Minor Bug: Encode->Overlay->Channel Title is NOT enabled, but text shows up on snapshots (not on video as it should)

[alekk]

I've got a IPC-T54IR-ZE-S3 running the very latest software ... although I believe it may have been in the previous version as I looked at snapshots from before the recent upgrade.

Encode->Overlay->Channel Title is NOT enabled ... but the default value of "IPC" was still there for Input Text.
This is NOT seen in the live feed/video (which is correct), but if I look at captured snapshots, it's there on the lower left.

If I change this text to something else, that shows up (so it is using that field) and if I make it blank, it won't accept that ... so I just put a single space in there! ;-)

So I thinking there is a bug in the snapshot code such that it looks at the "Input Text" and uses that ... WITHOUT first checking to see if it is even enabled.

Plus you should ideally be able to blank out this field.

 

[alekk]

Here's another oddity that I noticed and was able to duplicate on two different camera's.
I'd be curious if anyone else is seeing ... or maybe I'm doing something really stupid! ;-)

I have a time-stamped overlayed in the top right of the video/image/etc.
This is done via Camera->Encode->Overlay->Time Table->Enable (and then move it where you want it)
Note that while I can get it "all the way" to the top, I can't seem to push it quite all the way to the right?!?
Now go to Camera->Encode->Overlay->Font Properties->Font Size and changed from 32x32 to 16x16

The timestamp gets smaller on the video ... but when I look at the saved snapshots, it's not there?!?

This seems to further support my theory that there is duplicative/different code paths for the text overlay depending if producing video or images.
Ideally this should all the single/same code ... and then near the very end, branch to the appropriate output format.

P.S. Also per this post:
4. Speaking of snapshots, I think there is a bug in the code such that for these, the dynamic range is compressed and I'm guessing they are using a grey scale of [16-235] rather than [0-255] - see thread here. It's pretty easy to duplicate - compare a snapshot obtained via the API versus one using a snapshot from the web interface (or pulled from the video stream). I (mostly) use ffmpeg to pull the video feed (you don't see the issue there), but sometimes it's nice to pull snapshots (either in real-time or from those taken due to events - loving the new animal feature BTW!) and those look flat due to the reduced dynamic range. I show some histograms in post #23 and in #24 and hypothesize this might be due to using the YCbCr color space(?) Note this is NOT an image quality/sharpness issue - I would expect those to be different depending on the bit rate of the video stream versus the compression of the snapshot.