I have a Mavic Pro Platinum, actually, I've now had two (the first one I bought and the one DJI replaced it with), and both of them exhibit the exact same BAD behavior. DJI has alos replaced the remote controller, but didn't fix the problem, which I believe to be a FW issue. I have created a video that illustrates the problem, and the steps necessary to replicate it. Can anyone with a Mavic Pro Platinum PLEASE take a few minutes and do a quite test to see if this happens on your drone?

(the password is mavicpro)

The video:      (the password is mavicpro)

  

To replicate, all you have to do is the following:

  
  

(1)          Hover the craft

  

(2)          Yaw in either direction continuously

  

(3)          As you yaw, activate and release (let it jump back to center under spring tension) the gimbal tilt control periodically.

  

(4)          Whenever you release the gimbal tilt control, the yaw motion will ‘hesitate’ and the resulting video image will ‘jump’ (this ruins any cinematic footage where you are turning and tilting up or down, and you end the tilt while still turning - at least it has on two drones that I've had in my possession).

My aircraft firmware is v01.04.0300 (with RC, batts, and Go4app at consistent level).

Thanks in advance.

Shane

  

I will give this a try as soon as I get flying weather here.

What happens when you update the firmware to current release v01.04.0500, that would prove whether it's a firmware problem or not ?

Just me or the video is password protected?

Is this password protected?

Dan B-P4 Posted at 11-6 10:27
Just me or the video is password protected?

User states the password in the top post just below the video.

A CW Posted at 11-6 11:07
Is this password protected?

User states the password in the top post just below the video.

Aardvark Posted at 11-6 08:18
What happens when you update the firmware to current release v01.04.0500, that would prove whether it's a firmware problem or not ?

Did that. No change.

FlyDK Posted at 11-6 11:51
User states the password in the top post just below the video.

I gotta be blind!!!

Thanks

GrizzlyAK Posted at 11-6 12:05
Did that. No change.

I've not had a chance to check my Mavic Pro, but something else you could look at is disabling 'Synchronized Gimbal Pan Follow" in 'Gimbal Settings' / 'Advanced Settings'.

Aardvark Posted at 11-6 12:19
I've not had a chance to check my Mavic Pro, but something else you could look at is disabling 'Synchronized Gimbal Pan Follow" in 'Gimbal Settings' / 'Advanced Settings'.

That was the second thing I checked. I also tested it in WiFi mode to eliminate the RC. I even told DJI this, yet they still sent me a new Rachel this last round. They apparently aren’t willing to try and replicate the issue, otherwise they would see for themselves on the two drones I’ve sent back (as well as the two they’ve sent back to me).

Just so I know exactly what to do, your saying is as you yaw, tilt the camera down and then back up. this is causing a "sticky" movement as the aircraft is turning. Am I correct?

Mine does the same thing. Its when the gimble gets into its farthest position and you release the gamble control on the remote is when it has that "sticky" effect. I just purchased mine last month from DJI, brand new, not a refurbished unit.

Mine does the same thing. Its when the gimble gets into its farthest position and you release the gimble control on the remote is when it has that "sticky" effect. I just purchased mine last month from DJI, brand new, not a refurbished unit.

djiuser_g37OmqEDpSCy Posted at 11-6 15:41
Mine does the same thing. Its when the gimble gets into its farthest position and you release the gimble control on the remote is when it has that "sticky" effect. I just purchased mine last month from DJI, brand new, not a refurbished unit.

Good to know. Both of the ones I’ve had were brand new and both exhibit the issue. I suspect even more that it is an issue with the control SW and not many people have noticed it. I did pretty quickly and have been working with DJi for over 6 months. My drones have spent more time in transit to/from DJI support than in the air!

GrizzlyAK Posted at 11-6 16:50
Good to know. Both of the ones I’ve had were brand new and both exhibit the issue. I suspect even more that it is an issue with the control SW and not many people have noticed it. I did pretty quickly and have been working with DJi for over 6 months. My drones have spent more time in transit to/from DJI support than in the air!

well if you hear anything back from them or a solution, let me know.

Thank you,
Justin

djiuser_g37OmqEDpSCy Posted at 11-6 17:20
well if you hear anything back from them or a solution, let me know.

Thank you,

Will do, Justin. I’ll post back here. Thanks for testing it out. I’m hoping others will too. I’m really curious to figure this out.

FlyDK Posted at 11-6 11:52
User states the password in the top post just below the video.

Thanks for pointing out

Mavic Pro owners, beware. This issue has been identified and confirmed by other owners, and may affect you. I’ve been strung along by DJI’s warranty repair for 6+ months, and they just keep replacing it with new drones that have the same issue. They have also refused to either refund my money or replace it with comparable product that actually works. They just keep sending me new units with the same problem. After 6 months of their inability to repair their own product, they now say it’s outide the timeframe for a refund. How convenient. All drones sent to them were in perfect condition, never crashed. In fact,  the one I’m sending back now, I got back from them just last week WITH THE SAME ISSUE. They have now GUARANTEED ME IN WRITING that the next drone I receive will not have the problem. I’m not holding my breath. If they do, I’ll be filing a complaint with the Alaska Attorney General for resolution, and if I can find more owners willing to test their drones, a class action suit.

Anybody else given this a try? If DJI used the same codebase, it is possible this same defect has made its way into the Mavic 2 as well. Please give it a try. I've had DJI return to me yet a third NEW drone that exhibits the exact same issue, so by now we're pretty sure this is a flight control FW defect. Anyone who uses (or wants to use) their Mavic Pro for any serious filming will be affected by this. It will ruin much of your footage, and make your post work much more difficult trying to account for it.

If you don't care what your footage looks like, then you probably won't even notice it happening.

Mine is behaving exactly like yours
01.04.0300
iphone 6
app 4.3

Mgozer Posted at 12-10 01:07
Mine is behaving exactly like yours
01.04.0300
iphone 6

Thanks Mgozer for taking the time to test and confirm. The more this is identified by Mavic Pro owners, the better chance there is of DJI fixing it.

Just verified the same problem at 4:00 pm, EST, Michigan.  MPP, 240 flights, latest GO 4, latest firmware, all functions calibrated.

However, I did notice that the severity of the jerk was lessened if the rate of yaw was much less.  Even so, it does occur.  
I'm not sure where I'd use such high velocity yaw in any cinematic sense.  I like my pans to be slow.  But I suppose never say never.

RadioFlyerMan Posted at 12-10 13:47
Just verified the same problem at 4:00 pm, EST, Michigan.  MPP, 240 flights, latest GO 4, latest firmware, all functions calibrated.

However, I did notice that the severity of the jerk was lessened if the rate of yaw was much less.  Even so, it does occur.  

RadioFlyerMan, copy that. Thanks for the test and confirmation. The more the merrier, as they say.

Agree with the severity being greater at faster yaws, but the reason I first noticed this is when I was doing a slow cinematic tilt up while turning (yawing) flying over something. It was slow, but the subsequent 'jump' or hesitation in the video stream when played back was visible, and ruined the footage. One or two pixels on the video screen is noticeable. Granted, I could have cut the footage before that point in post, but that kind of leaves you wanting more.

Imagine flying over a spectacular waterfall in Iceland, flying up the river to the falls, then up over the downpour (or vice versa, off the cliff and down the river), and you tilt up or down as you follow the river's course (turning) revealing the fabulous landscape at sunset. You end your pan and BOOM the video jumps. End of take. Period. Not the smooth transition you'd like. This is exactly what happened to me time after time on my trip to Iceland with my Mavic Pro earlier this year.

Sure, one can plan their shots to be sure they end their yaw BEFORE the tilt ends, but be sure you don't have that left thumb on the control, because even a little yaw will do the trick. I salvaged some of my flights over Iceland by doing just that, but it's not a good work around. This needs to be identified and fixed by DJI, IMO.

Thanks again.

Just for info
I tested this with the mavic2pro todsy and it doesnot have this issue.
Will try this week with the mpp

Dan B-P4 Posted at 11-6 10:27
Just me or the video is password protected?

It is but further down the writer tells you the password is (mavicpro)

It indeed ruin the footage.
I had several of my shots trashed because of this behaviour last summer. I believed it was because of my mavic pro, but now I see it is a firmware issue.
DJI SHALL FIX IT ASAP.
How can we push this request to their attention?

GrizzlyAK Posted at 12-10 14:12
RadioFlyerMan, copy that. Thanks for the test and confirmation. The more the merrier, as they say.

Agree with the severity being greater at faster yaws, but the reason I first noticed this is when I was doing a slow cinematic tilt up while turning (yawing) flying over something. It was slow, but the subsequent 'jump' or hesitation in the video stream when played back was visible, and ruined the footage. One or two pixels on the video screen is noticeable. Granted, I could have cut the footage before that point in post, but that kind of leaves you wanting more.

Excellent point!

By the way, I executed a POI to see if the inherent yaw during execution would have any jerk while moving the gimble. I could not detect any at various travel speeds.

DO NOT count on DJI to fix this. It just won't happen. YOU must take action into your own hands and resolve this issue yourself.

Here is how: Roll back to V01.03.0700. (I used NLD) I have done this and do not have ANY flight control issues; as it pppears that DJI has slowly ruined the flight system software (firmware) with updates. Further updates will simply make it worse and possibly introduce new and improved issues. I can only guess that there is a system resource issue from the bloated firmware with all the cool features that RTH crash and dive bomb your face if you use active track. I do not have a use for those features on a professional drone product so they can really save that for the Spark, Tello and Mavic toy drones.

Good luck!

Since I use pano modes and low noise propelles, I cannot go lower than  V01.04.100
Any feedback related to 01.04.100 and this odd yaw behaviour?

dronybaloney Posted at 12-10 14:39
Just for info
I tested this with the mavic2pro todsy and it doesnot have this issue.
Will try this week with the mpp

Thanks, dronybaloney, good to know. Please do let us know what you find with your MPP.

Mgozer Posted at 12-11 01:08
It indeed ruin the footage.
I had several of my shots trashed because of this behaviour last summer. I believed it was because of my mavic pro, but now I see it is a firmware issue.
DJI SHALL FIX IT ASAP.

Mgozer, there are probably two ways to push this with DJI: You could open your own problem report with support.us@dji.com, or you could reference my Complaint Case that Support just created for me (after nearly 6 months of back and forth and 2 new drone replacements). You can email the address above, stating that you are experiencing the same issue reported in the following complaint case (which might be a more focused approach from DJI's perspective):

Complaint case ID: CAS-2564564-N1H6X3

They have acknowledged receipt of a link to this forum post, so they should also (if they read it) be able to see my videos as well as everybody's comments related to it.

Cheers

RadioFlyerMan Posted at 12-11 05:21
Excellent point!

By the way, I executed a POI to see if the inherent yaw during execution would have any jerk while moving the gimble. I could not detect any at various travel speeds.

RadioFlyerMan, that may actually make sense, since the drone itself is making the yaw calculations automatically without input from the user or RC. I do believe this is an issue with the interface between the gimbal tilt and yaw controls, perhaps either in the RC itself or in the Drone's FW after receiving those inputs from the RC. The code controlling the yaw may be completely different in a POI flight (and not affected by any tilt input), but without seeing the code, I'm speculating. This is a good data point though that should help the engineers trace this problem back to the source. Thanks!

Bob Brown Posted at 12-11 06:28
DO NOT count on DJI to fix this. It just won't happen. YOU must take action into your own hands and resolve this issue yourself.

Here is how: Roll back to V01.03.0700. (I used NLD) I have done this and do not have ANY flight control issues; as it pppears that DJI has slowly ruined the flight system software (firmware) with updates. Further updates will simply make it worse and possibly introduce new and improved issues. I can only guess that there is a system resource issue from the bloated firmware with all the cool features that RTH crash and dive bomb your face if you use active track. I do not have a use for those features on a professional drone product so they can really save that for the Spark, Tello and Mavic toy drones.

Thanks for the info, Bob. I may give that a try. At the very least, it is yet another data point that may help the engineers track this particular issue back to the source if, again, as you suggest, they actually do exert effort in resolving it. I doubt they'll remove any of the bloatware you mention, but they may be willing to fix this particular issue in a new FW release. Otherwise, the MPP is pretty useless as a serious filmmaking platform. it would also be bad juju if they didn't fix it because they now have the Mavic 2 and expect you to upgrade. However, I've never had a working copter (none of the three they've given me), so I'm not going to go away without either a resolution or a full refund or exchange to a device that works as advertised.

Mgozer Posted at 12-11 07:09
Since I use pano modes and low noise propelles, I cannot go lower than  V01.04.100
Any feedback related to 01.04.100 and this odd yaw behaviour?

Mgozer, .03 was the 'current' version when I got the first drone almost a year ago. Since then I've received two new replacements, which probably included a .04. The current drone I have is at .05 (just verified), and they've all exhibited this issue. Versions prior to that,though, I have no data on and have not tested, unfortunately.


Edit: However, I just see that my RC is still at .04.

GrizzlyAK Posted at 12-11 12:57
Thanks for the info, Bob. I may give that a try. At the very least, it is yet another data point that may help the engineers track this particular issue back to the source if, again, as you suggest, they actually do exert effort in resolving it. I doubt they'll remove any of the bloatware you mention, but they may be willing to fix this particular issue in a new FW release. Otherwise, the MPP is pretty useless as a serious filmmaking platform. it would also be bad juju if they didn't fix it because they now have the Mavic 2 and expect you to upgrade. However, I've never had a working copter (none of the three they've given me), so I'm not going to go away without either a resolution or a full refund or exchange to a device that works as advertised.

I was close to selling my drone last summer... NLD and Firmware downgrade saved the day. My bird is flawless now. Seriously... less seems to be better. Who would have thought?

which improvement did you notice going back to 03 firmware, apart from this annoying yaw/tilt stutter?

In addition, I am disappointed by the absence of the administrators/dji staff in this topic.

GrizzlyAK Posted at 12-11 12:38
Thanks, dronybaloney, good to know. Please do let us know what you find with your MPP.

I tested the mpp and i got it too. To be honest i dont find it a big issue.  if i yaw the mpp and at the same time pan the camera up i never take my finger of the turning wheel. Only after i finish the shot. I never use the scenario as shown in the video. I have flown the mpp for 1 year now and i never noticed this and never use this kind of movement as shown.

dronybaloney Posted at 12-12 11:14
I tested the mpp and i got it too. To be honest i dont find it a big issue.  if i yaw the mpp and at the same time pan the camera up i never take my finger of the turning wheel. Only after i finish the shot. I never use the scenario as shown in the video. I have flown the mpp for 1 year now and i never noticed this and never use this kind of movement as shown.

Dronybaloney, the movements shown in the video is only to demonstrate the easiest and most self-explanatory way of demonstrating the problem. It is not representative of the way I fly. I have hours of footage from Iceland that was ruined by this behavior, even times when I tried to “work around” the issue by ending the yaw prior to ending the tilt action. True, there are shots that can be accomplished flying as you describe, but there are other cinematic shots that do require a more coordinated ending that can’t be accomplished that way. Most of my original Iceland shots were of such nature, until I was force to work around the issue, doing exactly what you describe. In fact, that is when I knew it was a FW issue. I suspect there are a lot of MPP pilots who have not yet noticed the defect, and may never. But I believe anyone who uses the drone for serious cinematic scenes has at least noticed it, and perhaps discounted it to user error or a gimbal glitch, like i did initially