In this article reverb pioneer Michael Carnes shares one of his other passions: Recording classical music. With his detailed knowledge of the mechanics of sounds in spaces how does he approach capturing that most challenging of instruments, the grand piano?
Capturing The Performance
I was recently asked to record two pieces of piano music played by my friend Jason Hardink (pianist with the Utah Symphony). I’ve enjoyed recording Jason in the past—both in concert and in controlled sessions. This was our most ambitious project yet—2 pieces of very difficult music for video. I was to record audio with Ashkan Tabatabaie handling video. We booked the concert hall at Westminster College in Salt Lake City and had Jason’s piano delivered to the venue. Jason owns a very special Steinway D and it has to be my favorite of all the big pianos I’ve recorded.
In this article I’m going to describe the process of recording. I’ll list mics, positions and my reasoning. The pieces are Promontories by another good friend, Steve Roens and a melody which the air had strained by Jason Eckardt. They are both very difficult pieces to play, but this is the sort of stuff that Jason Hardink loves to sink his teeth into. These are both world premieres. Original in-person performances were scuttled by the pandemic, so Jason opted for an online performance instead. He’s looking forward to playing the pieces live in the same venue, probably in spring of 2022.
I should say first that because I am a composer (in the traditional conservatory sense), I hear like a composer. One of my goals in a recording is clarity. I want to hear all the lines—all the details—in a balanced way. If I’m recording music of the Romantic period, I may want some more distance from the instrument. But in music like this, I want you to hear things in detail, just as it was in the composer’s mind. So with that, let’s take a look at my mic setup:
The most focused set of mics is the M/S pair right in the waist of the piano. I use a Schoeps MK21 (wide cardioid) as mid and a Samar VL37a ribbon as side. For placement I used the tried-and-true method of covering one ear and moving my head around as Jason played. It’s usually pretty obvious when you find the right spot. A ribbon—if it matches well with the mid mic—is the best choice for mid-side. It has near-perfect off-axis rejection and will give you better control of the overall image. You’ll notice that the M/S pair is quite close to the piano. If the session was to use this pair as the only mics, then I’d back them off a couple of feet.
Next up is an A/B pair of omnis (Schoeps MK21) spaced on either side of the piano, from the audience perspective. These are quite important as well. They help to support a wide and robust image, with the omnis doing a great job of getting the bottom of the instrument.
Behind the piano is a pair of room mics: Schoeps MK2-S. They’re on a single tripod, separated by a couple of meters and are about 10’ up, with the mics pointing roughly toward the back part of the hall ceiling. These mics have the tiniest bit of HF lift to counter air absorption. In an ideal world, I may have placed them on the other side of the piano, but the mics needed to share the stage with 4 video cameras. I doubt this really made much difference.
That leaves an unusual pair of AKG 414XLS mics, set up as slightly-spaced omnis, off the tail of the piano. This was an experiment based on a technique developed by engineers at Decca. By themselves, the two mics give an interesting open perspective that’s nice for Romantic music. Since this was an experiment, it wasn’t important if I used them in the final mix. I did find a use, but probably not what Decca would expect. And to be honest I’d have used a pair of small-diaphragm condensers if I had a couple more. But the AKGs were fine for this test. If they turned out to be usable, so much the better.
The mics fed into my Focusrite Red 8Pre, running at 192K. Then into a MacBook Pro. I recorded into Logic (mixing itself would happen in Pro Tools). It was a hot day outside—about 100F (38C), so the hall air conditioner was going full blast. From setup to teardown, we ran about 6-7 hours.
Audio files were recorded as two long segments—one for each piece. The same is true for the cameras. This meant that it was easy to sync, since all we needed was a handclap. We did not involve timecode. Digital clocks are very good these days and we simply didn’t encounter any sort of slipping. This made it easy for Jason to choose his edit list. All we needed was a time offset and measure/beat in the score. I immediately shipped a rough mix over to Jason, Steve and Ashkan. Performer and composer had edits to select. Videographer had color grading and alignment for four cameras. I also had plenty on my plate.
Jason Hardink has made the first video public. You can see it here.