Orchestral Audio Mastering

[matthew82475]

I'm trying to see if anyone has information about mixing, EQing and, mastering orchestras. I'm using EWQLSO Silver and a VST Reverb Plug-in in CUBASE. My recordings sound okay, but often I'm fighting an issue of presence (for lack of a better term). Often the orchestra sounds way over there somewhere, rather than upclose. It seems to be most noticeable when the reverb is added. If I bump up the faders, though, I end up with a whole bunch of clipping problems. But without the reverb, the recording sounds flat. How do you get the reverb to sweeten the mix without making the orchestra sound far away and avoid clipping at the same time?

Also, I've noticed than on professionally recorded CDs, the mix usually sounds good on almost any combination of speakers. However, the audio tracks I'm getting from silver seem to need a subwoofer. With just a two speaker stereo, there is almost no low end. Is it just the nature of silver, or do I need to do some sort of EQing?

Thanks,

Matt

P.S. I've tried looking online for info, but most everything out there isn't relative to orchestras.

[BMoody326]

Are u using the reverb as a send on an aux channel or using it as a insert in the East west plug-in..using it as an insert tends to muddy it up..Using a convolution reverb will help also..when i wasn't using the convolution and just a reverb i noticed the sounds lost some of there bite..but when i used tthe convolution it actually brought more presence to the instruments...as for the low end u have to remember that you can write for tuba, contrabasson, etc in order to get the low end you might be looking for....youcan avoid clipping by using a limiter on the master fader....Remember there is no one way of achieving the right mix...just listen to a professional piece in the style you are composing and try to emulate the mix....

[Qjimbo]

I have exactly the same problem! I can never get my final bounces to sound as upclose and clear as I'd like... I'm also using a similar setup with the Silver edition and HyperVerb as the reverb plugin. Interesting the suggestion of a limiter though, hadn't thought of that, but yeah I really don't know where to begin on getting it to sound bigger, brighter and clearer.

[matthew82475]

BMoody,

I'm sorry I don't understand the terminology. In Cubase, there is a slot for Master Effects. I just dropped it in there. I know I could have set it up separately for every instrument, but that seems like an immense amount of work, since I was going more for ambience than authenticity. The problem is that as soon as I add the reverb, the orchestra sounds far away. If I turn down the volume fader for the reverb, the reverb gets shorter and shorter making the "room" seem smaller and smaller. There's a bunch of other faders and dials on the reverb plug-in with all kinds of cryptic numbers and descriptions, but I can't seem to find anyway of figuring out what they do. They are: Size (m2), Damping, Predelay (ms), High Damp, Low Cut (Hz), Early Ref (dB), Mix, and Level. Level is about the only one that makes any obvious sense. Does anyone know what all these others do?

As far as emulating a mix I like, I would do what you suggest except that I don't have the first clue what buttons to push, what faders to fade, and what sliders to slide. The quantity of options is pretty staggering. I've tried looking online for guides or tutorials, but they all seem to geared toward pop music (like how to make your kit sound better or how to EQ a bass, etc.) I'm wondering if there are some basic principles for mixing and mastering orchestras. I imagine there probably is, I just can't seem to find them.

In regard to the low end issue, I'm not referring to an orchestration issue, I'm referring to the fact that when I play my mix back without a subwoofer, there seems to be an inordinate amount of low end that disappears. The effect seems to be that there is a gap between the low end of a floor speaker and the subwoofer. So much low end is lost when the subwoofer isn't used that you can barely tell that there is a tuba, contrabassoon, and double basses playing. I can't simply bump them up, because that throws off the balance. Any ideas? For comparitive purposes, I've played professional CDs with both the subwoofer on and off, and aside from the obvious difference a subwoofer adds, the mix sounds pretty much the same either way; the subwoofer just adds a lushness to it.

Thanks,

Matt

[Qjimbo]

Ok I've done a bit of experimenting and found a few basic things

* Keep the Reverb having long trails just at the lowest possible wet mix you can stand :p
* Try using a MultiBand plugin, I found this can really crispen up the sound if used in roughly a pyrimid shape (there were a few good presets on the Sonitus:fx one included with Sonar 4)
* Use EQ. Try putting your songs though the EQ on you hifi or in winamp and then copy the good EQ into an EQ plugin. Just adding a bit of a bass boost seems to improve the sound somewhat.

Also another thing to improve the bass, non-production related, is if you're using a snare, double it with timpani.

[Stefan Podell]

I think your reverb recommendation is a good place to start. But as BMoody suggested, try using an Aux Send instead of a Master slot. That way you can send different amounts of each instrument to the reverb. Less on the featured instruments and more on the background instruments, for example.

The drawback of your other two suggestions is that it's particular to your listening environment, and may not translate well to other systems. You must really know your monitoring system to get good results out of it. Listen to CDs on it, and make your mixes sound as much like those as you can. But remember that if your room has troubles (dead spots, bass buildup, etc) or if your monitors don't extend into very low or very high frequencies, then those kinds of things will conspire against you. You won't be able to make good mix decisions because you won't be hearing the impact of the changes you make.

- Stefan

[Qjimbo]

Yeah you are right. I mean it's so frustrating when I play, for example, the Sky Captain and the world of tommorow soundtrack on my Monitors up here and it sounds clean, and I get my tracks sounding close to as good. Then I play the soundtrack on someone else's hifi and it sounds clean, and I play my tracks on their hifi and they sound fuzzy and unfocused >_< so annoying!

[David_Lucas]

You must also remember that the creator of the music often isn't the best one for mastering it. It's hard to be objective about your own stuff.
I've found that using the reverbs inside the Kompakt player for Silver work quite well for most situations. Generally I will apply a master reverb to each section of the orchestra after mixing down into my pre-final mixdown.
If you have a reverb that has controls for both Wet and Dry, try keeping the dry at about 90-100% and don't put the wet over 20% unless it's a slow piece. (and 20% is pretty high if your reverb is more than about 1.5 seconds)
Also, about the lack of bass- I find that there is ample bass in the sound library, but if you are sure it's too tinny, try cutting frequencies before you boost. I often apply a cut of about 1.5 db @ 440 Hz with a Q of 2, and then re-normalize.

[matthew82475]

Thanks for your responses, especially David. If you noticed in my last post, I listed the dials and adjustments I can make in my reverb plug-in and there isn't an option for wet or dry. So, I don't really know what you're referring to. Also, as far as the EQ suggestion, I got everything but the Q. What is Q? And what does "re-normalize" mean?

My frustration lies in the fact that the general consensus is always, "Do what sounds right to you." The problem is, I don't know what "right" is? Even if I tried to match a mix I heard elsewhere I wouldn't know how to go from point A to point B to end up at result C. To make matters even more frustrating, all of the online tutorials and guides I've found talk about making your drum kit sound bigger or fattening up the bass, and a bunch of other pop music related references. I was hoping that there were people here who have had experience in mixing orchestras who might be able to offer some general guidelines and approaches (and maybe a few definitions would be nice too) to getting a "good" mix.

Qjimbo, in regard to your last post, that is precisely the frustration I'm having. It seems like there's got be some "thing" that we're missing. Some setting, or effect, or adjustment, or something that fixes that issue. But what it is, I don't have the first clue.

Stefan, if using an Aux Send for reverb is what I need to do, I'll suck it up and do it. The question remains, though, how to set it for each instrument. Are there any basic approaches or techniques that could guide a complete beginner? Also, in regard to your mix/monitor comment, is there any imperical way of getting close to a good mix (and by mix I don't mean balancing each instrument, I mean an acoustic mix) that could compensate for limitations or variations with a monitor setup? For example, I've got a frequency analyzer plug-in, is there something I should be looking for that would clue me in to where I might be going wrong?

Thanks for your continued help all,

Matt

[Stefan Podell]

Stefan, if using an Aux Send for reverb is what I need to do, I'll suck it up and do it. The question remains, though, how to set it for each instrument. Are there any basic approaches or techniques that could guide a complete beginner? Also, in regard to your mix/monitor comment, is there any imperical way of getting close to a good mix (and by mix I don't mean balancing each instrument, I mean an acoustic mix) that could compensate for limitations or variations with a monitor setup? For example, I've got a frequency analyzer plug-in, is there something I should be looking for that would clue me in to where I might be going wrong?

Matt,

Well, putting reverb on an aux is definitely *typical* but you should do what gets you the results you want.

Here's how I do reverbs on an aux in Cubase (I use LE -- if you don't, some UI specifics might differ):

1. Open the VST Send Effects window.
2. In one of the slots, select the reverb plugin you want to use.
3. Set the Mix to 100% wet. We'll control the amount of reverb using each channel's send.
4. In Kompakt, set each instrument's slot output to a different channel. The channel is selected in a field with a little icon that looks like the business end of a pitchfork followed by 1/2 (the stereo channel numbers). Set each slot in succession: slot 1 to channels 1/2, slot 2 to 3/4, 3 to 5/6, etc.
5. Open the Cubase Mixer window.
6. Enable the "Show/Hide VST Instrument Channels" icon. It looks like the C through E keys on a piano.
7. Click the "e" for the VST Instrument channel return you want. You can label them in the Mixer window to make this easier, but they'll just look like "EWQLSO 1/2" etc.
8. You'll see all the available sends on the right side of the channel editor. One of them will have your reverb plugin in it. Enable it by clicking its "On" button (circle with a line in the 12 o'clock position).
9. Control how much of this channel will go to the reverb with the horizontal slider beneath the plugin's name.

Repeat steps 7 - 9 for each instrument's channel. Now to control the amount of reverb on a single instrument, all you need is step 9. To change the reverb type or anything, you only do that in one place: the VST Send Effects window.

I hope that helps!

As for handling your room and monitors, there are tons and tons of information (literally, probably, at the bookstore or library) as well as websites, mailing lists, forums, etc. that cover that kind of topic. It's difficult to summarize it, but you might find you want (1) different speakers, (2) to move the speakers you have, (3) to treat the room with absorbers and diffusers and bass traps, (4) bag it all and use really good headphones (but that has other issues).

Best of luck.

- Stefan

[matthew82475]

Thanks Stefan,

Now the primary questions, which kind of takes me back to where I began, is this: How do I know how much reverb is "right" for each instrument? I'm sort of back where I began, for whithout experience or guidance, I don't really know how much each instrument should have. Also, will this solve the presence problem? In other words, by simply moving the reverb to an Aux Send rather than a Master Effect, will that cause the mix to sound more forward? I wouldn't think so, but, hey, what do I know?! It would seem to give me more control, but I suppose my question is: what should I do with that control?

In regard to changing reverb type, that's related to a previous question I asked which was, does anybody know what all those buttons and setting mean. For example: size is m2, which I assume means meters-squared. If so, does that mean that the higher the setting the larger the "room" and thus longer reverb? Or is the length of the reverb determined by the predelay in mlliseconds? If so, what exactly would the size option effect?

Thanks again,

Matt

[David_Lucas]

Hmm.
Generally what you are looking for in your final mix is a sound which plays at fairly even levels across all frequencies, using as little compression as possible to acheive that end, and you'll learn to do that with a couple years experience.

As far as this 'presence' you seem to be having a problem with: in simple terms, the instruments that you want to stand out should have an EQ boost in their fundamental range- that is, whatever, when playing that instrument alone, is the first frequency you see (as opposed to the overtones). Also, the sounds requiring more presence should be more "dry" (less reverberated) than the other sounds.

Two of the most important things to pay attention to in the final mix are stereo seperation (which EWQLSO pretty much handles for you) and an even frequency response (which EWQLSO also pretty much handles for you).

[matthew82475]

Thanks David for that very specific post. I'm going to give your suggestions a try. I had noticed in my analyzer that I had more of a bell curve rather than an even response. I suppose I can EQ the mix in two ways. I can adjust the low end up in individual instruments, or I can adjust the low end up over the entire mix. If I chose the former, should I be focusing on bumping up the low end on the bass instruments only (contrabasson, tuba, bass trombone, double bass, etc.)? Or would it make more sense to simply balance out the frequencies of the final mix?

Also, should I be balancing the frequencies before or after the reverb is added? Could the "Low Cut" or "High Damp" (whatever it is those settings do) in the reverb be causing or adding to the bell curve shape of the frequency response?

Thanks again for that post.

Matt

[Stefan Podell]

Matt,

Like David said, in EWQLSO, the EQ is pretty much spot on for most general uses. So whatever EQ changes you make should probably be relatively minor.

One thing to watch out for: I don't know what kind of speaker setup you have, but most consumer-oriented speaker systems are hyped in the low- and high-ends. If that's the case with your speakers (or room), then you'll be compensating in your mix by reducing the amount of low- and high-end. And that might be what's contributing to the bell-curve you see.

As for the low cut and high damp in the reverb plugin, those will affect the reverb only. And generally those are a good thing: Too much low end in the reverb will muddy the sound. And too much in the high-end will sound strident or sibilant.

The target you want to have is to trust your ears over the frequency analyzer. But the analyzer can be a big help as one learns, especially if you're figuring out your listening environment.

As to your specific question about whether to EQ individual instruments or the whole mix... I think the goal should be to get the best sound at each stage. So try to get the mix sounding best by adjusting individual instruments first. For example, if you boost the low frequencies in the mix to compensate for thin-sounding contrabasses, you might make the timpani overwhelming in the process.

Also, you'll generally EQ before applying reverb. The same principle applies: make things sound their best before they go down the line too far.

Small tangent: In mixing, it sometimes happens that individual instruments sound not-so-great when soloed. But in the context of the mix they really work. That's not necessarily so true for orchestral work, but it definitely applies in pop/rock.

Still, my guess is that you're mixing in an environment that looks like an upside-down bell curve. And your mixes look like a bell curve to compensate.

- Stefan

[matthew82475]

Thanks Stefan,

That's great insight. Thanks to everyone who has contributed. I hope more forum members will add their experience and insight to this thread and allow us neophytes to learn from their wisdom.

Matt

[matthew82475]

If anyone is still watching this thread, I've tried to do what everyone has suggested and here is an excerpt from an orchestration I wrote for Song of the Open Road by Norman Dello Joio. If anyone has the score, the recording begins at the Adagio espressivo just after the first trumpet solo. Also, this is primarily a choral piece and so I did the choral parts in the only way I could with silver using the "ah" choral sound.

Song of the Open Road (excerpt) (http://music.matthewarmstrong.com/Song_of_the_Open_Road_excerpt.mp3)

Please give feedback. I think it sounds better than what I started with, but the mp to pp sections still give away that "presence" issue. The problem is, if I try and bump things up the balance goes out and it still peaks slightly just at the end, so I'm still at a bit of a loss as to what to do next.

Thanks,

Matt

[T. S.]

Hi Matt,

I'm no expert but I'll try give my understanding of some of these different perameters. Lets use a hall reverb here since I think it helps to better explain.

Size (m2): This would probably refere to the size of the room. Small Hall, Medium Hall, Large Hall, etc..

Predelay (ms): It's my understanding that this is the length of time it takes for the sound to leave the instrument untill it reaches the first wall or surface. I think also this would be called the "Early Reflection".

Early Ref (dB): I think this is the loudness or amplitude adjustment of the "Predelay" or "Early Reflection"

High Damp: I think this is sort of like an EQ adjuistment that lowers or "dampens" the higher frequencys. I also think this would act more like a "low pass filter".

Low Cut (Hz): Like "High Damp" this would do the opposite and cut the lower frequencies, more like a "High Pass Filter".

Mix: THis adjusts the "wet" and "dry" levels. Some reverbs have sliders or knobs for each of these but in this case with only one, with the slider totaly down (knob totaly turned left) the signal would be complletely "dry". Going the other way would be completely "wet". In the middle would probably be half and half.

Level: This of course would be the overall output of the reverb which you shouldn't have to adjust unless it's clipping.

Many reverbs also have an "RT 60" (Reverb Time) that controls how long the reverb is or rings which would also corespond to the Room Size "Size (m2)". Many reverbs have a control for both.

About the only two things I tend to adjust in the reverb is maybe the Room Size and/or RT-60 and the Wet-Dry level (allways full "wet" when used in an aux). I usually try to find reverbs from the "presets" to match what I'm doing and thier usually set up pretty well already (with the exception of the "reverb time" and the "wet/Dry" aspects).

Many here have already commented on some of the following but I'll try Echo or add a little to them.

I don't have Cubase but I did work with it many years ago. Back then I seem to remember it had "Sub Groups" where I put the Reverb. In other words I used it as an "aux". Personaly I would almost never put a reverb on the "master". In fact about the only thing I use on the master is a compressor, limiter, or maximizer and then only use these spareingly.

I then use my "sends" to adjust the amount of reverb for each instrument. As far as the amount of reverb goes, generaly the lower the frequency of the instrument the less reverb I'll put on it Lower freqs can muddy things up and end up being slurred. This isn't always the case but is a good thing to keep in mind.

As far as EQing goes I seldom use it on anything but the individual instruments themselves and there again very spareingly. Most of the libraries now-a-days have been recorded prety well and don't need much. I might add here that if you have a monitor system that you don't trust or haven't become intimit with, I think I would tend to trust the libraries or good Mics first. I've been in the recording field for nearly 40 years and back in the old days I used to EQ the crap out of everything. Now I use very little.

I know there are those here that have a better handle on all this than I do so if I stand to be corrected on any thing please jump in and do so. I'll be glad you did.

Incidently Matt, I tried to download and play your file but I got some kind of protection error.

T, S.

[matthew82475]

Thanks for that info T.S.

I converted the excerpt to an MP3, so hopefully that should help. I look forward to specific suggestions regarding what I've got thus far.

Matt

[T. S.]

Hi Matt,

Haveing reread your postings I noticed you asked about "Q" and "re-normalization".

"Q" stands for "bandwidth" and is normaly associated with a parametric EQ. A parametric EQ basicaly has 3 parts: Center Frequency, Gain, and Q (or bandwidth). The "center frequency" is the frequency you wish to alter, Gain is the amount you want to cut or boost, and "Q" is how far on each side of the center frequency you want to be affected. In todays EQ plugins "Q" can be as little as less than 1/16 of an octave to as much as several octaves.

As far "re-normalization" is concerned I'm not sure what exactly the author was refereing to. Normalization itself is to adjust the amplitude of a file (or portion of a file) to a pre-determined level. This is something I seldom use untill I've got the song mixed and I'm doing the final mastering for the CD or what ever. Most all music programs that deal with "wav" files have this feature and you can set the taget "peak amplitude" to what you want. Many people set it to 0.0 db or maximum but I prefere something more like (-.3 db) minus point 3 db (Just wanted to make sure you saw the decimal point).

One thing I'd like to mention about normalizing is that the better you get at auto mixing your tracks (midi and/or audio) and then learn how to apply compression (ever so slightly) the day will come when you may not have to normalize at all.

Hope all this helps.

Incidently I managed to download your file and will give a listen very shortly and get back.

T. S.