CanJam at RMAF 2011: Sennheiser

These guys are my heros. Sennheiser reliably delivers very good sound over a wide range of headphones, and they do it the right way: lots of science, and lots of listening. That's not to say they don't miss the mark sometimes --- they do --- but my ears and measurements tell me they miss it by quite a bit less than any other major maker. On quick listen my ears tell me they nailed it again with their latest wireless headphone.

I'll quote their press release:

In the RS 220, the transmitter sends the audio signals to the headphones without compression via a stable 2.4 GHz connection using the so-called Direct Sequence Spread Spectrum (DSSS) technique. “This is a frequency spreading process in which the output signal is spread to a width of 22 MHz by means of a specified bit sequence. If interference occurs at one point within this frequency range, the output signal experiences no interference as the data is transmitted with redundancy, in other words several times. As a result, only a minimal and inaudible part of the whole information is affected and the overall sound quality is not influenced,” explained Axel Grell, Senior Acoustical Engineer at Sennheiser."

I'm quite surprised they released a new one as I felt their previous top-of-the-line RS 180 and RS 170 were the best I've heard. In fact their entire line of wireless products are quite good. If you're looking for headphones to do some late night movie watching, peruse Sennheiser's offering for features that suit, then pull the trigger. If you need some serious music listening without the wire, the RS 220 would be a good place to start looking.

CutTheCord's picture

After previously owning both the RS170 and RS180, I was left very disappointed—the "Kleer" wireless technology did not sound lossless at all, despite their claims, and they were noticeably delayed behind my television.

I desperately want to get rid of the long headphone cable that I have trailing across my room though.
Is there any chance you will be able to get a pair of these to assess their performance?

Your reviews have been very useful to me, and I've been happy with my P5 purchase based on your recommendation. While I like how they sound though, I do find that they start to get very uncomfortable after a couple of hours wear—the memory foam is too soft and stops giving any real cushioning past that point once it's fully compressed. I'm surprised you didn't experience that.

mikeaj's picture

That description of DSSS makes absolutely no sense. If there's narrowband interference in part of the frequency range of your DSSS signal, then you're still going to get interference on top of the received signal (it just won't be as harmful as if you don't use spread spectrum and have your entire channel hit by interference). Anyway, spread spectrum is no big deal, being used by a lot of devices and systems. It's not like a breakthrough.

I'd really hope that the wireless link is using a forward error correction code for redundancy--and that can be done with or without spread spectrum. And if it's used, even if bits get flipped over the air, you can easily have ZERO data lost, so I'm not sure why they mention that the sound quality will be affected inaudibly (as opposed to not at all).

Also, 2.4 GHz is right in the ISM band with lots of cordless phones, most Wi-Fi, Bluetooth, and many miscellaneous wireless consumer devices. I guess that's to be expected.

In addition, the 22 MHz channel bandwidth after spreading isn't really that wide, since a Wi-Fi channel in normal mode is 20 MHz. So regardless of DSSS, you could very easily still get the whole band hit by interference? Let's hope they don't both use the same band...

I'm going to guess that communications is not Grell's forte, which is very understandable.

TyroneShoes's picture

I'm going to guess that reading and understanding are not mikeaj's forte, which is also very understandable.

My interpretation of Grell's comments were that it is due to the DSSS technology that interference is less likely to affect the signal, which is probably exactly why they use it. This does make sense, and for the very reasons you state, among others.

It is also pretty clear that forward error correction is used, as it is stated very clearly that redundant data is sent, not to mention that this is a ubiquitous primary way of preventing interference artifacts.

And it is not really a puzzle what "inaudible" means, which is that listeners will not hear a difference between a signal that is interfered with (under a particular but high threshold, of course) and a signal that is not interfered with. If a change to an audio signal is "inaudible", then that is the very definition of "not at all", from the point of view of perception, which is the only POV that matters.

FEC and spread spectrum are the reasons why; SS means all critical data is not sent at the same frequency and so the entire signal is not vulnerable to narrow-band impulse noise (which is what comprises common 2.4 GHz interference) at any point in time, just a small part of it, and FEC adds to the robustness of the signal so that any bits that are knocked out can be replaced by redundant identical bits.

This is common in digital delivery; ATSC uses techniques very much like this to broadcast OTA television. That is also not a "breakthrough", and I did not sense that this was the implication, just that this is the technology required and used to overcome such interference, which is likely very effective in most cases, and certainly probably significantly more resistant to interference than not using such techniques.

Interference degrading a digital signal is not like interference degrading an analog signal, which is one of the big reasons digital is commonly used. With analog, the information is impressed directly onto the carrier using the same attributes of the carrier to modulate that carrier, and if the carrier is degraded by noise or interference, the information will be degraded as well, because the interference and noise "modulates" the carrier in the same way the desired information modulates the carrier. Change the modulation (by interference and noise) and the information is changed (degraded) as well.

In digital delivery the information is not directly impressed on the carrier with the same attributes and vulnerabilities of the carrier, but with different attributes not affected by carrier degradation (the carrier and the information are fundamentally different from one another), so noise and interference that degrades the carrier does not necessarily have any affect on the information itself.

While the common historical methods of modulating an analog signal onto a carrier are very similar to what noise and interference can do to that carrier and its modulated information, digital modulation techniques are very different from what can happen to interfere with a carrier during transport through a hostile environment, and can't be mimicked by common events in nature.

For instance, lightning anywhere even close to the transmission path can manifest as static in an analog AM radio broadcast. The plasma RF generated by lightning can raise or lower the instantaneous aggregate carrier level effectively AM modulating the carrier on top of the desired AM modulation. Natural background thermal noise can do the same thing. One results in static, the other results in raising the noise floor audibly. Neither happens with digital modulation techniques, which are primarily mathematical in nature and quite removed from what can occur naturally within the transport environment.

A common problem with digital delivery can be loss of some bits of the information. But proper error correction can replace those bits as if they were never missing, which is why even degraded digital delivery can be fully losslessly identical (inaudibly different) from the equivalent signal with no degradation.

This is just one more example of taking advantage of the cliff effect; as distance increases or the signal to interference ratio increases, correspondingly more degradation happens to the carrier and more and more bits become corrupted (unreadable) or even obliterated, but thanks to FEC the signal quality remains identical all the way to the cliff, where it then drops out altogether (the decoder decodes perfectly up to a point and then mutes when the bit corruption threshold is exceeded). SS just pushes the cliff that much further away by minimizing the effect of the interference. And if you remain on the transmitter side of the cliff, reception is as if there never were any carrier degradation or bit corruption at all, and the difference is "inaudible".

Tyll Hertsens's picture
Great comments, thanks guys. Yes, I will try to get an RS 220 for review.