Minimalist vs processed mixes for classical/chamber?

[jnorman]

i am starting to mix the first set of new recordings for a flute/pedal harp duo. the isntruments were recorded separately, at the request of the players.

i have prepared several draft mixes so far ranging from very processed to very minimalist. the processed mixes include stereo widening effects and heavy early reflections on one side of the harp to spread it out wide across the soundstage, and putting the flute in the middle, along with some fairly significant EQ to round out the harp response, and 2 or 3 reverbs mixed together. the minimalist mixes avoid anything that would introduce more phase problems, so no stereo widening, no significant EQ, etc, - i am just panning the harp slightly left and flute slightly right to present them as if they were playing on a stage together, and generally just adding some med hall reverb on the master to tie the isntruments together in space.

i can see benefits and drawbacks to both approaches, but i am leaning more toward the minimalist mix approach, while the harpist seems to like the highly processed "studio" type mixes (the harp is more "glittering" that way...). i dont want to argue with the players about what they want, but i did want to get your comments on whether you tend toward processing your small acoustic ensemble work, or if you tend to leave it alone as much as you can.

thanks for any thoughts on this.

[MichaelPatrick]

jnorman,

My approach is like yours -- based on minimal techniques because less is usually best. The main production goal is to to deliver the music on a pleasing sound stage in a complimentary acoustic that sounds "natural." An ideal production process is to put up one main pair of mics in a nice acoustic, record, normalize, DSRQ and print. When more steps or effects are used it's usually to remedy deficiencies in the acoustic, such as blending-in spot mics that are there just to ensure that every performer get's a proper spot in the sonic picture frame. (I hope you don't mind the metaphor; photography is apt because it is also an art of preservation and presentation.)

I just finished mixing choir, Mezzo-soprano, piano, and orchestral ensemble. The narrator's words had a piano bed. The piano, singer and narrator had spot mics but only the narrator was amplified. The physical staging was awkward; the acoustic was dry and mostly poor. My challenge was to create one sonic stage and to make the acoustic pleasing and natural. I used panning, EQ, compression on the narrator, and some hall ambiance.

For small groups that want to record-only, I tell them their options and try to encourage that they play into a nice acoustic balance. That way I can mic with mains and use spots if I think they may be needed in post.

[boojum]

Jim, I am a "less is more" from way back in the Mercury "Living Stereo" days. Currently Gaston and Petrus have posted sweet samples of their two mic technique, wiith their sweet little DPA 4060's The '61's work, too. My last pull was that duo with an ORTF CMC64 pair and that came out nicely, too.

I wanted to spot mic the instruments but the violnist discouraged me saying, "We don't listen to the instruments close so why mic them close?" He was right.

The chorale I record has a director who loves the '61's at about 12 feet over her head. It has the chorale in a nice balance and is far enough that the mid-range bump is not apparent.

It works; I'm sticking with it.

Cheers

[BrianHanke]

I'm not a fan of the "minimalist" approach in classical recordings. I associate that aesthetic with a thin, dry sound such as is often found in BIS releases. I think their recording style works better with large ensembles, but for solo and chamber music I'm not convinced.

Although it's easy to admire the technical accomplishment in recordings like this, I find it hard to truly enjoy what I'm hearing. My position is that one should do whatever it takes to get a rich, three-dimensional sound. If that involves digital effects, then so be it.

[rumleymusic]

I absolutely try to get the best recording possible with as little digital processing as I can. The results are always better when the source is better. Though I never shy away from processing when it is needed. Reverb is needed in about 80% of the locations I record in. And since the mic positions in live recordings are usually a crap-shoot, corrective stereo imaging in post is almost always present. LP EQ, always. I rarely, if ever, use compression and 'Doppler-like' stereo effects in a normal classical situation. Though a limiter is always there to bring the levels up.

I think it was said here that recording itself is an illusion. It is a trick, a lie. Engineers are masters of sonic fakery. I believe it is a little hypocritical to call oneself a "purist." That, I am sure of, is another way of saying, "I do half a$$ed recordings." The tools are there to correct problems, and complete the illusion. But like any good illusionist, it needs to sound real, not processed.

That is my philosophy on that.

I like minimalist miking techniques, but I think they work better on large ensembles rather than chamber music.

[BrianHanke]

Originally Posted by rumleymusic

I believe it is a little hypocritical to call oneself a "purist." That, I am sure of, is another way of saying, "I do half a$$ed recordings."

Haha, I love it.

[MichaelPatrick]

Originally Posted by rumleymusic

...a limiter is always there to bring the levels up...

Daniel,

I don't understand how a limiter gets the levels up. Doesn't it just keep them from going over the top? Do you mean you limit in order to raise input gain, confidently knowing that it won't hit zero dB? There are better ways, like getting a sound check before the downbeat to set levels relative to the loudest fortissimo.

With 24 bits and a clean input chain you should not need a limiter unless the music you're recording, or the gear itself, is unknown.

[rumleymusic]

No, no, I mean in post production/mastering, using a limiter to raise the level. Not until it is well past peaking, just enough to reach approx K-20 for the average of the loudest piece of the concert, and control stray peaks of heavy percussion.

I don't use a limiter during recording, I always try to give myself 6dB of headroom at the loudest sections for safety, but that is too quiet for most people's liking in a final product, those volume knobs are too hard to reach I guess.

[MichaelPatrick]

Originally Posted by rumleymusic

No, no, I mean in post production/mastering, using a limiter to raise the level. Not until it is well past peaking, just enough to reach approx K-20 for the average of the loudest piece of the concert, and control stray peaks of heavy percussion.

I don't use a limiter during recording, I always try to give myself 6dB of headroom at the loudest sections for safety, but that is too quiet for most people's liking in a final product, those volume knobs are too hard to reach I guess.

Ah. I have a similar technique that's manual: Search for peaks and surgically reduce them by applying a volume envelope adjustment to one peak or, at most, to several zero crossings at a time. Then I increase overall volume in one step to peak at about -.2 dB. I'd go to 0 dBFS but some cheap DACs don't handle it well.

[boojum]

Originally Posted by MichaelPatrick

Ah. I have a similar technique that's manual: Search for peaks and surgically reduce them by applying a volume envelope adjustment to one peak or, at most, to several zero crossings at a time. Then I increase overall volume in one step to peak at about -.2 dB. I'd go to 0 dBFS but some cheap DACs don't handle it well.

Sounds a lot like manual compression and normalization. What am I missing here???

[MichaelPatrick]

Originally Posted by boojum

Sounds a lot like manual compression and normalization. What am I missing here???

Manual limiting and normalization. In a surgical approach the "limited" peaks are selected and you can check each one before and after reduction. Also, there's no processor involved that can affect the overall sound. Most often only 3 to 15 samples are affected in each peak. No limiter I know of is that focused. Usually the number of peaks I need to reduce in a 1hr live-recorded acoustic program is between 1 and 10, so the "limiting" is pretty quick, and you can be confident that nothing is damaged because you've listened to every peak individually before and after "limiting."

[rumleymusic]

No limiter I know about is that focused.

Well most good modern limiters with lookahead technology can easily handle that. Nowadays I use PSP Xenon, it has two limiting stages so it is perfectly transparent, and has oversampling to prevent inter-sample peaks, so there is no real need to maintain the -.3dB cap. And as long as it is using 32bit or 64bit internal processing, there is never a "sound" imparted on the material unless the process is actually taking place.

The only issue when using a limiting processor is user abuse, and making sure it is not diminishing the sense of punch on the clipped section. A couple years ago I was in post on a performance of Stravinsky's Firebird. There are a few major hits at the beginning of the infernal dance (and in the finale). I had to do most of the volume work manually, bringing up the quiet sections, because a master bus limiter just killed the sense of attack on the louder sections. Strangely enough a recent recording of Firebird I just made did not have the same problem.

[jnorman]

i do the same thing as michael - i use volume envelopes to control peaks and smooth out overall volume levels. i have tried a number of limiters, but found they had deleterious effects on other aspects of the mix, particluarly if i tried to apply any gain in the process (within the limiter).

however, i guess the main question was basically about going toward a "studio" type sound versus a "live" type sound when mixing. when you record in a live venue, you dont really have much choice, but in a studio, you can do anything you want.

the previous 5 CDs for this group have all been studio productions, with generally close micing, harp spread wide LR and flute in the middle. this time i am not using any close mics, but rather ORTF and spaced omnis at 7-8 feet out from each instrument, but since tracking them individually still have the ability to manipulate each instrument in any way i want.

[MichaelPatrick]

Originally Posted by rumleymusic

Well most good modern limiters with lookahead technology can easily handle that. Nowadays I use PSP Xenon...

Makes sense since you can trust your tool. Having grown up old-school, i.e. creating masters by riding faders and splicing 1/2 track tapes, I have no problem managing a few odd peaks manually; and it would be hard for me to give up that level of control.

Originally Posted by rumleymusic

...[it] has oversampling to prevent inter-sample peaks, so there is no real need to maintain the -.3dB cap.

I leave final prints -.2dB or -.3dB down because some cheap DACs are unstable at zero. They're getting better, but I still feel safer not touching ceilings because there's no control over how a playback chain will behave.

[Nobilmente]

Overall my best recordings have always involved a good acoustic and less rather than more.
I would say, to the best of my memory, without exception.

Some things with just a pair have been full of presence and a sense of being - just there, especially the live events.

Michael, your recording of the Faure Requiem that you put up on the site about 6 months ago, it had a lovely presence. That was the crossed fig8 ribbons through a DAV. That recording was tangible.

And I do agree with you too about the 2db to 3db under, as when I first started recording digitally, I used to try and use almost all of the bits, like saturating the tape, but things improved when I backed off, and I'm happier with manual control too for the things I do.

You can do the musical things too with manual control.

[boojum]

I thinnk we are going to get into a "how many angels . . . " situation here. There are the analog vs digital groups and the minimalist vs "maximist(??)" groups. In the end up, though, I think most of us are in both camps in varying degrees.

In a perfect world I would be 100% minimalist as Gaston seems to be along with Petrus. Their minimalist setups DPA 4060's and a good recorder have produced astounding and wonderful results. They have good rooms and good performers. The trick then is to place your mics and performers as well as possible, and that is an area for endless discussion, too. ;o)

I will tweak a bit, if I need to. I am cautious in this as I am a real tyro trying to crib from you guys "how it is done." Additionally there are different degrees of post work in the pop and classical sides of the house. There seem to be no hard and fast rules here.

And the idea of creating a realer-than-real effect has some merit, but being discrete in this is the secret. As in all things this calls for reasoned judgement. Each case will be a bit different. I will strive to get a pull that needs as little tweaking as possible, and tweak a bit, maybe just a tiny bit, if I must.

It would be wonderful if there were just a button which would make it all work just right. When I was a programmer we all wanted the "FIXIT" verb for our code. It never showed up and we had to do it the hard way, just like in recording.

Cheers

[rumleymusic]

It would be wonderful if there were just a button which would make it all work just right. When I was a programmer we all wanted the "FIXIT" verb for our code. It never showed up and we had to do it the hard way, just like in recording.

Don't worry I'm sure Waves will come out with a plugin bundle of Alan Meyerson presets that will claim to make even the worst live recording sound like a multi-million dollar scoring session.

Obviously, as that saying goes: "you can't polish a turd". There is no substitute for good engineering during the recording, and a recording that allows you to do as little as possible in post, usually nets you the best sound. But in my career so far, I have never recorded in a perfect acoustic environment, with a perfectly balanced ensemble. And no mics or equipment out there can be considered "perfect". I see post work as a way to polish those subtle imperfections on an otherwise very usable recording. Whether you do it manually, or carefully use (good) processors (I think it is odd that some limiters have been causing trouble simply as a volume slider), I think we can all agree that results are the only real concern.

[MichaelPatrick]

I love recording because it takes some knowledge of the science and tools, but nothing can replace the artistic skills we must employ to preserve and present musical artworks in their own deserving glory.

[rumleymusic]

I love recording because it takes some knowledge of the science and tools, but nothing can replace the artistic skills we must employ to preserve and present musical artworks in their own deserving glory.

Now you are getting all spiritual on us

The thing that attracted me most to recording, after getting a degree in composition, was the similar feeling that I was creating something unique, and that was very personal to me. I approach a recording just as I approach a composition, from an artistic perspective. Just as knowledge of theory and instrumentation are there to serve the composer, knowledge of the tools and science are there to serve the engineer, not the other way around. So I guess we agree wholeheartedly on that.

[Don S]

Originally Posted by rumleymusic

But in my career so far, I have never recorded in a perfect acoustic environment, with a perfectly balanced ensemble. And no mics or equipment out there can be considered "perfect". I see post work as a way to polish those subtle imperfections on an otherwise very usable recording.

Sums it up for me. I've labeled it "maximizing musical value". It's our job to present a musical representation of what actually happened. These days many artists expect the recording to sound "better" than the concert anyway!

[Plush]

I think we would have to hear examples of the work with and without enhancement. Then we could offer an informed opinion.

I think that either approach is valid, but gotta hear dat sound.

[tenor39]

Originally Posted by BrianHanke

[COLOR=black][FONT=Verdana]I'm not a fan of the "minimalist" approach in classical recordings. I associate that aesthetic with a thin, dry sound such as is often found in BIS releases. I think their recording style works better with large ensembles, but for solo and chamber music I'm not convinced.

Really? A bit off the topic, but BIS came over to record the North Carolina Symphony a couple of years ago and I counted 29 mic's on the stage! Maybe they've changed tactics?

North Carolina Symphony

[rumleymusic]

I leave final prints -.2dB or -.3dB down because some cheap DACs are unstable at zero.

Supposedly (important word there) upsampling in a limiter is supposed to prevent that problem from happening, something to do with the DA completing the waveform information between 2 samples near the threshold, but I usually leave it at -.3 anyway, out of habit.

0dB It is important because we need that extra .2 dB to make our masters as loud as possible to compete in the professional realm.

Really? A bit off the topic, but BIS came over to record the North Carolina Symphony a couple of years ago and I counted 29 mic's on the stage!

I have seen some "minimalist" label's recording sessions. They usually bring the kitchen sink, leading me to question just what the term minimalist is supposed to mean. There is probably a difference in methodology between "minimalist" recording labels and "broke" engineers, the TRUE minimalists.




that 0dB part was a joke.

[roonsbane]

Since often I am expected to help out on mastering pieces for "Performance Today" when I am not dealing with my own recordings, our team is constantly listening to recent recordings from around the world. I am sure these are often sourced from many fellow remote slutz, but often we don't know who recorded them. It's hard to imagine anyone having a better overview of the state of live classical recordings today since two hours of music must be prepped five days a week. A lot of what we do is manually prepping for consistent levels. We also are expected to do whatever it takes to make recordings sound decent. Of course the Concertgebouw recordings need little to no sonic work. The vast majority of these recordings however can be made to sound more listen-able with appropriate use of one or several of these sonic tools like a High pass filter/manual gain riding/light EQ/Limiting /Light Compression/Noise Reduction/Spectral Repairing/and imaging plugins like a waves S1. Today I prepped 3 Beethoven Symphonies from the Orchestra of the age of Enlightenment that sounded fantastic and needed no EQ-ing. What a space! What a perfect frequency response! When I recorded that orchestra last year in a crap sounding church here in Saint Paul, anything I could use to make it sound more like that London recording was fair game.

Generally I think it's fair to say that the European stuff sounds nicer. They just have better sounding halls or maybe only their best sounding stuff is making it to us.

Cameron

[BrianHanke]

Originally Posted by tenor39

Really? A bit off the topic, but BIS came over to record the North Carolina Symphony a couple of years ago and I counted 29 mic's on the stage! Maybe they've changed tactics?

North Carolina Symphony

I've been trying to remember where I read about their approach to recording and it finally popped into my head just now. Some of their albums carry the motto "a BIS original dynamics recording." Unfortunately I can't immediately clarify what exactly that means, but I remember seeing somewhere that they basically just hit record and that's it.

This is especially noticeable in Fredrik Ullen's recording of the Ligeti Etudes. The first Etude is fairly quiet and thin-sounding for the most part (averaging around -10dB), but there's a pretty big spike in the right channel on one very brief chord that peaks at -4dB. Overall there's a lack of resonance and it sounds just like you'd expect something styled "original dynamics" to sound.

I have another CD with some of Schnittke's concertos. The performances are fantastic, but I find the sonics to be somewhat clinical and thin overall. For the Concerto Grosso they list two TLM 170s and two U 89s as the only mics used. Seems like a pretty minimalist approach for a large ensemble. Perhaps they've changed now? The Schnittke was recorded in 1987 and the Ligeti in 1996.

I also have a BIS recording of the Hindemith Tuba Sonata that is unbearable. It truly sounds like somebody put up two KM 184s, pressed the record button and then burned the results to a CD and called it a day. No depth, no warmth, just a plinky-sounding piano and a huge amount of tuba. Switching to Glenn Gould's recording of the same piece is a breath of fresh air, even though it was recorded decades earlier.

[jnorman]

inre:
""broke" engineers, the TRUE minimalists." that's rather funny, but a little too true...

brian: hey now - you can do hella worse than putting up a pair of km184s... i have owned pairs of cmc64s, 4011s, 4006s, c481s, m300s, at4051s, km140s, r-121s, 4038s, etc, and from my experience, none of them are perfect either.

[BrianHanke]

Oh no, I love KM 184s! I just meant that the Hindemith sounded like little care or thought was put into it. Two cardioid mics plopped somewhere in front of the players, press record, call it a day. Maybe in some special cases that approach would yield stunning results, but most of the time I think it's going to need some extra post-production help.

[tenor39]

Originally Posted by jnorman

inre:
""broke" engineers, the TRUE minimalists." that's rather funny, but a little too true...

brian: hey now - you can do hella worse than putting up a pair of km184s... i have owned pairs of cmc64s, 4011s, 4006s, c481s, m300s, at4051s, km140s, r-121s, 4038s, etc, and from my experience, none of them are perfect either.

Hmmmm... I think I'd put the 4006's up against any of the rest of that field and get a pretty good recording using only 2 of them.

As to what is considered 'minimalist', I think it could be agreed on as the fewest possible mic's arrayed to give the best possible imaging, depth, impact, realism, and so on. It still comes down to mic choice and placement. In my experience the better the mic's are, the less of them you need.

I would bet that the BIS recordings that sound dull and lifeless are probably multi-mic'd and not time-aligned. Seriously, 29 mic's in the orchestra?? How about massive amounts of comb filtering and phase issues. Blech!

[Plush]

Originally Posted by roonsbane

Since often I am expected to help out on mastering pieces for "Performance Today" when I am not dealing with my own recordings, our team is constantly listening to recent recordings from around the world. I am sure these are often sourced from many fellow remote slutz, but often we don't know who recorded them. It's hard to imagine anyone having a better overview of the state of live classical recordings today since two hours of music must be prepped five days a week. A lot of what we do is manually prepping for consistent levels. We also are expected to do whatever it takes to make recordings sound decent. Of course the Concertgebouw recordings need little to no sonic work. The vast majority of these recordings however can be made to sound more listen-able with appropriate use of one or several of these sonic tools like a High pass filter/manual gain riding/light EQ/Limiting /Light Compression/Noise Reduction/Spectral Repairing/and imaging plugins like a waves S1. Today I prepped 3 Beethoven Symphonies from the Orchestra of the age of Enlightenment that sounded fantastic and needed no EQ-ing. What a space! What a perfect frequency response! When I recorded that orchestra last year in a crap sounding church here in Saint Paul, anything I could use to make it sound more like that London recording was fair game.

Generally I think it's fair to say that the European stuff sounds nicer. They just have better sounding halls or maybe only their best sounding stuff is making it to us.

Cameron

I am curious whether or not "Performance Today" has a "house sound?"

Do you adjust the material so that it will sound best in an automobile?

with thanks, PHLUsh

[roonsbane]

Hi Plush,

You asked:

I am curious whether or not "Performance Today" has a "house sound?"

Do you adjust the material so that it will sound best in an automobile?

with thanks, PHLUsh

Well I will say that we have a producer that likes a big ambient sound. He is formally of Saint Paul Sunday. We also have a window we try and make our dynamics fit into. This way the on air processing is not doing the bulk of the work. Also each of us has a fairly similar template with plug-ins already loaded but bypassed. Usually that includes several different verbs. We also must work quickly since there are essentially two folks that are assigned to the show but must also run the show roll between the two of them. Then there are three of us that help when we can. This adds up to a third person for half a day on average Monday through Friday. It is a giant black hole.
Best!
Cameron