Manipulating 4k input

General
41

No, this were two suggestions: Use embedded audio, that comes out of the Monitor card OR do the delay in the audio mixer. Do whatever is easier.

If you play the input, including embedded audio, on the 12G and then route that picture, including embedded audio, to the monitor card, the audio, that is embedded in the monitor cards SDI signal should still be in sync.

What version of Caspar do you use?

Can you key DIAG into the CasparCG console window and post a screenshot of the diag window?

42

Hi again.

I have sorted out the above matter. Now I have a new challenge. I’m wondering if it’s possible to extract only the audio from different layers and send it to the sound card. Is this possible in CasparCG? Thank you.

43

Create a channel with only a systemaudio consumer and route or play the layers on it, sync to other channels may be an issue.

44

Thank you!

45

How did you sort the matter above?

46

So, I developed a CasparCG client for recording news broadcasts. We needed to shift the camera feed (4K) that films the anchor to the side, in order to make room for graphics (OTS) to be loaded on the right side of the presenter.

The setup includes two computers: one for controlling the camera movement and another for loading the graphics. We connected the 4K camera to an input (Blackmagic Extreme 4K 12G) and output HD through a Blackmagic Mini Monitor.

In essence, we routed the 4K input to the output using the following commands:
PLAY 1-10 DECKLINK DEVICE 2 FORMAT 2160p5000
PLAY 2-10 route://1
PLAY 2-10 DECKLINK DEVICE 3 FORMAT 1080p5000

We then moved the input with:
MIXER 1-10 FILL 0 -0.1 1.15 1.15 50 EaseInOutQuad
or
MIXER 1-10 FILL -0.130208 -0.1 1.15 1.15 50 EaseInOutQuad

We encountered a few issues, especially with audio, but after replacing the SDI cables, everything worked fine.
Here’s a recording of how it looks: https://we.tl/t-6gdYLQxK8l

47

Interesting. Why don’t you play the graphics directly on top of channel 2, instead of using two computers?

48

The display of all graphics (picture in picture, …) is built using multiple layers. First, I build everything on an empty channel, and then route it to the active channel (to avoid the effect where you see it being assembled). When everything was running on a single computer, we had issues with stuttering, so now a second computer handles the processing.