Take a Listen...

[richthebeard]

Hi there,

I was wondering whether you guys could take a listen to some recordings I made recently of a band I play in and help me to improve them.

You can find the recordings at:

http://www.jazzupmyevent.co.uk/sample1.mp3

http://www.jazzupmyevent.co.uk/sample2.mp3

To my ears, they seem very muffled and clipped (or compressed?) - could this be due to the microphone, or maybe the settings on the recorder? I did try it with AGC set to "For Music" but this seemed to cause a lot more clipping as the band is very loud - does that observation seem reasonable?

Could you suggest settings to use in future for clearer recordings? Due to the nature of the sessions I don't have time to adjust settings for each track, so suggestions for a "best compromise" solution would be gratefully received. Obviously the quieter tracks will turn out quieter on the recordings, but I don't mind this as long as they're clear!

Do I have to bite the bullet and buy a new mic?

The process used was:

1) Microphone Sony ECM-DS70P Stereo "T" Mic

2) Recorder MZ-RH1

• Rec Mode: Hi-SP
• Mic Sens: Low
• Mic AGC: Standard
• Rec Level: Auto (AGC)

3) Import and convert to WAV with SonicStage 4.2.02.11020

4) Convert to 128Kbps 44.1kHz stereo mp3*

*I realise this compression isn't ideal, but the lack of clarity and clipping occur before compression

Edited March 12, 2007 by richthebeard

[A440]

Mostly I think you need a better mic. The DS70P doesn't have much bass response--it only goes down to 100 Hz--and it's got more self-noise than some other comparably priced mics. A mic with more bottom end will stop the music from sounding so tinny.

Two mics you can separate by six inches like your ears will also open up your stereo image a bit. I do a lot of recording with the very basic Sound Professionals BMC-2, mostly because I need to be stealthy, but to me they sound a lot richer.

Try this recording:

http://forums.minidisc.org/index.php?autom...si&img=3080

Since you're recording your own band, you don't have to be so stealthy, and could get bigger and better-sounding mics. Look at omnis, look for frequency response of 20-20,000 Hz, and choose low sensitivity if there's a choice since the music is loud. Don't get anything more sensitive than -30 dB (that is, not -20 dB--since it's negative numbers, a lower quantity is higher sensitivity.) Look through the microphone choices here:

http://www.soundprofessionals.com

Or if you have a Guitar Center, etc., in town, you might ask them about full-range recording mics. Don't break the bank--I'm sure a $1000 mic will sound wonderful, but you're recording in a live setting, not a studio and don't have to get crazy about it.

The reason the DS70P doesn't have that deeper bass is because deeper bass can freak out the preamp in the MD unit. So you need one more gadget: a battery module, like this one:

http://www.microphonemadness.com/products/mmcbmminminc.htm

which supplies extra power to the mic. I record live music Mic-->Battery Module-->Line-in at a level of 20/30, more or less depending on the music.

Terminology-wise, what you're getting is not clipping, which creates an unlistenable distortion. It is compressing, in the audio rather than the computer sense--limiting the dynamic range-- (shrinking a file), because every so often you can hear the AGC adjusting the volume.

You're better off with Manual than with AGC. Your music never gets so quiet that you need AGC to rescue a sudden quiet passage, and the minidisc can capture a very wide dynamic range.

You do any Hermeto Pascoal or Milton Nascimento tunes?

[greenmachine]

I think the problems you hear in the recordings are dynamic compression artifacts caused by the use of the automatic gain control (AGC). The "for loud music" (instead of the "standard") setting may perform a bit better for music, but for the most accurate results you may want to use manual levels. The "best compromise" setting depends to a great part on the sensitivity of the microphone, the mic position, the loudness of the signal to be recorded and the room acoustics (among other factors). You'll need to experiment a bit to learn to estimate a good setting with your current equipment. For similar gigs like the ones shown in the samples, start with a setting of approx. 15 or 16 (of 30) and work your way upwards or downwards, depending on if the recording is becoming too quiet or there is clipping. It's usually better to leave some headroom (peak level below maximum possible level) than to risk clipping. It's usually better to set the level before the session and leave it there instead of re-adjusting during a recording.

[boojum]

I think the problems you hear in the recordings are dynamic compression artifacts caused by the use of the automatic gain control (AGC). The "for loud music" (instead of the "standard") setting may perform a bit better for music, but for the most accurate results you may want to use manual levels. The "best compromise" setting depends to a great part on the sensitivity of the microphone, the mic position, the loudness of the signal to be recorded and the room acoustics (among other factors). You'll need to experiment a bit to learn to estimate a good setting with your current equipment. For similar gigs like the ones shown in the samples, start with a setting of approx. 15 or 16 (of 30) and work your way upwards or downwards, depending on if the recording is becoming too quiet or there is clipping. It's usually better to leave some headroom (peak level below maximum possible level) than to risk clipping. It's usually better to set the level before the session and leave it there instead of re-adjusting during a recording.

I think that GM is right. Drop all AGC, use PCM rather than SP, and record at a low level with plenty of headroom. The low level can be corrected on your computer after you have uploaded it. Or, that is how I do it. I have not been able to D/L your samples yet but that is my cursory advice.

And A440 is right about the mic, too.

Cheers

Edited March 12, 2007 by boojum

[richthebeard]

Thanks for the quick, comprehensive and helpful responses guys.

@A440:

The group is a Pat Metheny tribute group (the samples being of Have You Heard and First Circle). Sadly I've never heard of Hermeto Pascoal or Milton Nascimento, but I'll look them up! That recording sample you sent was fantastic - it sounds like an amazing venue and that must certainly help add to the recording quality.

I think i'll definitely look at getting a new mic. Could you make any more specific recommendation based on the following?

• $200 is my top limit for the mic (I'm in the UK, so it'll actually be about £100), but if i could get away with spending less that would be great
• Size isn't too much of an issue, but I really need it to be one unit rather than left and right on separate leads
• I don't mind it being a studio mic etc, as long as it sounds good!
• You'll probably hate me for this, but it would be great if it could cope with recordings as diverse as being sat on top of my bass amp and 200ft away at the back of the venue
• Something freestanding (or mounted on a small stand) is a must
• I play mostly jazz, so the recordings I attached are about as loud as it gets (ok, maybe this is: http://www.jazzupmyevent.co.uk/sample3.mp3 )

Greenmachine's battery box seems so neat (and I'm an electronics student) that i'd probably knock one of those up if needed. Could you advise whether I'd need it for any mic you recommend? Do you think i'd be better off going in through the line input?

@greenmachine

Thanks for the reply. In terms of the settings are you talking a mic-in level or line? I am definitely fine with leaving headroom as I can always bring the level up later. Listening closer, I can really hear the AGC adjusting the level, making for a slightly uncomfortable listening experience (the loud bits almost seem to get quieter). Thanks for the battery box design! If I end up making one I might try making a board and surface mounting the components to see how small I can get it.

@boojum:

Yet again, thanks for the reply. How much difference will using PCM make? I only ask because I would like to avoid it unless it will make a significant difference. I tend to like to keep my collection of recordings on the MDs themselves and PCM would use up a lot more discs and battery, especially for 5-hour rehearsals!

@All:

What difference does the "Mic Sens" setting of the RH1 make? I have it set low, but is that unnecessarily attenuating my input, especially if i manually set the record level lower?

Mic recommendations (based on the above) gratefully received.

[A440]

Ah, you're in Europe. I suggest that if you want to purchase mics, get in touch with greenmachine himself, who builds mics that all the Europeans who have used them love.

Maybe he can build you a one-point mic. Or you could take one of his pairs and clip them side by side--you can get a little desktop mic stand, a tripod or something, on Ebay (I just got one for $7). A good pair of omnis should be flexible enough for both situations you describe, particularly if you have a battery box. GM's battery box will do the same thing as the battery module I use, and if you're handy, you should definitely build it.

You can also use GM's instructions for DIY mics, since you're an electronics student. Research some capsules--if you DIY you can probably get great capsules for your budget. Sennheiser and Audio Technica are widely used capsules for better microphones.

PCM will be somewhat better, but at the cost of disc space. You jmight get in the habit of uploading after you record and reusing the discs--but sooner or later, of course, you'll either need to burn the recordings or get a humongous hard drive. Do a comparison and see if you think the sound quality upgrade is worth it.

Low Sens is a good idea for anything but speech.

Unlike boojum and GM, I think you can be a little more daring in your levels so you do less amplification later. The ideal (which they would agree with) is to peak just above the middle of the scale. If you record at a lower level, amplifying the sound will also amplify noise, so the ideal is to get close to the peak but not over it. I find that with line-in, it's harder to overload. Since your band is probably playing regular gigs in a certain environment, you should be able to find settings that work reliably for you with a little experimentation.

Metheny's influences on those tracks are totally the Brazilians I mentioned (and guaranteed he would name-check them). Listen and see. You also won't regret picking up "Native Dancer," Wayne Shorter's album of Nascimento tunes.

[greenmachine]

When I was suggesting a level setting, I was assuming a similar scenario. i.e. recording the same band in the same location from the same distance via the mic-in at low sensitivity setting (you want to use this setting most of the time, particularly for anything that is louder than an average conversation). As I understand it, the AGC sets the level approx. comparable to a manual level setting of 20 and automatically corrects downwards where necessary to avoid clipping, at the cost of destroying the natural dynamics (sharp attacks will be softened). Since the AGC seemed to be active only every now and then, I thought a relatively small correction of the level setting downwards from 20 might be sufficient (one step equals approx. 2dB).

The Sony mic you're currently using is a good beginner mic . It has the advantage of a relatively low sensitivity (at the cost of weak bass response and below average S/N), which is useful when recording loud sounds via the mic-in. With a higher sensitive mic, the preamp may overload no matter of the level setting and you'd be better off recording via a battery box through line-in (a more complex setup, can cause additional annoyances like and automatic track marking). Single point mics also have the advantage of being relatively easy to set up. You may want to stick with this mic for a while as you're learning the basics of live recording and upgrade anytime you feel confident.