Live gig CD and DVD mix, any practical methods?

[Gerax]

Hello

I'm mixing a live concert I recorded last month for 2 different releases: CD and DVD.
CD is obviously going to stereo.
I'd have a couple of questions about the DVD.

I recorded 32 tracks (along with the video shoot of 6 HD cameras) and I have started the editing and mixing process for the CD (I'm almost finished with it).
I have just received some finished cuts form the picture editor and I'm putting that mpeg file into my PT session to check sync and tweak the mix for the DVD.

Now, first thing: I really don't feel the need to have the DVD mix that much different form that of the CD, I just spread things a bit wider, got lead vocal a bit louder and followed a bit the various cuts on the musicians to raise some cues that you see and that you want to hear (which may be softer in the CD mix).
Is there any particular tweak you have to do for the DVD mix to be working, such as nominal levels etc..or it's just that I can go with my own taste?

I recorded several audience and room tracks that I've used in the mix. The recording wasn't a "real surround" so to speak, do you think that I can put them in my LS and RS speakers and pan them and have a "coherent" sounding 5.1 mix?
My idea of the surround mix is to have the band upfront, maybe a bit wider than the CD mix and have room and audience only in the back.

Last thing: stem delivery to mastering. Is there a rule of thumb to create the 6 stems as fare as track assignment go? What do you put in the C channel (that will go to the corresponding bus and final track)? I have panned things left and right, but I don't have any "flybys" or things like that, so can I keep things like Kick, Snare, Bass and lead vocal in my C and assign everything else to L and R? What about reverbs, they are panned in stereo...

Thanx

L.G.

Hi Lorenzo,

Surround setup:

First of all you will need a 5.1 surround mixing setup with 5 identical 2-way loudspeakers, in vertical position, tweeter axis 1.2m from ground with ITU 775 setup.

Make sure all loudspeaker have equal response within +/-1.5dB (when installed) or possibly apply very few, gentle, broad parametric EQ (with analyzer microphone at mixing position) especially in low frequency. Have overall pink noise level matching within less than 1dB.

ITU 775: http://www.ambiophonics.org/AES%20Mu...S_Munich_1.htm

Subwoofer in the room middle or slightly off-axis until it couples well with your control room without any boominess and being as tight as possible. Cross-over 120Hz with +10dB in level compared to LCRS.

Surround mixing

C: Keep the center channel for vocals and lead instruments, leave out any rythmic or bassline out of it, especially low frequency which would induce masking. Use L,R for short delay, panned pitched (harmonizer) to make the voice wider and thicker.

LR: use LR as in stereo mix. Discrete LCR mixing is fine but avoid panning sources only inside the box, we appreciate better phantom images.

Dry surround: use the benefit of 5.1 to mix things with depth for dry signals. Consider L and LS (left surround) as a new left phantom stereo image with panning from front to back. Similarly R and RS. Just some slight dry signal panning will provide great depth, use small delay to push back (Haas panning) in L, R panned instruments with respect to central vocal in C.

Wet surround: ideally use 2 identical reverb units with identical preset

- early reflections only (without tail): typically panned only in L, R
- reverb tail only (without ER): typically panned with 20% L, R and 80% LS, RS

This is most realistic to the way we perceived room acoustics : early reflections high in level from front sides and diffused reverb developing in front and blooming in back.

ER in front and slap back echo in back with low levels can provide outdoor live ambience with some touch of diffused de-correlated reverb.

You may want to use a short stage reverb only in LCR to mimic stage acoustics, then feeding into a long chamber, hall reverb in LS, LR to mimic audience room acoustics -- representing the coupling of 2 rooms stage/hall.

Do not use so much overall compression as this may restrict the surround space back to pentagonal mono.

Most engineers feel using the 5 speakers also for feeding lows (so that if subwoofer setup is poor in playback, it does not confuse everything). I would recommend using only L, R, LS, RS for bass and leave out C for clear vocal or lead information 8even panned in LCR).

Use subwoofer only for bass below 60Hz (channel doubling with LPF at 60Hz) in order to only feed very low frequency to the sub channel so that if the playback subwoofer is a poor boom box with lousy 120-160Hz cross-over, the surround mix will not be contaminated with 80-160Hz boominess that mask and confuse everything.

Use .1 sub only for real sub : kick thud, bass fundamentals, keyboards last octave -- only below 60Hz (very low frequency that makes air to move, not bass).

This way in a cheap 5.0 setup (without subwoofer) there will be still enough of a cleaned and controlled bass mix in LCRS (as you intended) playing well in balance.

Do not worry playback microcube systems with plastic subwoofer route all low frequency signal to the only driver able to do low frequency somehow. It will play but to listener boombox taste (1 to 11 :-).

In general, think of having 4 phantom stereo pictures available (LC, CR, LLS, RRS) and mixing with depth LCR versus back surround channel.

If you get comfortable and like surround mixing consider a surround mixing capable console. The Yamaha DM-1000 V2 is a great console for recording and live surround mixing (even if the pan pot is super tiny) to start with, on a small budget.

Have fun!

[Geert van den Berg]

All good points from HiFi.

Because 5.1 was originally conceived with film mixing in mind, there aren't any music standards/rules. Here are some extra things to consider.

Put only things in the subchannel, that could be left out mix and still sound good. All the 5.0 are supposed to be full range, and the sub is only intended as an FX channel, that doesn't mean you shouldn't use it, but use it sparingly, adding a bit of bass to it is, fine, but make sure there's plenty in the other speakers too. I mean spreading the bass to the sub makes for an easier mix, but if the stereo mix was already good, I'd go with that balance instead.

Also when an 5.1 is folddown into an LtRt (Dolby Stereo/Dolby Surround) the sub channel is usually being left out. (certainly when this folddown happens automatically in a player where the stereo outputs are used, this is specified by Dolby)

Like a couple of years ago, people still don't have good centre speakers in their home (I mean equal to their L-R speakers), so you might want to leave the centre channel out of the mix. The vocals are the most important, so you might not want that to suffer. Same goes for the bassguitar. While I think using the centre speaker sounds much better, much more realistic, it's something to consider. (Maybe make 2 mixes if you got time and then play them on some consumer setups to see how it translates)

So then what is 5.1 good for? Well like you said you could spread the ambiences to the surround speakers and also use 5.1 reverbs on the intruments to make it wider.

[Gerax]

Thanx guys. Some really good points.

I'm going to be pretty busy experimenting various solutions for a correct mix.
So there could be a couple methods for the C channel: vocals only or getting it out of the picture. My main fear about putting vocals only in the C channle is that nobody knows where people are going to put that speaker in their consumer/home setup, yes, I know it should be in the middle of the other two front speakers and under (or above) the screen...but, you know...Second, my vocal track isn't exactly "pristine" as the recording was a live gig: even though I have automated and edited the track I still have some bleed from drums and stage volume...and I fear that by listening closely to that C speaker alone with just vocals in it could spoil me a bit ; it sounds good in the mix, but if you solo it you can definitely hear it. The "C-less" solution would solve this issue, but I'd have to go for a 4.0 mix...if you know what I mean...is that still fine for a music DVD release?

Thanx

L.G.

[Geert van den Berg]

Originally Posted by Gerax

The "C-less" solution would solve this issue, but I'd have to go for a 4.0 mix...if you know what I mean...is that still fine for a music DVD release?

I know it kind of sucks to do this, but IMHO it's the best thing to do. People will be more focussed if there is sound coming from the surround speakers anyway, and even then there will not be much, some verb, and some crowd ambience. Just to make it seem wider.

You're just making sure the mix will translate to most homes, instead of only those with highend playback systems. And as you said yourself, you're not sure if the speakers of the listeners will be correctly placed.

I hope some other will chime into this thread, as this is an ongoing problem in 5.1 music mixing. The funny thing is, that if you buy a cheap consumer 5.1, the centre is now usually equal to the L and R speaker, so then putting the vocal in the centre could work.

You could spread some of the centre information to the centre channel, so it's not completely silent. The centres of the reverbs you're using if you're using real 5.1 reverbs for example.

[jahtao]

Hi, just quickly while i've got a sec.....

I am surprised you feel to have the dvd mix wider than the cd mix? I find the opposite - if the drums are panned too wide you expect to see a drummer with incredibly long arms playing a kit spread across the entire stage, so keep it reallistic, no? I find with a CD mix its easier to tend toward 'wider = better'. No?

I don't like much bass and kick in the center speaker if only because a lot of center speakers i've seen in people's homes are small, weak little things. Vocals, snare etc. i 'm currently favouring a 'mostly out of the center' approach - you gotta put something there right? I know lots of people love the harmonizer thing on live stuff, i've never got into it, not tried it much tho'

As for the rears and the general surroundness i reckon its basically about using what you've got in a way that sounds good to you, end of story - often (rock stuff, haven't done classical etc.) that's simply the right kind of verb. And if you've got some ambient mics that are working back there try and get the most out of them but do expect to use verb. Just make sure you do something with them, something definitely audible : )

5.1 verb is def preferable.

You can use the rears to make the front stereo image feel a little wider, but only a little in my experience (worth it?) but remember if you have a sound source coming strongly out of more than two speakers, you will hear phasing when you move your head around - cool, but not cool.

Remember to turn the audience up like crazy in between songs, and make a sound that works with or matches the picture - don't forget this one! But its tough becuase if you make it sound like you're really there in the mosh pit it can sound crap but if you make it sound like the radio, as soon as you see a shot of the mosh pit going crazy it feels kinda wrong and that kills the energy a bit i find. So, my concept is psuedo-super-real-sound -

We like to turn off the lead vox slapback in between songs (when singer talking to audience etc) and find a smaller (stage-like as someone mentioned) verb helps to set drums back into the mix. I reckon try to avoid really dry sounds that stick out from a sea of wet-ness. Like sampled percussion against live drums. I don't like that.

Remember with dolby it is possible for a consumer to downmix the 5.1 to 2.0 in the home (inadvertantly)so check that

Hows that sound everyone?
Matt

[jhg]

Though mentioned before, needs to be stressed, make sure the down mix works, make sure it's not a Lo/Ro or odd situation with the downmixer - the Dolby specs are kicking around this forum. Some awesome 5.1 mixes are destroyed by comb filtering, phase horribleness and other notfun in the 2 channel realm...