Seeing Sound Waves
Wednesday, January 21, 2009 at 08:00AM in Entertainment Technology, Live Sound Sucks by John Huntington I noticed something interesting during Garth Brooks' performance at the pre-inauguration concert on Sunday: I could actually see the sound waves. No, I wasn't hallucinating. To see the sound for yourself, check out this little video I made:
(If the embedding isn't working, try clicking here).
What does this have to do with sound systems or entertainment technology? Many people don't realize that sound, which travels only about 1,130 feet per second (depending on the temperature), actually travels way, way slower than light, which travels at roughly 186,000 miles per second. This accounts for why when you see a lightning flash, you generally hear the thunderclap seconds later.
And so, when we have a sound system where we want people to hear some distance away from the stage, sound system engineers have to take this into account. Speakers at a distance from the stage (like at the concert pictured in the video) must be intentionally delayed to get them to line up in time with the sound emanating from the stage. This happens because the electrons in the cables transfer the audio signal out to the delay speaker positions effectively, for our purposes, "instantly", while the sound waves, vibrating air molecules, can take many seconds to travel. And if the "delay" speakers, remote from the stage, are not aligned in time (even milliseconds matter), the audience will hear either destructive interference called "comb filtering", or perceive an echo. (Interestingly, from what I saw on screen, it appears that the large video displays on the mall were actually delayed to match the slowness of the sound.)
Keep all this in mind the next time you are at a large concert!
If you liked this entry, please check out the rest of my blog...
John Huntington
One of my favorite blogs , Urban Prankster, by Charlie Todd of Improv Everywhere, linked to this video! Before Mr. Todd's link we were at over 4,000 visitors to this entry already...
John Huntington
Wow, thanks to "saranowitz", who posted a link to this page on Reddit, this page has had over 6000 additional visitors and the video views, which had topped out at about 4000 visitors back in January, is now nearly 30,000!
View Printer Friendly Version
Reader Comments (10)
You can also see and hear a similar effect at large football games. The "O H I O" cheer at OSU football games in the Horseshoe (Ohio Stadium) is a great example of this - the north section yells O, then the west section yells H, south I, and east O, while throwing their hands up in the shape of the letter YMCA style. The north and south sections are at the ends of the field, so they are pretty much in time with each other. But the east and west sections are quite wide, so it looks and sounds like they are doing a wave. The northern end of the west section starts when they think the timing is right, which is a good half a second before the southern end of the west section. This makes perfect sense looking at the stadium on Google - those sections are about 600 feet wide. It's a neat effect when you're sitting in the middle of the east or west sections - the cheer starts to your left and ends to your right, and it pans right to left on the opposite side of the field. I've always wanted to sneak in a portable stereo recorder, but my HD-P1 doesn't quite fit under my jacket, and the VP88 isn't exactly inconspicuous!
Very cool way to show this "lag".
To geek you out further. Since heat rises, many sports arenas can have a 10degree or greater temperature difference from floor to ceiling during a show. Sound traveling through the top of the room to the balcony's gets to the last row before the sound on the floor.
This is not a constant. During sound check the room is much more balanced. But put 5,000 excited dancing people in the room. Now try and calculate your delays.
I know the effect well. When I mixed sound for outdoor rock festivals and they'd often times have the mixing booth a 200 - 300' from the stage.
I'd always have to slip the headphones on and mix that way when timing was critical. It was the only way you could time changes properly.
Most sound engineers didn't know this trick and they would always screw up the timing and couldn't figure out why.
The same engineers used to criticize my use of the headphones as they said you have to mix for the audience not for the headphones.
Well you can't mix for the audience if you have a 1/4 second time delay.
Stevie, you may just have become my hero... Over a year ago, an engineer on a show I do tech support for complained of fader "latency" or "lag" in the VCAs on his Heritage 2000. We all didn't deny that he was noticing something, but thought it was crazy, since, to the best of my knowlege, the only digital parts of the Heritage's automation are ancillary to the actual VCA control voltage, and that the VCAs are a fully analog signal path.
He eventually stopped complaining about it, so we kinda moved on and wrote it off as one of those "weird puzzles" to think about, but that we couldn't do much about at the moment. Although I'm very familiar with the concept, LOL, and also a big fan of using delay ON the headphone line to align solo'd channels with the stage sound when timing ISN'T an issue (once you've done it, you'll never want to go back!), it never occurred to me that this might be in play here. It's a fairly small room, but I'm pretty confident now that what he was hearing was actually the system delay used to align the speakers both as a system and to the actors onstage, and not any delay in the console. He thought the fader was lagging, but, since the sound to him was delayed, HE was actually lagging.
[smacks self in forehead]
Why didn't I realize this a year ago?
--A
Not a sound guy. Not even close. But many, many thanks for putting up something that demonstrates how this works so even an old stagehand like me can "get it". Wish the manuals were this easy to understand.
So not to burst any one's bubble but VCA's are not a fully analog system, they are in fact a voltage control amplifier and no sound actually passes through the VCA's. To continue, I disagree and believe that you should mix for the audience. Mixing through headphones doesn't give you an accurate image. If anything you should mix through a monitor placed next to you if you really don’t like mixing for the crowd.
Great presentation - it is so obvious now that you point it out!
I read long ago, in the 70s, that the system of delays for distant speakers was pioneered by a San Francisco sound company called Alembic Inc, who used to work a lot for the Grateful Dead. They're mostly known for building instruments, and created the very first guitars and basses with active electronics.
Actually Steven, VCAs are indeed fully analog devices, unless you have a dramatically different definition of the word than the rest of the world. And audio does indeed pass through VCAs, or they wouldn't function. What you meant, I believe, was that no audio passes through the FADERS on a console with VCAs. The signal passes through the VCA, which is controlled (that's the C) by the voltage (the V) that is regulated by the faders.
What we commonly refer to as VCAs on a console are, in fact, properly called VCA master faders. The VCAs themselves are ICs buried inside the console. And again, if no audio passed through them, they wouldn't work.
--Andy
could that be fixed by adding speakers at particular intervals. and delaying them