PPK data now has two camera timestamps from Dual 640T since FW 2.3.23

[vr-pilot]

Hi there,
Did anyone already notice this, too? (Most likely a "Dual 640T"-only-problem due to its two cameras)
Yesterday I flew a mapping mission planned as PPK due to possibly bad network connection on site.
When using Emlid Studio v1.3 for PPK-processing, I got a message pointing to an error that the number of photos does not match the number of timestamps in the .mrk-file.
Checking the respective Timestamp.mrk-file I noticed that there were doublettes for almost every image number, having different GPS "taken-at" times and positions.
By comparing the images' GPS locations from the MAX-camera (RGB) and the IRX-camera (IR) with the Timestamp.mrk-file entries, I realized that the first doubled number line was for the IRX-camera and the second for the MAX-camera.
If there was only one line (no image doublette) it was for the IRX-camera.
In order to get only timestamps for the MAX-camera (RGB), I manually edited the Timestamp.mrk-file by deleting all single number lines or only keeping the second line from the doubled numbers.
It was not very handy and can produce some "human errors".

This flaw is due to the "well intended" change of always taking two images (instead of one) even if IR or RGB only is selected. This came with FW 2.3.23.

Is it the same with the EVO2 RTK Dual 640T (not Enterprise) version?

 

[jmason702]

next time I fly my 640T in RTK, Ill let you know

 

[vr-pilot]

next time I fly my 640T in RTK, Ill let you know

Thank you, that would be great!
IIRC you also fly the EVO2 RTK(not Enterprise). Is it the "EVO2 RTK" with "Dual 640T"or "PRO"?
For sure the timestamp doublets derive from the "Dual" cams of the "640T". Unfortunately they are not handled as different cameras by i.e. having their separate Timestamp.mrk files.
On the "positive" side (as I mentioned) is the fact that now RGB and IR images are taken simultaneously not only during PIP mode: this means that RGB imagery is no longer limited to UHD but is shot at QUHD along with the "NTSC" IR images. That also applies to (all) mission modes.

Also interesting: I was able to switch between the two cameras' views during polygon mission mode while both their images were continuously saved to the SD card at QUHD and NTSC respectively.
But: if you change the camera view to "both" (PIP) the images are saved in the "usual manner" as RGB-UHD, IR-NTSC and PIP-HD.

IMO the most valuable way of "debugging" the Timestamp.mrk-file's doublet image listing would be to store them in two useable separate files.
(Allowing for precise PPK of RGB and IR imagery!)
Or (as before) let the app decide for one of them i.e. depending on the camera selected for the survey pattern (flight path "fixation").

The aim should be to have reasonable data again inside the resulting FTASK folders...

 

[DroneRat]

Sadly all is UNCHANGED as of 7/1/2023!

 

[vr-pilot]

Sadly all is UNCHANGED as of 7/1/2023!

What I noticed some weeks ago that if you set all exposure options to AUTO, the timestamps file gets 100% consistent with two values for every dual shot of the 640T.
(IIRC it only happened once that on the first "sharp corner start' of a mission the entry for the RGB camera (second line) was missing)
As I always used Shutter Priority before for manual shutter speeds regarding ground speed vs. hight/resolution, I am now relatively sure that this extra EV processing could cause the timemark inconsistencies...