[cconcepts]

I recently bought a secondhand Original DJI Mavic Pro for chump change. For someone who knew nothing about drones it staggered me. For something so technologically advanced to be considered 'old' tech (manufactured 2017) is just unreal.

It is now my frame of reference for all sorts of technology. Automatically not crashing, auto return to home, insane maneuverability on sport mode and all at a bizarrely low price for something that would have seemed science fiction to many punters 20 years ago.

By commoditizing this kind of consumer tech that interfaces with the real world, DJI has changed my perception of what existing technology could be capable of and changed the way I consider usability in my own work.

[ehnto]

I had a similar epiphany except it has driven me to consider a new industry, as I can't possibly hope to solve the state of my current one and frankly don't want to, they made their own bed. The capabilities present when you own the whole stack and make a point of offering safety and robustness really makes the behavior present in other industries almost embarassing.

Drones are science fiction dreams come true. But more than that, they offer a really tangible way to work through software and hardware to as you say, interface with the world.

It's a really exciting space and I was initially really skeptical, but as their practcal uses outside of just a camera in the sky become more achievable it became clear to me that I want to get involved.

[ur-whale]

> outside of just a camera in the sky

This sounds a little dismissive ... a camera in the sky is a tremendously useful tool.

There are of course the art application, the "make memories" application that can be perceived as nice but not that important.

But how about: search and rescue, automated land surveying, 3D reconstruction of buildings and landscape, understanding the impact of new architectural projects on the environment, etc ... ?

Combine the "just a camera in the sky" with ML and that's a lot of new applications (and therefore business opportunities).

[ehnto]

Oh I didn't mean to be dismissive, I am extremely impressed with the possibilities said eye in the sky can provide, I guess I meant that it's more useful than a hobby photography tool. I of course think that is really rad too since it's what I do with my drone now. But the commercial side of the industry is getting really exciting with exactly the kind of tools you're talking about, some of which are just new ways of applying the camera like you're pointing out.

Ag spraying, high definition cinematography, industrial inspections, affordable localized LIDAR scanning, even underground mine surveys are all super interesting to me.

[fsloth]

And it's incredibly robust! Mine has spent total I think a week on a tree. First time it was climbed down. The second time climbing was not feasible. Luckily a storm blew it down after a week or so. The bad part was it landed on sharp steel debris. But it did not have a dent in it. I presume it landed on it's battery, which was missing, which probably then absorbed most of the energy from the fall.

After all this abuse it's still flying like a champ.

I think DJI probably had some quality control issues on at least on the original batch (first I got had lots of problems) but their service reps simply sent a replacement for no charge. The second one has had no problems.

[Filligree]

If you do ever damage it, you can send it to DJI and they’ll give you an itemised repair bill, which will be much cheaper than a new drone.

This isn’t the usual throwaway consumer hardware. The drone is designed to be repaired.

[TedShiller]

So you don’t think littering the environment with a battery encased in plastic is a big deal?

[fsloth]

Why would you think I abandoned it on purpose?

The crash site was a pile of steel beams and other debris from a demolished building.

Do you have suggestions what else I could have done except search for it?

[darknavi]

I have a Mavic Pro from a few years ago. Still runs like a champ.

The biggest draw for me to get a new one at some point would be the flight time per battery. My Mavic Pro gets ~20 minutes and the newer ones seem to get ~30 or more.

Not a huge deal with a bank of batteries, but it will be nice ones my drone eventually kicks the bucket.

[WrtCdEvrydy]

Wait till someone shows you a Skydio 2 drone (look around, there's probably a drone pilot that uses one).

[subhro]

Only 23 mins? That is not terribly useful at least for any serious matters like, say, search and rescue operations.

[dylan604]

Sounds like youre using the wrong tool for the task at hand. However, it also sounds like you've never personally experienced the ~23 minutes of flight time and the amount of ground that can be covered. It's way more impressive than you are dismissing.

[amcoastal]

You're getting downvoted because your complaint makes it seem like this is a problem with the particular model. 23 minutes is probably even above average for drone flight times. Don't be so kneejerk negative.

[comeonseriously]

>... say, search and rescue operations.

Oh good grief. If that is the use case, you get something designed for that.

[some_random]

That's incredible, it's almost as if it's not optimized for search and rescue.

[latchkey]

If you really want to blow your mind, buy a used DJI FPV next and then put it into full manual mode...

[InvertedRhodium]

It'd blow their mind how easy it is to crash in manual mode with no experience.

[latchkey]

One neat thing is that it has what I call the "oh shit button".

If you feel like you're going to crash, hit the button and disables manual mode, rights itself and hovers in place. Near immediately.

As long as you have your finger on that button (and practice using it), it is a pretty safe way to learn.

Best to go in manual mode with a bit of height and in an open area first too.

[FPGAhacker]

Definitely this. Use a simulator for a while first.

[stavros]

Don't, you'll certainly crash pretty soon and you'll basically have to buy a new one. If you want to try FPV, get an FPV quad, which is an open ecosystem and much cheaper (though not nearly as plug-and-play, granted).

[Steltek]

DJI and FPV are two different hobbies, IMO. Either you want a slow-and-steady flying camera or you want to feel like you're actually flying an insanely maneuverable aircraft.

And yeah, the FPV ecosystem with interchangeable parts and open source code is eye opening. It's like building a desktop and installing Linux yourself vs buying a tablet. Mind you, your desktop won't fall out of the sky and shatter into pieces. But it's fun!

[lebed2045]

+1 to that. I started by FPV hobby about a year ago (just regular analog tinywhoop). So it's 30% flying hobby, and 70% tinkering around and learn:

1) how to solder,

2) how how charge and use lipo batteries,

3) how antennas work, polarisation, connectors etc,

4) what all these buzzwords in betaflight means

5) binding, expressLRS setting,

6) 3d printing protection parts,

7) even there are different connectors standards!,

etc so it's whole umbrella of little hobbies that comes with it. At least for me it's lots of fun, because it takes a few hours of learning time and you get working result. But it's not for everyone for sure. Drastic different with PnP experience of DJI drones.

[stavros]

> DJI and FPV are two different hobbies, IMO

Yes, but the DJI FPV that the GP mentioned is an FPV quad.

[Steltek]

DJI has a quad called "FPV" but that doesn't mean it's the FPV hobby. I can replace an arm on my quad for $2 vs some absurd money for DJI Care plus time spent waiting for shipping back and forth. Now let's blow up some ESCs or crack the camera on a rock...

[jmacd]

I have both. DJI FPV 2 and many hand built racing drones (lumineer, TBS).

The DJI is an absolute stunner with fantastic range and video transmission. The build quality is far superior than anything I have seen home built.

I would never recommend racing it, as I am ok with sacrificing my own drones because I can repair them easily, but as an introduction the DJI is an amazing platform.

[stavros]

> The DJI is an absolute stunner with fantastic range and video transmission.

Hmm, what's the range like? ELRS solves the RC link range (it's functionally infinite) and analog has longer range than DJI video (though DJI's quality is obviously leagues ahead).

I have the DJI video system in my home-built drones, I definitely recommend it but analog goes farther (though you get much better quality for 13km with DJI).

> The build quality is far superior than anything I have seen home built.

If by "build quality" you mean "it feels good", then yes, of course something factory-built is going to be more sleek than something you build in your garage. If you mean "sturdy", though, the carbon fiber quads will take much, much more punishment.

[latchkey]

This is why I bought the insurance.

[stavros]

Do they even sell insurance for it, when it's nearly 100% sure it'll be used?

[latchkey]

Yes. Also statefarm covers it too. =)

I'm not an expert flyer and I really have had no issues with it. It isn't that hard to fly at all, even in full manual. The "oh shit button" is a godsend.

[stavros]

It's not so much that it's hard, it's that if you want to learn freestyle or racing, you're going to crash. I guess many people just fly it like I fly my quads, cautiously and away from any obstacles.

[mclightning]

I'm all on board with DJI FPV except for the fact, it is an acrobatic drone with proprietary parts which will be costly to replace...

I want same feature-set on a DIY frame, or at least DIY-able.

[ryanlol]

>acrobatic

aerobatic, surely?

[latchkey]

DJI FPV is acrobatic, the thing can do flips when it is in full manual mode. It is pretty amazing, but definitely requires some experience flying drones.

[ryanlol]

Excuse my pedantry, but in an aviation context this is referred to as “aerobatics” while “acrobatics” refers to feats of the human body.

Just some meaningless trivia.