Are they holding us back?

[NeoNewbie]

Howdy!

Let me start out by saying that I absolutely LOVE my neo. It is my very first drone, and I am enjoying the hell out of it.

However, I do have one gripe about it. The quality out of the camera.

I have the RCN3 remote, so I've already changed the noise and sharpness settings)... but I think the neo can do more.

The camera sensor size is the same one the Air 2 had (half-inch) - but the quality is not the same. And I suspect it's a software issue.

I would love to see DJI put out a firmware update that removes the digital overprocessing and gives us 24fps and Dlog-m.

I know they can do this. And hopefully they will.

Is there a way we could all come together and ask them nicely as a group?

 

[Droning on and on...]

You're forgetting the image processing SoC, which is expensive, and a key component in quality of media.

I strongly suspect the SoC in the neo is a much cheaper, less functional and capable image pipeline than what's found in the Air 2, to keep the price down at $199. Good chance the sensor, although the same size, is cheaper (and noisier) than the Air 2.

We'll have to wait (and hope) DJI decides to follow Zero Zero Robotics and come out with a competitor to the Hover H1 Pro MAX (ignore the Pro). The neo so far has exceeded DJI's expectations, so there's hope.

Dlog-m? Stuff of fevered dreams on the neo, my friend. Take a look at the Avata 2.

 

[Kilrah]

Sensor size and resolution being the same means nothing, it can be completely different with various tradeoffs and probably is, along with the processing as mentioned.

 

[ryguy80]

There’s also a crop factor with the electronic image stabilization, so there’s some upscaling of the image to fit the 4K frame.

I don’t think we’ll ever see DLog-M on this Neo, but hoping we at least get D-Cinelike, which has been offered on other DJI 8-bit color sensors.

 

[NeoNewbie]

There’s also a crop factor with the electronic image stabilization, so there’s some upscaling of the image to fit the 4K frame.

I don’t think we’ll ever see DLog-M on this Neo, but hoping we at least get D-Cinelike, which has been offered on other DJI 8-bit color sensors.

D-Cinelike would be a HUGE improvement. And 24fps, too; there's no reason they can't give us 24fps.

 

[Droning on and on...]

D-Cinelike would be a HUGE improvement. And 24fps, too; there's no reason they can't give us 24fps.

I agree, D-cinelike would be awesome. The cheap image processor might not support it, however.

As far as 24fps goes, probably not as the target customers, Kaden Cristil and Mirengue, really don't care for their TikTok channel

 

[4 Inch Pistons]

You have to remember what it is. A $199 entry level FPV capable flying camera with more than average bells and whistles. Mini 3 or 4 Pro with goggles may be more what you’re looking for.

 

[spacewizard]

While we're on the topic of "why do t they give us the software for this" I sure wish the app would let us capture 360 panos (without having to turn and take the photos manually).

 

[NeoNewbie]

While we're on the topic of "why do t they give us the software for this" I sure wish the app would let us capture 360 panos (without having to turn and take the photos manually).

Let's add waypoints to the wishlist. (Since it's all software based anyway)

 

[spacewizard]

Or... while we're dreaming big... expose the controls to the drone via an sdk (for Android AND iOS) so that we can get all that stuff via a third party app like Litchi.

 

[spacewizard]

I mean, here's the thing. They absolutely *are* "holding us back" because there are features that could 100% be implemented in the controller app (DJI Fly) which are not in the controller app, and there is no third party app so nobody can implement them.

It's a way of artificially giving your more expensive drones "more features" where the *features* in question aren't in the drone, they're in what the app will do with the drone.

I've got a Mavic Mini, and it works with either DJI Fly or with Litchi. If I fly it with Litchi I get things like 360 panos, waypoints, other stuff that I can't have in the native app, which demonstrates that DJI could perfectly well implement those features in the app, they just choose not to.

This is one of the problems with proprietary, closed-source, vendor-lock-in software: they provide arbitary limitations based on their business need, not technical issues.

Unfortunately, they're *also* kinda the best, hardware-wise, so you are stuck with the software-based limitations if you want the best hardware.

They're kinda like Apple that way. You want the iPhones, you accept that you're locked into the Apple App store and the developers have to obey their rules.

 

[Droning on and on...]

Let's add waypoints to the wishlist. (Since it's all software based anyway)

Now THAT is truly a pipe dream, given what appears to be DJI's current product strategy of keeping this feature to the top-line models. Also lacking a 3-axis gimbal would make wayoint flights little more than novelty.

It's really a camera drone feature, not a whoop FPV.

 

[Droning on and on...]

Software has introduced this gap in understanding among the public that leads to this logic, "it's just software", and they're "holding back" for marketing and/or product strategy reasons.

While there is truth to that, there's no more truth to it than for cars, recliners, cookware, or table lamps. Or anything.

I spent my career dealing with software from the R&D side. The fact that software allows for easy deployment of new features fools people into thinking it's cost-free to the manufacturer, and they're screwing us by "holding back".

Nothing could be further from the truth. Software development isn't cheap. I suspect the software engineering for the neo was on par with the cost to develop the hardware, possibly more.

Then, there's the cost of support, which ain't cheap. Those people on the other end of the phone are being paid, but not by you. Your call is free.

So, to add and support waypoints, for example, would require a not-insignificant investment, and then an on-going support expense for years. DJI has limited resources... where best to spend them? Will waypoints on the neo yield significant PR? Will it drive significant neo sales? At the price-point and profitability of the neo, would that engineering cost be well spent, or would those engineers be better spent bringing a full-autonomous SDK to the Air 3S?

Software really isn't any different than hardware when it comes to all the business considerations involved, including the basics of ROI, which is an aspect of bringing a product to market we should all respect, as no one stays in business losing money.

 

[spacewizard]

Software has introduced this gap in understanding among the public that leads to this logic, "it's just software", and they're "holding back" for marketing and/or product strategy reasons.

While there is truth to that, there's no more truth to it than for cars, recliners, cookware, or table lamps. Or anything.

I spent my career dealing with software from the R&D side. The fact that software allows for easy deployment of new features fools people into thinking it's cost-free to the manufacturer, and they're screwing us by "holding back".

Nothing could be further from the truth. Software development isn't cheap. I suspect the software engineering for the neo was on par with the cost to develop the hardware, possibly more.

Then, there's the cost of support, which ain't cheap. Those people on the other end of the phone are being paid, but not by you. Your call is free.

So, to add and support waypoints, for example, would require a not-insignificant investment, and then an on-going support expense for years. DJI has limited resources... where best to spend them? Will waypoints on the neo yield significant PR? Will it drive significant neo sales? At the price-point and profitability of the neo, would that engineering cost be well spent, or would those engineers be better spent bringing a full-autonomous SDK to the Air 3S?

Software really isn't any different than hardware when it comes to all the business considerations involved, including the basics of ROI, which is an aspect of bringing a product to market we should all respect, as no one stays in business losing money.

There's probably some of both factors. An unwillingness to spend the time and money to create and support a capability in the app, and a positive wish to introduce extra (not strictly necessary) differentiation between the machines at different price tiers.

When I look at my old Mavic Air and see that, despite much older technology, it had many more features than modern minis.... And when I look at my Mavic Mini and realize that it has *way* more features when you run it with Liitchi than with DJI Fly, I conclude that they have chosen to make the hardware do less with DJI Fly than it is capable of doing, with newer cheaper models.

But it's also true, as you said, that in addition to forcing people who want those features onto higher priced hardware, they're also avoiding the costs of supporting and developing those features, which does contribute to making the cheap stuff cheap.

 

[NeoNewbie]

I totally understand the costs of software development. But let me give you a clear example:

Setting the camera to record in 24fps (when it can already do 30fps and 60fps in HD) wouldn't carry any significant development costs to them. The software, hardware, and capability are already there. It would also not generate any additional support costs. And yet, it's not available for the neo. This is why I feel like they're holding us back quite a bit with this drone.

I also understand that "this is a $200 drone". And while that may be technically true. It's also not. If you want to use it as a drone (and not just a selfie camera), you need a remote. And when you add that, the price is quite similar to the mini 4k and the mini 3 (which have far superior visual features).

 

[Droning on and on...]

I totally understand the costs of software development. But let me give you a clear example:

Setting the camera to record in 24fps (when it can already do 30fps and 60fps in HD) wouldn't carry any significant development costs to them.

I agree 100%. I'm a software engineer at my roots, have written boatloads of code, and can guess at what it would take to enable 24fps.

Can be as little as uncommenting a section of a switch statement in code base that was without a doubt leveraged from existing drones. The engineer would spend more time finding where in the code to enable it than the actual code changes.

OTOH, we don't know the hardware in the neo has 24fps capability. Cost savings might have driven the selection of a cheap sensor that only reads out at 30/60fps, and 24 is impossible.

Or, it can do 24, but it's an entirely different programming API than the sensors in more expensive cousins, and the sensor readout module had to be completely rewritten for 30/60. 24 might require roughly 50% of the 30/60 effort to add it.

The point is, we don't know. Assuming the first possibility I cite above is in fact the case is a mistake.

The software, hardware, and capability are already there. It would also not generate any additional support costs.

I ran the technical support group at HP for the HP 1000 mini computer. I can assure you that this is very, very wrong. Add any feature or capability to a product, and you will get support contact because of it, even though it may have no bugs.

Given the target market for the neo, calls simply asking what "frames per second" means, and "why would I use 24fps" will occur, in addition to the ones they're already getting from the Cristil sisters, Kaden and Meerangue.

And they have to provide that support for the life of the product.

So DJI gets requests for 24fps from people like you (and me), and must do a basic ROI stacking forecast increased sales against development and support costs. Based on my experience, my opinion is 24fps doesn't move the needle at all in any sales of the neo. Adding it will incur costs over the life of the drone. Easy decision.

DJI would respond to people asking for 24fps to get a Mini 4K. And they'd be right, in my opinion.

 

[NeoNewbie]

Damn, you're right. In the end, 24fps wouldn't move the needle on sales (and given the target audience for this particular done, neither would a flat color profile.)

 

[Droning on and on...]

Damn, you're right. In the end, 24fps wouldn't move the needle on sales (and given the target audience for this particular done, neither would a flat color profile.)

Both features I drool over having on the Neo. Not counting on it ever.

I am hopeful for a higher end version to compete with the Hover X1 Pro MAX thatcwould have these and other features.

 

[NeoNewbie]

I am hopeful for a higher end version to compete with the Hover X1 Pro MAX thatcwould have these and other features.

Well, apparently that's what the "flip" will be.

 

[Droning on and on...]

Well, apparently that's what the "flip" will be.

I'm not convinced of that yet, but I am worried