Poor weather = days for experimenting and trying things out. Here’s a few of things I’ve learnt:
Remote ID
A Mini 4 Pro can broadcast its RID over at least 160 metres in open country.
An Air 3S can broadcast its RID over at least 252 metres in open country. If your DJI Fly app has a Remote ID section in the Safety tab, it can broadcast RID (e.g. V 1.20.0); if it doesn’t have anywhere to input your credentials, it won’t broadcast RID (e.g. V 1.19.4) - even if your credentials had previously been entered into a different version of the app. In other words, the ability to broadcast RID depends on the software currently in use in the controller, not the drone’s firmware. So keep an older version of DJI Fly available (or use Litchi Pilot where that is possible) if you don’t want to broadcast RID.
Similarly, the limiting of maximum altitude is set by the controller software, not the drone firmware.
Turning off your KarenPhone’s WiFi hotspot seems to make RID reception easier. DJI and Potensic drones use WiFi Beacon mode for RID, not Bluetooth. The WiFi scanning rate can be increased by going into Android’s Developer mode and turning WiFi throttling off.
OpenDroneID picks up the drone quicker and seems to refresh more frequently than DroneScan.
Noise
A Mini 4 Pro is quieter and less intrusive than an Air 3S (no surprise there)
Master Airscrew props are not reliably, measurably, quieter than DJI props.
To my ears, above 45 metres an Air 3S isn’t intrusive; above 50 metres you can’t locate it by sound and above 60 metres it is inaudible.
Cinematics
Adopt one set of Gain & Expo settings, then stick to them so you can learn what to expect.
If you use multiple drones, replicate camera and flight settings across controllers and drones so everything works the same way.
Fixing white balance by setting it to auto then to manual to lock it works well and holds the value as the drone pans around in varied lighting conditions.
Fixing shutter speed in Pro mode and then letting the ISO auto-adjust can work, but is often no better than Auto exposure mode.
If you shoot video with H265 coding, it is 10 bit, even in Normal colour mode.
If you use a LUT to colour grade your footage, it often becomes recognisable as a particular DJI LUT. If you use a less aggressive LUT, you might just as well have used 10-bit Normal colour mode and saved yourself a grading job. LUTs are fine for producing consistent colour effects. If you just want natural colour, use Normal colour mode.
Telephoto lens autofocus is better if you are not fully zoomed in – 8x is better than 9x. Better still avoid digital zooming if you can.
I don’t like the 180° shutter speed rule. I think it simply perpetuates the failings of twentieth-century cinematography, and we have moved on since then. For most of the motion I am filming I want to be able to analyse the motion, not blur it.
Filters
For combined Neutral-Density-and-Polarising filters, I think the ND16PL performs best. The ND64PL and denser filters are pretty much a waste of time even in strong midday sunlight, and are only of use in special circumstances.
The NDxPL filters are less good at looking through water than the straight CPL filter.
Most polarising filters are set by default to reduce reflections from water surfaces. To optimise the reduction of reflections from glass, they need to be turned 90°.
Waypoints and 3D models (using free software)
Litchi is a better waypoint tool than DJI Fly.
MipMap is slow (it runs on my home PC) but it gives the best 3D models.
Polycam is reliable but limited to 150 stills or 3 minutes of video as source.
Skyebrowse is expensive in order to get it to do anything
@bryand When flying a drone low over grass in bright sunlight without ND filters (leading to a high shutter speed and breaking the 180-degree rule), the resulting “pixelated,” shimmering, or sharp, choppy look produces at best, aliasing, moiré patterns, and/or shimmering foliage and looks quite frankly… crap.
That’s an interesting read - even moreso the educated comments afterwards. Damn, I love this site!! Thx for putting the effort in and making my little grey cells work up a sweat….
You don’tsay how low “low” is, but here’s a clip taken so low the camera couldn’t focus properly on the grass. Nevertheless, it’s not pixelated and has no moire patterns.
I don’t normally fly that low or in Sport mode, but it works for me. Air 3S with a CPL filter.
The filter was a CPL, not a Neutral Density. No real effect on exposure.
Shutter speed varied between 1/800th and 1/1600th sec, with ISO between 100 and 110,
Hmmm…CPL filters act like light sunglasses for your drone’s camera, typically reducing the light entering the lens by 1.5 stops.
Because they block a portion of incoming light, they force your drone’s camera to compensate in order to maintain a proper exposure. How this affects your shutter speed depends on the drones settings:
In Auto Mode the drone will automatically detect the drop in light and will slow down the shutter speed to let in more light. Therefore you may unintentionally get some, or more motion blur, in your footage anyway.
Maybe test in strong sunlight in auto without any filter for a true result?
As I said, the slowest shutter speed was 1/800th, freezing motion. I can’t think of a situation where I wouldn’t want to use a CPL filter, however much it absorbs, and anyway even if it absorbs 1.5 stops, on a camera with a 14 stop dynamic range the effect isn’t enough to worry about.
Fair enough. But there are an awful lot of pro’s out there that have a different view.
Wouldn’t 1/8000 shutter speed not cause juddery, choppy, or stuttering motion effects and will almost entirely eliminate natural motion blur?
The camera sensor is exposed for such a tiny fraction of time, each frame becomes a perfectly “frozen” snapshot as you say. When these snapshots are played back, your brain doesn’t receive enough blurred information to bridge the gap between frames, resulting in an unnatural “stop-motion” or strobe-like appearance.
I eagerly await your video, taken at 1/8000th sec (or even 1/800th if that was a typo) showing the artefacts you describe. I’ve shown you mine: now let’s see yours.
I’ve done about 340 videos on YouTube and have used ND filters on the majority of them, using close to the 180 rule. Motion blur, galore! Why would I not use 'em! I’ve showed examples of the issue using more learned guys than myself and could show you more. It is real, I’m not making it up!
If your happy what your doing then rock on, why should I worry about it! Just don’t mislead others into thinking ND filters are waste of time.
I’ve showed examples of the issue using more learned guys than myself and could show you more. It is real, I’m not making it up! 
