CsvView/DatCon version 4

[BudWalker]

Both CsvView and DatCon had a bug which was causing some users to not be notified if a newer version existed. That's been fixed in CsvView/DatCon 4.0.3. Hopefully, this thread will be seen by those users.

DatCon 4 is a substantial reorganization of DatCon 3. Most of the changes will also affect CsvView 4 when used to display a .DAT file. The current version, 4.0.3, is an alpha version. Bugs can be reported here or email to [email protected].

Just wanna know the bare minimum to use DatCon or CsvView 4.0.3? Check out the Signals section (page 5) of the new DatCon User Manual here
DatCon 4
And, check out this document which shows how several signals have been renamed.
old2new

DatCon 4 is more explicit about which data is the result of the reverse engineering that was necessary when the DatCon project started. Assumptions and guesses were made that have been overly simplistic, partially incorrect or totally incorrect. The new organization makes the remaining assumptions and guesses explicit.

DatCon 4 also relies on a better understanding of the .DAT structure. Converting a .DAT for the Mavic 2, Mavic Mini and Mavic Air 2 platforms will yield much more data than DatCon 3. If you get a chance check out the AdsbEscape data found in the .DAT posted here
Crashed MA2. Too much damage??

 

[slup]

Both CsvView and DatCon had a bug which was causing some users to not be notified if a newer version existed. That's been fixed in CsvView/DatCon 4.0.3. Hopefully, this thread will be seen by those users.

DatCon 4 is a substantial reorganization of DatCon 3. Most of the changes will also affect CsvView 4 when used to display a .DAT file. The current version, 4.0.3, is an alpha version. Bugs can be reported here or email to [email protected].

Just wanna know the bare minimum to use DatCon or CsvView 4.0.3? Check out the Signals section (page 5) of the new DatCon User Manual here
DatCon 4
And, check out this document which shows how several signals have been renamed.
old2new

DatCon 4 is more explicit about which data is the result of the reverse engineering that was necessary when the DatCon project started. Assumptions and guesses were made that have been overly simplistic, partially incorrect or totally incorrect. The new organization makes the remaining assumptions and guesses explicit.

DatCon 4 also relies on a better understanding of the .DAT structure. Converting a .DAT for the Mavic 2, Mavic Mini and Mavic Air 2 platforms will yield much more data than DatCon 3. If you get a chance check out the AdsbEscape data found in the .DAT posted here
Crashed MA2. Too much damage??

Have some ... let's say observations, can't say it's a bug, perhaps more of a inconvenience. But thought I mention it if it's a way to correct it (... or perhaps I need to get used to it )

I't's about how the stick commands is shown in the DAT logs ...

In a .TXT the signal group still says RC & all stick commands is shown on the same Y-axis with 0 in the middle ... like below. It's easy & quick to see where & how big the the inputs are just by a glance.



In a .DAT log the inputs can apparently be in different named signal groups depending on from which AC they come ... from a Mini the stick commands are in the Mr_controller group & it looks like here below. All stick commands are on separate Y-axes & 0 isn't in the middle ... if it's only positive inputs 0 will be low down & opposite if only negative inputs. All much harder to quickly take in.



From my MA1 the stick inputs are in the Controller signal group ... but equal to the Mini, all signals on it's own axes & not 0 in the middle.

 

[BudWalker]

Have some ... let's say observations, can't say it's a bug, perhaps more of a inconvenience. But thought I mention it if it's a way to correct it (... or perhaps I need to get used to it )

I't's about how the stick commands is shown in the DAT logs ...

In a .TXT the signal group still says RC & all stick commands is shown on the same Y-axis with 0 in the middle ... like below. It's easy & quick to see where & how big the the inputs are just by a glance.

View attachment 108515

In a .DAT log the inputs can apparently be in different named signal groups depending on from which AC they come ... from a Mini the stick commands are in the Mr_controller group & it looks like here below. All stick commands are on separate Y-axes & 0 isn't in the middle ... if it's only positive inputs 0 will be low down & opposite if only negative inputs. All much harder to quickly take in.

View attachment 108518

From my MA1 the stick inputs are in the Controller signal group ... but equal to the Mini, all signals on it's own axes & not 0 in the middle.

View attachment 108519

I think most of what you're describing stems from these signals not having a common axis. You can always get those 4 signals onto a SigPlayer and then hit the Multi Axis/Single Y Axis Button. That will result in all 4 signals having the same axis.


You'd probably want to CsvView to do that for you since it's known that these 4 signals have the same scale and bias. But, what to call that axis? In this case I don't think it means stick input. Looking at the Mavic Mini there, is in addition to mr_controller:ctrl_pitch, the signal RC_Info:E. Putting aside the scale difference, they are close but not the same

Up until version 4 one or the other of these two signals have been used to represent stick input. That's a mistake. Even if there is only the mr_controller:ctrl_pitch signal it's not clear that it represents stick input. In this incident
MA2 (RC Disconnected) No RTH, Drone gone
the pilot claims that no stick input was given yet the log indicated otherwise. Judging from the discourse I believed him. The possibility that he was correct was summarily dismissed. Maybe that could have been avoided if the log analysts had known about the stick input subtleties. @Scott

Version 4 is all about reducing the problems stemming from the guesses made by the DatCon developer - that would be me. Let's do this. The mr_controller signals will be Engineered so that they have a common axis with name "Mr_controller:Input". Likewise the RC_Info signals will have a common axis with name "RC_Info:Input". IMHO, it's a pretty safe assumption that each group has a common scale and bias. The term "Input" is sufficiently vague so as to avoid any particular interpretation.

 

[slup]

Hi @BudWalker , any way to get the IMUCalc track into the GeoPlayer in CsvView 4.0.3?

In earlier versions you had an alternative like this:



But that isn't there anymore ...

 

[BudWalker]

Hi @BudWalker , any way to get the IMUCalc track into the GeoPlayer in CsvView 4.0.3?

In earlier versions you had an alternative like this:

View attachment 109202

But that isn't there anymore ...

ImuCalcs are done automatically. So there is no need to select them.

It's a bug that ImuCalcs aren't shown in the GeoPlayer. That's fixed in the next version.

 

[BudWalker]

Beta version 4.0.4 is available. There is also documentation on the IMUCalcs package.
http://datfile.net/Doc/IMUCalcs.pdf