I was bored enough to research this. These fellows are actually somewhat unusual in that they filed for a patent, which covers the claim of putting magnets inside connectors for some physics-free claim of performance improvement. What's great about patents is the applicant is pretty much required to pull down their pants and let us look at them naked. This is probably why most audiophile companies don't file for patents.
So, reading this gem of a patent:http://www.google.com/patents/US8272876
We find the following experiment:
The signal voltage for standard inter-component audio signals is typically between 0 V to 2 volts root mean square (V RMS), and generally with negligible or little current. The frequency range is typically 20 Hz to 20,000 Hz, with some systems requiring a frequency range up to 50,000 Hz. The test signal was a 1,000 Hz sine wave generated with a notebook computer using a TrueRTA real time analyzer (RTA) Tone Generator. A first test was run using a 1 kHz sine wave at −10 dB, or about 0.2 volts amplitude. Because a standard signal voltage level is generally about 1 volt, this test was determined to be run at too low a voltage level to provide meaningful results based on the capabilities of the test setup. Accordingly, a second test was run using a 1 kHz sine wave at a more realistic −3 dB, or about 1.0 volt amplitude, which better represents a typical inter-component voltage level. For completeness, however, the results of both tests are provided below.
OK, if they think a 7dB difference is meaningful in testing cable distortion, because at the lower signal level they can't show an improvement:
The results were then analyzed for total harmonic distortion (THD), total harmonic distortion+noise (THD+N), Intermodulation Distortion (IMD), and signal-to-noise ratio (SNR), using SpectraPLUS software. The two tables below provide a summary of the results.
−10 dB Amplitude Test
−10 dB THD THD + N IMD SNR
Sine Computer 0.03498% 0.03610% 0.2549% 68.849 dB
(Linear Phase Filter)
Sine DAC 0.01813% 0.15582% 0.2553% 56.148 dB
(Minimum Phase Filter)
Magnet Connector Cable
Sine Computer 0.03505% 0.03720% 0.2559% 68.589 dB
Sine DAC 0.02014% 0.15623% 0.2557% 56.124 dB
−3 dB Amplitude Test
−3 dB THD THD + N IMD SNR
Sine Computer 0.02181% 0.02199% 0.2554% 73.157 dB
(Linear Phase Filter)
Magnet Connector Cable
Sine Computer 0.01798% 0.01842% 0.2553% 74.694 dB
So if you listen to your music slightly quieter than we expect, our magnet cable actually makes distortion worse. Or, stated another way, these results are somewhat random, we don't understand why, we don't advance any plausible physics to explain this phenomenon. No, we just cherry-pick the good result and file for the patent.
What's my personal favorite though is the test rig--all well-controlled, audiophile-grade stuff here (this might explain why their THD figures in all tests were rather appallingly high):
The test signal was a 1,000 Hz sine wave generated with a notebook computer using a TrueRTA real time analyzer (RTA) Tone Generator. The computer generated test signal was in a digital format. The digital test signal was then converted to analog internally within the computer using a linear phase reconstruction filter. The digital-to analog converter (DAC) within the computer outputted an analog signal via a ⅛″ female plug. A ⅛″ to RCA adaptor was inserted into the computer analog output, and the test cables were alternatively inserted into the adaptor.
So they used *a notebook's built-in soundcard* into a bog-stock adaptor cable.
The analog signals were then recorded using a Tascam US-122L recording device.
MSRP: Not $4,500!!!
The Tascam US-122L accepted only either a ¼″ plug or a balanced/X-series, Latch, Rubber (XLR) input. Therefore, a Cardas RCA female-to-male XLR adaptor was use at the end of the analog test cables to connect to the Tascam US-122L.
Oooh, Cardas, now they are using the good stuff. No mention of how the unbalanced-->balanced conversion is done, or what impact that has on the Tascam's noise and distortion.
The analog test signals were recorded at a resolution of 24 bits at a sampling rate of 96 kHz.
Because you need every bit of 24 to record a signal with a rather horrible 75dB SNR.
Seriously guys, if your cable is so great you need to spend the dough to get a patent, rent a real lab for a weekend, maybe something with an Audio Precision analyzer
But then they have to go on:
Observation has shown that in preferred embodiments, magnetic poles preferably are aligned with the direction of signal flow within a cable, between pairs of cables, and into and out of a device.
Oh dear please turn off the stupid. *Alternating* current here, people!
I skimmed the rest; there were no more experiments, just a long list of unsubstantiated claims.