxiety’ in an electric car. However, I feel like all the discussion of mitigation, and acceptable risk, dances around the core question that plagues this entire push to cloud-based software: what is the benefit to the customer of a cloud-only solution?

            The benefits to the vendor are clear.

            One code base for multiple platforms.
            No piracy concerns.
            Steady subscription revenue stream.
            Lower support cost.
            Higher user lock-in.
            The ability to shunt all low-level technical problems onto the browser developer.
            Simplified license management.

            Likewise, the downsides for the customer are clear as well.

            No perpetual licenses.
            Dependent on connectivity.
            Your data on someone else’s server.
            No practical way to automatically backup data.
            Security concerns.

            But the less clear question is what is the benefit to the customer of doing everything in the browser? Technically speaking, there is nothing you can do in the browser, that couldn’t be done in a dedicated local application, with network connectivity. From a user perspective, I don’t see a single downside to a local app that uses network features, over a browser.

            Let’s face it, the push to have everything “in the cloud” is about business models, not solving customer problems. From a customer perspective, I would always rather have both local access and storage, and the option to use cloud features as appropriate, and there is no technical limitation that keeps that from happening, it is an business choice.

            That is why, sure, there are solutions to get around possible connectivity issues, but every one of those comes with a cost, either in money or time, and it is hard not to ask, as a customer, what am I getting for those additional costs? Now, as it stand as the moment, the answer is usually “free or really cheap software” but I think we all know, that doesn’t last forever. At some point, the company has to turn a profit.

            Perhaps for a younger generation, this is a lesson they will have to learn for themselves, but for those of us who are old enough to remember the mainframe time sharing client/server days before the PC revolution, it is immediately obvious how expensive being dependant on another company’s computing resources can get.

          • Fair points, but I think you shortchange the advantages to the customer. The other day I picked up my wife’s Macbook Air to take the the coffee shop and work with Onshape. No install required.

            As a teacher, one of my biggest annoyances with SW was that students had to pirate copies to use on their own laptops (I discouraged this vigorously, btw, but that didn’t stop them from doing it). Not only that, but SW was only available on certain workstations in certain labs, since the IT guys found it really irritating to install. In-browser solutions get around those issues, allowing people to work from any machine, anywhere.

            When was the last time you had to re-install your laptop from scratch, just because Windows was being stupid? It sucks up a full work day at least twice a year for me, and installing CAD tools is at least half of that day.

          • Lee Lloyd

            But how many of those advantages are served just as well by Fusion 360, with a local client? No pirating, use on any computer, free for students.

            I am not yet convinced, in fact I am fairly convinced of the opposite. I don’t think it is actually very controversial at all to say that a browser is a poor substitute for an optimized local client. I know that when I have the option of a local client, or a web front end, I almost always find myself using the local client.

            I understand, this is how they have chosen to do business, and if you want to use this software, that is what you agree to do. I’m just saying that, as a potential customer, it is my job to look after my business interests, and use the software that will best help me get my work done, not my duty to help them prove their business model will work.

            I am not resistant to change, but by the same token, it isn’t to my advantage to make change, purely for the sake of change. I think we all agree, the node-locked single seat, geographically restricted model of software licensing is not going to be missed, but from what I’ve seen so far, the future looks a lot more like what Adobe and Autodesk are doing, than running everything in a browser through HTML 5.

          • I think we’re in a cycle of aggressive agreement here. I agree completely, I just think there is an argument to be made for both.

            Food for thought: if Google Docs required an installed local client to function, do you think adoption would have been as widespread as it was?

            I’m not arguing right or wrong here, just trying to show the nuance.

          • Lee Lloyd

            But isn’t that the question, how widespread IS Google Docs? I mean, I use it, but I have yet to have a single client actually share a Google Doc with me. The one time I tried to share a Google Doc with a client, they asked me why I couldn’t just send them a “regular” Office Doc! Everyone still uses MS Office. Hell, I’ve actually had a client send me an Apple Numbers document, but I’ve never had one share a Google Doc.

            This is one of the oddities of the intersection of the professional world and the web. It isn’t hard to generate buzz, and get everyone on the web talking about your product. A good marketing campaign, and giving it away free, and you can bet on a wave of good press, and huge sign up numbers. It is, on the other hand, very hard to get businesses to change how they do business.

            I can’t help but feel that a large part of the success of Adobe and Autodesk in this regard, is that they were just honest with their customers about moving to software as service being purely a business decision. You keep doing things like you’ve been doing them, but we are going to change our billing model. Nothing to think about workflow wise, just a straight financial decision.

            The other side of that coin if the dismal failure of Dassault with their new product offerings. It comes across as a trick, a trap. You want these new features? Then you need to change software, change to software as a service, and increase your budget.

            Back to the topic though, I still don’t know how this will pan out with a program that has no legacy. As it stands, I am not sure they are going to get much interest from people actually currently using another CAD program, but I could be wrong.

        • Yeah, he said that the internet on his “area” was the issue so I imagined a local Starbucks would help much… Another thing that appeals to me from an Admin and Consumer stand-point is the ability to run it from any device… I’m thinking getting rid of the impracticality, cost and overhead that takes to run “CAD Workstations”, Now designers will be able to run Dell XPS, Lenovo Yoga, Surface Pro… yes and even Macbooks! This is just awesome! coming from someone that has owned a “Mobile Workstation”

      • But, as I asked below, what benefit do I get as a customer, out of software that requires me to incur additional expenses for a 4G hotspot? I have software right now that works just fine without any net connection at all.

        I feel like this is getting all twisted around. It is not my responsibility, as the customer, to figure out ways to work around the limitations of how a company chooses to do business, unless it is for some reason advantageous for me to do that. How is it advantageous for me to give up software that works with or without a network connection, for software that requires me to maintain a backup connection or schlep to the nearest working public hotspot?

        • Additional Expenses? like requiring a Workstation type computer which tipically cost 2 or up to 3 times more than other computers with similar CPU/Memory. Or like 20% subscritption just to get Bugs Fixed or new functionality…

          • Lee Lloyd

            Browsers aren’t magic, and WebGL isn’t even as efficient as OpenGL. If your part/assembly required a hefty workstation to get good speed and interactivity with before, it still will in Onshape. In fact, if Autodesk’s attempt at a browser-based CAD program is any indication, you will actually need a more powerful workstation than you do for a local app.

            That said, computing is cheap, and in an age where even phones have quad core processors and 3D acceleration, I don’t know why you think you need some special machine to run any CAD program. I’ve seen plenty of people running Solidworks, Maya, Modo, Inventor, and other 3D programs just fine on commodity hardware like Dell notebooks, Macbooks, and Surface tablets. Just because there are people out there selling workstations with dual Xeons and ECC memory, doesn’t mean those are required to run a CAD package.

            As far as the subscription goes, if you think that you will have a long successful career, never using anything but free software, you have a rude awakening ahead of you. Companies need to make money, and Onshape is not a non-profit. I have no doubt that they will find some way to charge you, just like every other software company on the planet.

          • Sorry Lee, but now you’re just contradicting yourself…

          • Well you say that WebGL won’t handle as well as a “Hefty Workstation for good speed and interactivity” for Complex/Large files; then on your second paragraph you say that you can run Solidworks efficiently on a consumer type Notebook. I’m sorry but that is not the case. Believe I would like to have a Surface Pro 3 or a Lenovo Yoga instead of my current Dell Workstation, but performance does not compare!

            And then you go on and justify the current Subscription scheme, but complain about the “Additional Expense” of a 4G Hotspot. Which in my boat is a lot less expensive than maintaining a backup server for SolidWorks Data.

          • Lee Lloyd

            1: I don’t know anything about your work, so I don’t know how complicated your assemblies are, so I can’t say what your personal processing requirements would be. However, I have seen my wife design plenty of pieces of jewelry on her HP laptop in Modo and Fusion 360. I know a machinist who does all his designs in Inventor on an old Dell, and I can’t even keep track of how many artists I know using Macbooks with various 3D and CAD programs. So obviously for whatever reason, your results vary (maybe your video drivers?), but it is quite possible to get decent performance out of CAD packages on commodity hardware. People do it every day.

            2: I am not complaining about expense, I’m complaining about value for expense. Setting up my own file server for data redundancy has enormous value to me. Paying for an extra net connection, because some vendor couldn’t be bothered to design with intermittent connections in mind, has zero value to me. It is a kludge to get around a flaw in their design.

          • Thanks for the insights, guys! If you have anything new to add, please do. I think you’ve both made your points on this topic.

  5. Kevin Quigley Reply

    Spot on Adam. I hold my hand up and say I have been looking at Onshape for over 12 months now as I was one of the early invitees (thanks to you I believe). Ironically enough I was also one of the early invitees to the Fusion 360 product as well (something like 2 years before it was released). In the same time frame (last 3 years) I have also purchased another license of SolidWorks for my business. Odd? No. At this time – note – at this time – neither Onshape or Fusion are suitable candidates for production tools for my business.

    This is not a pedantic Kevin Quigley, stuck in the mud old fart “I only use SolidWorks” (and anyone whom knows me will know I am happy to fork out for something that is better) saying this. No. This is down to the work we do. The jobs that bring in the dosh.

    At this time, neither Fusion or Onshape can touch the surfacing we can do in SolidWorks. Yes Fusion has Tsplines and very nice it is as well, however, sub d tech is rarely used for production level surfacing. We have PowerSurfacing in SolidWorks, and TSElements for SolidWorks and TSplines for Rhino – and Modo – so we know the tech OK.

    Of the two products Fusion does more, but the Onshape modelling approach is more approachable for a SolidWorks user, because it has the full history. Fusion has history of sorts but it is convoluted at present with a rather odd timeline approach. This alone rules it out as a product for 80% of our turnover.

    I’ve used both and I prefer the Onshape approach – the browser CAD with simple log on is very compelling. I’ve been using the iPad app version for a few months now and that is even more compelling for viewing and simple edits.

    Now the future.

    AutoCAD have Fusion and are developing it rapidly. The question for me is when Inventor ceases to exist and Fusion takes over (and introduces proper history modelling at the same time). This has to be the master plan for Autodesk.

    Onshape are financed by people looking for a return. They have created a compelling product and shown it can be done using probably the industry standard modelling kernel. They have created the infrastructure and suggested a pricing structure that is unique. Or is it? Actually no it is not. SketchUp was and remains free for many. Fusion is free for education/hobbyists and BUSINESSES turning over less than $100k a year (and that is a lot of small design businesses), and start ups. Then if you do want to pay you get it all for $100 a month (based on annual charges). That “all” includes (or will include) a hell of a lot of functionality.

    I cannot see how Onshape can compete with that. Everything they have they license, so I cannot see how it can scale and remain at $100 a month. Once they add sub d, FEA etc will that all stay in $100? No chance.

    So I see Onshape as primarily the first in a new breed of product, where they have proven the technology and are now waiting for a bigger fish to license it or buy them. You need to get a return on $64m somehow, and no matter what people say there are just not that many people in the world interested to pay for a 3D CAD system to generate IPOs on a social media scale.

    Interesting times ahead for sure and I will happily be the first in the queue to buy the pro versions if and when they start to deliver the efficiencies that will translate to my bottom line. For me neither are there yet, but I am very sure both will be there in a year or two. But in a year or two who knows what the rest of the market will do. The whole online offline thing is irrelevant now – that ship sailed. The cloud is the future whether you like it or not. What will matter from here on in will be efficiency and value.

  6. Great post, Adam! I love this passage –> Even if Onshape itself turns out to be a dud (it won’t), mark my words: the game has changed.. Couldn’t agree more. Game has changed. Period.

  7. Adam, put in a word for me! Tell them I’ll start my blog up again (i’m still in contact with my 3 readers). I’ll do a demo project. Just get them to approve my email for access! I want to play!

  8. It sure would be nice to know some meaningful specifics about Onshape such as:

    Does Onshape have local direct editing like whatever DS/SolidWorks is now calling SolidWorks Industrial Conceptual?

    I’d call local direct editing innovative even if IronCAD has done something similar for over a decade. ;>) Tell me again why Autodesk Fusion 360 tacks direct edits onto the end of a history tree if it’s so innovative?

    While it’s clear that Autodesk Fusion 360 has a lot more features than Onshape does are they really well thought out features that break new ground? In the case of Autodesk CAM, I can assure you the answer is definitely not.

    The race to the bottom to destroy $5,000 up front CAD cost has most certainly begun. Not sure if this is a good thing or not because it’s not radically new CAD modeling that’s leading the way. It’s a new sales model and the ability to run on tablets, phones, etc.

    Jon Banquer
    CADCAM Technology Leaders group on LinkedIn

      • Ric,

        Notice how Philip Thomas starts with a “dumb solid” to begin with in the link you gave. I’ve watched this video (I actually downloaded all of the Onshape videos to watch off-line) and I’m not able to determine if a solid that has a history can be direct edited without appending all the direct edits to the end of the history tree. Did I miss something?

        Jon Banquer
        CADCAM Technology Leaders group on LinkedIn

        • I could be wrong, but last time I used them direct editing in Onshape is just like traditional feature based direct tools: it appends a feature to the end of the tree. Changing features to direct features is something unique to SWMD, AFAIK.

          • From pouring over the Onshape videos I downloaded today, I believe you are correct, Adam and that Onshape can’t do it. I’d love to be wrong.

            It’s not unique to SWMD, IronCAD can also do it.

            I’d like to know if any other CADCAM systems can do this? I think it’s very important that users understand that adding a ton of direct edits to the end of a history tree is bad modeling practice.

            Jon Banquer
            CADCAM Technology Leaders group on LinkedIn

          • What you want can easily be done by exporting/re-importing a direct edited model to loose the direct editing history

          • I don’t want to trash the history!

            When I use direct editing, I want the direct edit to appear in the history tree in the place where I did the direct edit, not at the end of the history tree where it can be next to impossible to see what the direct edit affected.

            Jon Banquer
            CADCAM Technology Leaders group on LinkedIn

          • Actually this is a feature I’d never heard of before you brought it up, John, and its interesting. Having never tried it I can’t pass judgement, but I kinda wish I could have the best of both, since I like keeping around the original design intent in many cases. I wonder if it would help to put the direct feature right next to the original feature in the tree, and possibly even group them together? Just brainstorming.

          • Here is my proof that IronCAD can do direct modeling on a history based model without tacking all the direct edits onto the end of a history tree.


            It shouldn’t take much to realize that the IronCAD and SWMD approach is a much better approach than the approach that Autodesk Fusion 360, Onshape, Solid Edge, Autodesk Inventor or legacy code SolidWorks, etc. are taking.

            Jon Banquer
            CADCAM Technology Leaders group on LinkedIn

        • Jon, here is a lot of functionality not exposed on the product tour. I encourage you to play with it and ask Philip directly. Of course the cool thing about this architecture is that they release features multiple times a month as there is no software to download. If there is something missing that you want you can request it.

          • Ric, I’ve spent days of my time emailing Philip back and forth before Onshape was ever shown where I conveyed what I thought was important in a CADCAM system. I don’t have anything new to say to Philip that I haven’t said to him in the past or on my LinkedIn group. An example of what I mean would be that I think CADCAM suffers from an over dependence on using a history tree and that in this regard not much has changed since Pro/E was released decades ago.

            I also think SolidWorks is a very dated program. One way that we can see just how dated SolidWorks is, is that SolidWorks requires a feature for everything where as Onshape has what I’ll call “Super Features”. Many people will see “Super Features” in Onshape as new and a big step forward because it equates to a much better workflow for the user than SolidWorks has. The sad fact is that much better workflows have existed in other CAD programs for decades. Here is one that hasn’t really been developed since 2001! The person who is responsible for this program is an active member of my LinkedIn group and a CADCAM developer who I have always had a lot of respect for. Notice how in this CAD program shelling doesn’t have to be a separate feature and how it automates creating Assemblies:


            Onshape has what I would describe as a good foundation that I think is much better than Autodesk Fusion 360. I would describe Onshape as true browser based CAD. I’m sure Onshape will meet with some success because it runs on everything and because of its excellent marketing. Besides being true browser based CAD, having some good ideas and what appears to be an excellent foundation, I don’t yet see anything that’s really radically new CAD wise in Onshape.

            I see Onshape as the beginning of the CADCAM market badly splintering. I’m not really a “designer” CAD guy. I’m much more of a manufacturing guy and sadly we are not considered to be anything but a minority by “designer” CAD people. After decades of manufacturing people having to take a back seat and given lip service by designer CAD guys, I believe it’s time for real change.

            I believe a major split in the market has recently started and what I like to call CAD For CAM is going to emerge very strongly.

            Jon Banquer
            CADCAM Technology Leaders group on LinkedIn

  9. scottmoyse Reply

    I just noticed how brilliant that feature image is! nice work.

  10. samscholes Reply

    Onshape is impressive but it’s also very limited. It has some great tools for basic part modeling and assembling modeling and is lacking functionality in a lot of key areas.

    – No drawings (yet)
    – No sheet metal
    – No weldments
    – No mold design
    – Limited surface modeling
    – No photo rendering
    – No Tool Box or libraries
    – Etc, etc.

    No doubt, Onshape is a cool product and it offers a lot of promise. It’s exciting. But it also has a LONG way to go before it rivals the power and functionality of SOLIDWORKS and similar 3D CAD products.

    • Sam sells legacy code SolidWorks for a living.

      He works for GoEngineer.

      I get a very strong sense of lack of objectivity as well as fear of new and better technology than legacy code SolidWorks has.

      Jon Banquer
      CADCAM Technology Leaders group on LinkedIn

      • samscholes Reply

        I’m afraid you’re just an Internet troll Jon with a lack of perspective yourself.

        • Everything I said about who you work for and what you bias is, is true. I strongly suggest you identify yourself as working for a SolidWorks reseller in every comment you make as this way others will realize why you post what you post about your competitors.

          Jon Banquer
          CADCAM Technology Leaders group on LinkedIn

          • This branch of the thread ends here, guys. No more.

  11. Dries Vervoort Reply

    Regarding Fusion 360 vs. Onshape…

    To me, Fusion 360 isn’t necessarily better value than Onshape.

    1st argument:
    I think feature richness doesn’t say it all. Onshape immediately clicked
    with me. Fusion 360 didn’t and doesn’t. There’s something about the
    Onshape UI that makes it incredibly intuitive and accessible. I think
    this is key for adoption. Also, it is extremely similar to SolidWorks in
    a lot of ways. For this reason alone, I think Onshape will attract more
    SW users than Fusion will.

    2nd argument:
    No matter what Autodesk says (“our customers wanted offline”), there’s
    no denying that being able to access Onshape from any platform/any
    browser/… is very powerful. Maybe if Autodesk do their research again
    in 2015 the outcome would be very different? The Times They Are A-Changin…

    3rd argument:
    Onshape has been rock-solid since day 1. Can the same be said about Fusion? How valuable is stability?

    4th argument:
    True, Autodesk has all tech in-house. However, I expect Onshape to be a
    much more open platform for partner implementations. They seem to have a
    very strong emphasis on API development. I think this, too, will be a
    driver for success. Just as it has been for SolidWorks.

    Just saying there’s a lot more to value than the list of features for a
    given price.
    Value can mean very different things to a designer, QA dep., product

    In the end, I think Fusion 360 and Onshape will both be winners. They
    are clearly leading the way, leaving others clueless, stunned, with
    their head in the sand, backed into a corner…


    • Onshape will need Sub-D modeling. Not having Sub-D modeling puts Onshape at a major disadvantage for any designer who wants/needs to design aesthetically pleasing shapes.

      Onshape not having their own CAM could be a major problem in the near future if Autodesk ever figures out the advantages of CAD and CAM sharing the same database like Open Mind has. Open Mind has hired ex-think3 employees to design their CAD For CAM. Proper CAD For CAM is going to revolutionize how toolpath engineers see integrated CAM. Up till now I would consider CAM running inside of SolidWorks to be lots of unfulfilled promises because it runs on top of a SolidWorks Assembly and CAM integrated inside the SolidWorks user interface doesn’t share the same CAD database. Not many toolpath engineers currently understand why CAD and CAM sharing the same database is critical but that’s going to change:

      It seems to me that the Onshape user interface is also going to need a ton of work. I definitely see Tab management as a major problem in Onshape that needs to be addressed right away.

      Autodesk Fusion 360 also has a lot of issues with its lousy user interface. The Fusion 360 user interface isn’t very innovative and its Timeline isn’t anywhere near powerful enough. The Fusion 360 Browser lacks basic functionality that’s well documented on the Autodesk Fusion 360 forum. I question whether you need both a Browser and a Timeline. The Fusion 360 Timeline could be innovative if it was made more powerful. As it stands it’s right now the Fusion 360 Timeline is pretty limited and not really much better to use than a history tree.

      Jon Banquer
      CADCAM Technology Leaders group on LinkedIn

  12. Charles Clarke Reply

    Adam, my greatest joy was teaching my daughter that engineering was not just for boys — you have many treats in store. Onshape is yet another point of inflexion on the CAD development curve — the old band’s still got it!!

      • Charles Clarke Reply

        I got her into McLaren for ‘work experience’ last summer and she hasn’t stopped smiling since!!

  13. Craig Rochester Reply

    This make my mentoring First Robotics teams CAD work so much easier! Up yours PTC.

  14. I fear this must be a joke. I hate the SW licensing model with a passion, and I desperately desperately what this to work … but, even when free, it doesn’t.

    I don’t live in the US. In fact I am in New Zealand. Yes we have electricity down here and broadband. We also have clean air and water.

    I have been designing computer systems for 30+ years and it quietly amuses me how the pendulum swings from mainframe to desktop back to central processing back to distributed processing.

    So I open my browser and load a sample file. The cool little model engine.

    And I wait. And I wait. And I wait. And I think, gee I should have timed this. No point doing it again because it has bound to have stored a local copy of the files in my cache so the second time will be much quicker. Won’t it? So I close the document and open it again. No! No quicker the second time. 3:52 later I get a list of Instances (19) on the left of my browser and still a blank screen. 5 MINUTES 14 SECONDS later, the document opens and the image appears. Completely and utterly impractical!!

    This is why no one I know seriously uses Google Docs. Or indeed any cloud software that I can think of. When it is hosted on my machine I am in control. I can blame no one but myself if I mess it up. But if I need it faster I can upgrade, or close background apps or something. When it is in the cloud I am completely powerless to do anything.

    And so, I try to rotate the image, or zoom with my mouse wheel, something that would be near instant with desktop cad. And my screen hiccups and staggers into a new orientation. Quite disconcerting and disorientating. Not even being able to control the model in space.

    I desperately want this to work, but it doesn’t when the servers are on the other side of the world, or perhaps the servers are just busy or perhaps the ISP is having a bad day or perhaps …. who knows? I can’t control anything that is happening.

    I feel sorry for these guys, they have clearly made some really nice software and they are trying to do the right thing. But somewhere along the line someone sold them a lemon. It was labelled “Cloud Based …” and had bells on it and they said it was the future. The Venture Capitalists love it. The media love it. The business analysts love it. Maybe even the students will love it.

    Sadly the end user is left out of the mutual group hug and ends up suffering the consequences. A business tool, this isn’t.

    • Hey Kram!

      Thanks for the review. Definitely hit the guys up and let them know the issues you’re having. They’re really responsive.

      I do suspect part of your problem is the distance from their servers. Over here it’s lightning quick.

      I use Google Docs every day too, though. In fact, pretty much all of my data is stored in the cloud one way or another (Dropbox,, gDrive, GrabCAD, etc), so these days it’s not much of an issue for me.

      My internet is down less often than my computers. How many times have you lost data to a hard drive crash, a bad motherboard, yadda yadda?

      • Hi Adam,

        I know I am probably the consumer from hell, but here’s the thing. 10 or 15 years ago our business purchased CAD software for electronic design. We built products using this software. We developed part libraries. We integrated it into our manufacturing and ERP systems. It is stitched into our manufacturing robots. We stopped paying support for it yonks ago. It is not perfect, but we know the bugs and we work around them. We made the decision NOT to upgrade because we couldn’t afford to “learn” the new bugs. For better or worse, it is ingrained into our business.

        So … what happens if we embrace any cloud solution? Doesn’t have to be Onshape. We build up a business and products and libraries around this software. Now we are assured we will never lose the data. Great. But what if …. what if next year Onshape thinks “gee, that didn’t work, let’s go do something else”. And suddenly we have no basis to run our business. No “software” any more, because we never owned it, we just rented it. Suddenly ALL our models become obsolete overnight.

        Unless we can keep local copies in a format that is universal, as a back up, in case the unthinkable happened.

        The thing is, the unthinkable will happen. It might not be Onshape. They might go from strength to strength. But someone is going to back the wrong horse. And then their customers are going to curse the “cloud” model.

        For that not to be a picture of the future, there needs to be “open source” data exchange, so I can access my models on anyone’s software platform. Then you have a cloud future that might be viable.

        So, perversely, I would argue the cloud is fine for non mission critical applications (students fooling around making models) but a serious business risk for mission critical applications?


        • Let me be clear up front: there is currently no cloud solution that is anywhere near production-ready for an operation like yours. It’s just not there yet, and as far as that’s concerned, we’re in complete agreement.

          I think there’s real validity to what you’re saying in the longer term as well, though I think saying that cloud solutions are only good enough for “students fooling around” doesn’t quite capture it. Your business may well have perfectly legitimate reasons not to go cloud just yet, but much of the SMB market will. Not everyone has the same uptime requirements that you do.

          You’re of course correct that there is an inherent risk built into using a tool like this in a production environment, but I think you’re not giving enough thought to the seriousness of the limitations you accept by going with legacy-version boxed software. What happens if, for example, you want to add more seats to your legacy installation? You can’t buy the old stuff, so you have to buy newer stuff. Now your team is on a variety of different versions, each with their own problems.

          There are lots of problems with the cloud, but there are lots of problems with boxed software too. We’ve learned to work around those problems, so they feel safe and familiar. I’d argue that the problems with the cloud are not fundamentally worse, just different.

          The fear that a software company effectively has your entire operation by the short hairs is a very legitimate one, but I’d argue that that’s the case with boxed software too, especially for enterprise customers. Once a firm has invested millions in a full PTC, DS, or Siements PLM solution, extricating themselves from said solution is going to cost millions more. A certain leap of faith is intrinsic to the way these large, integrated platforms work.

          So I’m with you, Kram. These tools are not ready for prime time, not by a long shot. But the industry trajectory is clear, IMHO, and in ten years’ time I expect this whole debate will seem a bit quaint.

  15. OrangeMonkey Reply

    Who in their right minds would tie their business productivity to their internet connection? I mean, I have been in software development for years and have worked remote myself as well as with many remotely connected groups. We would never ever even for a brief nanosecond ever think of using a set of software development tools which would stop the developer’s progress if his internet connection were lost. That would be absolute insanity. So, how are these tools different? I don’t see that they will work on the designer’s box without an internet connection, so if an asteroid, or China, hits one of our satellites, how many businesses are going to be brought to their knees due to stupid software like this? I cannot imagine using even the free version of this stuff for this reason. Why get addicted to something that will eventually kill your business? I guess I am married to the big beige box concept. And for good reason.

  16. I was intrigued by the Power of Fusion 360 at first glance. However, I live in a place where all I can get is DSL, and Fusion 360 left me incredibly frustrated – waiting and waiting and waiting. I’ve gone back to DesignSpark Mechanical – the best truly FREE 3D CAD design tool that I’ve found for rapid prototyping. If not for the missing loft/blend function, I would have had no motivation to try anything else at all.

    Blab about how great “the cloud” is all you want. For me, it’s useless hype.

    • That sounds like a great story. The PocketNC is just adorable. Was a bit out of my price range, but would be so cool to have. You already have yours?

      • Rebel Razer Reply

        I did get my Pocket NC the middle of December. I’ve done some test cuts and it’s really nice. Very well made machine for the money. I’m designing a bear head necklace in Fusion 360 and using Sculpt with Lofts and sweeps. I’m about 80% complete and it was a stiff learning curve getting it to do what I need. I plan to machine a wax mold and use PMC3 Silver Clay to form the necklace. It should be pretty stellar when it’s done. Dang a girlfriend that likes bears is costing me about $4000+ πŸ™‚

  17. Can Onshape be used to generate 3D models for 3D printers?

    If it can, what product would be used to change it into G codes for the 3D printers?

    • Josh Mings Reply

      Yes. I would export, then import to a slicer/g-code software like Slic3r.

Write A Comment