Comparative Controlled Evaluation
"Major Top of the Line" speakers vs EdSets "Thump Pump" Speakers
10 March 2007
ABSTRACT: Using standard acoustic engineering principles and studio quality test elements, perform a comparitive evaluation of speakers reproducing full-spectrum, logarithmic power distribution pseudo-random noise (aka "Pink") and typical musical content. Show the relative and the absolute frequency response of each speaker under identical conditions and signal input. Plot frequency-domain power spectrum and determine overall speaker fidelity to original noise and musical signals.
FREQUENCY
In the first picture below you see a graph. This graph represents power (defined as amps x voltage much like your bikes battery cranking power: 150A at 14.2V...or 'sound pressure per volt' in audio...) and this power is determined at audio frequencies from deep sub-sonic bass (like elephants communicate with) all the way up to sounds even a dog can't hear. In other words, if you give a speaker a signal with even power across all frequencies (pink noise does this trick for us), it's interesting to see the fidelity with which it reproduces the input signal, and this is where the term 'High Fidelity' comes from. Humans hear well between about 50-60 Hz up to 5Khz or so and your mileage may vary! These are the frequencies you see on the bottom of the graph.
POWER
The decibel scale you see on the left side is logarithmic (based on powers of ten), not linear like a ruler. Therefore, a small increase in decibels represents a great increase in intensity. For example, while 10 decibels is 10 times more intense that one decible, 20 decibels is 100 times more intense (10 X 10, rather than 10 + 10), 30 decibels is 1,000 times more intense (10 X 10 X 10), and so on. The sound intensity multiplies by 10 with every 10-decibel increase. The reason for such a scale is simply that the human ear is sensitive over such a wide range of acoustic energy that the numbers involved had to be compressed for convenience. In some ways, the decibel scale resembles the Richter scale for earthquakes. A small numerical increase represents a great increase in intensity. The ear can detect a very slight change in noise intensity. Even a small reduction in decibels then can make a difference.
So, the scale is in negative numbers because each horizontal level is 'lower than' the maximum of "0" or Zero. In other words NO attenuation is a lot louder than -150dB attenuation. So this is a 'loudness' measure.
AN IDEAL WORLD
In a perfect world, the speaker will reproduce the input signal perfectly, but we'll leave that for Jordi La Forge to figure out. Here on Earth, we strive for 'High Fidelity'.
TIME TO SEE AND LISTEN
The Green area below represents the 'Other' headsets performance, the Blue area, the EdSets. EdSets speakers produce between 10 and 20 decibels more sound in the same controlled experiment than those other speakers do. You can see that in the 'blue' cap over the left side of the graph. After about 5Khz where humans don't hear well at all, the performance of both speakers is about the same. On a noisy motorcycle, you want power in the area of low end and midrange, typically considered 50-1Khz and that's exactly where EdSets speakers excell. Tired of looking at pictures and reading? Scroll down...time to listen!
Turn on your best computer speakers, and click here (MP3 1Mb) or here (WAV 7Mb) to listen to this comparison
You are hearing 4 sections of sound, and they may repeat on your player. First, you hear "Their" headsets reproducing Pink Noise, then EdSets reproducing Pink noise. Even on a laptop speaker, you will notice a difference in these 2-second sounds. Then a minute of "Their" spkeakers reproducing an average piece of music and then EdSets reproducing the same piece of music. In both cases, you should be hearing a smoother tone and better bass response from the EdSets speakers. Your experience with BOTH speakers will be better in a closed helmet with the speaker directly on your ear, but the performance of the basic speaker and the governing physics won't change, you will still find there is a significant difference in performance, and the work here was to give all our customers some engineering background on WHY they like our sets so much!
HOW IT WAS DONE
The speakers were placed in an acoustic foam 'horn' to simulate the padding of a typical motorcycle helmet and a Shure Type 57 vocal microphone was aimed 1.5" from the center, on axis of the voice coil of each speaker to simulate the ear.. The audio test signal came from an Apple iPOD at 75% volume into a Behringer 16-Track Studio Mixer with no EQ applied and no Audio Effects enabled in order to take advantage of it's ultra-clean amplifiers. This signal was then passed directly into the speaker while the microphone was fed into a Behringer 12-Track mixer connedted to a hard-disk recorder. The MP3 file was generated by Sony Sound Forge software and the WAV file saved in raw format, uncompressed, no eq or effects, by the same program. If you have questions about the test, you may contact me at any time.
Enjoy!
Ed