Official DJI Response to the GPS Issue

[Suren]

If this problem could have been fixed with a firmware mod, it would have been in the last big firmware release. They've known about this from the day it was released. It's very possible it's a hardware problem, and they selected a bad chip, potentially because other chips were not readily available. Since we know DJI doesn't test their drones adequately before production, this GPS lock issue could have very well gone unnoticed by DJI. Now they have a pig to put some lipstick on. It will be interesting to see how this ends up. I hope I'm wrong.

That explanation does not make sense. The Mavic worked properly with Full GPS lock prior to the December update. If this was a hardware issue then it would have been so since day one and I had my drone since late November

 

[Meta4]

I think it was Jeven Dovey on YouTube said that DJI had chosen not to use GLONASS and that was the reason for slow and low satellite acquisition issues, can any one confirm this? Slow and low is best for smoked pork...

That doesn't explain why the drone is so slow to acquire GPS.
It uses satellites from 3 different systems, not just one.

 

[PompousWindbag]

That's a lump of coal from Santa, right there. He needs to stick with reindeer and sleighs.

 

[mswlogo]

That explanation does not make sense. The Mavic worked properly with Full GPS lock prior to the December update. If this was a hardware issue then it would have been so since day one and I had my drone since late November

Again, people are assuming “it worked fine” before the update. Maybe for you but maybe not everyone. Also I don’t know how licensing works with GPS systems but they might have elected to only go with China’s GPS system and the initial release was never intended or intended long term.

If it could simply go back to the way it was, they would have done it by now. I’m guessing too, just like you did stating that EVERYTHING was fine on initial firmware.

 

[Meta4]

Again, people are assuming “it worked fine” before the update. Maybe for you but maybe not everyone.

You are assuming (incorrectly) that I don't know what I'm talking about.
I've examined the recorded flight data from plenty of Mavic flights and the G{PS was working just as it should have.

Also I don’t know how licensing works with GPS systems but they might have elected to only go with China’s GPS system and the initial release was never intended or intended long term.

All GPS satellite systems are open access.
There is no licensing.

If it could simply go back to the way it was, they would have done it by now. I’m guessing too, just like you did stating that EVERYTHING was fine on initial firmware.

There's no need to guess that I was guessing.

 

[mswlogo]

You are assuming (incorrectly) that I don't know what I'm talking about.
I've examined the recorded flight data from plenty of Mavic flights and the G{PS was working just as it should have.

All GPS satellite systems are open access.
There is no licensing.


There's no need to guess that I was guessing.

I wasn’t speaking to you.

 

[Suren]

Again, people are assuming “it worked fine” before the update. Maybe for you but maybe not everyone. Also I don’t know how licensing works with GPS systems but they might have elected to only go with China’s GPS system and the initial release was never intended or intended long term.

If it could simply go back to the way it was, they would have done it by now. I’m guessing too, just like you did stating that EVERYTHING was fine on initial firmware.

This is not an assumption as I talk from my experience and we also never had any, and I mean ANY complaints about long GPS acquisition times before the December updates.

 

[Suren]

For what's its worth, I tried Ken (Originaldobo - I think) his suggestion of turning the drone on, waiting around 45 seconds to get 5 sats, turn drone off and back on. My drone connected to full sats in 2,5 minutes and did so every time I did that method. It sounds silly but beats me standing like a dope waiting around for 6 minutes and 50 seconds to get a GPS Lock. Can someone else try that method and see if it helps you as well and post back.

 

[mswlogo]

This is not an assumption as I talk from my experience and we also never had any, and I mean ANY complaints about long GPS acquisition times before the December updates.

You mean how fast a Drone gets a GPS lock is the only metric of the Drone running perfect? I didn’t know that. I thought how reliable and accurate RTH worked was. Silly me. I stand corrected. I really thought RTH had to really be consistent and accurate to labeled as “running perfect” and not only how fast it locks. Phew thanks. Now I have hope they can go back to fast locks and possibly sloppy RTH and all will be good again.

You should suggest that DJI I’m sure they have no idea.

 

[Suren]

You mean how fast a Drone gets a GPS lock is the only metric of the Drone running perfect? I didn’t know that. I thought how reliable and accurate RTH worked was. Silly me. I stand corrected. I really thought RTH had to really be consistent and accurate to labeled as “running perfect” and not only how fast it locks. Phew thanks. Now I have hope they can go back to fast locks and possibly sloppy RTH and all will be good again.

You should suggest that DJI I’m sure they have no idea.

Did I say the M3 was perfect before the December update? No I did not, what I did say it was getting GPS lock much faster prior to this update and there were no complaints about GPS only about the missing features and over sensitive OA sensors. You seem to be on some other sort of planet making up your own theories. Good Luck with that.

 

[mswlogo]

Did I say the M3 was perfect before the December update? No I did not, what I did say it was getting GPS lock much faster prior to this update and there were no complaints about GPS only about the missing features and over sensitive OA sensors. You seem to be on some other sort of planet making up your own theories. Good Luck with that.

But you’re saying that one metric was better and ignoring possible reasons that might have been not so perfect and why it *might* have changed. Data you don’t have.

 

[Meta4]

I really thought RTH had to really be consistent and accurate to labeled as “running perfect” and not only how fast it locks. Phew thanks. Now I have hope they can go back to fast locks and possibly sloppy RTH and all will be good again.

What are you getting at with this talk of "possibly sloppy RTH"?
Do you have any data, or is this some fantasy that you've made up?

The only GPS issue that I've seen evidence for is that the M3 is not receiving enough sats, even after waiting a long time.

 

[Don Testme]

Occasionally, I just take off when I only have 5-6 sat links. Or when it says it has enough links to hover in a fixed poistion, which is usually less than a minute. I fly to the max of 100ft. At That height the GPS links occur much quicker and I get a home point lock in about 1.5 minutes total. Yes, the landing accuracy isn't as good, but I get out and into the air much quicker. Most of the time, I don't really care how accurate the landing is. I perform my landing manually most of the time. As long as the drone gets within a few feet of the home point i'm happy.

 

[langcpht]

For what it's worth, my experiences aren't the same as many others here. I fly very infrequently (particularly in winter). Mine seems OK, and although definitely slower than the Mavic 2, I'm not experiencing the 5+ minute wait times. I just performed the most recent firmware update to both the drone and RC Pro via the DJI Assistant. I took the aircraft into my front yard this morning and got enough satellites for a home point (11) in less than 2 minutes. I launched, hovered about 10 feet, moved in all directions another 10 feet, and had 15 satellites another 30 seconds or so later. There's likely other variables at play here as it's clearly not affecting everybody in all situations.

 

[mswlogo]

What are you getting at with this talk of "possibly sloppy RTH"?
Do you have any data, or is this some fantasy that you've made up?

The only GPS issue that I've seen evidence for is that the M3 is not receiving enough sats, even after waiting a long time.

The exact same evidence that claims everything was fine on early firmware. I have none, just as Suren has none.

You can’t have your cake and eat it too.

Nobody knows everything was fine or not on the first firmware. DJI made a change, for some reason and has not reverted it. Now going on 2 months !!

Do you think they are just stupid and don’t know they broke it and will eventually fix it?

Do you think they knew exactly what they were doing. Since RTH is probably one of the most critical features that depends on GPS I’m guessing it had something to do with RTH (like the release notes mentioned in firmware .400). Which is another small piece “evidence”. Funny how they mentioned improvements in RTH and now Home Point takes longer. It may not be RTH, but I claim DJI fixed something that wasn’t “perfectly fine”.

Which do you think is more likely?
Neither is great for DJI in my opinion.

If there is a 3rd option I’m all ears.

Pretty simple question. DJI is either clueless or knows exactly what they are doing.

Because if it was a coding error, this is an absolute joke in resolving it. I’m at least giving DJI credit that it was intentional change to fix something. All be it with a costly compromise DJI knows about.

 

[Meta4]

The exact same evidence that claims everything was fine on early firmware. I have none, just as Suren has none.

So if you have no evidence that there was some non-specific problem with RTH, why suggest there might be?

Nobody knows everything was fine or not on the first firmware. DJI made a change, for some reason and has not reverted it. Now going on 2 months !!

It's quite clear that there was no problem with GPS earlier.
There was nothing about it that needed fixing, but DJI wrecked it.
By accident, not for "some reason".

Since RTH is probably one of the most critical features that depends on GPS I’m guessing it had something to do with RTH

It sounds like you don't understand much about how GPS works.
There was no hint of any RTH problem.
Guessing is non-productive, it's best to stick to things there is evidence for.

 

[sar104]

The exact same evidence that claims everything was fine on early firmware. I have none, just as Suren has none.

You can’t have your cake and eat it too.

Nobody knows everything was fine or not on the first firmware. DJI made a change, for some reason and has not reverted it. Now going on 2 months !!

Do you think they are just stupid and don’t know they broke it and will eventually fix it?

Do you think they knew exactly what they were doing. Since RTH is probably one of the most critical features that depends on GPS I’m guessing it had something to do with RTH (like the release notes mentioned in firmware .400). Which is another small piece “evidence”. Funny how they mentioned improvements in RTH and now Home Point takes longer. It may not be RTH, but I claim DJI fixed something that wasn’t “perfectly fine”.

Which do you think is more likely?
Neither is great for DJI in my opinion.

If there is a 3rd option I’m all ears.

Pretty simple question. DJI is either clueless or knows exactly what they are doing.

Because if it was a coding error, this is an absolute joke in resolving it. I’m at least giving DJI credit that it was intentional change to fix something. All be it with a costly compromise DJI knows about.

Have you been experiencing poor RTH accuracy with your M3 with this or previous firmware? There don't seem to be any complaints about RTH in this thread - it seems to be purely about the length of time it takes to get an acceptable GNSS solution.

In terms of "evidence", there are fairly clear reports from users that the position lock was much quicker on the previous firmware. What evidence have you seen that there was poor RTH performance with the previous firmware? Speculating that DJI changed the firmware to address a problem that no one encountered is not evidence.

It's not without precedent for DJI to release firmware, including upgrades, that break or degrade functionality.

 

[drjmax]

In Austin, TX my acquire time for >15 sats with the Mav 3 is less than 1 minute.

 

[dirkclod]

 

[zinko]

Hey

I try this riht now and by me is even faster like one minute. Before i always wait around 5. Minutes.

Just turn on controller and then mavic 3 wait 45 secound (in this 45 secound i get 7-8 satelits) turn drone of and then on and the home point is imediatly updatet! With 8. Satelites!!

I confirm thats work.