TAS 194: Meridian 808.2 Reference Signature CD Player

I'm reminded of an article I read eons ago (it seems) by one of the lead Philips engineers who helped developed the CD (as well as being instrumental in vinyl pressing and playback technology). He mentioned something very akin to pre-ringing on vinyl, which he claimed was the source of many listeners' preference of that format over CD. Vibrations in the grooves before and after the groove being currently followed by the stylus were being created, and so information from all three grooves blended together and created a false sense of openness and warmth (he claimed).  Wish I could remember the magazine. What HiFi, maybe...?

• reply

Mr. Harley, First I want to thank you for your continued dedication to the improvement of digital media.  Now my question - Have you auditioned Tim Paravincini's EAR Acute CD player? 
Everything you wrote about the sonic brilliance of the Meridian player is there in spades.  In fact, it sounded EXACTLY like you were describing what I hear from the Acute - Hi-Res-like clarity & detail, long natural decay, huge 3D soundstage with lots of air but no 'etched' outlines... plus the most natural and realistic tonality & timbre I've heard fom any digital source, including SACD.  (And I DO love good SACDs!)
I've not yet heard the Meridian.  I will get to audition it extensively in October.  But several respected listeners I know have already heard it side-by-side with the EAR Acute, and they preferred the Acute 3-to-1 (and the 1 is a Meridian dealer).  The kicker is that the Acute costs about 1/3 as much as the 808.2.
If you haven't heard the EAR Acute, you owe yourself a listen.  And I hope you'll agree that you owe all of us a review of the best player I've ever heard.

• reply

The 22.05 kHz as the max recordable information in the CD standard has proven to be a limitation not very well understood at the formation of the standard.
Although less than 1% of us mortals can actually hear a static signal over 18 KHz, more of us can hear transient sound pressure effects that goes beyond 50 kHz.
The ideal, phase-linear, anti-aliasing filter actually will produce a symmetrical sin(x)/x function with the "ringing" frequency of the cutoff frequency. This is how it should be. The Dirac pulse ("Spike") contains all frequencies and cannot be represented with any system. Journalists in particular have a tendency to misunderstand sample theory. It is for instance not that the pre-ringing occurs "before the sound" (no exciting temporal relativistic effects here, sorry).... The actual sound is linearly delayed and in the case of the pulse, the peak occurs slightly later.
As some of us remember, the first CD players that came out did not have phase-correct filters and thus only had a post-ringing. Testing showed that the players that had phase correct filters (i.e. pre- and post-ringing) sounded better than those with only post-ringing! So, the pre-ringing does not necessarily mean worse sound. This is just a consequence of the math behind sample theory. Now, with complex musicand transient material alot remains to be researched.
The minimum-phase type filters are not phase correct and will delay various parts of the frequency spectra differently (and possible correct a speaker defect.... It is commonly known that most speakers have a high-frequency pre-ringing as the tweeter often is located closer than the bass).
What no one really knows is how to best try compensate for the 22.05 kHz limit. Wadia tried for many years in applying various filter techniques, but since you can't add any meaningful information (oversampling just spreads out the aliasing spectrum), I figure we will see companies like Meridian come out with various tricks that will exploit certain psycho-acoustic effects and tradeoffs of introducing beningn distorsion, but you can't create more information from nothing.  Actually, to remove the pre-ringing, just means adding phase distorsion!
Regardless, the phase alterations, frequency limitations and ringings introduced by speakers and the listening room are arguably orders of magnitude larger than what any CD player may or may not do. In fact, it would be interesting to compare the electrical pulse output of a bunch of CD players with the recorded acoustic output of the same pulse through a microphone compaing a number of speakers and listening rooms! Or ideally a music piece. I still wonder why this is not done on a more regular basis. Some HiFi mags in Europe and China have started to do that. We now have the compute power to sample a complete CD test disc and compare the input with the electric or acoustic output and assess some psycho-acoustic distorion metric, and not just perform the much less useful static frequency sweeps. This is where I believe TAS and others have a huge task in front of them. Or, will this remove the "magic"... and just spell high-end audio's eventual demise, similar to the lack of double-blind A/B testing?
Still, many of us can probably hear (or think they can) differences between CD players, and Meridian may have something here, but I would challenge anyone that even the most modest DSP room-speaker correction system would make a much bigger difference.
In hindsight, for us audio geeks, the CD format should probably have had a higher sample rate and resolution, combined with loss-less compression - and still fit Beethoven's 9th. But honestly , 95% of the public would not be able to hear a difference.
 

• reply