[FFmpeg-user] Understanding ffprobe output

Simon Roberts simon at dancingcloudservices.com
Sat Nov 28 17:18:33 EET 2020

Thanks for the input Carl :)

On Fri, Nov 27, 2020 at 5:05 PM Carl Eugen Hoyos <ceffmpeg at gmail.com> wrote:

> Am Fr., 27. Nov. 2020 um 19:50 Uhr schrieb Simon Roberts
> <simon at dancingcloudservices.com>:

> > One camera feeds to a monitor that's only capable of a 60 Hz vertical.
> The
> > monitor just says "1080 at 60 Hz". The other camera feeds a different type
> of
> > monitor that runs at 30 Hz vertical. (Note, I don't really know what
> > "vertical" means, frame? field? can't tell. I also don't know if the
> higher
> > rate one is running interlaced. I have some vague recollection that I had
> > noticed hints that it's progressive, but I don't recall what those hints
> > were)
> Just compare the depth of the screen with its diameter;-)
> (Only CRT's know about fields.)

So... the term "field" isn't used to describe one half of an interlaced
frame? 'coz I have devices that claim to output progressive and interlaced
as a choice, and they're not CRTs..

> > Anway, everything works, and the recordings are fine. But I noticed that
> > the file size for the "60 Hz" camera was nearly double that of the other
> > camera. Both cameras have identical settings internally, and are actually
> > set at 1080p output.
> The bit rate shown in ffprobe's output confirm what you write here, I
> believe there is an endless number of reasons why the cameras
> produce streams with different bitrates.

Well, it's unfortunate if there are really that many, because that's the
information I need.

Can anyone tell me the most likely features I should look for in a monitor
that would create/avoid this effect?

I had believed it was happening because of a 60Hz source frame rate
something (since that's the only documented difference between the two
monitors, and as I mentioned in opening, the cameras are identical, make,
model, and firmware version)

However, having counted the frames in two files that exhibit this effect, I
can say categorically that they're both running at a constant 30 frames per
second. That does not, however, tell me what the input rate was.

Is it still possible that if the source were at double the frame rate with
duplicated frames, this would increase the resulting bitrate this much? I
know for a fact the source generates duplicated frames when running 60 fps,
since a) the shutter speed on the camera is 1/30th and b) I've looked at
the recording I made "raw" and saw 60 frames per second with each frame

If it's not the source framerate, what else could create this higher
bitrate? (Note again, these are identical cameras, and they're aimed at
essentially the same scene, which is mostly--in the test cases--static
chromakey green).

Thanks again!

More information about the ffmpeg-user mailing list