Hacker News new | comments | show | ask | jobs | submit login
AutoCAD Coming to Mac (and iOS) for First Time in Nearly 20 Years (nytimes.com)
97 points by gphil on Aug 31, 2010 | hide | past | web | favorite | 73 comments

A step in the right direction, I suppose. But as a hacker and an architecture grad student, no industry infuriates me more than the architectural software industry.

If you're looking for a new startup idea and you can build a cross-platform 3D modeler that:

    * is compatible with existing file formats
    * can cut real-time 2D plans and sections
    * provides a well documented API in multiple modern languages (please no more BASIC)
I will hand you $1,000 right now. And that's coming from someone who has $3,000 in the bank thanks only to his latest loan check.

Seriously, I think there is real money to be made here.

Incidentally I've been working on a toy BIM software package for the last few months, just for fun. The amount of work to make anything that even remotely resembles production quality is staggering. The problem with this software is that it's hard to build incrementally. You need all the functionality to design 90% of all houses or people won't buy it. It needs 3d for the design + renderings and 2d for the cutouts / details. And you're asking for an API - that would be at the bottom of an implementor's list after all the basic functionality! Plus the cross-platform thing, which doesn't make software companies any money either (are you talking just Mac or also Linux?)

But yes there is real money to be made, and Autodesk is reaping it in. There are contenders for the small time market like Chief Architect but they're not bleeding money it seems. This is a winner takes all market and Autodesk has a 30 year advantage.

Awesome! As I replied further down, I think we're at different ends of the industry. BIM needs are amazingly complex, and you're right about all of the functionality that is needed.

But for my end -- designers that depend almost entirely on Rhino or similar products -- those functionality needs aren't there.

But yes, we then have to have our designs "reimplemented" as construction documents if they are going to be built. It would be great if I could work solely in a BIM package, but none of them come close to meeting my (or most progressive designers') needs yet.

What if you inverted the product idea and started with an online object repository that was presented as an API for search, retrieval and instantiation. You would include the basic design methods for parametric designs in your API, but you wouldn't be directly challenging AutoCAD's desktop business. It seems like the missing piece these days is not the design tools, but the marketplace and the process of taking a design and validating and manufacturing it. If you could network a number of machine shops of varied capabilities and material specialties into a single point of contact for rapid assembly of completed designs with a marketplace for designers to sell their work, that could be a powerful combination.

There are already a few companies dabbling in this space http://www.shapeways.com/ being the one I'm most familiar with. The most interesting approach would be if you could find a market niche that had idle capacity and a need for creative custom solutions that was currently going unmet.

I'm not sure how that would work, but anyway I was talking about architecture and CAD for that only. How would you design something in your proposal? By writing code to glue together a bunch of objects? You need WYSIWYG editing, otherwise people could just write povray scripts and do everything locally. Maybe I'm just not understanding what you're proposing?

I'm thinking of something that acts as a catalog of pre-designed parts, that you can import into autocad, CATIA or even blender or maya. You could probably even put together a web-based interface to the geometry. It would act to some extent as a translator between different formats (.dwg,.stl,.vrml97, etc.) but the main purpose would be to act as a marketplace for designers and production bureaus. The ideal use case would be a construction project manager who's been tasked with remodeling a space on a tight time budget and shops for custom fitted design elements purchasing several designs from different sources and ordering production from one or more production bureaus within a 1-day delivery radius.

The real chicken and egg problem with this idea is that you won't have enough designers and qualified production bureaus to form enough of a critical mass of purchasers until you have a critical mass of purchasers to make the service attractive to designers and production shops. It would probably work best if you start off with a small regional network of production houses that employed their own designers.

Ah I see, yes this is also part of the 'BIM grand vision'. It's in the final stages of implementing an industry-wide system though. This would be hard to do without a solid toolchain in the beginning of the process, with a client to visualize how those parts look/work together. I don't see this as a replacement for 'traditional' tools, more like an add-on.

Big manufacturers already provide construction blocks, in standardized formats, for importing into design tools. From ICF forms to wallpapers. The 'marketplace' part is missing though, that's true. I don't even think it's a matter of technology, it's just that the industry isn't ready for it IMO. Construction is a very conservative industry.

If I were a betting person, I'd guess that the 80/20 rule/myth applies here: http://joelonsoftware.com/articles/fog0000000020.html . Everyone uses 80% of the features, and then another couple of pet features they CAN'T LIVE WITHOUT.

Speaking of Basic, didn't AutoCAD use Lisp for "macros"?

It still does. As a younger programmer, it was my first taste of Lisp. Still, it would be nice to have more language choices.

Up until the mid 1990's yes. Then they switched to Visual Basic and C++ which were considered easier to integrate with the Windows environment or something.

I'm still using AutoLisp and I don't think our version of AutoCAD is that old?

ISTR that it was R14 that was all c++/VBA but it's been years since I touched autocad.

autolisp is still around, with VBA going the way of the dodo in favour of .NET

Hell yes. The same is true for the mechanical CAD market. Autodesk (the makers of AutoCAD) and Dassault (the makers of Solidworks) are intensely irritating to deal with. Their pricing is obscured, their marketing is heavy-handed, and for the price, their software is not that good. If I were starting a new software company now, these are the clowns I'd love to compete against. There are literally decades of pissed-off customers who want something better, but are stuck with these dinosaurs.

(Yes, Blender, Sketchup, and so forth. Not tailored for the industry well enough.)

I remember hearing that if you called up Autodesk and asked them if they were working on a Mac/Linux client, they would literally laugh at you.

Sounds like you are describing Revit, apart from the cross platform thing.

- can link in/import a number of existing file formats - can cut sections as you described - API is well documented and in .NET which lets you program in C#, F#, VB, C++ and any other language which runs on the CLR.

Great product, which improves dramatically in each release (so if you only looked at it a few versions ago, try again). It will cost you more than $1000 though!

I'm not familiar with 3D modelers, so could you explain what you mean by "cut real-time 2D plans and sections"?

Architectural plans are drawn as a series of 2d plans: top-down views ('overviews' or 'blueprints', like http://www.the-house-plans-guide.com/images/floor-plan2.gif) + sections (colloquial for 'cross sections', a 'view' on the inside of a building, like http://www.the-house-plans-guide.com/images/cross-section.gi... .

The 'traditional' way of CAD for architecture is to draw a bunch of lines and circles, much like you'd draw a plan by hand, but on a computer. This is very error prone and labor intensive. Your client asks you remove a window, you need to remove it in several views etc.

The holy grail of architectural CAD is in a part of what is loosely called 'BIM', Building Information Modeling. You make a 3d model of a building and generate separate views on that, like cross sections and plans, but also use that same model to do e.g. earthquake simulations or energy efficiency analysis or cost estimating. Autodesk has a product called Revit that purports to do this, but it has a long way to go to really fulfill the 'BIM dream'.

This is a subject near and dear to my heart - working with architects and contractors who run around with A0 format paper copies of plans, who do cost estimates by measuring, with a rule (!) each room on the paper plan and calculate areas with a calculator on a piece of paper, it makes me die a little bit inside each time I get a quote and I see the areas or dimensions or whatever different for each fricking ocntractor, instead of using software that would do pretty much everything that they just spend a day on in a few seconds.

Most 3D modelers are solid modelers (i.e. I construct a solid cube or sphere). When I aggregate many of these solids together (and possibly do some boolean operations to them), I have a design.

Now I would like my 2D cuts. By this, I mean I would like to look at a 2D cross-sectional drawing of my design. This may be at the XY plane, YZ plane, or some other arbitrary plane.

Most 3D modelers can do this, but when you make the 2D cut, you're often exporting the cut to a new file. If you change anything in the design, the cut does not update, and you must "re-cut". This is what I mean by real-time.

Good 3D modelers don't do this well, and poor 3D modelers can do this exceptionally (I'm looking at you Revit). Some can do both, but they're incredibly complicated and cost tens of thousands (Digital Project)

But for architectural design, you want to work at a higher level than solids. You want to draw a 'wall' and then parameterize it - 8cm of outside brick, 2 cm of air, 5 cm of insulation, 14 cm of brick, 2 cm of plaster. Plus this needs to have the right hatch in 2d views etc.

Have you tried Chief Architect? It works rather well in this respect, it's really made for ease of use. I don't know how well it would scale for large projects though, I've only toyed with it for small residential design. Plus the estimating and electrical/HVAC tools are poor to non-existent.

I'm taking the view of an avant-garde designer in a firm that has a separate team of drafters to implement designs and construction documents. My focus is on the conceptual design phase, so I simply have different needs.

There are a number of programs, but most designers I know prefer Rhino (rhino3d.com) and the Grasshopper (grasshopper3d.com) plugin. Rhino is now adding Python (and OSX) support, but they still have a ways to go.

Never mind my other reply, I hadn't seen your response above. I had a quick look at Rhino, I hadn't heard about it before. I'm a bit surprised that you work with the same tools as e.g. an industrial designer, although when I think of it I can understand, with the lack of expressiveness in the specialized packages and all.

But do you do that high-level design in e.g. Solidworks? I don't have any experience with that side of the industry, I'm just curious and a bit surprised.

(I'm not the OP.)

"plans" means 2D overhead views of a 3D model. "sections" means 2D side views of a 3D model, like a cross-section. I don't know what "real-time" means here, but I'd bet that it means you can grab onto the plane that defines the cross-sectional cut, move it in the normal direction, and have the model update smoothly. (The non-real-time alternative is that the location of the plane is specified by a value in some dialog box.)

$1000 would barely cover 25% of the price of a current seat of AutoCAD.

Good call, maybe I should have gone to the source. I just linked to the article I came across myself.

There is a link to the NYT article in the article you submitted. The article you submitted seems to be nothing more than a rewording of the NYT article.

What possible use case is there for autocad on the iPhone? This seems unnecessarily silly to me.

Unless I missed something from that article...Is this just a client for viewing renderings or something? I cannot imagine a designer sitting down at an iPad to do any real work... (although as a sort of digital paper, it works)

From the NYT article (the source of this blog spam):

"Autodesk will soon introduce a free mobile version of the software that will run on the iPad, iPhone and iPod Touch. That version has more limited capabilities, Mr. Hanspal said. But with it, an engineer, for example, could bring drawings to a job site on an iPad, rather than on a big roll of paper, and make annotations on them."

It will probably be something similar to Autodesk Design Review:


isn't the ipad screen very reflective and hard to read in sunlight (like that which would be on a jobsite)?

In practice it's not that big a deal. Sometimes you have to tilt the screen or shade it.

But that's more than made up for in the added usefulness. For reference, i have a far harder time reading an HTC Incredible in the sun.

The main use case is that you've been trapped in an insane asylum, for various reasons we won't get into, with nothing but an iPhone and access to the Autocad iPhone app, and you really want to impress the administrators with your pitch for a new swimming atrium on the south wing.

There's many cases where you might not want to create technical drawings but might want access to them.

I worked on a project years ago for the US Navy. At the time the average warship had something like 5 tons of paper documentation on board so they can do wartime repairs at sea if necessary. But 5-tons of paper docs are risky on a warship since they're combustable and easily destroyed.

Putting all these docs in digital form reduces the weight and risk (doesn't burn and easily replicated). Having that information available via an iPad, iPhone or other portable tablet device would be very handy, indeed.

From my understanding the iOS versions of AutoCad are in no way meant to serve as a replacement for the desktop version. But rather a very simplified version of the tools and features available.

As a designer, I can imagine being out for coffee or dinner with a potential client and not having my Macbook with me. Being able to pull out my iPhone or iPad and do a quick and dirty sketch of my thoughts/ideas to better convey the message could be incredibly useful. Much better than a standard cocktail napkin.

AutoCad is seeking to simply fill a void in a complete cross-hardware line. This addition will enable users to "CAD" wherever and whenever within certain constraints. These applications will work seamlessly together - so a quick sketch on your iPhone can later be redrawn and upgraded on your desktop.

Architects will love using this on the iPad once the construction phase begins on a design and they have to do their Punch Lists.

It certainly could be great as a viewer- an interactive blueprint- but are you suggesting they will do actual design on the iPad?

No, I don't think there is much value during the design phase. Punch Lists are created when the building is near completion. The architect walks around the building, taking notes of what's not acceptable, and sends photographs + descriptions of an issue to the other contractors. With a portable version of AutoCAD this process could be made more efficient by allowing the architect to enter notes in the drawing file while at the site (and maybe attach images to it taken with iPad with the help of a macro or some other add-on).

No, but AutoCAD isn't used by designers anyway.

I know several designers who use AutoCAD (and other CAD programs). There are more types of design than doing web pages you know.

Go to any medium to large architecture firm.

You will find 20% of the employees doing "big idea" design. They don't use AutoCAD.

You will find 80% of the employees (who wish they were the other 20%) implementing the designers "big ideas" in AutoCAD.

But by people who design something. Using the verb is entirely appropriate.

It's to save on trying to look at an A0 drawing outside on a windy day :-)

Or worse, somebody on site redlines an A0 drawing, the changes doesn't get back to the office and 18months later you are trying to fit an elevator into a shaft that is 2" too small in a finished sky scraper.

As mentioned at the end of the article, "The NYT reports that the idea behind the mobile version is to allow an engineer to bring digital drawings to a job site and make notations on them from the field."

As an architect, what would be really helpful is a great pdf viewer with enough speed and storage capacity to hold large sets of drawings (which can approach 50 to 100 megs). The way to do it, of course, is via a webapp so you don't have to locally store the data. Having autocad on an iphone is silly as it would be radically difficult to try to use a cad package with such a limited screen. But being able to mark up drawings on site - that is useful - with a red pencil and text, etc.

Besides supplemental uses on the current iOS devices, if there were to be an iMac with touch down the road, this would be a good first step getting ready for it.

I know AutoCAD is 2D, but can you imagine how awesome Autodesk Inventor or Solidworks on iPhone would be? They could make use of the Gyro to rotate the 3D model on screen. That way, if you modeled a building, you could put the camera 'inside' the building and look around just by moving and turning the device. You could completely visualize and walk through your house before building it

Ironically I first used Autocad on SUN Sparcs. We bought a Sparc 5 just for Autocad because PCs of the day (386) couldn't really handle the graphics needed.

Then they dropped Unix support, then dropped OpenGL support.

Back in the day Sparc was the development platform for AutoCAD and the DOS, Mac and other UNIX versions were ports.

Once Windows took off in the marketplace the writing was on the wall and the anemic sales of the Unix ports in contrast to Windows made keeping those ports unprofitable.

AutoCad is a dying technology - the real thrust is toward Building Information Modeling away from 20 year old 2d drafting software. When Autodesk's Revit is available on the Mac, that will be an impressive announcement.

Is this endgame Linux ? For all intents and purposes, this is the market that Linux should have. High performance *nix, highly stable, can run huge programs on it.

Coming from the silicon IC design industry, there are huge IC layout tools which run exclusively on Linux. They work a lot similar to AutoCAD-like software - they do a lot of automatic layout, place-and-route, 3D design, etc.

These tools run for weeks at a time on linux machines and are highly optimized for Linux.

That existing market could have been so easily translated into software like Autocad for Linux. But something went wrong somewhere - probably the OS graphics API was not good enough (a frequent enough complaint for Linux)... I really dont know what.

But even after a proven 4.4 billion dollar market (EDA industry market cap) for Linux based high-performance design tools, Autocad went after Mac. Knowing fully well that TCO for a Linux based system is far, far lower than a Mac based.

It's also possible that Apple offered to help them. If Apple has been putting together dev tools to help track down Windows dependencies.

On the other hand Maya already runs on Linux and Mac OS X. So they may be using Maya numbers to calculate sales and difficulty porting. Since it's used for video work you'd expect more Mac sales.

Mac is a more commercial software friendly platform. While Apple isn't quite as friendly as Microsoft, they're going to do a lot more to keep things stable and friendly for developers than the wild wild west that is graphics and/or GUI-intensive development on Linux.

Most commercial software for Linux is standardized for RHEL or UBuntu LTS platforms.

I think stability is not the main reason - it may be perceived market or the graphics API problems that I was talking about (especially with third party drivers - nvidia and ati)

>Autodesk will soon introduce a free mobile version of the software that will run on the iPad, iPhone and iPod Touch.

Why do I get the feeling that that was the motivation for bringing it to OSX?

Touch interface and CAD has been a powerful combination for a long time. When I work in simulation, Wacom tablets (while not traditional touch) where used extensively in both 2D and 3D work. I think the iPad is the huge motivator for the porting work. If they don't get there they could loose a whole market segment to an upstart.

I agree AutoCAD might be hard to use on a tiny screen per se, but the more important message here I guess, is that 3D object modeling/manipulation capability on mobile device (hardware and software) has come closer to what a desktop could offer to do something slightly more sophisticated than just bubble drawing and lame gaming.

Also, with the introduction of OpenGL ES 2.0 support in Android earlier this year, there should be no surprise the Android version should also come out very soon.

Autocad has advocated DirectX over OpenGL for years, and OpenGL support has been removed from Autocad 2010 all together. The OpenGL support in earlier versions was emulated on top of DirectX. See for example this paper from an Autodesk engineer: http://archicad-talk.graphisoft.com//files/autodesk_inventor... .

So I don't know how they're implementing this, I hope they didn't make some poor sod reimplement the whole rendering engine or worse yet write a new DirectX -> OpenGL layer that works in a Java stack...

I think they have to be using OpenGL, as they are "coming back to Mac" and AFAIK directx is no where to be found on Mac. Also with the simultaneous release for both the Mac and the iOS version, I'd think there must be lots of reusable code around OpenGL that'd made the porting to android not an extremely daunting task.

Yes I expressed myself wrong, they must use OpenGL. I assume they have different rendering engines as the paper I linked to mentions an option for software-OpenGL (MESA-like). Also my 'OpenGL is emulated on top of DirectX' is wrong it seems from re-reading that paper, I misremembered.

Will be interesting to see what variant of AutoCAD is released.

My Dad is an architect, very long-term AutoCAD veteran, long term disliker of Windows. IIRC the previous cross-platform AutoCAD was the stripped down LT (is that right?) version. Useful for some things but not a viable choice for heavy AutoCAD users at the time.

Does this mean AutoCAD is going to be shipping with Mono? Quite a bit of AutoCAD functionality is now written in .NET.

As a .NET developer for Autodesk Revit (and soon to be Autocad) I am interested to see if they are going to make us rewrite our applications in a different language to be compatible with the Mac.

Unlikely Acad is heavily DirectX

No. Recent version of AutoCAD are customizable via .NET and core libraries that ship as part of AutoCAD are written in .NET too. If you have AutoCAD installed look at HKEY_LOCAL_MACHINE\SOFTWARE\Autodesk\AutoCAD\R18.1\ACAD-9007:409\Applications, all of the sub entries that have a entry MANAGED of Type REG_DWORD with a value of 1 are .NET extensions. With that being said they either have to ship Mono with the Mac version or rewrite a ton of code in C++.

And please don't forget, Autocad was originally written in Lisp :)

NO. AutoLisp has been used only at scripting level

Actually, it was written in C.

Why? The other choice was Pascal and that required swapping floppy disks to do the compile on a dual floppy PC: http://www.fourmilab.ch/autofile/www/chapter2_112.html#11286

AutoLisp was added on in a later (albeit pretty early) version.

Guidelines | FAQ | Support | API | Security | Lists | Bookmarklet | Legal | Apply to YC | Contact