DJI Mavic, Air and Mini Drones
Friendly, Helpful & Knowledgeable Community
Join Us Now

Litchi on Mini 3 Pro with new controller?

That might be why they want to ‘mainstream’ the RC controller with screen built in: no more phone updates breaking compatibility. Providing mission-critical drones even sub-1000 dollar
 
That might be why they want to ‘mainstream’ the RC controller with screen built in: no more phone updates breaking compatibility. Providing mission-critical drones even sub-1000 dollar
Ummm, But they still support (and are selling) the "regular" standard controller with the Mini 3 which means they will still need to support IOS and Android "forever" as long as they sell them. I'd hate for the ONLY option to be a sucky 5" screen (don't care how "bright" it is 'casue I barely can read it) so I doubt they will ever rid themselves of IOS and ANDRIOID apps and updates unless of course they require a vision test before purchase. It's highly likely they will sell more WITHOUT the (allegedly) smart controller due to price.
 
This may go more quickly than I originally anticipated. This afternoon I was able to get the code running on Android 12 and the latest 4.16 DJI SDKs quite quickly. I haven't tested the emulation yet for the Mini 2, but assuming it works like it did under 11/4.14 this suggests that it might also work for the Mini 3 with minor changes. The code emulates drone type as Mavic Air 2 and camera as Mavic Mini. I'll test this if I ever have a Mini 3 to test with (which I'll note is not at all likely in foreseeable future at the predicted price for even the drone without a controller).
 
  • Like
Reactions: Faster
Emulation and waypoint simulation flight tests look good. So this went much better than expected, and now running on two Android 12 devices with the Mini 2. I have the hooks in place to test with a Mini 3 if that ever becomes practical.
 
Emulation and waypoint simulation flight tests look good. So this went much better than expected, and now running on two Android 12 devices with the Mini 2. I have the hooks in place to test with a Mini 3 if that ever becomes practical.
Start a GoFundMe to subsidize a Mini 3 since your work will benefit everyone.
That said it appears the Mini 3 really is an upgrade to the Mini 2 so may features would not likely be supported such as vertical orientation, manual camera control, touch screen adjustments. Not to mention O3 and new transmission protocol.
It does look like a fun little beast and I ordered one for an upcoming trip to the EU since there will be few if any restrictions - unlike the last go around where the Mavic and Spark were virtually useless.
I'll likely sell off a couple of other drones and settle on this one.
 
  • Like
Reactions: Coskier
Much of what you just listed are likely not relevant to this app. For example, even though the app considers the Mini 2 camera to be a Mini 1 camera, you can set the more advanced Mini 2 camera features in Go Fly and they'll keep working when you switch back to this app (though you wouldn't get the higher Mini 3 resolution through the app itself, because it's transcoded so it can display on the GCS like QGroundControl -- even on a different device via TCP/IP -- but the SD card should have the proper recordings at the selected full resolution). O3 should not be relevant either, that's just a protocol decision between the controller and the drone, and I've heard that apparently even the standard Mini 2 controller will supposedly switch to O3 if paired with a Mini 3. No way for me to confirm these Mini 3 aspects yet, of course.

I don't do GoFundMe or Patreon -- I frankly do not like their business models.
 
Addressing the question as posed by the O.P.

Unlike the original smart controller and crystal sky units which can exit your flight control app and drop into the operating system it does appear (at this early stage) that you are locked into the DJI Fly app which is going to make loading third party apps difficult. I’m sure some enterprising folks will find a way but it may be a deeper dive than most want to make and I suspect (like the original smart controller) that hardware limitations will make it a less than stellar experience.

That notwithstanding, almost every one of my commercial colleagues who run a P4Pro Plus ended up buying a standard controller on the second hand market as well precisely for ease of running third party apps, likewise I have both the SDK controller as well as the standard one for my P4RTK for the same reason.

If you are going to buy a Mini 3 Pro and want the controller pro option (I know I will) and will want to run 3rd party apps (I know I will lol) then there is always some poor soul who’s lost their Mini 2 or Air 2(s) selling the controller off to recover some of the loss. Pair that with your Mini 3 pro and favourite smart device and problem solved.

Regards
Ari
 
Last edited:
  • Like
Reactions: Coskier and TT3023
And notably, everything I've seen so far indicates that when using the standard RC-N1 controller the M3 Pro will have exactly the same characteristics as with the new phoneless controller. That is, it will reportedly use O3, have the same range and return video resolution, etc.
 
And notably, everything I've seen so far indicates that when using the standard RC-N1 controller the M3 Pro will have exactly the same characteristics as with the new phoneless controller. That is, it will reportedly use O3, have the same range and return video resolution, etc.
TT3023 Thank you to the original poster and TT3023 for your insight. Together you've answered my biggest questions about this drone.

TT3023, is there a platform or method we could set up a fund to buy you a Mini 3? It'd take 100 contributions of €10's. I expect the number of users wanting to use Litchi sooner with their Mini 3 will far exceed the donations required. All extra will keep you filled up with coffee. Given it's all on trust, perhaps a thread on here or a word press to keep everyone updated on the fund, and then your progress? No pressure!
 
Hi. Thanks much for your thoughts. However, I am not interested in and will not accept monetary donations related to this.

On the technical front, I want to be clear about a few details. Assuming that the technique used for the Mini 2 prior to its support in the MSDK will also work for the Mini 3 prior to its support in the MSDK, this does not mean you can run commercial programs like Litchi, that depend on direct support in the MSDK to function. They have to wait for the formal DJI support.

What this technique does do is provide an interface to MAVLink applications, e.g. QGroundControl or (in theory, though I've never tested it) Mission Planner, such as are commonly used in the PX4 drone world. There are some limitations. Manually designated waypoints works, though currently the "automated mapping" functions of QGroundControl are not implemented, for example.

Another issue is that while source and instructions for building the app yourself could be provided, it would not be practical to distribute prebuilt versions of the app, due to the limitations and potential costs involved for the Google Maps and DJI API usage required (both effectively free at individual use levels, but distributed the costs could pile up).

So I wanted to be clear on these points. Thanks again.
 
If you are going to buy a Mini 3 Pro and want the controller pro option (I know I will) and will want to run 3rd party apps (I know I will lol) then there is always some poor soul who’s lost their Mini 2 or Air 2(s) selling the controller off to recover some of the loss. Pair that with your Mini 3 pro and favourite smart device and problem solved.

Regards
Ari
Lol! I did just that. Picked up a used RC231 off eBay. Didn’t want to wait 3 extra months or whatever it was for the RC231 kit. Figured I could eventually easily sell the new remote. The screen on my Samsung S21 Ultra (which is a spare phone) is bigger and brighter than the new remote screen.

6DAF5C15-5BF7-4865-9A1C-0262248D992C.png
 
Hi. Thanks much for your thoughts. However, I am not interested in and will not accept monetary donations related to this.

On the technical front, I want to be clear about a few details. Assuming that the technique used for the Mini 2 prior to its support in the MSDK will also work for the Mini 3 prior to its support in the MSDK, this does not mean you can run commercial programs like Litchi, that depend on direct support in the MSDK to function. They have to wait for the formal DJI support.

What this technique does do is provide an interface to MAVLink applications, e.g. QGroundControl or (in theory, though I've never tested it) Mission Planner, such as are commonly used in the PX4 drone world. There are some limitations. Manually designated waypoints works, though currently the "automated mapping" functions of QGroundControl are not implemented, for example.

Another issue is that while source and instructions for building the app yourself could be provided, it would not be practical to distribute prebuilt versions of the app, due to the limitations and potential costs involved for the Google Maps and DJI API usage required (both effectively free at individual use levels, but distributed the costs could pile up).

So I wanted to be clear on these points. Thanks again.

Hi. Thanks much for your thoughts. However, I am not interested in and will not accept monetary donations related to this.

On the technical front, I want to be clear about a few details. Assuming that the technique used for the Mini 2 prior to its support in the MSDK will also work for the Mini 3 prior to its support in the MSDK, this does not mean you can run commercial programs like Litchi, that depend on direct support in the MSDK to function. They have to wait for the formal DJI support.

What this technique does do is provide an interface to MAVLink applications, e.g. QGroundControl or (in theory, though I've never tested it) Mission Planner, such as are commonly used in the PX4 drone world. There are some limitations. Manually designated waypoints works, though currently the "automated mapping" functions of QGroundControl are not implemented, for example.

Another issue is that while source and instructions for building the app yourself could be provided, it would not be practical to distribute prebuilt versions of the app, due to the limitations and potential costs involved for the Google Maps and DJI API usage required (both effectively free at individual use levels, but distributed the costs could pile up).

So I wanted to be clear on these points. Thanks again.
Testing the Mini 3 should be relatively straightforward. I don't need the new RC (and I'm seeing dismal reports regarding range using it currently in any case). And I already have an RC-N1. Unfortunately, the Mini 3 without a controller seems to have among the longest lead times, so I don't think this is going to happen very soon.
 
So I'm pretty satisfied with testing so far on the Mini 2, and I'll proceed to test with Mini 3 when I can get my hands on one. As I mentioned above, I only need the drone itself, not a controller since this testing of the app would be with the RC-N1 anyway which of course I have. If anyone sees a source for the M3/with one battery FCC version (no controller) that is available earlier than the end of August (the sooner the better) please let me know. Thanks.
 
So I'm pretty satisfied with testing so far on the Mini 2, and I'll proceed to test with Mini 3 when I can get my hands on one. As I mentioned above, I only need the drone itself, not a controller since this testing of the app would be with the RC-N1 anyway which of course I have. If anyone sees a source for the M3/with one battery FCC version (no controller) that is available earlier than the end of August (the sooner the better) please let me know. Thanks.

Don’t hold your breath. :) You likely won’t be testing Mini 3 Litchi for maybe 9-10+ months which of how long it generally takes DJI to release the SDK for a new AC and about how long it took the Air2s and Mini 2 SDK to be released. Or do you mean with your own hack?
 
Yeah, it took over a YEAR for the Mini 2 support to get into the MSDK after the M2 was released. No, I'm talking about the app I work on that supported the Mini 2 *before* the MSDK support, by permitting use with QGroundControl (even on a different computer). My goal would be to get the Mini 3 working the same way now, as the Mini 2 was supported before the MSDK supported the Mini 2. It spoofs a supported device.
 
Actually it should in theory work with pretty much any MAVLink compatible system, including DroneKit (python) for AI type functions. But personally I've only tested using QGC so far.
 
But it appears DJI is holding back the supply of the standalone drone version of the Mini 3, because it has the latest of the delivery dates listed so far, which is a bummer.
 
I would be extremely grateful if you guys could find that hack, in several ways I’ve found “maven” to be a superior app and nice compliment to the standard DJI app. For me, my huge dream is to be able to use the rc controller with the mini3 and the air 2s Using both maven and the fly app. A year or so ago I had a similar set up with the mini2 and air2 since they both used occusync 2 and it was great. Carrying one controller on big climbs and having two drones was great for me. Did one of my best videos ever that way. This new set up would only be better using occusync 3 with both birds on the all in one controller eliminating the PITA of phone set up. It is stressful juggling electronics cliff side! It also discourages me from getting good phone footage in that moment since I’m tempted to just leave the phone in the old controller. If you guys haven’t tried Maven and have any use for a very wide range of autonomous tracking options, Maven is very versatile. The on screen virtual home point and controller location is a nice touch too, way easier to get accuarate manually flybys of yourself. . . thanks! And good luck!
 
Maven's big limitation is being iOS only. Nor is it likely to work with the Mini 3 until DJI provides official support for it in the MSDK. As I mentioned, that took DJI over a year with the Mini 2 after the release of the Mini 2 itself. They could be faster this time. Or not.
 

DJI Drone Deals

New Threads

Forum statistics

Threads
131,066
Messages
1,559,498
Members
160,049
Latest member
kramme