Punchbuggy
First Officer
Flight distance : 573054 ft
Australia
Offline
|
Aardvark Posted at 2017-7-27 03:31
So it could be the firmware updating procedure that may be a bit flawed rather than the firmware. Is everybody using latest version of Assistant 2 as opposed to an older version installed sometime ago ?
Yeah, I agree with you. I've been wondering why the update process has such variable success, and that would explain it. Of the firmware releases captured in Icer's Dumbledore collection, the only one I have real confidence in is the .602 which Duc provided from his new P4 (if I got that right, Duc). The others were captured, to my understanding, as update downloads.
Two interesting points of note from this amazing thread; firstly, each firmware release is a collection of individual controller updates applied one-by-one, as guided by the master .cfg file. Secondly, the flash process Must be done twice, as the first flash doesn't apply all individual updates on the craft. My take from the last is that the first flash may not apply drivers already existing at the same (or more recent) version level, but doing it twice is like a 'brute force' approach - desirable should the existing driver have been corrupted on download.
Personally, I'm thinking that the best approach for DJI from here is to change their whole approach to craft updates. DJI should simply make available their firmware release as a single .bin file, like their P4P+ controller, which can be downloaded with integrity checking, then applied locally. The current issues are most likely caused by corruption on downloading from highly conjested DJI servers, with further latency introduced via internet links, home WiFi, etc. I wonder what file integrity checking is done during the current DJU Assistant 2 process.
But, of course, I'm just making observations, and my conclusions could be cr@p... hehe |
|