@ufcfan: Thanks for the intel on comparable packages. I don't know this for sure, but perhaps some of capabilities from that other Autodesk program you mention formed the basis of Project Photofly -- who knows.
I think a lot of people will share your thoughts and reservations about putting design IP up in the cloud, but as it becomes more commonplace, I think those fears will dissipate, at least for some types of applications and some industries. You're right that many larger firms will have these kind of capabilities in-house. B ut there's a huge number that don't and an even bigger number of DIY makers (do-it-yourself), which Autodesk and many other CAD/3D printer vendors seem to be targeting. My bet is this is part of that effort.
Autodesk no longer produces new versions of the software, but this is very similar technology to their ImageModeler 2009 software. I use it and it's quite good software, the models become are more accurate with more pictures from different angles. Key difference is you have to match the similar points in the pictures manually with ImageModeler 2009.
There is also another program called Strata Foto 3D. Works well, requirements and function very close to Photofly. To me, Photofly is Autodesk's ImageModeler tech upgraded, and put on the cloud. What I don't understand is what is the need for the cloud in this application, honestly. Anyone who uses CAD/CAE/DCC software of this caliber have computers with the horsepower to do this themselves, and say it is a prototype - I wouldn't want proprietary information being sent/processed/stored in the cloud; for however long they say it will be. I think Autodesk should release this as either part of one of their packages, existing programs, or rerelease a standalone.
This strikes me as one of those technologies that starts out as fun, then finds a thousand different applications. I could see this having loads of consumer applications, as well as applications in industry for, say, reverse engineering.
You got it, and there are a bunch of other Software-as-a-Service-type capabilities in the works in Autodesk Labs. There's Project Cumulus, which leverages the cloud computing model to deliver more computational horsepower for MoldFlow plastic design simulations, and Project Centaur, which lets Inventor users offload simulation tasks to the cloud. I believe there are others as well. It seems like Autodesk is really moving forward with this concept.
The idea is a good one. You invest in Autodesk, then draw on its processing power to go beyond what could be done on your resident computer for particular tasks that are needed occasionally -- such as creating the virtual 3D model. Then you go back to your resident computer with the sophisticated image.
The pricing model would probably be similar to smartphones, where you buy the device, subscribe to the service, then add apps on a per-drink basis. There's a whole generation accustomed to that pricing model.
Most likely there will be some sort of pricing model like you suggest, Rob, but this is only a technology preview. Autodesk does this all the time, experimenting with technology being developed in Autodesk Labs and then making it available to the public to try out, comment on, suggest, etc. Pretty cool stuff, I agree.
I can see the ingenuity now. All it will take is for a PIC to be programed and attatched to a camera to snap the pictures at certain intervals. Then a pedestal to revolve, or the camera to revolve around the object, and voila! Instant 3d aquisition for the service.
Cool stuff. Since this involves using processing in the cloud, I would imagine this includes a service-based pricing model. For users, is this structured as a hybrid, where you buy Autodesk, then add on services such as the one you describe in the cloud on a per-use basis?
For industrial control applications, or even a simple assembly line, that machine can go almost 24/7 without a break. But what happens when the task is a little more complex? That’s where the “smart” machine would come in. The smart machine is one that has some simple (or complex in some cases) processing capability to be able to adapt to changing conditions. Such machines are suited for a host of applications, including automotive, aerospace, defense, medical, computers and electronics, telecommunications, consumer goods, and so on. This discussion will examine what’s possible with smart machines, and what tradeoffs need to be made to implement such a solution.