Audio Summing

  • Thread starter Thread starter danny.guitar
  • Start date Start date
D

danny.guitar

Guest
When two or more sounds are summed (mixed) in mono, how does it work?

Is it:

New frequency = Freq 1 + Freq2 + Freq3 / NumberOfFrequencies (3)?

Do both the amplitude and frequency get divided like this? Or just the frequency?

A guy I know is making an audio app using DirectSound and he said when mixing two or more signals together the volume decreases. That's why I ask.
 
I bet it is a phase issue.

Think of it like this


you have a wave..... it has hi parts and low parts. When the high parts line up..... the wave gets taller.... (louder) however, when the wave doesnt line up, you get a high part and a low part, and they cancel each other out..... hence less volume when things are mixed.

Also, if he is "mixing down" and it gets quieter, it could be that his settings will automatically render at a certain level, which may be lower than his monitoring level??

Simon
 
Thanks for the replies.

The audio he is working with is mostly people talking. It's a live voice chat/streaming program.

Although from my experience with using VOIP/voice chat software, there has never been any noticeable phasing going on. At least not to the point where the volume seems to go down.

Thanks for the links, I'll pass them on to him. :cool:
 
danny.guitar said:
Although from my experience with using VOIP/voice chat software, there has never been any noticeable phasing going on. At least not to the point where the volume seems to go down.

it's not the same kind of phasing you're familiar with...i mean, it IS...but it's not BAD phasing.
when we usually talk about phasing we hear people say "the top and bottom snare are out of phase, flip one of them."
But did you also know that when you record a vocal or guitar double, those two instruments are phasing too? If they played the EXACT same way perfectly twice (pretty much impossible), then all that would happen is it would get louder. But since there are slight variations in tone, timing and amplitude...it gives us the distinct sound that two instruments are playing the same thing. This is GOOD phasing. We like this.

Phase or phasing of waves is actually a natural component of sound. It's happening all around you right now. An easy experiment is to just play a pure sine wave through your speakers. Now, this is just one sound. However, when you walk around your room that one sound is bouncing off all the walls and colliding with itself. So now it's not just a single sound, but several. Technically it's the same sound hitting itself at different points in the cycle of the wave. This we understand to be how reflections work. But while you're walking around your room you'll notice that in some places it's louder and in other places it's much softer...almost canceled. At the points where it is canceling, the waves that are colliding right there are subtracting from one another. In other words, one wave is in it's positive cycle and the other in it's negative cycle. Simple math tells us when you add a positive and a negative number together...the resulting number is closer to zero.
And in the parts of the room where it is louder, the waves are summing together in similar phase with each other. Again, this is a simple explanation. It gets more complex when you're talking complex (pun intended) waves. But the theory is still there. When summing sounds, it's the amplitude and timing (phase) of each one that matters more than the frequency itself.

with the math he gave you:
New frequency = Freq 1 + Freq2 + Freq3 / NumberOfFrequencies (3)?

are you saying that for example:
x= (1000Hz + 900Hz + 700Hz)/3 = 866.6 Hz?

this is simply not true. Adding multiple frequencies together doesn't just give you one frequency. You start achieving a complex wave. It's actually theorized that any sound can be created by an infinite number of sine waves.
If it were true that mixing two or more signals together makes the volume get softer...then mixing a 32 track band would make out for a very quiet CD wouldn't it ;) Generally as you add instruments together, the LOUDER it gets. This plays into why we harp so much on gain staging and knowing how to mix properly without clipping your stereo buss.

another great, simple example:
http://www.kettering.edu/~drussell/Demos/superposition/superposition.html

and this fun little web based program:
http://www.udel.edu/idsardi/sinewave/sinewave.html

More physics on sound (great site)
http://hyperphysics.phy-astr.gsu.edu/hbase/hframe.html

(side note) I'm not a physicist or a programmer ;)
 
Last edited:
Is he summing the waves himself or using channels of direct sound. Their mixer might be doing some back end volume adjusted if the latter. If he is just adding the two wave together himself he shouldn't hear any noticable drop in volume.

You don't get any new frequencies AFIIK when summing waves. I think it's theororetically impossible actually, but I'm not totally sure.
 
Back
Top