Question / Help Control overall audio volume in stream

Yes. Here it is:

- YouTube

Enjoy the videos and music you love, upload original content, and share it all with friends, family, and the world on YouTube.
youtu.be youtu.be

The metering to the right is a german Steinberg Wavelab ("PEGELMETER"). The audio is stuttering a little bit due to accelerating the video 3x (each test tone was 20 seconds originally).
So i ACK: the levels are at least accurate.
 
konsolenritter said:
Yes. Here it is:

- YouTube

Enjoy the videos and music you love, upload original content, and share it all with friends, family, and the world on YouTube.
youtu.be youtu.be

The metering to the right is a german Steinberg Wavelab ("PEGELMETER"). The audio is stuttering a little bit due to accelerating the video 3x (each test tone was 20 seconds originally).
So i ACK: the levels are at least accurate.
Click to expand...
Good to know! Though I don't see the OBS meter itself in that recording anywhere. OBS can absolutely output at full 0dB, it's the display in the in-program meter that (historically) had the -2dB offset on the top end.
 
Yes, its two videos. =D
The meter in OBS seems to be right, too. Shure... there can be an one- or two pixels difference. Sometimes difficult to decide. But in general terms the metering shows the fitting figures. That i checked at first, then i wanted to know if the real output would be unity gain, too. That's what the video shows. In OBS i definitely not checked the dot metering. It's almost to little for my eyes. ;)

While making the second video (window capturing obs + desktop audio capture) i made another disturbing experience: While the playback was stereo with exactly the same playback levels at left and right (as seen in the wavelab), the desktop audio capture was slightly different. I checked in the advanced audio settings that the panning of that channel was in the middle. Mysterious! =D That i have to check again, i'm afraid.
 
So here is a final version (still 3x speed timesaving), but with window-capture of the obs' soundmixer in the middle. Sorry for blurriness. Was made with MagicWindow magnifiered).

- YouTube

Enjoy the videos and music you love, upload original content, and share it all with friends, family, and the world on YouTube.
youtu.be youtu.be

The obs meter shows even a single pixel below/against the scale, i think.

(The audio track in the actual youtube version is lowered overall by 2dB (by me) to accomodate for the last two test tones. Even the -1dB peak tone brought up strange behaviour in the audio encoder.)

That last thing is the real difficult one for going youtube. Target loudness should be -14dBFS, but as experienced here again its strongly advisable to keep the outgoing levels (true-peak) below -2dBFS before entering the encoder. So there remain 12 dB at all in between rms/avg loudness levels and total maximum, including all transient material, seldom peaks and headroom at all. That's not much to play with... ;)
 
Last edited:
I still don't understand why it's not seen as good practice to have a master audio output meter (OBS tracks 1-6 selectable) and possibly a master filter bank in OBS.

The notion of "just set your inputs nicely and get a good, safe, but properly loud mix" is insufficient.

What if someone sums together 12 channels of audio, all at nice safe levels? Well, summing happens, and that information is missing.

Shouldn't we have the luxury of knowing what is happening at the master output, visually?

For an example let's use 12 channels:

"Summing 12 channels of identical, in-phase -6dBFS sine waves will cause the master output level to increase by approximately 21.58 dB, resulting in a master output level of +15.58 dBFS."

There are ways to intercept the OBS output, but man what a hassle when the info is so near-by, but obfuscated by an odd decision.

Thank you for attending my TedTalk Necro
 
You must log in or register to reply here.
Back
Top