Surround Mix Tips & Tricks

Hi everyone!

I would like to hear from people here working on surround project. If you could give some tips & tricks you are using in surround. Not the technical aspect of it, but the artistic aspect of it.
Like what's your taste, what do you like to hear or to create, a mood or a perspective effect.

Ok in example, on the Nine Inch Nail albums in surround (Downward Spiral and With Teeth), Trent Reznor seems to like to do someking of built-up with his surround mix. He almost always starts the song in the front like a regular stereo mix and the he slowly bring stuff in the rear until it gets very surround intensive. I like this, it makes you wonder at first "mmh is this in surround or not" then bam he surprises you with something in the back. He also seems to like the ping pong effect within the 2 rear speakers. That's just one example.

Something I noticed earlier when I start to make tests in surround, when I want something positionned right behind me in the center, if I put the green dot right there, it seems to comes right from the front center. I think I corrected that by moving the green dot slightly to the left or right, then it's seems in my back as if I had a speaker right behind.

Any tricks like that would be nice to know. Share some with us all please!

Something I noticed earlier when I start to make tests in surround, when I want something positionned right behind me in the center, if I put the green dot right there, it seems to comes right from the front center.

Are you monitoring with a 5.1 speaker setup, or have you chosen the 2.1 mixdown in Options→Project…

Regards,
David

I hate to say it. It will depend on the song. If its hip hop then ill eather use drums or what ever beat source to follow the music around the room. If its a female vocalist in a group I would prefer to have her voice off to ether next to me on the left or right as if singing or whispering in my ear. If its a duet female/male that would depend on who's the primary vocalist and who the song is written for.

ORIGINAL: Ethan Winer
I have but one "trick" - vocals and other centered sound sources should come mainly from the center channel, not from the left and right only panned halfway.

I've tryed that before and it seems to me that it sounds weird to have the vocals right in the center, it sounds kind of, isolated and not part of the mix. I'll try again see if I can come up with a good result.

Thanks

ORIGINAL: CarcPazu
...
Something I noticed earlier when I start to make tests in surround, when I want something positionned right behind me in the center, if I put the green dot right there, it seems to comes right from the front center. I think I corrected that by moving the green dot slightly to the left or right, then it's seems in my back as if I had a speaker right behind.

...

I've had that problem when I forgot to assign the speaker channels properly in SONAR's surround setup (Project|Surround, I think). This setup is per-project, which is an annoyance - I'd like it to be able to be a Global choice as well.

Panning anything to Center Surround is problematic in any 5.1 system. Our ears don't hear a phantom image from behind us very well. This is especially true in a movie theater setup where the surround channels are played with an array of speakers - pan something (imagine a jet flying) from front to rear, and the image is sharp and focussed in the front and diffused in the rear. The mixers at Lucasfilm demanded a Cs channel, and Dolby responded with EX, which adds a discrete Cs channel, so things can accurately be located right behind us; this is DolbyEX, and is used in movie theaters and is distributed on DVD and decoded on many home theater systems sometimes using one speaker (the right way) and sometimes using two.

Center Channel should be treated as a great resource (some music mixers avoid it altogether, which is an ignorant position). James Taylor Hourglass on SA-CD and David Bowie Heathen on DVD-A are two examples where lead vocals are primarily mixed to C with great results, and both surround mixes (James Guthrie's modern one on SA-CD and Alan Parsons' early QUAD mix, not available commercially) of Pink Floyd Dark Side of the Moon use phantom center channel for the vocals (I guess we all grew up with that particular sound). Any given sound panned directly to C will sound one way, and the same sound panned to phantom C (equally to R and L) will sound different- a phantom image will be louder (depending on pan law, don't get this started) because it plays from two speakers, it will have a notch at around 7k Hz (how many great mics, like Neumann U87s, have a peak right there? and how many of us have EQ'd a vocal there to make it stand out because we panned a mono source to phantom C?), and it will have phase and time distortions (because of Head Related Transfer Function - HRTF caused by the distance from one of your ears to the other, and the time it takes for sound to travel that distance). Generally, the more we pan stuff to discrete speakers, the more direct and clear the mix will be, and the better it will translate from one system to the next.

Surround channels should be treated with respect, too. They are tools for diffusion, envelopment, distraction, environment, or arrangement, and many cool concepts.

Remember that with 2.0 stereo you have one phantom image and two discrete images available, but with 5.1 you have ten phantom images and five discrete images available, so have fun.

Never, ever put any content into the LFE channel. Your five main channels already are full-range - in both frequency and dynamics - and don't need the Low Frequency Effects headroom boost that a movie might have.

It's very clear - read it again!

You can get all the punch you need by putting bass content in one or more of the main channels.

Use of the LFE channel in a 5.1 music mix guarantees that the mix will not translate well to a variety of 5.1 systems.

Never put any content into the LFE channel.

I said nothing about monitoring the LFE channel - in-fact, if you are going to play any 5.1 mix with LFE content, you'd better be able to monitor it correctly. This is generally done by properly integrating a subwoofer with main speakers using Bass Management.

Do not mix any sound to the LFE channel. You don't need it, and it will harm your mixes.

ORIGINAL: seriousfun

I said nothing about monitoring the LFE channel - in-fact, if you are going to play any 5.1 mix with LFE content, you'd better be able to monitor it correctly. This is generally done by properly integrating a subwoofer with main speakers using Bass Management.

Do not mix any sound to the LFE channel. You don't need it, and it will harm your mixes.

I don't want to stir some hit over here haha. I'm just asking because I received so many contradicting opinions from various sources that I don't know what to think anymore, especially about the center and sub. But my lack of experience/knowledge on the subject combined with my artistic taste want to argue with you on this one.

You might be right, it can surely harm a mix (especially if it's not monitored correctly without bass management) for conventionnal music. But I have difficulty to imagine (let's say) a Drum & Bass song without using the subwoofer for a massive kick in the butt feel. Same for techno/electronica/dub music which often feature some very deep and low bass that even full range speakers couldn't deliver.

So basicaly if I listen to everyone, music in 5.1 should be call Quad Music instead? lol

I red the article, mmh interesting, I guess this wrapped it all up.

Thanks for the info and the link.

seriousfun is giving very good advice RE: the LFE channel-- that channel should basically be reserved exclusively for movie sound effects (explosions, dinosaurs stomping, etc).

This brings up one of the central dilemmnas of surround mixing, which is that playback systems tend to be extremely inconsistent. If you are doing surround music mixes for home listeners there is simply no way to know what the actual listener's setup is going to be like, and most of them are going to be pretty suboptimal as a general rule, with everything from the decoding algo to speaker placement to speaker type. Very many home surround setups are configured with weird delay or "embiggening" effects, and/or all the lows sent to the subwoofer and only directional tweeters, and/or bizarro speaker placement and so on. This is very different from doing surround mixes for movies where you should expect the theaters to be fairly consistent.

There are two broad approaches to surround mixing, that I'll call "quad" and "stereo plus."

"Quad" here refers to a genuine full-surround mix, where you might have say the bass in left rear, the lead guitar in front right, the organ in front left, the vocals centered, the drums panned all over the place, and so on. They don't all have to be hard-panned to one speaker, it's just the idea that you are using the entire environment to locate sounds more or less equally.

"stereo plus" basically uses a more or less conventional stereo soundstage in front and uses the rear channels as a sort of sweetener or "special effect," e.g. for ambience or backing parts or swirly echo effects or doubled guitar parts or whatever.

Quad-style surround is actually extremely forgiving and fairly easy to do. The elimination of a lot of masking effects means that the listener is usually going to hear all the channels pretty clearly even if they are not sitting in the "sweet spot" or if they have improperly-positioned speakers, and it creates a very immersive listening experience where the listener can focus on different elements easily. It is not a rigid, sculpted soundstage so much as a genuine surround environment in which the listener can turn their head and hear a different mix. The major downside is that the listener really has to have genuine full-range speakers and no weirdo special effects on their decoder for the effect to work, and also that this type of mix is problematic in stereo foldown situations or in environments where some listeners may be outside the surround setup.

"stereo plus" is a bit trickier to do, but also better-suited to the ways in which many home theaters are set up, and infinitely more forgiving for stereo folddowns. Like stereo, this basically presupposes a listener who is more less centered in the speaker configuration and it also allows for the various kinds of playback system wierdness that people often have in the real world, e.g. mismatched front and rear speakers, delay or stereo widening effects on the rear channel, unequal amplification between channels, etc. This approach additionally allows the mix engineer to sculpt a genuine soundstage and a more focused and controlled experience. Note that "stereo plus" type mixes should ideally be done in a very good monitoring environment with a calibrated system.

One of the things that overarches any discussion of surround is that people hear in something like "stereo plus." We've only got two ears, but their shape creates phase anamolies that allow us to distinguish up-down and front-back directionality to some degree. However, these directional perceptions are always relative. I.e. we need to be able to move our head a little and to have some acoustical frame of refernce in order to distinguish whether a sound is coming from directly behind us or directly in front of us. because we don't actually hear in "true surround," the surround mix engineer is often somewhat counter-intuitively using reality to create illusion, sort of the opposite of what many of us are used to with stereo mixing.

The most important consideration for any surround mixer is to decide up-front how much and what kind of comprimise they want to make for the benefit of listeners with suboptimal setups. For instance my living room amp has a tendency to default to a "hall" setting where the rear channels are turned down 6dB and have some kind of delay effect on them. I know enough to turn it off, but for a lot of people this would seriously comprimise the experience of a "quad" stye mix. Similarly I think we all know people whose primary listening position is in a corner couch or easy chair where their head is right next to one of the rear surrounds, and they may even have their backs to a side wall, so they are hearing front-back as left-right. This is actually not that bad for quad-type mixes but can be really weird for "stereo plus."

Consumer surround sound systems are just starting to become widespread, and there are a lot of different ideas about how to use them. Stereo is certainly suject to bad speaker placement and such but still tends to be relatively forgiving, as we still basically hear an intact stereo soundstage, just off to one side or maybe with an unfocused center or whatever. The big difference with surround is that playback variables can make a much bigger difference in what the listener hears compared to what the mix engineer intended.

Cheers.

yep

Very nice and well laid out and I being somewhat new and and still learning. But Just for the sake of discussion I agree for the most part if not all usually the quad surround is the best set up. There are a couple of key factors regarding if it will translate well from one system to another is the initial setup IE larger front monitors and maybe one or 2 size smaller for the rears that way you will for the most part will be duplicating what is a typical home set up. The second part is how well you have your monitors calibrated. third part involves for the lack of a better term how discreet the monitors are paned from from front to right to rear right as well front to left to rear left and on to rear left to rear right. If set up right especially in sonar where the monitor cuts off during a pan lets say from left front to left rear. If the front left monitor cuts off about 3 to 5 degrees in front of your left ear or it set so low that the left rear is the most dominate at that point it is quite notable. Now all this will be base on the type of song/music and how it is laid out. where others will have very little noticeable effect. Because of our inherent traits of selective listening.

Hey thanks yep


Right tool for the right job. I don't know it maybe me it just seems easier to set up a quad mix. But on the other hand I have not explored all the other options. When I started all this my goal has always been to incorporate surround sound as the main part. Its more of a matter of what goes where when to achieve the desired effect. Ill explore the Stereo plus and see how it turns out.

Thanks

ORIGINAL: CarcPazu
...I honnestly don't know where is the fine line between the gimmick and the too conservative mix...

There is none, in my opinion.

We only have two ears, and only ever hear two discrete signals at a time. In that sense, all surround is a gimmick.

At the same time, real sound in the real world comes from everywhere, and although our central nervous system is at any given point in time only sensing two mechanically summed signals, our ultimate perception is quite three-dimensional in terms of percieved hearing experiences.

All multitrack music production is a gimmick. The relevant question is not whether something is a gimmick, but whether it's a good one.

Movie special effects that are well-done fly right by our conscious perceptions and we simply accept them as part of what we are seeing. Even if considered reflection would obviously tell us that they are impossible or unreal, context and artful execution and integration with the rest of the material causes us to simply accept them and go along with our perception that they are part of the reality that we are witnessing. Badly-done special effects, on the other hand, or forced or gratuitous special effects are jarring and/or distracting, or are at least "noticed" as special effects, and are percieved as "gimmicks." There is no quantifiable "line" that seperates them other than the subjective experience of the audience.

Clumsy production and mixing techniques are revealed as gimmicks even in mono. Surround merely gives the unsubtle engineers more ways to make themselves obtrusive and distracting. It's not a list of techniques or dos and don'ts, it's the art of creating sonic illusions in its entirety.

If you have never done so, I heartily encourage all sound engineers to listen carefully to Dusty Springfield's "Son of a Preacher Man." It's probably a song that everyone has heard a hundred times, but did you ever notice that everything in the mix is hard-panned in very unnatural ways? The drums are all in the right speaker, the guitar is all in the left, the reverb is in the right-- it's really quite bizzare when you pay attention to it. But the balance and the performances are so good that the sum and total of the listening experience is a very natural, "smooth" and realistic one. In fact, the overall experience is of a very expansive, sweeping, "modern"-sounding mix, in spite of the fact that the stereo placement is one that is in theory hopelessly archaic and "gimmicky."

Surround becomes a "gimmick" in the bad sense when the audience percieves it as such.

Personally I think a ton of the stuff that is mixed as a "stereo mix that spread to the rear a bit" is frankly just annoying and sort of gratuitous. Why not just make it stereo? What is the point of reverb coming from the rear speakers except to further obscure and muddy up the sound and create phase problems?

Some things to consider:

There is nothing wrong with *not* using the rear channels, even for entire songs. Way too many surround mixes just seem like a solution in search of a problem in this respect, just a gratuitous use of extra speakers to little or no apparent purpose that ultimately obscures rather than refines the sound field. Similarly, there is nothing wrong with having the rear channels carry exactly the same signal as the front channels. Resist the compulsion to simply put something different in the rear for its own sake. Not all songs require organ, and not all mixes require 5 speakers. You can amplify the dramatic surround effect a hundredfold by only having it kick in at select times. Listen to movie theater mixes for clues.

There is nothing wrong with completely arbitrary channel assignments and pan positions, so long as the listening experience is a good one.

Swooshy surround effects and other such gimmicks are fine, so long as they contribute to the listening experience. We're not trying to pass a test graded by audio engineers, we are making popular entertainment recordings for regular people.

It is unlikely that your audience is going to be listening in an environment that is the same as or even similar to yours. All mixes, but particularly surround mixes should tend to err on the side of being dryer rather than wetter in terms of ambient effects. Sadly we cannot control the acoustical space that our recordings will be heard in, and we generally should not try to.

Use dynamics. With surround, almost all of the need for technical and practical dynamics compression goes out the window. Your available space and dynamic range is huge. For instance during loud or powerful passages where a conventional stereo mix would require compression or fader-riding, you can just the pan the "extra" signal to the surrounds. This can have a very dramatic effect if used with discretion. Alternatively the ability to place discreet sources anywhere in the surround matrix means that we can leave plenty of headroom with little need for the dynamics and eq comprimises that are frequently used to overcome masking effects in conventional stereo mixes.

In all cases, be wary of over-thinking. As we add more tools and more speakers and more complexity to the system, there is a temptation to drift into theoretical or academic approaches that lose sight of the material and try to make the music fit some kind of technical ideal of how it should be mixed and presented. The mix should serve the music, not the other way around. The engineer should be a secondary support role, a delivery person.

My two cents anyway.

Cheers.