minimizing annoying acoustical beating

[lanterin]

Hi folks - I am new to the forums and have a question for you ... thank you in advance for your time.

I'm recording some synthesized sustained sawtooths with a low pass 4 pole filter tuned relatively on the low side, say 2 octaves above the fundamental ... so the fundamentals are fairly exposed. I am noticing that when I play chords whose notes cluster in the range right below middle C down to the c below, most especially the chord DGB right below middle C, I get some really annoying beating and difference tones created.

I have tried many things to minimize the annoying quality of the beating ... equalizing these tones down in the mix, raising the filter a little to add more distraction, compressing that particular range ... and the only thing that works is to route the middle tone to the opposite speaker, G in the above example, while leading D and B in the other speaker ... also turning down the G by about 12 dB works, in combination with the panning.

I am know that the beating and difference tones are just an acoustical fact, and I am assuming they are particularly noticeable in that range ... and that the more cover you have in the rest of the mix the less noticeable it becomes ... but I am wondering ... what if I wanted to solo some chords in that range ... is there anything I can do in addition to the panning, decibel lowering to reduce the annoying beating via mastering techniques or otherwise.

Thanks again for your time!

Lanterin

[finetuner]

Hi Lanterin,

It sounds a bit like intermodulation distortion to me, not sure.

It's an unpleasant interference of tones that can result in popping sounds.
Sometimes heard in recordings of violin- , flute- or soprano solos. So mostly higher pitched notes.
Once recorded, there's nothing you can do about such distortion.

In your case, where you seem to have total control over each seperate note (the way i understand it) you may try to experiment with phase/delay.
Your remark about channel routing or speaker placement hint towards that direction anyway.

With a very short delay, well - more like phase shift - on one of the tones, you can change the interference pattern. For better or worse.

That's all i can think of.
What kind of music is it anyway ? minimal / Zorn / Stockhausen ?

Peter.

[lanterin]

Hi Peter, and thanks for responding...

It's new age space music! ... along the lines of Michael Stearns or Steve Roach, but different of course.

So I was in a scientific mood yesterday, and I got out some organ albums and some space music albums in search of tones clustered in that range, with minimal upper harmonics ... and sure enough, when they were loud enough, and especially in the G chord formation, the beating/difference tones bothered my ear. Cover from some other sound source helped in each case, but I got out my FFT filter and removed the cover to try and maximize the annoyance, and was able to get it in every case ... so it's not just me, lol!

What it is is the speakers I am using. I test things out on my crappy computer speakers, because I don't yet have an easy connection to my good stereo speakers ... I can plug them in, and do so when I am ready to really make something, but it's a pain. These computer speakers must emphasize that range in particular, and since they are very close together, they max out any beating that you hear, even between the speakers. When I played those particular moments on my good stereo, what you got was an intensity of sound, but it didn't translate into discomfort for the ears.

However, I did find some things which covered the beating well on the crap speakers ... lower tones, in the bass range. Probably that is where the difference tones are occurring, any real notes down there lower your ability to hear it. Perhaps there are two difference tones that occur with the chord DGB (I'll have to figure it out some time), and they happen to be close enough together to "beat" themselves. Weird.

Anyhow, thanks again ... if anyone else has any ideas feel free

Lanterin

[Kiwiburger]

I think what you are doing is exposing the weakness of the Equal Temperament tuning. I think you should experiment with alternative tunings - most midi synths can be tweaked extensively.

A major influence on my musical education when I was growing up was Switched On Bach by Wendy Carlos (formerly Walter Carlos). W not only brought the Moog synth to public attention, but also was a major user of alternative tunings. It's this attention to detail that made the synth a true musical instrument, equal with string instruments (which by nature have infinately variable tuning, and are not locked into the problems of equal temperatment.

It's easy to think that playing Bach on a synth is matter of sequencing some notes - but if anyone tries to duplicate music with the majesty of Switched on Bach, it becomes obvious that this was a painstaking effort - a masterpiece. And all with a monophonic synth ... inspirational stuff.

[Bobro]

Originally Posted by Kiwiburger

I think what you are doing is exposing the weakness of the Equal Temperament tuning. I think you should experiment with alternative tunings - most midi synths can be tweaked extensively.

A major influence on my musical education when I was growing up was Switched On Bach by Wendy Carlos (formerly Walter Carlos). W not only brought the Moog synth to public attention, but also was a major user of alternative tunings. It's this attention to detail that made the synth a true musical instrument, equal with string instruments (which by nature have infinately variable tuning, and are not locked into the problems of equal temperatment.

It's easy to think that playing Bach on a synth is matter of sequencing some notes - but if anyone tries to duplicate music with the majesty of Switched on Bach, it becomes obvious that this was a painstaking effort - a masterpiece. And all with a monophonic synth ... inspirational stuff.

Yes this is it.

Mabe more synth musicians use alternate tunings than you might think at first- Robert Rich for example, and actually anybody who does big creamy pads by ear is working with intonation whether they're aware of it or not. And if you study the early history of synthesizers, you'll find that flexible and alternative intonation was a motivating factor in their development.

Using twelve-tone equal temperament for ambient, new-age, spacey, or for that matter most of the music in the world is actually kind of pointless, (personal opinion) as 12-tET evolved largely for different purposes- transposition, and a simple structure for modulating from any key center to another (historical fact). In the process, both calm harmonies, which don't beat, and countless wildly beating or otherwise biting intervals were lost (historical fact) 12-tET is kind of all things to everyone and not really just right for anyone. (personal opinion).

Try this- lay down a drone. Above that, play a fifth. Tweak the fifth upward just a hair, just 1 to 2 cents. It will start a very slow creaming thing, then it will do a "thing" when it's pure, you'll know it when you hear it. At a certain point just a hair above the 12-tET fifth, the third harmonic of the upper tone is at the same frequency as the second of the lower tone- that's the point where it happens. The proportion of the frequencies is 3:2.

Record that fifth on top of the drone. Then play a major third above the drone and slowly detune it, downward. The beating will slow down, then start to do the creamy thing, then at about 14 cents flat of the standard 12-tET major third, it will become still. When I demonstrate this for people, they say it's "ancient" or "ethnic" or "sweet" or "church-like" or "spiritual". The proportion of the frequencies at this point is 5:4, and so the fifth harmonic of the upper tone is at the same frequency as the fourth harmonic of the lower.

Now you have a "true" major triad. In the case of 3:2 and 5:4, many more (theoretically infinite) partials of the tones involved are coinciding in frequency than the ones I mentioned above, those are just the lowest ones. In direct digitial synthesis like Csound, where you can tune and render with deadly precision, you can actually make the upper tones "disappear" so the major triad sounds like the timbre of a single tone. It is not very exciting sounding. In real life, due to instrument imperfections or synthesizing to taste, the little imperfections really make it kind of "glow".

This is "Just Intonation". There are infinite variations and tunings. You can have
13 tones of the same size in an octave, or 43 of different sizes (like the composer Harry Partch, who is a cult hero among synthesists as well as acoustic musicians).
You can lose the octave altogether, or continually adjust the intervals so that it's not worth the bother to even count how many intervals you have in an octave.
Beating, not beating, both, etc.

Also, as you have already percieved I am sure, we're talking about harmonic tones. With metallic sounds- FM sounds for example, there's plenty of beating even in pure intervals like 5:4 or 6:5 because those harmonics are at things like 5.131 times the fundamental, not 5x. It can get quite complex, but strangely enough there seems to be a kind of "ideal" harmonic series wired into us which tolerates a great deal of inharmonic action in the partials, so purely tuned sounds with "impure" partials still "work". There has always been debate about where the borders actually lie- even the piano has slightly inharmonic partials.

If you get into this, beware- many people who are into tuning are, well, kind of nutty and can be pushy about their own interpretations, tastes, and terminologies. I'm probably nutty but I won't try to convert you to my personal tunings or whatever, LOL.

Mostly just trust your ears as you know already!

take care,

Cameron Bobro

[lanterin]

Dear Cameron,

Thanks for your lengthy and detailed reply. I'm sorry I did not respond sooner, but I got involved with stuff, and only recently was able to check the boards. I did some reasearch on what you said ... it does indeed work. The beating, say from a major triad tuned to equal temperament, is caused by the fact that the first and third, when combined together at equal volumes, create an "ghost overtone" of a fifth. This fifth is caused by the individual sine waves going in and out of phase, creating amplitude peaks and dips at approximately the frequency of that fifth. This fifth, however, is out of tune with the real fifth, and so beats against it. Very funky. Tune everything to just intonation however and nothing beats - well not that much anyhow - and very slowly. Very cool - and of course a string section would play more in just intonation than they would in equal temperament, especially if they were playing by themselves, as would a choir.

Also, since the beating that you hear when they are equal tempered is manifested as reocurring amplitude peaks/dips, you can also band limit compress it down.

Fun stuff - and thanks very much for your input ... and also Kiwiburger who suggested it.

Lanterin