Processor Shootout Arcam AV9, Lexicon MC12, Proceed AVP2, Upgrade Company Onkyo SC885

[BachToRock]

Originally Posted by welwynnick

What I haven’t figured out yet is where the D to A conversion should be done for best system-level SQ. This was the question I pondered when I had both the player and the processor upgraded, as no-one seemed to be able to answer it for me. So I decided to cover both bases and try to answer it myself. The answer isn’t really clear cut, and I’m still trying to work it out.

The biggest differences are in the ability of the equipment to preserve small details like decays and reverberations while the recording is busy with lots of other loud things. Small signal fidelity helps to reproduce a large, three-dimensional soundstage and to flesh out the subjects into solid, tangible elements of that stage, rather than just point sources. It also allows the phrasing and interaction of instruments to be captured better, so that each note flows into the next the way it was played, rather than being treated as a boring, metronomic sequence of independent sounds. That’s difficult to explain, but listen to the Lexicon side-by-side with the Pioneer, and it’s the difference between wanting to play all night and have songs running round your head the next day, and wanting to go and watch the TV instead.

Where a player or processor loses fidelity, it’s usually the smallest details that go first, and you can hear loss of resolution rising upwards from the ground as if the small detail is just truncated or dropped off the bottom. The bass and the treble and dynamic peaks are usually intact and the “sound” is the same, the reproduction isn’t complete, the soundstage is flat and the reproduction unconvincing.

Having said all that, the differences between analogue out from the 3800 turbo and bitstream decoding in the Onkyo turbo are more down to style. Processor decoding is more immediately impressive, and it’s what you would do as a demonstration to make a quick impression. Analogue however, tends to grow after extended listening, and is more natural. It has a deeper soundstage with a different perspective. Bitstream sounds more forward, with the subjects pushed in front of you, and LPCM sounds much the same, but with fog everywhere. Analogue sets everything a bit further back, less in your face and a bit more natural and relaxed. It’s what I preferred after extended listening, and while the processor comparisons are fairly straightforward, I can see different people coming to different conclusions with the analogue question, which I haven’t really answered yet. Of course, with a different player, the conclusions might be different again.

Nick

Nice thread!

Your observations are right on the money...

Digital Data is best decoded at the source as long as the analogue signal path that follows is clean. The accurate transmission of digital data and negative effects of Jitter are more sensitive in a multi-channel enviroment than a 2-channel system.

A player like the 3800 is ideal... the D/A conversion and circuitry are better than most receivers and processors. Digital data is comprimised in a noisy electrical enviroment... the kind that exists in receivers and processors. It is the reason companies like Bryston omit video circuitry from their surround processor.

The depth and warmth that Nick describes are the low level details that are lost due to Jitter and Error Correction. There are comprimises in every type of setup, but if you own a player like the 3800 you should set it to Pure Direct and seek out a SONY TA-P9000ES http://www.stereophile.com/solidpreamps/700/ it's the most impressive 5.1 going... no op-amps in the signal path... all discrete circuitry.