4. From Stereo To Surround

The CMSS 3D Surround system has existed for some time, but was designed for "upmixing" a stereo source to four, five, six, or seven channels (plus a subwoofer). Admittedly, it hasn't been really convincing until now, at least not with a lot of sound sources that require very precise reproduction of the soundstage. That was also the case with all systems of the same type (Dolby Prologic II, Neo:6), since they're all based on the same principles: extraction of the different channels from the sum and the difference between the two stereo channels. This is the matrix-decoding principle that's been widely used in many products for some time, but whose successive improvements have obviously never eliminated the fundamental defects.

Typical diagram of matrix decoders in use for several years

While this type of decoder gets good results with encoded sources, i.e. the 4-channel MP Matrix used in cinemas and known under the names Dolby Stereo or Dolby Surround, that's not true of classic stereo sources. With X-Fi, Creative - thanks to the power of its new processor - uses a very different and obviously more complex approach, through processing that separates direct sound from ambience sound by processing the signal in the frequency domain. That makes it possible to identify and extract ambient sound, such as the reverberation of a hall or applause, in order to use it for the surround channels, while extracting the center channel without altering the other components of the stereo image.

The principle used by X-Fi for upmixing (here two to five) makes use of a transformation from the time domain to the frequency domain, and then back in the other direction.

Creative's claim, aside from the obvious advantage of "surround" output, is that the stereo listening area is extended via the presence of a center channel that's perfectly stable and well defined. It's the solution to a problem that's not often discussed, yet is essential once you go beyond the egocentric model of an audiophile sitting alone in front of his speakers.

From stereo to surround... The X-Fi version of CMSS 3D promises a high-quality listening experience.

Aside from reproduction by a multichannel loudspeaker system, CMSS 3D offers a Virtual mode that claims to be able to create the illusion of multichannel sound using a simple stereo system. This of course has been a long-standing dream of many, and there have been numerous attempts to make it a reality - but none of them have been very convincing, despite the arguments of their promoters. We haven't had a chance to really experience this mode yet, but the techniques developed by Creative seem to have moved in a promising direction compared to what we've heard up to now.

Virtual CMSS 3D uses HRTF (Head Related Transfer Filtering) to simulate the side and rear speakers. Added to that is a stereo crosstalk canceller to improve the illusion of the virtual loudspeakers, without changing the acoustic coloration of the original recording. The user can calibrate the system to achieve optimum rendering using the THX configuration panel.