4 ms·

As an emerging EE, I fully agree. I honestly believe it could be done with solid quality. Early today I saw this very old video of Elon at SpaceX with their des

As an emerging EE, I fully agree. I honestly believe it could be done with solid quality. Early today I saw this very old video of Elon at SpaceX with their design software [1]. It's amazing to me that this video was six years ago. Revit is the prized gem. Whomever creates it, focus on Revit and focus on the pain-points. [1] - https://www.youtube.com/watch?v=xNqs_S-zEBY https://www.youtube.com/watch?v=xNqs_S-zEBY

SenHeng 6y ago
I saw that Leap Motion [0] video almost 10 years ago and have wanted to have one ever since. Sadly, it seems they didn't sell enough to sustain theirselves and have mostly concluded their incredible journey [1]. I had a few friends that worked in 3D modelling them and they really wanted to try it out for sculpting. I guess actually swinging your hands around in air like Tom Cruise in Minority Report was too tiring for most people. I wish Apple had bought them and integrated them into the macbooks instead of the touchbar. [0]: https://en.wikipedia.org/wiki/Leap_Motion https://en.wikipedia.org/wiki/Leap_Motion [1]: https://www.ultraleap.com/product/leap-motion-controller/ https://www.ultraleap.com/product/leap-motion-controller/
- zingplex 6y ago
  The Leap Motion is still on sale. https://www.ultraleap.com/product/leap-motion-controller/ https://www.ultraleap.com/product/leap-motion-controller/ I had one when they first came out and while the tracking accuracy was quite good, it wasn't good enough for doing any precision work. Every so often the tracking would skip and glitch.
- johnwalkr 6y ago
  I remember being really annoyed at the reaction to that video. Here's 2 headlines from the first page of hits: "Elon Musk Waves His Hand and Designs Rocket Parts Out of Thin Air" and "Elon Musk to build rocket ship parts with hand movements through the air, and 3D printing". His tweet before that video was "We figured out how to design rocket parts just w hand movements through the air (seriously). Now need a high frame rate holograph generator. Will post video next week of designing a rocket part with hand gestures & then immediately printing it in titanium". That's really disingenuous. First of all, it looks like a typical demo made by an intern and some toys. Second, mechanical engineers don't need or want that type of interface. Actually manipulating and rotating CAD is such a small part of the process the time sink (aside from all the design and documentation aside from CAD) is adding dimensions and relations to geometry and relations. This video is the equivalent of a software developer's manager showing off a visual programming language and declaring text-based languages and keyboards are obsolete.
- vsareto 6y ago
  >try it out for sculpting. I bet the Index has potential for that with the finger control
- tootie 6y ago
  Leap Motion was recently acquired by a company that builds ultrasonic haptic devices. The standalone USB Leap Motion sensor is still available for like $80. I've done some work with it (and the haptic device) and it's extremely underwhelming. It's equivalent to voice dictation versus typing. It feels miraculous at first but there's a lot of edge cases (like hands going vertical or occlusion) and there's simply no good UX patterns for gestures. Keyboard and mouse is faster and more accurate in 99% of scenarios. Google actually one-upped Leap Motion with Project Soli that's in the new Pixel phones It can track minute finger gestures in a much smaller form factor using radar. It similarly doesn't have any killer app that actually makes it useful for much of anything.
StandardFuture 6y ago
As someone who was working in the physical engineering world when that video came out ... I am not convinced that design software specifically actually empowers engineers to be any faster at designing new and novel engineering (over pen and paper -- with isometric style drawings, etc.). Both software and paper design require a huge amount of training upfront and a huge amount of effort per project. Anything that is standardized (engineering-wise -- and many things are) does not actually need a designer working with design software. It needs a generation algorithm (a script) with input parameters. The biggest problem of why this area does not get more attention? We just don't build shit anymore. So, there is less money and opportunity than ever in it. Like really. We don't build shit. We need to fix corruption in politics long before we fix the lack of a proper marketplace to create incentive for maximizing the engineering potential of our society.
- fock 6y ago
  and next year, chinese companies have that cheap Autodesk-compatible-software out there (like the online-office featured recently on HN) and everyone jumps there, because it's sooo cheeeeap. It seems like communist-painted authoritarism finally wins.
- heimatau 6y ago
  > I am not convinced that design software specifically actually empowers engineers to be any faster at designing new and novel engineering I don't adhere to a luddite attitude. It's my opinion Autodesk's tools are deeply lacking and this is based on my experience and knowledge on software. When I first starting learning AutoCAD, I was very annoyed at the slow snail pace to designing things (as to pen/paper). I used to be a web designer/developer for a handful of years, those tools are much more refined and allow a person to do more with less time. Revit is a step in the right direction for the building industry but it has many drawbacks too. I strongly believe this is due to lack of competition and innovation on Autodesk's part. Adobe should be seen as the model. They've developed many tools that empower it's users and they continue to push the envelop forward. Autodesk will be unseated, if they don't continue to push their tools forward like Adobe has with many of it's products. The software isn't terribly hard, just needs a lot of investment upfront for a MVP competitor that chips away the market, then reinvest back into improving the tools with AI/DL/ML. I.e. One agency may use specifically a specific contractor, the AI recognizes revisions and highlights them to the architect. Or using specific materials. Etc. Etc. Construction industry is stuck in the 1940s, IMHO. Hopefully a competitor arises to bring it into the 21st century.
tootie 6y ago
Musk is espousing a UX fallacy by saying that a more "natural" interface is superior. Natural interfaces have a shallower learning curve for beginners, but CAD designers are all going to be expertly trained in their software. This is why pros always know the hot keys even when the mouse is simpler. Hot keys are faster. Learning how to pan, tilt and zoom with a mouse and keyboard is an 1 hour lesson that they will apply for 20 years and never worry about again.
- MiroF 6y ago
  I think that many commentators here on HN will be very unused to the market that Autodesk serves - mostly trained engineers who have been doing this stuff for decades, not so easy to "disrupt."

SenHeng 6y ago

I saw that Leap Motion [0] video almost 10 years ago and have wanted to have one ever since. Sadly, it seems they didn't sell enough to sustain theirselves and have mostly concluded their incredible journey [1]. I had a few friends that worked in 3D modelling them and they really wanted to try it out for sculpting. I guess actually swinging your hands around in air like Tom Cruise in Minority Report was too tiring for most people. I wish Apple had bought them and integrated them into the macbooks instead of the touchbar. [0]: https://en.wikipedia.org/wiki/Leap_Motion https://en.wikipedia.org/wiki/Leap_Motion [1]: https://www.ultraleap.com/product/leap-motion-controller/ https://www.ultraleap.com/product/leap-motion-controller/

zingplex 6y ago
The Leap Motion is still on sale. https://www.ultraleap.com/product/leap-motion-controller/ https://www.ultraleap.com/product/leap-motion-controller/ I had one when they first came out and while the tracking accuracy was quite good, it wasn't good enough for doing any precision work. Every so often the tracking would skip and glitch.
johnwalkr 6y ago
I remember being really annoyed at the reaction to that video. Here's 2 headlines from the first page of hits: "Elon Musk Waves His Hand and Designs Rocket Parts Out of Thin Air" and "Elon Musk to build rocket ship parts with hand movements through the air, and 3D printing". His tweet before that video was "We figured out how to design rocket parts just w hand movements through the air (seriously). Now need a high frame rate holograph generator. Will post video next week of designing a rocket part with hand gestures & then immediately printing it in titanium". That's really disingenuous. First of all, it looks like a typical demo made by an intern and some toys. Second, mechanical engineers don't need or want that type of interface. Actually manipulating and rotating CAD is such a small part of the process the time sink (aside from all the design and documentation aside from CAD) is adding dimensions and relations to geometry and relations. This video is the equivalent of a software developer's manager showing off a visual programming language and declaring text-based languages and keyboards are obsolete.
vsareto 6y ago
>try it out for sculpting. I bet the Index has potential for that with the finger control
tootie 6y ago
Leap Motion was recently acquired by a company that builds ultrasonic haptic devices. The standalone USB Leap Motion sensor is still available for like $80. I've done some work with it (and the haptic device) and it's extremely underwhelming. It's equivalent to voice dictation versus typing. It feels miraculous at first but there's a lot of edge cases (like hands going vertical or occlusion) and there's simply no good UX patterns for gestures. Keyboard and mouse is faster and more accurate in 99% of scenarios. Google actually one-upped Leap Motion with Project Soli that's in the new Pixel phones It can track minute finger gestures in a much smaller form factor using radar. It similarly doesn't have any killer app that actually makes it useful for much of anything.

StandardFuture 6y ago

As someone who was working in the physical engineering world when that video came out ... I am not convinced that design software specifically actually empowers engineers to be any faster at designing new and novel engineering (over pen and paper -- with isometric style drawings, etc.). Both software and paper design require a huge amount of training upfront and a huge amount of effort per project. Anything that is standardized (engineering-wise -- and many things are) does not actually need a designer working with design software. It needs a generation algorithm (a script) with input parameters. The biggest problem of why this area does not get more attention? We just don't build shit anymore. So, there is less money and opportunity than ever in it. Like really. We don't build shit. We need to fix corruption in politics long before we fix the lack of a proper marketplace to create incentive for maximizing the engineering potential of our society.

fock 6y ago
and next year, chinese companies have that cheap Autodesk-compatible-software out there (like the online-office featured recently on HN) and everyone jumps there, because it's sooo cheeeeap. It seems like communist-painted authoritarism finally wins.
heimatau 6y ago
> I am not convinced that design software specifically actually empowers engineers to be any faster at designing new and novel engineering I don't adhere to a luddite attitude. It's my opinion Autodesk's tools are deeply lacking and this is based on my experience and knowledge on software. When I first starting learning AutoCAD, I was very annoyed at the slow snail pace to designing things (as to pen/paper). I used to be a web designer/developer for a handful of years, those tools are much more refined and allow a person to do more with less time. Revit is a step in the right direction for the building industry but it has many drawbacks too. I strongly believe this is due to lack of competition and innovation on Autodesk's part. Adobe should be seen as the model. They've developed many tools that empower it's users and they continue to push the envelop forward. Autodesk will be unseated, if they don't continue to push their tools forward like Adobe has with many of it's products. The software isn't terribly hard, just needs a lot of investment upfront for a MVP competitor that chips away the market, then reinvest back into improving the tools with AI/DL/ML. I.e. One agency may use specifically a specific contractor, the AI recognizes revisions and highlights them to the architect. Or using specific materials. Etc. Etc. Construction industry is stuck in the 1940s, IMHO. Hopefully a competitor arises to bring it into the 21st century.

tootie 6y ago

Musk is espousing a UX fallacy by saying that a more "natural" interface is superior. Natural interfaces have a shallower learning curve for beginners, but CAD designers are all going to be expertly trained in their software. This is why pros always know the hot keys even when the mouse is simpler. Hot keys are faster. Learning how to pan, tilt and zoom with a mouse and keyboard is an 1 hour lesson that they will apply for 20 years and never worry about again.

MiroF 6y ago
I think that many commentators here on HN will be very unused to the market that Autodesk serves - mostly trained engineers who have been doing this stuff for decades, not so easy to "disrupt."