Slashdot Mirror
Voices in Your Head
ceejayoz writes "MSNBC/Newsweek is running a story about a 'Hypersonic Sound System' that can 'can take an audio signal from virtually any source and convert it to an ultrasonic frequency that can be directed like a beam of light toward a target up to 100 yards away.' Sounds like something that advertisers will love - Minority Report just got a little closer." These guys (and the Audio Spotlight guys) have been hyping this technology for years with nothing much to show from it. But now, Newsweek promises, it's going to change the world as we hear it.
For some reason I feel this is a double post, but no-one here seems to have noticed, so I must be nuts.
Here's how it works in laymans terms. I am no science wizard, but this sounds good to me..
There are things called beat frequencies that occur when you have two frequencies present. For example, if you play 20Hz into one ear, and 25Hz into the other, your brain can be 'tricked' into thinking it is hearing 5Hz (the difference between the two frequencies).
This is all well and good, but 20Hz soundwaves don't travel too good. Ultrasonic frequencies do though. Remember those TV remote controls in the 70s and 80s that used ultrasonics? You could control your neighbor's TV. (See the start of Poltergeist 1 if you forget)
But how does sending 50Khz sound waves through the air help you hear anything? Ay, well there's the rub. The concept of beat frequencies is used once again.
If you send a 50Khz sound wave from one source and pinpoint it at a certain spot, and then send a 51Khz sound wave from another source to the same spot, anyone at the place where those two beams join up will hear a 1Khz sound, thanks to beat frequencies.
That's how you can pinpoint sounds to a single place. It just took a genius to get the connection between beat frequencies and ultrasonics to work this one out. I think it's cool.
mogorific carpentry experiments
You got a few things right, but a few things wrong.
First, beat frequencies are quite real - there is no "tricking" your brain into hearing something that isn't there - the signal is there.
Specifically, whenever you feed 2 signals f1 and f2 into a system with any non-linearities, you will get four frequencies out - the original f1 and f2, and two new frequencies (f2-f1) and (f1+f2). So, if I feed 51kHz and 50 kHz into a system, you will get 1 kHz, 50 kHz, 51 kHz, and 101 kHz. This is the same principle that all modern radio receivers work on - it is called heterodyning, and a modern radio is a superheterodyne receiver.
Now, in terms of propagation, low frequency sound does better than high frequencies - hence why thunder goes "CRACK" when it is close and "rummmblee" when it is far away - all the high frequencies have been attenuated by the air. Also, this is one of the reasons why all you hear of the assholes with the ThunderThump 3000 car stereos is the low frequencies - what little high frequencies they produce are attenuated by the car's body and the distance.
However, to get any directionality from a sound transducer, it must be large with respect to the sound frequency. The problem is that the bulk of the frequencies humans hear have very long wavelengths - it is possible to make a directional beam of 20 Hz sound, but you would need a speaker system the size of a football field. Somewhat impractical if you want them all over the place, pumping out your "BUY ME NOW" message.
However, by translating the frequencies up to 50 kHz, you reduce the wavelengths down to the point where the speaker needn't be much larger than a paperback to get the directional gain you want. So, you upconvert the signals to ultrasonic frequencies, and you use the fact that just hitting a surface acts as a nonlinear mixing element.
However, I have always wondered how much of the signal is going into the (f2-f1) component, and how much of the power is in the other three frequencies you cannot hear? What kind of damage will this energy do over the long run?
Not to mention that, with the steady erosion of the respect of the right of people to be left alone, how will this be abused? Will we see "reality TV" shows freaking people out? (say, by beaming "LOOK OUT! HE'S GOT A GUN" to one person in a crowd). Let alone the targeted advertisments ("Hey lard butt! Yeah, YOU. Get your fat ass into Fred's Gym, across the street. NOW!")
Personally, if this sort of thing gets deployed in public places, I want to start carrying one of the boxes you used to downmix bat echolocation down to audible, locate the speakers, and use my Leatherman on them... Or my Browning...
www.eFax.com are spammers
One of the provisions of the alliance was that Microsoft would publish Newsweek.com. The first reaction I heard from my coworkers was a concern for Newsweek's objectivity, or at least the appearance of objectivity. The Post's top brass assured everyone that the deal would have no impact on Newsweek's objectivity. I'm not saying it has; just putting a little insider kerosene on your fire.
One of the other provisions was that washingtonpost.com would feature Windows Media clips of Washington Post reporters on MSNBC. We were instructed to embed the video in our templates and also call a .js file hosted on MSNBC's site. The sole purpose of the .js file was to weed out non-Windows browsers. For instance, I visited the page (on our site) with a Mac and was redirected to an MSNBC page stating "Windows Media Player is not available for the Macintosh." I showed this to one of the top editors, who replied our users should know better than to buy a substandard computer. I then downloaded Windows Media Player for Mac from Microsoft and demonstrated that the video works perfectly if you take the .js reference out. The next morning the multimedia editor was waiting at my desk to get the details, and later began re-encoding the video files in Real format -- over Microsoft's loud objections.
Bottom line: Microsoft tried to use its deal with The Washington Post to prevent non-Windows users from viewing Post reporters on The Post's own site. I can only imagine what goes on at "newsweek.msnbc.com."