Downmixing 6 channel AAC to 2 channel?

[hello_hello]

I always use MKVToolnix for muxing and extraction of tracks as well. What about MP3 track? Is there any means to know the audio delay and actually where it is stored. What encoding parameters do you recommend for MP3 track and music files? I have all my music files encoded as flac with mode-5 and MP3 CBR 320 Kbps, not to sacrifice the quality and might be due to ignorance.

According to this, for MP3 it's written to the Xing header at bytes $B1-$B3, and it should be find-able with MediaInfo in debug mode, but I haven't tried digging it out yet. Then you'd have to translate it into human-speak.
Foobar2000 definitely accounts for it. If you check the number of samples in an audio file, encode it as MP3 and check the number of samples again, they'll be exactly the same, but it doesn't tell you the amount of padding, it just skips it.
As far as I know MKVToolNix only accounts for it for AAC, although I kind of remember VirtualDub accounts for it for MP3, but that could also be completely wrong. I know nothing about AC3, but honestly, it's probably not enough to care about as a rule, just something that if a muxing program accounts for it, that's better than not accounting for it. My brain doesn't notice an audio de-sync till it's around 100ms, and I tend to be more likely to notice audio that's early rather than late. I think that's fairly typical, although I've walked into our living room while family members are watching something and immediately noticed an audio de-sync that apparently nobody else does. Practice, I guess. Free to air TV seems to have a mandatory audio de-sync where I am.

So. this means that if I save the ReplayGain information and play it in any player the volume will get adjusted to the same level of 89. Is it right?

No. The player has to be able to read the ReplayGain info in the tags and use it to adjust the volume, but very few hardware players do, and of course Apple re-invented it for their players and called it "Soundcheck", used different tags and increased the target volume to -16LUFS. There's an option in fb2k's Advanced preferences under Tagging to also write soundcheck tags. It's disabled by default.
So if a player doesn't support tags then physically changing the volume of the audio (which is what the right click "ReplayGain/Apply track gain to file content" option does for MP3 or AAC, or adjusting the volume when re-encoding does), so then it doesn't matter if the player supports ReplayGain because the audio is already at ReplayGain volume. Of course to do that you still have to scan the audio and save the ReplayGain info first.

For soundtrack audio, it should be 83 as per European standard. Again, if we do peak normalising the volume will naturally increase by 20dB. It seems like contradictory, if I put it in other way, I am not quite clear in this aspect. Would be please throw some more light?

Hopefully my previous post helped, but ReplayGain's 89dB actually refers to sound pressure level, not volume as a player's output meters display it.
There's a SMPTE standard for calibrating a cinema sound system that says if you run pink noise (I think) recorded at a certain volume, when it's played back in the cinema the sound pressure level should be 83dB SPL. That's where 83dB comes from and after complaints about it being too quiet it was increased to 89dB for ReplayGain in the early days, and we still refer to ReplayGain volume that way, but to mere mortals it's effectively meaningless. 89dB is the same as an average volume of -18dB according to your player's output meters, which provides headroom of 18dB for peaks before you hit 0dB and start clipping.
The EBU standard still uses -23dB, or -23LUFs for soundtrack audio. Hopefully that makes sense. It's a pity the ReplayGain spec refers to volume in a counter-intuitive way, because unless you know where it comes from it's confusing, but I guess we're stuck with it.

[hello_hello]

Damn! Of course you can peak normalise to -3dB (or minus any dB) using foobar2000. At least when converting. Why had this not occurred to me before? I might even do it now to keep pandy happy.

Configure the ReplayGain processing in the converter setup to peak normalise, follow it with a DSP to reduce the volume by the desired amount, and you're peak normalising to minus "x"dB. Easy peasy. The Amplify plugin is the perfect DSP for that. That's all it does. Adjust the volume up and down. Pity it's not called the volume DSP given it adjusts both ways, but you can't have everything. I've attached it as it's hard to find.

[shans]

hello_hello- Thanks again for your time and patience. I think I need to do some more exercises / practice to learn more. One more thing I would like to know from you as to what encoding parameters you recommend for MP3 movie sound track and music files?

[hello_hello]

All I use for MP3 music tracks is LAME's V2 variable bitrate mode. I think it's both the LAME and foobar2000 default.
Have read of this. Recommended encoder settings.
If you're only putting MP3s on a portable player there's no point getting carried away, in my opinion. Low quality headphones, lots of background noise.... the quality of the audio is the least of it.

For MP3 audio in an AVI some people swear constant birate MP3 is more compatible with hardware players (old DivX DVD players etc), but I've not had trouble with VBR MP3 in AVIs.
We're out of the AVI dark ages now anyway, so if I was going to use MP3, which I rarely do for soundtrack audio, I'd use the same V2 setting and mux it into an MKV.