I have been working on a few dynamo scripts that almost work as intended utilizing Generative Design for Revit however I am worried to put too much more effort into this should there be something else happening behind the scenes with Autodesk that will be more applicable.
I will be interested to hear the answer to this also.
Personally I feel generative design in general struggled to satisfy mainstream expectations as it does tend to suit complex tasks (simple problems tend to have obvious solutions). Like many hype cycles in AEC I feel it was oversold by many software platforms just how much work goes into codint a GD solution (it’s by no means a task for a noncoder).
GD was and continues to be a broader field regardless of Autodesk’s tool in practices though.
That’s a shame to hear Lili moved on. If i recall correctly she was at Autodesk for a long time!
Ok, so I am not on the product team so I know not much more than anyone else even though I work for Autodesk, and I am not a spokes person or someone who can speak to roadmaps directly. And so don’t let my words sway any investment or purchasing decisions. You want to buy your software based on the capabilities found today nkt what might happen tomorrow.
But with that said I can give my personal opinion and some insights which I do have.
No and no. The core technology is still a must for so many aspects of AEC tech today, the Revit add-in wrapper is such a tiny bit of the larger process.
In my opinion there isn’t much need for the tool itself to improve, but the aspects of leading up to the use of the tool are where the issues lay. Not in the GD UI or optimization tools.
Lilli left for her own reasons, not related to GD’s future. Same can be said of every PM I know who has left Autodesk oddly enough. Plus she was also the PM for Dynamo Player, and that tool is not going anywhere. Drawing a parallel to Ian leaving ages ago, yet Dynamo is still going strong. The AEC tools produced today are not built by a single person or two like AuroCAD was back in the 80’s, and the talent on the team is significant.
And some not so good ones, many of which have been disproved by users in the world.
I think forma is a a great fit for Generatove Deskgn. There was an AMAZING AU session outlining the early Dynamo for Forma beta at the very start of the conference (I may be impartial though).
Reading into any bit of marketing will get you a big headache. The ‘big moves’ are what get the attention, and stuff like bring GD closer to other platforms (Civil 3D, Forma, FormIt, etc.) will have a LOT of benefit to end users.
Ok, enough with responding to bullet points.
if you ask me, the beauty of Generative Design is that it allows computational designers to make informed decisions regardless of the platform they are in. If you look back on my work in this spaced I have always advocated for leveraging Generative Design in context well beyond Revit. To me this was always the end goal for the tool, but it has to start somewhere in the market. And as Dynamo is the backbone of GD, and the largest user base of GD was Revit it made sense to start there. And users felt the same - I get asked once a month about Generative Design coming to (insert product name here) formally. It’s one of the most frequent questions I get, and I get a LOT of questions.
So to me the next logical step is to push stuff into the other platforms with a Dynamo integration - Civil 3D, Alias, FormIt, Forma, Robot, Advance Steel, and of course Dynamo Core (there is a log-in now after all). While that is under way (it is a LOT of work after all), the team can collect feedback and decide what comes next.
Rhino and its zoo has technically provided a majority of generative design for 14 years since Galapagos landed in 2010. Wallacei and Octopus see a good amount of use in major architectural practices. A lot of legacy still functioning and extending there.
One could go as far to argue that GD for Revit heavily draws on Wallacei UX, which probably draws on other solvers in and out of industry. No one company has the right to stake GD as theirs I think.
These firms may happen to use Revit, but whether that translates to GD in Revit I’m unsure of.
Hello @SteveDFT - My answers to your questions as follows, expanding upon some of the good thoughts from both @jacob.small and @GavinCrump above
Generative Design in Revit, as an expression of the concept of Generative Design that uses genetic solvers to work towards an optimal design largely has it’s core competency complete.
Does that mean we can still improve it? Definitely! But is the return on investment there greater than that of the other, I would argue bigger, blocker to Generative Design uptake - Building a successful graph to run? No.
We are putting calories into making the Core Dynamo experience better to enhance all Dynamo users lives easier, inclusive of those building graphs for optimization inside of Generative Design.
Lilli had been at Autodesk for 21 years… a long time by any count! She reached a point in her life where she wanted to try something new, as we all do from time to time. This is not a reflection in any way around Autodesk’s desire to continue to invest in Computational and Generative tools.
I’m curious what you mean here, specifically - More advanced AI exists, sure. But it’s somewhat inverted to Generative Design in that you can start and get to a result extremely quickly, but getting the result you want is harder. Whereas Generative Design is harder to start with, but gets you to highly quantifiable results.
As mentioned above, we see the immediate steps in unlocking Generative Design better, and to more people is in improving the actual writing of the scripts themselves rather than UI features. We’ll still put some calories into that expression but the immediate focus is in dramatically lowering the barrier to entry inside of Dynamo, seeking to allow you to build graphs at much faster speeds and to a much higher fidelity, to enhance your ability to create Generative Design based graphs.
Generative Design in Revit (And in time elsewhere too) is still an active part of our portfolio of Computational and Automation tools alongside Dynamo and Dynamo Player. We have over the past year spent considerable effort on the back-end making it easier to integrate and merging it with the Player CLI runner that will enable us to put it in even more places beyond Revit.
Essentially the Player and GD back-ends are one and the same now, with different expressions in the UI. This is in service of the Ecosystem of Dynamo tools all working harmoniously together that lets us move faster and deliver greater value. Walk before you run kind of thing.
Largest user base of Dynamo was in the context of Revit.
As far as ‘user base’ is concerned, I’m guessing that a specialized tool outweighs both user bases by an order of magnitude. I’ve dug into history on this once a year or so, and every time the first architectural application of a genetic algorithm I have seen dates to the mid 90’s had either a stand alone app or an excel plug in… wasn’t specific intended for AEC but there are demos for optimizing stud walls (cost vs waste vs insulation vs construction time vs something else). If you told me onenof the alphabet soup companies had 1000’s of monthly users ebery year since 1998 I would’ve be surprised.
