Q&A
Welcome to our Q&A area, affectionately known to readers as "Talkback". Here, you can see our Editors', Writers', and Industry Specialists' answers to your recording-related questions -- we'll look to answer as many questions as we can. We're here to help!
Click on Question to view Answer.
Q: Hi, I wonder if you can help me. I am looking for a pair of pro-level instrument mics to handle almost EVERYTHING! Percussion (hand-held drums, tambourine, triangle, glockenspiel, clave sticks, chimes – but not side drums per se as I have a Roland TD-20 kit), acoustic (guitar, flute, violin, trumpet, etc.).
I don’t care what the cost is, as long as it is versatile mic with an extended range and pretty flat. I was interested in Earthworks QTC 50, but then I read that small diaphragms, which react very fast to transients, exaggerate the difference between the level of the initial transients and the level of the subsequent sustaining sound, making it very difficult to set gain levels.
What would you recommend?
Thanks for your help—Chris Dixon
A: Hi Chris, Well, that's a tough question. There are a lot of versatile mics out there, especially if you can afford something in the range of the QTC50. Here are some questions to ask yourself:
1. You want a pretty flat sounding mic, but what about coloration? Would you like a very neutral sounding mic, or something that might be more flattering? Earthworks are very nice microphones--not really my personal cup of tea, but many people like them as they are excellent devices. To me, they can sound almost too clinical and neutral, to the point of being cold. Again, however, many people like this.
2. Do you want a truly flat mic? Many "flat" mics are smooth through the lows and mids, yet have some high frequency boost as many sources really need some sparkle to not sound flat and lifeless. This all depends on what you're recording, the room and your tastes--but consider it. Truly flat mics sometimes lack the sonic curb appeal that you really need. The Neumann KM 184 is a very smooth microphone, yet has a fair bit of HF sparkle that makes it work really well for a lot of sources.
Here's how I break down the major players in small diaphragm mics. These are generalizations based on my personal experiences and there are many other brands. Flat + no coloration, sometimes described as "clinical" or "cold" or "like a microscope": - DPA (formerly B&K) - Earthworks Flat + no coloration, yet a little sweeter and more musical - Schoeps - Sennheiser (MKH series of mics--I'd say a little sweeter than Schoeps) Flat + no coloration but VERY musical - Specifically Sennheiser MKH 800 (expensive, unfortunately) Smooth and Musical - Neumann -AKG - Many others fall into this category, I'd say. If you can try any mics before you buy, that would be ideal. Get a pair of Neumann KM 184's as a place to start. If those are too bright/to much HF lift, then look at Sennheiser MKH series (not the MKH 40, though--it's too hard sounding for what you're doing) or Schoeps. Other mics to check out that aren't as neutral but very functional for what you're describing (without knowing your music): AKG C 414, Shure KSM 141, Neumann TLM 127 (sweet if you can afford it). Oh, and to answer your question regarding the small capsules of the Earthworks. It's true that really small diaphragms makes for really fast transient response. I've never had a problem setting levels, however. The difference is much more noticeable when going from a dynamic or ribbon to condenser. Never really a problem, though! Hope this helps! Sorry for being so long-winded, but microphones are fun and there's a lot to talk about! Let me know what you try/buy and end up liking. I'd love to hear about it! Thanks for reading...
Cheers, Justin Peacock
1. You want a pretty flat sounding mic, but what about coloration? Would you like a very neutral sounding mic, or something that might be more flattering? Earthworks are very nice microphones--not really my personal cup of tea, but many people like them as they are excellent devices. To me, they can sound almost too clinical and neutral, to the point of being cold. Again, however, many people like this.
2. Do you want a truly flat mic? Many "flat" mics are smooth through the lows and mids, yet have some high frequency boost as many sources really need some sparkle to not sound flat and lifeless. This all depends on what you're recording, the room and your tastes--but consider it. Truly flat mics sometimes lack the sonic curb appeal that you really need. The Neumann KM 184 is a very smooth microphone, yet has a fair bit of HF sparkle that makes it work really well for a lot of sources.
Here's how I break down the major players in small diaphragm mics. These are generalizations based on my personal experiences and there are many other brands. Flat + no coloration, sometimes described as "clinical" or "cold" or "like a microscope": - DPA (formerly B&K) - Earthworks Flat + no coloration, yet a little sweeter and more musical - Schoeps - Sennheiser (MKH series of mics--I'd say a little sweeter than Schoeps) Flat + no coloration but VERY musical - Specifically Sennheiser MKH 800 (expensive, unfortunately) Smooth and Musical - Neumann -AKG - Many others fall into this category, I'd say. If you can try any mics before you buy, that would be ideal. Get a pair of Neumann KM 184's as a place to start. If those are too bright/to much HF lift, then look at Sennheiser MKH series (not the MKH 40, though--it's too hard sounding for what you're doing) or Schoeps. Other mics to check out that aren't as neutral but very functional for what you're describing (without knowing your music): AKG C 414, Shure KSM 141, Neumann TLM 127 (sweet if you can afford it). Oh, and to answer your question regarding the small capsules of the Earthworks. It's true that really small diaphragms makes for really fast transient response. I've never had a problem setting levels, however. The difference is much more noticeable when going from a dynamic or ribbon to condenser. Never really a problem, though! Hope this helps! Sorry for being so long-winded, but microphones are fun and there's a lot to talk about! Let me know what you try/buy and end up liking. I'd love to hear about it! Thanks for reading...
Cheers, Justin Peacock
Q: Love the mag, long-time subscriber. Perhaps you can clarify something for me. I record in Cubase LE at 16 bit resolution. I apply 16 bit dither (as an insert on the master bus) during mixdown. My reasoning is that Cubase LE handles all audio at 24 bit internal resolution. So if I don't apply dither during the mixdown, Cubase will truncate the 24 bit mixdown file to a 16 bit file. Is my process correct?
Thanks—Mark A.
A: Hi Mark, I spoke to Paul Vnuk Jr., one of our Cubase gurus, and he confirms that your process is correct... But it's not your only viable option. Dither is always better than truncation, and it should be the very last thing you apply when mixing down a song for 16-bit distribution, as on a CD.
So if you're creating a CD from mixes created in Cubase, putting dither on the end of the master bus processing is the right thing to do (see Bob Emmet's article in the December 2008 issue for more on this). Alternatively, if you're going to have your recording professionally mastered, it's best to create a final mix at the higher resolution you've been working with, and present that to the mastering engineer without dither or final eq tweaks, so he has the simplest and best-quality mix to work with. Similarly, if you have a separate two-track editing/CD preparation program like WaveLab, Sound Forge, or Peak, then it's best to take your file into that program at higher resolution and do your dithering there.
Thanks for reading!—MM
So if you're creating a CD from mixes created in Cubase, putting dither on the end of the master bus processing is the right thing to do (see Bob Emmet's article in the December 2008 issue for more on this). Alternatively, if you're going to have your recording professionally mastered, it's best to create a final mix at the higher resolution you've been working with, and present that to the mastering engineer without dither or final eq tweaks, so he has the simplest and best-quality mix to work with. Similarly, if you have a separate two-track editing/CD preparation program like WaveLab, Sound Forge, or Peak, then it's best to take your file into that program at higher resolution and do your dithering there.
Thanks for reading!—MM
Q: Thanks for the reply Mike. I don't think I was clear in my question. Let me try this another way: I record audio in 16-bit resolution into Cubase LE (so ALL my audio files are 16-bit). I mixdown to a 16-bit wav file. The final destination will be CD (16-bit). The song will not be mastered, I will take the Cubase mixdown file (16-bit) and burn it right to CD.
In this scenario, do I need to dither in Cubase at mixdown? Some of my friends say I don't need to dither because all files are 16-bit. Others say I MUST dither because even though I specify my mixdown file to be 16-bit, Cubase mixes with 24 bit internal resolution.
Confused,
Mark
A: Mark, No problem, allow me to clarify.
Short answer: You should dither. (MUST is a strong word; it's not 100% required, just a really good idea.)
Long answer: All DAWs in fact convert everything to, and work with, 24- or 32-bit internal resolution (depending on which DAW), so producing a final 16-bit file requires dither.
Why do this? Let me show you with a simple example, working with 2-bit numbers rather than 16-bit.
- Let's say we have two numbers that we want to add together: 2 + 2 = 4. Okay?
- In binary, that's 10 + 10 = 100.
- We have added two 2-bit numbers and gotten a 3-bit number.
- If we were only capable of keeping two bits, we'd throw away the least significant bit, giving us a final value of 10, or 2.
- That gives us 2 + 2 = 2. Not very accurate math, right?
The same principle holds when you're working with digital audio inside a DAW. Digital signal processing is just math being done with 16-bit binary data... math like adding numbers together (which is what you do when you mix). When you add two 16-bit numbers, you'll get a 17-bit number sometimes. And if you keep adding them together as you mix and sum your audio signals, that sum will grow...
By working with a 24-bit data path, you don't have to keep shaving off the lowest bit every time you add numbers together, a process which makes hash of your audio in very short order.
When you're all done, you have a 24-bit file that has to be turned into a 16-bit file. Even truncating the last eight bits at the end of the process will sound better than truncating anything beyond 16 bits every time you do some math (44.1 thousand of these operations per second!), but dither will sound better still, and if you can do it as your last step, you should.
Thanks for reading and sorry for the confusion.—MM
Short answer: You should dither. (MUST is a strong word; it's not 100% required, just a really good idea.)
Long answer: All DAWs in fact convert everything to, and work with, 24- or 32-bit internal resolution (depending on which DAW), so producing a final 16-bit file requires dither.
Why do this? Let me show you with a simple example, working with 2-bit numbers rather than 16-bit.
- Let's say we have two numbers that we want to add together: 2 + 2 = 4. Okay?
- In binary, that's 10 + 10 = 100.
- We have added two 2-bit numbers and gotten a 3-bit number.
- If we were only capable of keeping two bits, we'd throw away the least significant bit, giving us a final value of 10, or 2.
- That gives us 2 + 2 = 2. Not very accurate math, right?
The same principle holds when you're working with digital audio inside a DAW. Digital signal processing is just math being done with 16-bit binary data... math like adding numbers together (which is what you do when you mix). When you add two 16-bit numbers, you'll get a 17-bit number sometimes. And if you keep adding them together as you mix and sum your audio signals, that sum will grow...
By working with a 24-bit data path, you don't have to keep shaving off the lowest bit every time you add numbers together, a process which makes hash of your audio in very short order.
When you're all done, you have a 24-bit file that has to be turned into a 16-bit file. Even truncating the last eight bits at the end of the process will sound better than truncating anything beyond 16 bits every time you do some math (44.1 thousand of these operations per second!), but dither will sound better still, and if you can do it as your last step, you should.
Thanks for reading and sorry for the confusion.—MM
Q: I’ve been a subscriber for a few years. I came across this sound sample (link below) and I am at a loss to explain the huge width of the sound field. It was recorded in 1952 by Hugh Tracey (inventor of the Kalimba - interesting story) on a Lyrec or Nagra recorder. I would like to understand what Hugh did to capture this. Please feel free to forward this and/or reply.
Thanks—Mark T. Chard
http://www.smithsonianglobalsound.org/trackdetail.aspx?itemid=41104
A: Paul Stamler and Mike Rivers responded to this with nearly identical replies.
Paul writes:
Judging by the streaming-audio sample, the recording's two channels have opposite polarities—in popular (if inaccurate) terms, they're "out of phase" with each other. A field recording from 1952 was almost certainly mono; for whatever reason, the streamed version is in stereo, with one channel's polarity backwards. Listen to the recording in mono (with the two channels combined), and it almost disappears.
Usually when a recording has "super-wide" stereo sound, something like this is going on. Engineers occasionally do stuff like this for special effects, generating sounds that appear to be "beyond the speakers", but they cause all kinds of oddities when the recording is played back in mono, and for various reasons they make the recording almost unplayable on the radio.
Peace—PJS
Mike writes:
Well, this one is easy. It's a mono recording with the same program material on both channels of a stereo file with a problem. The problem is that the polarity is reversed on one of the channels, putting them 180 degrees out of phase. This is what creates the big hole in the middle.
If you have a DAW program that easily allows you to invert the polarity ("phase") of one channel, do that and it'll sound like it's supposed to sound. You can accomplish the same thing by reversing the wires going to one speaker.
Hopefully only the sample file on the web site is screwed up like this and not the CD. Incidentally, in 1952, the recording was almost certainly mono (Nagra didn't introduce their first stereo recorder, the IV-S, until 1971).
In this sample, the two channels aren't identical, which they should be in a proper stereo representation of a mono program. This could be a result of the data reduction process used to make a small web-friendly sample (MP3 compression) or it could be the result of the the original full track recording being played with a stereo playback head and re-recorded as two channels. This is suggested by the bit of "swishiness" you can hear, resulting from the worn tape wobbling a bit as it passes the playback head, causing the same sound to be reproduced at a very slightly different time in the two channels of the playback head.
By today's standards, this would be considered pretty sloppy transcription work, but it may have been the best the low budget company could do when the record was originally issued. Smithsonian Folkways may not have had the original recording when they reissued theit on CD. There's really no excuse, however, for the inverted channel on the web sample other than that someone just wasn't paying attention. It's easy to fix.--MR
Paul writes:
Judging by the streaming-audio sample, the recording's two channels have opposite polarities—in popular (if inaccurate) terms, they're "out of phase" with each other. A field recording from 1952 was almost certainly mono; for whatever reason, the streamed version is in stereo, with one channel's polarity backwards. Listen to the recording in mono (with the two channels combined), and it almost disappears.
Usually when a recording has "super-wide" stereo sound, something like this is going on. Engineers occasionally do stuff like this for special effects, generating sounds that appear to be "beyond the speakers", but they cause all kinds of oddities when the recording is played back in mono, and for various reasons they make the recording almost unplayable on the radio.
Peace—PJS
Mike writes:
Well, this one is easy. It's a mono recording with the same program material on both channels of a stereo file with a problem. The problem is that the polarity is reversed on one of the channels, putting them 180 degrees out of phase. This is what creates the big hole in the middle.
If you have a DAW program that easily allows you to invert the polarity ("phase") of one channel, do that and it'll sound like it's supposed to sound. You can accomplish the same thing by reversing the wires going to one speaker.
Hopefully only the sample file on the web site is screwed up like this and not the CD. Incidentally, in 1952, the recording was almost certainly mono (Nagra didn't introduce their first stereo recorder, the IV-S, until 1971).
In this sample, the two channels aren't identical, which they should be in a proper stereo representation of a mono program. This could be a result of the data reduction process used to make a small web-friendly sample (MP3 compression) or it could be the result of the the original full track recording being played with a stereo playback head and re-recorded as two channels. This is suggested by the bit of "swishiness" you can hear, resulting from the worn tape wobbling a bit as it passes the playback head, causing the same sound to be reproduced at a very slightly different time in the two channels of the playback head.
By today's standards, this would be considered pretty sloppy transcription work, but it may have been the best the low budget company could do when the record was originally issued. Smithsonian Folkways may not have had the original recording when they reissued theit on CD. There's really no excuse, however, for the inverted channel on the web sample other than that someone just wasn't paying attention. It's easy to fix.--MR
Q: Hi there: Just a quick question for Paul Stamler. I’ve modified my cable for my Shure SM58 (“The Taming of the Shure”, May 2006) to better match the impedance on my TASCAM FW-1082 as written in Recording, and it’s working great.
My concern is that the FW-1082 has only one phantom power switch for all XLR inputs. I’ve used the SM58 before with other phantom powered mics with no problems, but haven’t tried it yet since modifying the cable. Is it safe to use the SM58 still with my other phantom powered mics?
Thanks for the great info.
Franz Peters
Ontario, Canada
A: Hi Franz:Yep, it'll work fine, no problem. The phantom voltage is the same on both pins 2 & 3, and with no difference in voltage, no current will flow through the added resistor.
Enjoy! Peace—PJS
Enjoy! Peace—PJS
Have a question that you would like our staff to answer? Then email us at:
talkback@recordingmag.com
