Apis Cor builds world's largest 3D-printed building in Dubai | Hacker News Reader

Apis Cor builds world's largest 3D-printed building in Dubai

dezeen.com

30 pointslelf19 comments

It's almost poetic how the robot uses one of the most primitive forms of building a 'hut': stacking layers of mud.

I believe a combination of printing prefab pieces at the site in a controlled environment could also be a great possibility.

Apis cor had some dealing with Samsung I remember. To me, they felt almost like marketing shop for their partners than a genuine engineering company.

When I voiced something about their engineering during an event, they weren't really able to reply with anything other than recital of their marketing.

IMHO, in its current iteration, 3d printed building are completely nonviable commercially.

The main point of 3d is to expensive, and error prone manual labour with a single pass automated process. 3D printed buildings as of now require more labour, more operations, and more materials than most conventional building techniques.

If we are still thinking of a house that has hidden wires in the walls and plumbing that is hidden in walls.... Then yes, printed homes as they are currently are not going to work.

If we embrace exposed conduit, exposed ducting and exposed plumbing lines, printed homes could be an order of magnitude cheaper.

Plus easier maintenance!

What about getting a house with conduits for hvac, electric, and plumbing, and still get it cheaper than 3d?

Construction technologies to make lowrise housing super duper cheap are already here: panels, lgsf, large prefabs, light concretes

And for really tall highrises, they are already coming within 50-40% of their materials costs in Asian countries. So for them, it will make no difference even in the most optimistic case.

You can't do much physically about price of steel and cement, other than finding ways to use less of them.

From the pictures it looks like there’s more than enough space for that in the walls, they are mostly empty.

But also, isn't great that someone is working on it? I am sure it will get better with time.

I don’t understand why these are considered notable. A building is a lot more than simply the concrete. In fact I would say that pouring the concrete is not that difficult or time-consuming. There’s a lot of work that comes before, such as excavation and compaction and grading. There’s a lot more work comes after to rough in plumbing, electrical, HVAC, etc. And then of course the finishing.

So what is the net benefit of this process?

As you can read in the article this is mostly a test case for new building methods. It's new, it will probably be widely used in the future so it is notable.

If you could own a $10000 dollar machine where you add concrete and a sd-card... then it printed a new room for your house, wouldn't that be revolutionary?

I don't know how much these machines will cost in the future, but I do know that I would love to have one. Construction is very expensive even for houses that are basically sticks and stucco. I would like to buy some land and print a house with modules for plumbing, electrical, etc. This enables a complete rethink of how we currently build homes.

I would note that this “machine” is 10% extrusion apparatus, and 90% a CNC crane. CNC + crane = very expensive (since you can’t just retrofit a regular crane; regular cranes efficiently rely on momentum, whereas this thing presumably needs to step linearly into place.) You could own the extrusion setup, but I don’t think you’d enjoy owning the crane.

Which is not to say you couldn’t get a hold of one. I’d guess that, when the tech matures, you’d be able to rent one from your local maker-space :)

For $10,000 you could easily hire someone to build a cement cinder block building. The high costs come from the other trades: electrical, plumbing, HVAC. Making all of these pieces modular would be revolutionary.

I think a prefab 4' x 8' section of steel frame and foam wall that could be bolted together with channels for interconnecting utilities would make truly plug-and-play style housing.

No. In that sense, you can already just give a guy at the Home Depot $500 to “add a new room to your house”. Are you living the revolution?

If construction is labor-intensive and expensive, stick framing is the cheapest and easiest part. If you come up with an automated solution that replaces the need for expensive interior finishing processes, you’ll have a huge market... finishing out existing stick-framed construction. No one selling houses wants this.

If you want to use modern technology to mass-produce homes with less labor, that’s called prefabricated construction and it’s a process that’s only about a thousand years old. Someday robots will do it all automatically on-site but we’re no where near doing that in the foreseeable future, in part because the demands on houses are predictably rising at pace with the capabilities of robots.

I don’t mean to be snide, but this is a little like dropping by a homebuilding forum and they’re all talking about how some day soon desktop PCB fabrication is going to replace all of Silicon Valley, because they don’t quite know the difference between a chip and a circuit board. It’s just embarrassing, y’all.

> If you come up with an automated solution that replaces the need for expensive interior finishing processes

There is already a large panel prefab process, which is highly automated, including finishing, but it doesn't seem to be anywhere near "revolutionising" the industry.

The thing is, in North America the preference for single family houses is so strong that making people care about extra $100k is really hard.

You cant even get a decent forklift for under $10K. Try $100K if said machine would be manufactured in volume, like medium sized excavators are today.

Although the work that comes before would certainly continue to be a separate process (though theoretically possible to automate more than we’re currently doing), the “part that comes after” could become something other robots reach in and do during, with modular lengths of wiring/pipework/venting placed by other robot arms in between the layers of concrete being added, their placement being part of the command-stream of the 3D model being printed. I would say 80-90% of the in-wall and through-wall work could be theoretically automated here, leaving only the fiddly surface work (faceplates, connecting fixtures) to human hands. It’s already done that way for the wall components of pre-fab homes; it just needs to be “generalized”!

Patents. Profit, hopefully, somewhere in the distant future. If you manage to make part of the process cheaper the real estate developers make more money and/or government can charge more for a permit.

I spoke to one of the founders to explore opportunities and partnerships with my professional network.

They are not consumer facing. They are looking for partnerships and investment.

Lots of cynicism in the comments here, but what the potential applications of a technology like this on the Moon or Mars?

Reply on news.ycombinator.com