Your post has been here for a while now, unanswered, so I thought I'd give it a go.Originally Posted by rfreez
You are wanting to record Indian music trios using a minimalist and 'purist' approach, ideally with just two microphones. Given that information, I think the best person to answer your questions would be David Lewiston. He has had considerable experience recording music similar to that which you're recording, and he always uses two microphones. To the best of my knowledge, his rig consists of three different stereo miking options: an MS pair with cardioid M capsule, an XY pair of cardioids, and a pair of omnis. Considering the quality of his recordings, I would expect that the solution you seek lies with one of those rigs. David sometimes visits this forum, maybe he will chime in. Are you there, David?
In his absence, I'll throw in some suggestions and anecdotes from my own experiences (see further down). I am currently in Nepal, where I have spent considerable time recording numerous types of seasonal, ritual and regional music. In fact, I am currently resting my legs in the lakeside town of Pokhara after a recording trek to a hillside village so far off the tourist and backpacker trail that I was the second white person to ever visit the place (and hence a great source of curiosity, especially when trying to wash my pale, skinny and hairy body!). Made some great recordings of women singing accompanied by mahdal (two headed drum that is strapped above the knees and played with the hands while sitting), flute and handclapping...
Having a big room with a great natural reverb is a good start, especially if you intend on making purist recordings with no added reverb.Originally Posted by rfreez
Going off on a tangent here, I'm curious about your definition of 'purist'... Do you mean 'purist' as in simply using two microphones and recording direct-to-stereo, or do you mean 'purist' as in Chesky Records or, more appropriately to the music you are recording, Water Lily Acoustics?
With a 3rd world budget, I assume you are referring to simplicity, rather than having custom-made tube equipment and so on. But I thought I'd refer you to Water Lily Acoustics anyway - if you don't already know of them, their recordings may provide some good references for you. They record a lot of Indian music, and use only a Blumlein pair in a good sounding room...
That depends on your definition of 'thin'. If you mean lacking low frequncy body and power, it is important to remember that all directional microphones (basically, anything that isn't an omni) suffer from low frequency issues related to distance.Originally Posted by rfreez
Up close they suffer from proximity effect, which creates an exaggerated LF response. This effect is used in pop music to create a vocal sound that is bigger, fuller and more intimate than actually exists in real-life.
At a distance they suffer from a LF rolloff, which can make things sound thin, especially if you're expecting to get that up-close vocal sound of popular music.
At about 30cm away from a cardioid, the proximity effect and the LF rolloff tend to balance each other, and the LF response is flat. But that is rarely, if ever, the ideal position to capture the voice and the other instruments with a stereo microphone system. Usually the rig will need to be further from the voice than that...
When recording direct-to-stereo indoors, the foremost priority is to get the right balance and sense of ensemble between the musicians, along with a good direct/reverberant ratio. If, after achieving that, the vocals (and/or the other instruments, for that matter) sound thin, then your only option is to apply some corrective EQ, chosen to complement the mic's natural LF rolloff. Using EQ has always been a no-no for the purist, but with linear phase EQ I don't think there is any problem. I'm a very purist engineer at heart, but I am happy to use linear phase EQ because the benefits outweigh the degradations. I cannot say the same for standard EQ, however...
An omni M capsule, when mixed with the S capsule at a 1:1 ratio, decodes to the equivalent of two cardioids back-to-back. It will have a 360 degree sound field and the only LF rolloff and proximity effect will be due to the S capsule. If the musicians are situated in a triangle as shown in your illustration and all facing each other, this could be a good choice. It will offer a lot of freedom of movement to achieve the right balance and ensemble, but it will also pick up a lot of room sound.Originally Posted by rfreez
But, from the position of the stereo microphone rig shown in your illustration, I am assuming the musicians are all facing the front. In which case, an omni M capsule probably isn't a good choice.
For something like your illustrations, I'd be using an MS pair with cardioid M capsule, and moving the vocalist forward and the other instruments backwards. It looks like a simple set-up, really... I take your point about the musicians being old and set in their ways, but sometimes all you have to do is play them two comparitive recordings (one how they traditionally set-up, one how you'd prefer them to set-up) and they'll understand. Then, after a little bit of practice, they'll get used to it.
For this kind of direct-to-stereo recording, you are faced with many factors that you cannot control. The positioning of the musicians is one of the few factors that you *can* control, so don't worry too much if they're set in their ways. Start pushing them around a little. Whoever said "you can't teach an old dog new tricks" wasn't using a big enough stick.
This actually depends on which polar response you choose for the M capsule and where you place the musicians relative to it, because, when summed to mono, the M capsule is all you will hear...Originally Posted by rfreez
Depends on how far it is from the XY pair. According to your illustrations, the vocal will sound quite a bit more distant than the string instrument shown to the right (saranghi?), because the string instrument is considerably closer. Nothing you can do will change that, by the way, without either moving the sound sources or using a spot microphone. This is a problem we face from time to time with direct-to-stereo recording. We have two sounds, one is considerably louder than the other, so 'common sense' tells us to simply move it further away. In doing so, we might get a better balance in terms of volume, but now we have the problem that the louder instrument is too far away. In other words, a good volume balance but a poor depth balance.Originally Posted by rfreez
I was faced with a variation of this problem about four days ago while recording the women in the village mentioned at the start of this reply. The mahdal is quite a loud and dominating percussive instrument so I moved it behind the singers. Because the hut was small and not too reverberant (walls and floor made from pounded earth mixed with straw, I am guessing), it didn't sound too bad, and I thought I'd found a good solution. BUT, the problem was that the woman who played the mahdal was also leading the singing for the other women - and now she sounded too distant. If anything, she was supposed to be in front of the other voices, not behind them! In the end I settled on a three-part solution. Firstly, I arranged two of the singers so that they were sitting directly in front of the mahdal, with their backs absorbing some of its direct energy and blocking the direct paths from each end (where the skins are) to the microphone. Secondly, I sat the player on a small step so that her head was just above all the other singers, bringing her voice a bit more to the fore again. Thirdly, I experimented with the microphone height and position until it was in a modal null of the room, reducing the level of the mahdal's 'boom' even further but with no adverse affects on the voices. It worked well for the first hour or so. Then, because it was a village celebration rather than strictly a recording, we all got a bit intoxicated and things started to fall apart. Before I knew it, the mahdal player/lead singer was back on the floor in front of the other women, the mahdal was booming again and... well, what the heck, by this time I was dancing around the fire with the other women, having moved the Nagra and microphone rig back to make more room, and the recording didn't seem to matter any more. The sobering moment came when my guide/porter placed the AA cells from the camera in the coals to warm them up (they were going flat) and one exploded, showering hot coals across the floor and burning the side of my right foot quite badly.
(Hence, one of the reasons I am resting my legs here in Pokhara - I can hardly walk! Actually, I'm going through a very accident-prone phase at the moment - I twisted my left knee quite badly the second day in the village when I slipped going up a clay embankment to record some children in the village school. Then I burnt my right foot while dancing, and because of my twisted left knee I had to keep my weight on my right foot while trying to remove the hot coal which tended to just stick to the skin and burn away. The five hour descent down from the village to the 4WD a day or so later was hell! I had to keep my weight off my left knee, so my right foot was doing all the work - step down with the right foot, follow with the left, crabwalking the whole way. But I was trying to use one side of my right foot only, to avoid putting pressure on the burn. As a result, I developed a blister on the other side of my right foot, which continually took all of my downward momentum. Doh! So, I'm here in Pokhara keeping my feet up. But Punam, my beautiful Nepali wife-to-be, wants me to take her to the Hindu temple in the middle of Phewa Lake in about 30 minutes from now. How do we get there? Paddle boat...)
But I'd do it all again, of course.
Is a stereo pair enough? Crikey! You're only recording three instruments... ;-) If you add a spot microphone you'll have three microphones. Maybe you could close mic each instrument and throw your purist intentions out the window!Originally Posted by rfreez
But seriously, if you are able to move the musicians around a bit, you ought to be able to do a fine job with a single stereo pair. But this also assumes the vocalist is not playing an instrument.
As for stability in the stereo image and anchoring the voice in the centre: the closer the vocalist gets to the stereo pair, the more sensitive the image becomes to movements of the singer. Most singers in these types of music tend to be sitting cross-legged on the floor and, consequently, rock side-to-side and/or back-and-forth while singing. The closer you get with a stereo pair, the more movement you will get.
Another problem you might face is with singers who not only rock from side-to-side and back-and-forth, but who also tend to turn their head from side to side, rhythmically, while singing. This can be disastrous for any coincident technique, especially when up close and indoors, because the voice is continually going on- and off-axis and changing tone and reverberation accordingly. The solution for this problem is to use a pair of spaced microphones, probably omnis but cardioids would do. I have had this problem a few times while recording a Tibetan monk reciting mantras. Sitting cross-legged on the floor, looking down at the scriptures in his lap, he rocks from side to side and back to front, often turning his head left to right and back as he does it. Very difficult to capture with a coincident pair unless it is handheld so you can follow his mouth, in which case you find your elbow is etching a figure of eight in the air! (Um... no, a figure-of-eight response is not the solution to this problem.)
I have made quite a few recordings of Newari ritual music in the Kahmandu valley, and the big problem with this type of music is that the lead vocalist, who leads all the other voices (which may number anywhere from a handful to a few dozen) also plays harmonium and therefore is sitting cross-legged on the floor and always looking down and leaning over the little keyboard (they pump it by hand) while singing. I can get a reasonable result using an MS pair positioned down low and facing upward, but I suspect the real solution is a spot microphone on his voice (probably a lavalier because he always rocks back and forth), blended in with a stereo pair. And, for this music, given a choice I'd probably use ORTF or a spaced pair of omnis for the stereo pair because, with an ensemble so large, a sense of 'bigness' is more important than pinpoint imaging. This will also let me adhere to the 3:1 rule for the moving spot microphone, minimising comb filtering problems.
It might be wise to invest in a matched pair of microphones that offer switched polar response (omni, cardioid, bidirectional). Then you can do *any* stereo technique you need to. A matched pair of AKG 414s might be a good choice from this point of view...Originally Posted by rfreez
I hope this has been helpful.