Recording New Classical Music

Recording Nikolai Kachanov Singers. Kind of brilliant piece by Podgaits.

Chorus, cello, and flute. I was gonna do a tree with Schoeps on left and right and an AEA ribbon 84 in the center. Then I was gonna put a Blue ribbon on the cello and an Oktava 012 on the flute.

The Blue, although an active ribbon, picked up a world of electro-magnetic nonsense and was very buzzy. I think that's from the lights inducting some sort of blerg into the signal. So I swapped it out with a Rode NT-1A.

The room the performers were in is small. Nikolai pushed for the bloom position -- which was technically in the next room over. About 10 feet apart and looking through the sort of portico into the room with the singers.

But the AEA started picking up EMI. And that far away, it wasn't doing that much anyway (meaning: it was doing nothing.) And the whole tree wasn't doing a "tree" thing. So I took it off.

The floors at the museum are mightily creaky. And the actual "performance" version had the flautist walking from the upstairs down to our level (which involved a videographer clearing out of her way.)

So I put some de-crackling on the main left and right microphone tracks. And, as always, pulled out a bit of 500-ish Hz (I think it was in the 460's somewhere frequency-wise.) I also put some dynamic EQ on the thing in life which bothers me so much -- the grinding sound of 2.6kHz coming off of parallel walls.*

* Whether this sound is just in my head or actually exists may be open to debate, but it's nice to get rid of it.

Wireless Antlion microphone.

Squigglewhumpity

The Lisp plugin is a free VST de-esser. So far I like it muchwise.

I've never seen any evidence this thing was ever actually manufactured. This may have been a teaser ad to see if they could get enough interest or orders in order to produce them.

Experimenting with the trigger gate plugin.
Paulxstretch is a time-stretching application for sound design.

Windows 10 download tool. Apparently it's still free for Windows 7 and 8 owners.

Coggle -- for making mind maps.

Cross-cultural emotional recognition.

More on volume. Moron volume.

So essentially there are 3 values to remember:

• The Program Loudness
• The Loudness Range
• The True Peak level

Mixing for broadcast loudness. 

At least this is what TC Electronic says.

"Studies have shown that 95% of the viewers will change the volume on their TV or receiver if the loudness jump or difference is an increase of 5 LU or decrease of 8 LU."

That seems like a pretty important thing to keep in mind I haven't heard anyone else mention.

For the real details check out this paper. 

I really need to get all of this information into my head, and figure out how to actually do it. 

Learning or not

On the East West piano patches the "DYN" patches allegedly have a more concave dynamics curve. That's the difference. So the Internet says.

99 Dollar Orchestra.

Frequencies of notes based on A=440.

Flute making.

Python script for calculating dimensions in flute making.

Opinion in academic writing.

Netflix has new delivery specs. What's nice is they publish them.
79dB SPL at -20db FS at the mix position. Well, or 82dB SPL.
Folding down 5.1's? Bring the center channel down -3dB.
But the big deal is -27dB LUFS on gated dialog for the entire program.

Stuf and such.

Cakewalk is back.

Unfilter tries to un-EQ stuff.

BBC's 50 Modern Classics podcast. About modern classical music.

An audio jobs blog.

Netflix loudness delivery spec.

Adagio For Strings chord progression.

I don't know if I posted this before. These be the books I must of read.

INDICATIVE READING

Required

Cooke M (2008) A history of film music (1st edition). New York: Cambridge University Press

Hill J and Gibson P (1998) The Oxford guide to film studies (1st edition). Oxford: Oxford University Press

Karlin, F. and Wright, R. (1990). On the track. 1st ed. New York: Schirmer Books.

Rona, J. (2000). The reel world. 1st ed. San Francisco: Miller Freeman Books.

Recommended

Adler, S. (2002). The study of orchestration. 1st ed. New York: W.W. Norton.

Jones C and Jolliffe G (2000) The guerilla film makers handbook (1st edition). New York: Continuum

Monaco J (2009) How to read a film (1st edition). Oxford: Oxford University Press

Jacob, G. (1940). Orchestral technique. 1st ed. London: Oxford University Press, G. Cumberlege.

Mancini, H. (1977). Sounds and scores. 1st ed. Greenwich: Northridge Music Inc.

Piston, W. and Piston, W. (1955). Orchestration. 1st ed. New York: W. W. Norton & Company.

Read, G. (1979). Music notation. 1st ed. New York: Taplinger Pub. Co.

Rimsky-Korsakov, N., Shteinberg, M. and Agate, E. (1912). Principles of orchestration. 1st ed. Berlin: Edition Russe de Musique.

This list of required and recommended materials applies to the whole course. Where additional specific recommendations are made, they will appear in the module handbooks. For the latest complete recommended resources, please visit our website: www.thinkspaceonline.com

MonoPrice 15W

The Monoprice tube guitar amp sounds disturbingly good for less than $200. Or maybe less than $600. Or maybe infinity if it's workin' for ya. (H/T my big brother Big Bells.)

One from the other

MIDI Machine Control.
Loopbe is a virtual MIDI port which I've been using to sync two instances of Samplitude together so I can write music while editing audio.
Bills Recording services and repairs Oktava microphones.
Libsyn is seemingly the CDBaby of podcasting.

Faith. And a new faith.

My faith is still crisis-y regarding preamps.

Maybe the whole theory that the accumulation of subtleties in multitrack recordings is what makes the very small differences in preamps important is actually correct.

I'm tempted to test this in a way that has one scientific flaw. But just you watch. I'm gonna do it anyway.

If you run a microphone into more than one preamp at the same time, you're loading down that microphone and it doesn't behave the way it ought to. So that's why preamp tests on the Interwebs are always one microphone:one preamp.
But uh. Modern preamps are made to bridge. And when one actually runs the numbers, the loss and weirdness of running a normal modern microphone into multiple preamps actually ain't that much. So I am going with the notion that one can record to more than one preamp simultaneously. There is a science-loss here because there is a mathematical difference. But it's worth trying because until we get robots who will exactly replicate all moves on all instruments, it's all we've got.*

1. Install and register Samplitude, Sibelius, and the most recent version of ReWire (the one that comes on the Sibelius CD is fine)
2. Start up Sibelius on its own, go to the Play > Playback Devices dialog, select Audio Engine Options and, if the "Repair" button in the resulting dialog is enabled, click it [Drew's note: I believe this option has been moved to another place in the latest version of Sibelius.]
3. Shut down Sibelius
4. Start up Samplitude
5. On the VST/DirectX/ReWire page of the Options > System dialog (which is inside the Effects group of pages), tick the box that says "Activate ReWire"
6. Restart Samplitude
7. Open a Samplitude project, and create/select a MIDI track
8. With the MIDI track selected, go to the MIDI panel (by default this is on the left of the screen, under the Track Editor); from the "Out:" drop-down menu, select New Instrument > Rewire > Sibelius
7. Sibelius should now start up (and Samplitude will give a warning message if you're not using ASIO)
8. Open your desired score; if you're using Sibelius Sounds Essentials you'll probably want to leave it a while for Sibelius to load all the required sounds. Now starting playback in Samplitude should also start it in Sibelius.
9. If you want to record Sibelius' output onto a track in Samplitude, choose a stereo Audio track, and, in the "In:" drop-down on the Audio panel, choose Instrument Outputs > Sibelius Mix-L Mix-R


More tests forthcomingwise. 

*So far those robots only exist on pianos.

2.6kHzI

Rightmark makes audio testing hardware which is interesting, and there's a free version.
Izotope RX for post. Way not free.

My ears have been clogged and it's really irritating. I've got these drops, but I may just go have them irrigated. Bleh.

FFT images of my three Oktava 012 mics with hypercardioid capsules aimed at an air purifier.

Oversampling in digital equalizers.

§

I am super irritated with  2.6 (or so) kHz. All my life the harshness of parallel walls or sopranos or something in this region has really bothered me. And other than notching it with a multiband compressor I just don't know what to do. Sometimes I think "Egads! Is that sound really happening in the space?" And... it is. Ugh.
I don't know why it takes me so many years to finally "get" what's bothering me. I have a specific memory of hearing a soprano sing at the little barn-theater at my high school and being bugged by it. I remember touring with the Wooster Group and being irritated by it once the volume got too loud. We could pull frequencies out, but it would get too muffled. Of course at the time all we had were sloppy Klark Teknik 31-band graphic eq's. So you could make the sound right when it was loud, but then when you got quiet again the sound was very muffled. It sorta sucked. A narrow-band de-esser might have worked but I don't know if there were any commercially available ones which went down that low (this was the early 90's.)
Anyway, I want some nice corrective eq's or something in the way of a phase-coherent hand-limited and frequency-variable compressor (without makeup gain.)
For classical music I'm really digging the preamps in the Zoom F8. I know, an unpopular opinion. But they're really great. There's just no inserts or EQ's available.
The new version of the F8 let's you record to your computer and to the SD cards simultaneously.
I would b interested in knowing what the preamps on (say) a Midas M32r sound like.

Pro Audio is Malarkey

Or: grousing for fun and profit.

The stereotype of sound guys is that they're lazy jerks. Just because that's true doesn't make it so, of course. But yeah.
One problem is the sheer amount of religion in sound. It's virtually impossible to do a real A/B comparison that's double-blind. But it can be done.
Firstwise is the infamous Sound On Sound preamp test where they could really do a double-blind listen to the same source. Remember this one? It's the one that showed that the cheap $300 ART tube mic preamp and the cheaper Mackie VLZ preamps to beat out Neve, API, and SSL. Yeah. That one.

Over at JWSound a poor user dared to test the Rode NT5 against the Schoeps MK41. Foolish mortal, he. (And don't get me started on how the nominal competence level of the production sound mixer is so far below average studio intern as to simply be embarrassing. I mean, they as a group have zero clue about what they're doing and the kind of audio they/we do is very simple. Not always easy, but always simple.)

And then there are these blind preamp tests.

I did a really un-scientific test of my three Oktava 012 mics with hypercardioid capsules.

You know something's up when the fear of a double-blind study is so great that people try to deliberately skunk the results.  Remember Ian Shepherd's test of humans vs LANDR and Aria machines?  (I know, it wasn't his test but he did work on it and I can't find any links to it now.) Well it turns out the "blind" listeners were super-affected by comments other people made. So the differences between man and machine weren't really all that significant (although they were there... we think.)

I have a notion that much of the actual differences in audio gear is that equipment has become on average vastly better than it used to be. A/D converters, for instance, are much better now than they used to be and that's true of mic preamps and analog signal paths as well as microphones themselves. I do have a great deal of difficulty caring about the difference between a U87 and a Rode NT1. I just... I don't care. An EQ will do whatever it is you need to an NT1 to make it a U87.
There's still differences between microphone types. So big-diaphragm multipattern mics have that big dry sound I like so much in live sound reinforcement (warning, danger, that is a radical opinion in and of itself, nobody else uses them that way.) And small diaphragm condensers will tend to sound the way they sound (very broadly speaking, I hate the off-axis nonsense of most all small-diaphragm other than Schoeps or Oktava, but I challenge you to distinguish in a recording which is which.)

Lost Faith

So I'm more than willing to admit that I'm going insane. I just need to put that out there, up-front and so on.
But I'm having a crisis of faith in preamps.

I have a pair of Neve 1272's (modified by Brent Averill.) I liked them better than the old 1073 I had (I know, let's just list this as apostasy #1). But although I've gotten some great guitar sounds with them, I've really started to prefer the Lindell preamps. So, I figure, maybe as I've gotten older I've been feeling more API than Neve. That's fine.

So the other night I was recording Russian Chamber Chorus with my Zoom F8 and I figured I would use Lindell preamps. Because, you know, more better, right?

No. No, not at all. There at the time I hated the sound of the Lindells on this ensemble. Hated.
Let's review. I have a custom ribbon microphone in the air above the conductor, and a pair of spaced Oktava 012's with hypercardioid capsules in the sweet spot in the audience where we get a "bloom" of the sound.

There's also a backup microphone on that stand with the ribbon. Oh, and the ribbon has a cloudlifter on it.

Now while I was recording it was immediately and imminently obvious that the Lindells were not as detailed and nice for this material. Counter-intuitively, however, when I got home I couldn't tell the difference between the Lindells and the internal preamps on the F8.
Which is odd because you'd expect that differences would be more apparent listening on Sennheiser HD 600's or my Blue Sky system. So why how what wut? How could something be so obvious in the field but not under more controlled conditions?
Well, on location I'm using lower impedance headphones -- Sennheiser HD 280's. And the F8 does not have the world's best headphone amp. So at first I thought that maybe the TRS line inputs took a different path for monitoring off the F8. That doesn't seem to be the case. Maybe it's because of the phase relationship between the line inputs and the mic inputs and the actual sound which was bleeding into my headphones from the actual singers?
I have no idea.
But in either case, the F8's preamps are very tres excellent for classical music. Does this mean I should just get rid of the Lindells?
Well, maybe not. Maybe for rock and roll the API-like sound of the Lindells is better. What I noticed was there was less "air" and less detail. But for a fat blues guitar, that might be just the ticket. I don't know.
I do know, however, that the Neve's which I'm not using probably need to go to someone who will love and make records with them. Because I'm not.
So today I recorded RCCNY with a pair of ART Pro MPA II tube preamps. In the field I could not tell the difference between the ART's and the internal preamps to the F8. So I figured "At first, do no harm" and recorded the concert with the ART's.
Listening back to the rehearsals where I was switching between the ART's and the internals can I hear a difference? Uh. Maybe? It's not a big one. And it's certainly under the threshold of placebo effect. So I'm not super into lugging the ART around.
For classical music it seems that the internal preamps to the F8 are more than fine.

§

But wait. That's not all. 2.6kHz. It drives me nuts. Wait 'till next time.

Animal Poems

So this is the first draft of the first movement of a horn concerto. I enjoy working in Sibelius. I do not enjoy the viola's C clef. Nor do I enjoy instruments which are written in other keys (although Sibelius makes up for this and lets me write the F horn in C so that it's less irksome to look at in the score.)
I think I'm not linking the themes together enough. And I don't modulate. I should probably modulate. I'm guessing that's one of my bigger failures as a classical composer.

Recording Chorus

I got to record the Nikolai Kachanov singers last weekend. They did two pieces -- an all-men's ensemble, and a piece for chorus and classical saxophone quartet.
The New Hudson Saxophone Quartet is really fantastic.
So we're at the Good Shepherd church (famous for Robert Shaw recordings, among others) and a couple things: there is a "bloom" of sound that starts at about the 5th pew. That's the sound the music director likes so I put a pair of spaced (about 2 meters) hypercardioid mics there as a stereo pair. Those are Oktava 012's.
But I also put up a center microphone. Well, I put up three.

Center cluster -- handmade ribbon mic, Oktava 012, handmade U-48 clone.

 I like to have appropriate mics ready because different pieces frequently must needs different setups. But also, I was experimenting with different sounds. So for the rehearsals I put up a cluster to listen to different sounds. I discovered that the ribbon just did some magical something. All the things everyone says about ribbon microphones? Yeah. Creamy yet beautiful and articulate high-end. Noisy, sure, but beautiful. I listened to an Oktava. I listened to the U48 clone. I listened to a Rode NT1-A. None of them did the "thing" the ribbon was doing.
The other thing about the ribbon is that it's inherently in a figure-8 pattern, which means it picks up sound behind as well as in front. So as a "close" microphone it sounds about twice as far away. So I put it close to the action. Almost right over the conductor's head.
But the sound of that mic was more like a couple meters away from the conductor. And it was being blended with the two mics behind it which were several meters behind.

But during the rehearsal for the men's ensemble (not pictured) I felt I had too much distance on the men and I wanted to place some large diaphragm condenser further downstage and pointed left and right.
So this is how it went on performance day: Ribbon mic in center. But I was a little worried about the delicate nature and ostensible crankiness of the ribbon, so I put an Oktava 012 up on that stand as a backup. And then I put the U48 clone up there in figure-8 mode but sideways, in anticipation of possibly doing a mid/side recording in the center cluster.
I also placed a pair of wide-spaced Rode NT1-A's just ahead of the first pew. The notion here was to better grab the men's ensemble.
Well, this is why we have options.
Nikolai told the group that the sound would dry up during performance because of people in the room. I looked at the room and thought "Well it can't dry up that much. The ceiling is like 15 miles away." But heck if I haven't recorded at this church a dozen times, yup. It dried up significantly down there on the floor.
What this meant was that the microphones at the 5th pew ended up being better than 50% of the stereo mix. And the ribbon just added a bit of prettiness and presence. The backup Oktava was not needed (thankfully). The U48 was also not doing anything for anybody. And the two Rode NT1-A's also didn't end up in the mix.
Seven mics in the air. Only three played.
The church has a bit of build-up at about 3kHz. Only once you get past a certain volume level. So back in the studio I put a bit of frequency-specific compression with a relatively narrow band at 3k.

There should be a new YouTube video of one or the other of these pieces. Asaf Blasberg was the videographer. Interestingly he also had a Zoom F8, so we jam-synced timecode to his F8, then he brought his F8 over to his camera (a 4K Panasonic) and jam-synced the camera. That way my files just blooped right over into his timeline. Kinda cool actually.


YouLean loudness meter. Oh boy. This is a nice loudness meter.

The Elephant's Dream

I'm learning Sibelius. And doing so I decided to write the world's worst string quartet. Here it is.

The Residents

The show last night was all old hippies and the most beautiful 50-year-old women you've ever seen. It took place at a really nice venue with lots of leg room which was an active temple. The menorah on the side of the stage was a nice aesthetic touch.
So it was a big surprise when the band just stopped and said the neighbors complained about the noise and the cops were there. Concert over.
I'd never even been to a *party* shut down by the cops before.
The weird comic bit I think is just how mellow an audience this was. I mean, it's a San Fransisco band. And basically anyone under 35 was carded and told they needed better ID*
The costumes were great. I was expecting more costume changes but maybe those were going to happen later in the show? I enjoyed how the musicians sometimes had to adjust their beaks between songs. 

The singing was more Cookie-Monster than I'd expected. And the guitar was more '80's finger-tapping than I would have figured.
The backdrop matched the costumes. We met the tour's LD and I told him he did a great job. He told me he only had 8 instruments.
Almost exactly 37 years ago Pete Cenedella made me listen to the Residents in his dorm room at Solebury School. Which is kinda awesome.

*This is a lie but one that amuses me.

Microphone tests with acoustic guitar

I test an Aurycle kit microphone (a u47 clone, all stock), the Austin ribbon kit (with Cinemag transformer), a Rode NT1-A, and an AKG C12-A.

Left to right: AKG C12-A (with the "Utopia" logo), Rode NT-1A, Austin Ribbon kit (Cinemag transformer), Aurycle U-47 multipattern clone kit.

They all go through the preamps on a Focusrite 18i20. I lined them up with a bit of distance from the guitar just so they shouldn't sound terribly different from one another just due to the positioning.
It's virtually impossible to do a real blind A/B comparison, so I generally don't even try. But there are some clear differences betwixt each of these mics.

Art Rock

I've had the very pretentious idea of "art song on the top, rock song on the bottom" for, well, since I was a teenager.
I've never quite been able to pull it off.

A baritone guitar would probably help.

The idea being to have a groove underneath, but to continue to vary the melody and harmony on top like classical Western music.

I dunno. Maybe I can pull it off eventually. I'll start with a concerto for electric guitar. That sounds appropriately pretentious. No?

And then what

Master no louder than -9 LUFS short-term at the loudest moments
(with True Peaks no higher than -1)

That's what Ian Shepherd says at least. 

SimulAnalog guitar suite has a pretty decent Univibe free VST plugin.