Recording Tips for The Theremin

Coalport said: "I record the theremin the way I would record a vocal. I monitor the accompaniment tracks with headphones and I hear the theremin both live AND mixed with the tracks."

Ah. This is what confused me. You're not wearing headphones in the videos - so you've already recorded it (unless some of them record direct)? Then I assume you're sticking an ear-bud under your headphones if you use pitch preview when recording? That must be something to get used to.

Thanks

Rich

"Something like TOSLINK would be the best for recording work as it gets one DA / AD middleman out of the critical path and isolates like nobody's business." - Dewster

I absolutely agree that optical wins hands down on every count (pun not intended ;-).. And yes, if one has a digital engine, your route of having the D/A out of the theremin is the only sane future..

But with an analogue engine its less clear cut - you then need an A/D and Toslink driver in the theremin, pipe the (superb/isolating/immune) optical cable out either to a D/A or direct to some digital device.

For an analogue instrument, I dont see that one is getting "one DA / AD middleman out of the critical path" if one is going to an analogue mixer/amp or whatever - But certainly this would be true for a digital instrument.

When it comes to analogue stuff though, balanced signals are really the only Pro way to go IMO.. But it does come with a high price tag - and probably more fiddle-factor than most people want.

Fred

I don't ever wear large headphones (which cover the ears) when making videos - regardless of what instrument I happen to be playing. I always try to wear earbuds. When making a theremin video, one of the buds is a pitch preview for the theremin, and the other is the theremin mixed with the accompaniment - and I can hear my instrument through both ears because earbuds do not cup the ears and shut out extraneous sounds.

If I am singing, I usually try to avoid large microphones that get in the way of the image, and prefer to use a "headset" mike or a lavalier, and earbuds. 

Yeah, I know, it all seems like a big pain in the ass - AND IT IS! But once you're plugged in and you don't need to move around, it's all quite comfortable and natural. It is no more difficult than driving with a headset telephone.....and considerably less dangerous!  

rkram53 said "Playing well is only half the battle. If you can't record properly - it's all lost. I need more experience in this area an am open to anyone who wants to share their experiences and ideas."

Absolutely, Rich.  I would love to contribute to this thread when I get a moment free. I'm a bit swamped now. There is so much that goes into capturing a good performance.  People have really gotta realize that a recording can make or break a performance. Audio is just as important as video, and there's plenty of ways to mess up. 

Clara Rockmore's recordings are a great example of masterful audio engineering.   I liken it to a great photographer. The subject can be stellar, but the person taking the photo could totally waste the moment or capture the moment perfectly based on their ability to balance light/shadow/color, create interesting composition between foreground/background and with framing the shot.  Quality recording is much more important that the average person realizes because of the subtle effects it has on the listener experience.  

Anyways, no one here needs convincing. I'm looking forward to hearing everyone's point of view on this subject.

One of the difficulties with the recordings of Clara Rockmore and Nadia Reisenberg became apparent when Bob Sherman decided to take the theremin track, isolate it, and add new accompaniment (which he did with two of the selections on "CLARA ROCKMORE'S LOST THEREMIN ALBUM").

Because the original recorded performance was live, and the two instruments were in close proximity, you can hear piano on the theremin track, and theremin on the piano track. Consequently, on the two pieces that were re-mastered with new accompaniment, you can clearly hear Nadia playing away in the background!

I know that part of my problem is recording level vs. mastering level. I'm pretty sure I'm recording things incorrectly and I'm losing dynamic range and shaping in the process. Then I'm also typically raising levels when mastering things. 

Can some of you perhaps share what levels you like to see on tracks when recording (I realize there is not an absolute answer here) and do you master things up in the final mix? And if so what mastering levels are you looking for and what are your techniques there?

Thanks

Rich

"I know that part of my problem is recording level vs. mastering level. I'm pretty sure I'm recording things incorrectly and I'm losing dynamic range and shaping in the process. Then I'm also typically raising levels when mastering things. 

Can some of you perhaps share what levels you like to see on tracks when recording (I realize there is not an absolute answer here) and do you master things up in the final mix? And if so what mastering levels are you looking for and what are your techniques there?"  - rkram53

I have some experience but am not an expert my any means so grain of salt.

I assume you are recording things and manipulating them in a format that has excess bit depth (>16 bits)?  In Audition I use 32 bit float (though from my DSP filter research 32 bit fixed point might be better). 

Levels typically get raised at mastering, though a mix can exceed 0dB due to the adding of many sources, so how you look at gain here can be relative.  It's probably best to master the mix directly in the mixing software so as to avoid a resolution bottleneck.  I've used the iZotope Ozone plugin in Audition with some success here, though I can see why one person generally doesn't do recording, mixing, and mastering on a project as I start to lose my sonic bearings at some point.

For Mastering I use Wavelab. Cubase for recording. And of course plug-ins of all sorts (especially reverb).

64 bit computer - but I think 24 bit audio (have to check my options there).

But I think its more level issues than bit resolution.

When recording a theremin track in a DAW, what level range would you normally expect for a solo line?

Let's start there.

Mastering classical music to 0 dB can be tricky in terms of dynamic range depending on the instrumentation without any compression (for the way people listen to music these days). 

Thanks

Rich

I know the question was addressed to Peter - This "answer" is not directly to this question, but (perhaps like Dewsters reply) a related technical branch to the question..

IMHO.. (as in, my opinion only! ;-)

For an electronic musical instrument, particularly (but not exclusively) if recording direct (DI), life is a LOT easier than recording acoustic instruments -

And the theremin should be about the easiest electronic musical instrument to record.. You know what the maximum level will be simply by staying away from the volume loop and doing a quick pitch sweep, set the peak level to whatever you usually set max for recording (anything perhaps as close to 0db as you dare ) - doing this your recording should have the full available dynamic range (16 bit, 24 bit - whatever).

When one has recorded an instrument with maximum bit resolution without clipping (as in, the loudest 'sample' is just a fraction below digital saturation) one has a signal which can be attenuated into the mix with minimal loss of resolution..

Remember that for every 6db reduction into the mix, one is effectively losing 1 bit, so an 18db reduction on a 16 bit signal one ends up with a 13 bit reproduction of the waveform, but if you recorded at -6db, you actually end up with effectively 12 bits (I think this is why I prefer to take my tracks through an analogue mixer, and re-record the mix-down from this mixer - it always sounds better to my ears to do it this way).

I strongly suspect that the whole digital audio scene is one of the most elaborate technical frauds / cons of all time - for years we have been listening to mixes done on early digital desks where one is sometimes getting parts of the mix (the low level background / backing) which has actual resolution sometimes less than 9 bits! .. The reason why bit depth has been steadily increasing as technology advanced isn't because we can hear the difference between say 16 bits and 24 bits - we cant!... But we certainly can hear the difference between 8 bits and 16 bits, and anything below 12 bits can be heard by many people.. 24 bits gives those extra needed 8 bits so that the digital mangling process at mix-down doesn't produce too many signals with resolution below 12 bits...

But IMO, recording full-scale at 16 bits, and mixing down with a good analogue desk, and the results kick the sh*t out of any 24 bit all-digital DAW.

Fred.

I must just say this - the above is based perhaps on out-of-date understanding of digital audio.. I was only heavily involved with digital audio right at the start of the "revolution" - pre 1980.. At that time I was the senior engineer at a large studio and CC duplicating factory (FPA in Wimbledon, now defunct like almost every company I have ever worked for - even the NHS is now defunct, I must be to blame.. )-:

We got one of the first digital recording / duplicating setups in the UK - data was recorded onto video tape (Betamax) and re-recorded onto 1" analogue R-R on a large Studer - This tape was then placed into a loop bin where it was looped at high speed synchronously with a stack of "pancake" recorders - these "pancakes" were spools of cassette tape, which were then run into cassette winders, churning out hundreds of compact cassettes ever hour..

But even on such a system, the quality degradation on digitally sourced content was noticeable - We were all puzzled, and did a full evaluation - What we saw was horrific.. All the low-level content was degraded - and with CC noise levels, this "low level" wasn't really that "low" - really "low" was lost below the noise floor.

After examining the source material and consulting with the digital equipment manufacturer and record company who had commissioned it, the reasons I disclosed above were revealed - the digital masters had been digitally 'engineered' prior to us getting the tape - and the process completely screwed the content. The digital equipment was removed .... I know things have advanced hugely since then - but AFAIK the underlying fundamentals haven't changed.   

Hmm. I'm not recording (connected to Mackie Analog Mixer that has a Firewire digital interface to computer) anything close to 0 dB. 

Rich