can you explain why this is happening?: https://www.youtube.com/watch?v=tJ1XNrEqPaI
i think it is automatically downmixes imoprted 5.1 audio to stereo.
i need to downmix 5.1 to stereo. but when i do this exported stereo audio sound is low. but when i extract all 6 channels from 5.1 audio and add them to 6 tracks and mix them to stereo sound level is not reduced. will it make any difference if i use multitrack(6 channels) mix to stereo? will it sound different from the original?
+ Reply to Thread
Results 1 to 11 of 11
Does the program have a "normalise" option for the audio?
Read about downmixing - especially about Lo/Ro and Lt/Rt
Side to this it can be that to prevent clipping overall level is reduced (for downmix matrix coefficients check AC-3 standard - http://www.atsc.org/cms/index.php/standards/standards/48-atsc-a52-standard ).
Well, it has options:
seems it does auto normalize. but i don't know how to stop it.
and it has an option:
Personally i would use: Front +Rear or even Front + Rear + LFE (i don't want to start discussion about LFE - i know is not recommended by Dolby).
And i would set normalize Max Peak to -3dB.
You probably want to normalise. If I had to guess, it's reducing the volume of the individual channels as it downmixes to stereo, so when they're combined, the peaks won't be louder than maximum and distort. That's normal, but to be absolutely safe, the volume needs to be reduced a fair bit. After the audio is downmixed to stereo, it's then "normalised" which adjusts the volume until the peaks are at maximum. Normally that'd increase the volume again by a reasonable amount. If it's not normalised after downmixing it wouldn't be unusual for the downmixed audio to be fairly low in volume.
I'm still guessing, but looking at the bottom picture from your previous post, there's an option called "automatic peak......" Unfortunately it's not possible to read it all but I suspect that's the normalising option when downmixing and it'd probably be the one to check.
As for the downmix type..... I don't use the program but the centre channel doesn't rate a mention. I guess it's always included when "front" is referred to, so you'd probably want to select the "front + rear + LFE" option to downmix all 5.1 channels to stereo. The LFE channel tends to be optional when downmixing (you could look at it as just adding extra bass) so "front + rear" would give you all the other channels without the LFE channel.
The first "audio gain" picture in your post. It also has a normalising option but I suspect it normalises the original audio so it would have it's peaks at maximum, but if you're downmixing you probably only need to check the "automatic peak......" option. Hopefully that's a good guess. What does the "automatic peak......" option actually say when you can read it all?
As I said, that's mostly guesswork as I don't use the program, but it might point you in the right direction.
Edit: Keep in mind, when 5.1ch audio is downmixed to stereo by your sound card or by a media player, by default it's usually just downmixed to stereo. There's no volume reduction to prevent the peaks from being too loud when the individual channels are combined. So even if the 5.1ch audio is normalised after it's downmixed and converted to stereo, the stereo version might still sound a bit quieter than the 5.1ch version when you compare the two using a media player. The difference shouldn't be massive, but generally there'll be some difference.
Last edited by hello_hello; 19th Nov 2014 at 07:54.
why when the individual channels are combined the peaks are being too loud?
and actually it is not too loud in media player but too quiet after downmix to stereo. And again when downmixing to stereo it reduces the volume when all channels in one track. with mono multitrack no any volume reduction.What does the "automatic peak......" option actually say when you can read it all?
I'll bite: I think the reason people aren't replying has to do with the fact that you've got a whole mess of difficulty here and WAY too many confounding variables, not the least of which is trying to explain the fundamentals of surround mixing & monitoring & downmixing.
Should we be talking about:
2. Downmixing coefficients?
3. Monitoring mix mapping?
4. Gain structure & consistent program level?
5. Variable playback of different formats with different dynamic range adjustments?
6. Differences in default settings in various editing or playback apps?
7. Differences in how you've set things vs. best practices?
See? It's all across the board! And without known-good reference material and demonstrated common settings, how could anyone isolate & troubleshoot these with you?
you made my butt hurt, Cornucopia.
If the front left channel has peaks at maximum volume and the rear left channel is added to it to make the left stereo channel (which happens when the audio is downmixed to stereo) there's a good chance the left stereo channel will then have peaks above maximum, or the peaks will be clipped or distorted. Plus the centre channel also needs to be added to the left stereo channel (as well as the right stereo channel) and maybe the LFE channel too, which can also increase the volume a little. In order to prevent the peaks from exceeding maximum, the volume of each channel is normally reduced when downmixing. After downmixing, "normalising" would normally adjust the volume so the peaks of the stereo/downmixed version are at maximum again.
Hopefully there's a "normalising" option somewhere which needs to be enabled. Have you tried the "normalise peaks" option from the previous pic you posted? (1st pic, post #4)