What do you Guys make of this

[Skywatcher2001]

While it's not in the user manual preflight checklist (really should be) it seems you should always make sure your drone icon on the map matches where in fact your drone is pointing.

Yes!! We have beat this up for years since the first mini release and a slew of fly/blow aways with compass and IMU issues. SAR104 has a big write up on Compass and Magnetic Fields . . . unfortunately DJI doesn't listen (or read) this forum or even its own . . .

 

[Bussty]

Yes!! We have beat this up for years since the first mini release and a slew of fly/blow aways with compass and IMU issues. SAR104 has a big write up on Compass and Magnetic Fields . . . unfortunately DJI doesn't listen (or read) this forum or even its own . . .

Thanks James might try and find SAR104's write-up, from memory he is a very knowledgeable chap! Cheers

 

[Skywatcher2001]

Thanks James might try and find SAR104's write-up, from memory he is a very knowledgeable chap! Cheers

Here it is .. . . A short explanation of compass function, calibration and errors.

 

[Bussty]

Here it is .. . . A short explanation of compass function, calibration and errors.

Good man! Thank you!

 

[Dougcjohn]

Until sales decline because of it they have no reason to NOT ignore it.
Would you care to bet that there will still be a metaphorical queue of people waiting at the door to buy the next model of drone that DJI release and or shouting about "oh I just placed my pre-order"?

No argument there, Buyers will still form lines... look at the huge reception for the M30 by Reviewers. It's already receiving the Best Drone praise before it's even in the hands of Users. Run out and drop $14-18k, great M2EA or M300 alternative.

When the rumored Inspire 3 is released, I'm sure it'll be Reviewed & Expressed as the best Cinematography craft, a great replacement for the aging Inspire 2.

I will agree, presentation of these new drones gets many anxious to aquire... I'm guilty & in some of those groups. I read positive of M3, and some expected negative... dismissing some and knowing FW corrections will arrive. Then later with M3, began receiving more negative and questioning if corrections will arrive after return expires.

And... that repeats. Reading & watching the M30 reviews and Inspire 3 rumors. Buyers will be anxious to aquire!

It takes more than 1 or 2 negative products to change public opinion. Many companies, auto industry, camera, appliances, etc have made bad or poor performing products and their market share continues. The Samsung Note model failure, iPhone iOS issues, Ft Load Washers, etc... market shares still strong.

The M3 may sting DJI a little... but if they Bring out a M3-S model that corrects and improves... Buyers will continue. The only ones that will feel really agitated will be the M3 Owners. Similar to the Air2 vs the Air2S Owners... some will declare being Done with DJI. Final outcome, it'll encourage anxious Buyers not to be in First Wave Buyers... many sworn never to buy DJI again will eventually aquire and DJI market share survives.

 

[Chaosrider]

So how do we explain the early firmware that we had that had no issues for many including myself ,, somehow something has gotten lost in the translation , and I am sure it is more complicated than that but maybe not.

Sounds like poor regression testing to me.

Regression testing is a dying art. A lot of people writing SW now don't even know what it means.

Regression testing is what you do after you make changes to a SW system, to make sure you didn't unintentionally break something that worked before, which you assumed was unrelated to the changes that you made.

As SW gets larger, more complicated, and more internally integrated, testing in general and regression testing in particular has become much harder.

I suspect there was an "Oh Sugar!" moment at DJI when they realized that in their attempt to fix thing X in the firmware, they unintentionally busted thing Y.

If that's true, it means that there are linkages within the SW that they hadn't previously understood. Frustrating though it is for users, it's better to actually figure out how that happened, than to just throw possible fixes at it. They could try to roll back to the previous version, but that has its own complications.

Not to mention that the whole system has to integrate with a bazillion different phone types, which will have their own update schedules, any one of which might have the potential to break the DJI--Phone interface.

I got my first programming job in 1974, writing output subroutines for what was, at the time, the biggest computer in the world, the CDC 7700. Military, of course. I think it was running a staggering 500,000 lines of code...

;-)

When I was Director of IT for the Nevada state government 30 years later, in 2004, I coined the smartbutt phrase:

"Interfaces = Death"

When the Governor asked me at one point what the biggest IT problems would be in the next 5 years, I said without missing a beat:

"Interfaces and integrations"

Nothing since then has suggested to me that I should modify either of those pronouncements!

;-)

Has anyone thought to correlate the GPS lock failures, with the type of phone being used?

Just a thought...

TCS

 

[Dougcjohn]

Sounds like poor regression testing to me.

Regression testing is a dying art. A lot of people writing SW now don't even know what it means.

Regression testing is what you do after you make changes to a SW system, to make sure you didn't unintentionally break something that worked before, which you assumed was unrelated to the changes that you made.

As SW gets larger, more complicated, and more internally integrated, testing in general and regression testing in particular has become much harder.

I suspect there was an "Oh Sugar!" moment at DJI when they realized that in their attempt to fix thing X in the firmware, they unintentionally busted thing Y.

If that's true, it means that there are linkages within the SW that they hadn't previously understood. Frustrating though it is for users, it's better to actually figure out how that happened, than to just throw possible fixes at it. They could try to roll back to the previous version, but that has its own complications.

Not to mention that the whole system has to integrate with a bazillion different phone types, which will have their own update schedules, any one of which might have the potential to break the DJI--Phone interface.

I got my first programming job in 1974, writing output subroutines for what was, at the time, the biggest computer in the world, the CDC 7700. Military, of course. I think it was running a staggering 500,000 lines of code...

;-)

When I was Director of IT for the Nevada state government 30 years later, in 2004, I coined the smartbutt phrase:

"Interfaces = Death"

When the Governor asked me at one point what the biggest IT problems would be in the next 5 years, I said without missing a beat:

"Interfaces and integrations"

Nothing since then has suggested to me that I should modify either of those pronouncements!

;-)

Has anyone thought to correlate the GPS lock failures, with the type of phone being used?

Just a thought...

TCS

Good points, as an old IT myself I concure on development pains.
The M3 GPS isn't dependent on Phone... you can turn on before App starts or connects and it'll be acquiring. Many are using the RC Pro with Internal screen and Android OS without cellular connectivity.

 

[Chaosrider]

Or this might be a great time to purchase a steeply discounted, lightly used M3 from one of the many disappointed M3 owners.

That's an interesting thought, depending on how deeply it's discounted.

Has anyone done a look-see market survey about what's available now in the used M3 market?

I just bought my first used drone, and it's opened my mind to a new range of possibilities...

;-)

Thx,

TCS

 

[Dougcjohn]

That's an interesting thought, depending on how deeply it's discounted.

Has anyone done a look-see market survey about what's available now in the used M3 market?

I just bought my first used drone, and it's opened my mind to a new range of possibilities...

;-)

Thx,

TCS

Just my Opinion... until a model comes out that replaces the M3... I don't think many will sell M3 at any big discount.

 

[Chaosrider]

It seems to me that the technical issues DJI is having is made far worse by the lack of candid communications with its fiercely loyal customer base. It really is no wonder that the lack of transparency is fomenting all of this speculation and ill will.

I’m sure this manufacturing giant is capable of issuing an official statement detailing the problem and it’s plan for resolution without compromising industry secrets. We deserve nothing less. Many reputable companies send products to market which fail to live up to all expectations. Today, however, we usually don’t let them get away with it without an explanation - nor should we.

Customers don’t expect perfection, just reasonableness.

The best that can be reasonably expected in this regard would be a statement acknowledging the problem, and that it's being worked on.

And the essential question is....how common is this problem? If it impacts 50% of the M3s, that's a big problem, and an acknowledgment would be appropriate.

On the other hand it it's just 5% of them...that's just something you fix in the next release.

Is there any quantitative data about this available?

Just out of curiosity, it would be interesting if all of the M3 owners here who have had a problem, would post which specific phone that you're using. Maybe there will be a correlation there, maybe there won't be, but it would be interesting to see.

Thx,

TCS

 

[Chaosrider]

You do realize that some frighteningly large fraction of the people who read this will believe every word?

What's the problem...you don't believe it??

;-)

TCS

 

[Chaosrider]

Received my Mavic 3 February 8th. Typically took just a little over 1 minute to ID enough sats to launch. After the latest firmware it still takes a little over 1 minute to launch. The discrepancy I'm reading from different people is what is most perplexing. I guess I'm one of the lucky ones!

There actually seems to be a natural variation in acquisition times. I fly three Mini-2s, and their GPS acquisition times are very different. The best one gets the "Go!" from the Fly Lady in just 15-20 seconds, and the worst takes over a minute.

Same flying location in every case.

I have no idea what might cause that variation, but with three different Mini-2s flying from the same location, it definitely exists.

I update them all to the most recent firmware as soon as it's available.

TCS

 

[Chaosrider]

The Samsung Note model failure, iPhone iOS issues, Ft Load Washers, etc... market shares still strong.

What's the Samsung Note model failure?

I have a Samsung Galaxy S20 Note, which I bought specifically to run DJI FLY (neither my existing phone nor tablet at the time would run it). It's performed flawlessly.

Just curious...

Thx,

TCS

 

[Chaosrider]

Good points, as an old IT myself I concure on development pains.
The M3 GPS isn't dependent on Phone... you can turn on before App starts or connects and it'll be acquiring. Many are using the RC Pro with Internal screen and Android OS without cellular connectivity.

Interesting.

Not (yet) being an M3 owner, I'm not familiar with the common practices involved. I was just drawn to this thread by the SW puzzle...

;-)

However, while your new-to-me data suggests that the problem may not be phone dependent, it doesn't rule that out. Perhaps, not using a phone contributes to the problem...or prevents it.

Without data, there's no way to know. I hope some of the M3 owners will post whatever configuration they're using, and whether or not they have a problem. For this purpose, reporting non-problems is as important as reporting problems. If your situation is "I use XYZ configuration and I have no problem", that's useful information!

Thx!

TCS

 

[Chaosrider]

Just my Opinion... until a model comes out that replaces the M3... I don't think many will sell M3 at any big discount.

That would be my guess too.

Used Mini-1 prices have stayed surprisingly (to me) high, given the release of the Mini SE.

It did finally find the Mini-1 that I was looking for. I paid more than I was looking to pay, for a bigger package than I was looking to get. I was looking for a base Mini-1 kit...aircraft, remote, 1 battery...for $150. I ended up buying a full Fly More Combo...which has three batteries, a charger, and prop guards...for $230. Not long before I bought this one, I lost an eBay auction for the same Combo to a $232.50 bid.

A reasonable price, and I'm very glad I got it (or it least, it's on it's way here), but in the presence of the Mini SE at $300 for a basic kit new, I would have expected used Mini-1 prices to be lower.

My best guess for why, and a guess is all it is, is that the capital and space cost of just keeping a Mini-1 are both so low, that people have no real motivation to sell for less than they'd like to get for it. If that's the case, Mini-1 prices will continue to creep down, but very slowly.

Thx,

TCS

 

[Dougcjohn]

Interesting.

Not (yet) being an M3 owner, I'm not familiar with the common practices involved. I was just drawn to this thread by the SW puzzle...

;-)

However, while your new-to-me data suggests that the problem may not be phone dependent, it doesn't rule that out. Perhaps, not using a phone contributes to the problem...or prevents it.

Without data, there's no way to know. I hope some of the M3 owners will post whatever configuration they're using, and whether or not they have a problem. For this purpose, reporting non-problems is as important as reporting problems. If your situation is "I use XYZ configuration and I have no problem", that's useful information!

Thx!

TCS

There are multiple threads on the GPS matter... more posts than probably desired. General consensus of User Experience and Technical references does indicate it's a Craft Technical issue... if that's FW or HW, that's not been identified and doubt will unless by DJI, and that won't happen based on past company experiences and hasn't created a safety concern... like the TB50 battery issue with M210v1 a few years back.

Their is a large latitude of User experiences regarding the GPS issue... that too hasn't been specified in production lots, electronic supplies, etc... again not a resource a non-compamy forum will reveal without DJI statements.

There is a quantifiable problem with the GPS.
That will either be corrected w FW or new model.

 

[RangerEd]

Received my Mavic 3 February 8th. Typically took just a little over 1 minute to ID enough sats to launch. After the latest firmware it still takes a little over 1 minute to launch. The discrepancy I'm reading from different people is what is most perplexing. I guess I'm one of the lucky ones!

Or maybe you're in the right part of globe.

 

[Dougcjohn]

What's the Samsung Note model failure?

I have a Samsung Galaxy S20 Note, which I bought specifically to run DJI FLY (neither my existing phone nor tablet at the time would run it). It's performed flawlessly.

Just curious...

Thx,

TCS

Few models back... Note 7
Note battery failures, fires and eventually a recall that received Telco companies participating by killing the connection of the Note if Owners didn't send back in for partial refund or trade-in allowance.

Mine worked flawless, loved it.. but eventually had to surrender... moved so S model until new Note arrived.

Point was, even that level of failure... Samsung continued strong and many not even aware of a serious problem.

Little history:

A Brief History of Samsung's Troubled Galaxy Note 7

Some considered it the best smartphone ever made — now it's no longer being made

time.com

 

[Chaosrider]

There are multiple threads on the GPS matter... more posts than probably desired. General consensus of User Experience and Technical references does indicate it's a Craft Technical issue... if that's FW or HW, that's not been identified and doubt will unless by DJI, and that won't happen based on past company experiences and that it hasn't created a safety concern... like the TB50 battery issue with M210v1 a few years back.

Their is a large latitude of User experiences regarding the GPS issue... that too hasn't been specified in production lots, electronic supplies, etxc... again not a resource a non-compamy forum will reveal without DJI statements.

There is quantifiable problem with the GPS.
That will either be corrected w FW or new model.

I'm sure that's true, that it will be fixed on way or the other, but as a once and future geek, I'll still be curious to know what the cause was.

;-)

TCS

 

[Chaosrider]

Or maybe you're in the right part of globe.

Is there any quantitative data about whether location matters? I'd consider that a plausible contributing factor, but only data can actually answer the question.

TCS