SST's MelodyWing: Does It Make Wireless Audio Sing?

Last night, I told you about a wireless A/V streaming transmitter-and-receiver set. In previous days, I’ve discussed wireless audio streaming products from Amphony, and those based on silicon from Avnera. This time, SST’s MelodyWing SP wireless audio chipset goes under the microscope (or is that over headphones?).

A qualifier before continuing; whereas in the past I tested full-production systems, this time I’m looking at a silicon supplier’s evaluation kit. Considering the functional variance I’ve already seen between systems based on the same silicon supplier, Avnera (see my positive results on the Rocketfish system re-test, and my subsequent poorer results on the Acoustic Research headphones), I’d encourage you to not draw any definitive conclusions about how MelodyWing SP will perform in one of SST’s customer’s systems.

The kit consists of a transmitter module:

and two power amp-inclusive receiver modules:

Even though each receiver module includes both right and left channel outputs, for some reason it’s necessary to drive each speaker with an individual module (trust me, I tried connecting both speakers to the same module: it didn’t work). The MelodyWing SP transmitter module only accepts line level inputs, so I wasn’t able to replace my existing Rocketfish transmitter (driven by the speaker outputs of my JVC A/V receiver) with it. However, to clarify, I did power off the Rocketfish transmitter and receiver before testing the SST-based setup.

Instead, I connected the MelodyWing SP transmitter to the line level outputs of an Apple Airport Express sitting right next to it, thereby not using the MelodyWing SP transmitter’s subwoofer signal input. Again to clarify, my MacBook wasn’t connected to the Airport Express’s integrated 802.11g access point. But in addition to the fact that the Airport Express was the only piece of gear I had handy with line-level analog audio outputs (my PS3 and Xbox 360 both connect to the JVC A/V receiver via S/PDIF digital audio connections), I thought that the Airport Express’s close proximity to the wireless audio transmitter would present a particularly interesting spectral challenge to the MelodyWing SP system.

To wit, the feature I found particularly intriguing about MelodyWing SP is consisely captured in one line of the product brief verbage:

MelodyWing SP Smart Channel automatically selects the un-congested ISM bands (2.4, 4.9, 5.1, 5.2, 5.4, or 5.8 GHz) to avoid interference with other wireless equipment.

Whereas Avnera’s chipset is 2.4 Ghz-based and Amphony’s system employs the 5.8 GHz band, SST’s ICs reputedly harness all unoccupied ISM spectrum between 2.4 Ghz and 5.8 GHz. How’d MelodyWing SP perform, and did it degrade my MacBook’s 802.11g link to my router in the process? Let’s find out.

My router, as I described last night, is about 15 feet away from my home theater stack, containing the Airport Express (and therefore, for purposes of this test, the MelodyWing SP transmitter). Six feet beyond the home theater stack is my cloth slipcover-clad leather couch, on which my MacBook and I resided. And behind the couch were the two MelodyWing SP receiver modules. I streamed audio to my router from my Dell Inspiron 700m (located 15 feet and two walls beyond the router) over 802.11g via Airfoil; from the router, the audio got to the Airport Express via a HomePlug AV spur.

Functionally, the wireless audio setup worked great. Since this was an eval kit, not a polished retail product, volume control wasn’t offered; the ‘mute’ button mentioned in the SST documentation also wasn’t present. But the transmitter and receivers immediately paired up, and I noticed no audio drop-outs through a lengthy listening session that also included plenty of 802.11g-based LAN and WAN activity via the MacBook.

What about the 802.11g connection between my MacBook, positioned in the ‘line of fire’ in-between the MelodyWing SP transmitter and receiver modules? The news here isn’t quite so stellar, although it isn’t too bad. Below, you’ll find my standard ping-to-Yahoo and BroadbandReports bandwidth tests; I ran the latter three sequential times to verify the results:

And, as a reminder, below are the ‘control’ ping and bandwidth results with no contending spectral interference:

The degradation wasn’t significant by any means; much less than with some other wireless multimedia links I’ve evaluated, and upstream bandwidth curiously appeared to be unaffected. But, as the ping test shows, degradation does exist. I have many unanswered questions about MelodyWing SP. How power-thrifty is the chipset, for example; could the transmitter and/or receiver be battery-powered in a portable application? What audio frequency range does MelodyWing SP resolve? How much of the 2.4-to-5.8 ISM band suite is required to transport each audio channel, and how does this vary with audio sample size and sampling rate? And how does MelodyWing SP perform in more challenging enviroments; at longer transmitter-to-receiver distances, for example, or with intermediary walls and other attenuators? Stay tune for any feedback that SST provides after the company reviews my published results.

p.s…speaking of feedback, I got some interesting news from Avnera’s PR agency representative earlier today. Company representatives were understandably concerned when they read my disappointing testing results on the Avnera-based Acoustic Research wireless headset, particularly in contrast to my earlier Rocketfish testing. Here’s what I was sent today:

We think we’ve figured out the cause of the problem you experienced with the Acoustic Research headphones. (It appears to be just a problem with those headphones and not any other Avnera products – we’ll explain to you why and the fix AR is going to implement to rectify the problem going forward.)

When I asked if AR was planning on pulling existing product off store shelves, here’s what came back at me:

I’m meeting with AR next week and will find out what they’re planning to do with the headphones and let you know. From what I understand, they just did a small run of the first lot and will be producing future batches of headphones with the proper engineering specs, which is where it went wrong last time around. I’ll keep you posted on that.

Stay tuned for more information as I receive it, including a possible AR headphone re-test…