sunbeamgls wrote:There are a few points on this page that I would like to comment on - only my opinion of course, but based on being at the factory for the launch and hearing Exakt at the factory and at House of Linn:
- I don't think, during all of this, that I've heard Linn say that digital is now better than analogue vinyl. If these claims are being made because someone has heard these actual words directly from Linn employees, then fine. If these claims are begin made because of digitizing of the vinyl signal in the Exakt system, then there is a mis-understanding of why Linn have done this.
- The point of digitizing the analogue signal in the Exakt is mainly to ensure time and phase alignment of the drivers, produce a more accurate splitting of the frequencies being fed to each drive unit and to allow for the compensation of manufacturing tolerances in the 3K array drive units. Now, I don't know the exact numbers on this, but there's a very high chance (I would think almost a certainty) that converting the analogue signal into digital in the EDSM (24/192 remember) is likely to do much less damage to the signal than is done by not using Exakt benefits at the speaker end. So, although some would consider it sacrilege to convert analogue to digital, if it results in a better time / phase-aligned signal, better active crossovers and greater accuracy between speaker pairs than would ever be possible in analogue crossovers and loudspeakers then I suggest you really need to listen for yourself before deciding if an all analogue KLP12 / KK / 350a sounds more musical than an KLP12 / KEDSM / E350a sounds ;-)
As an equation then what I'm suggesting (and why I think Linn have gone down this route) its this: Distortion Caused By Analogue Speakers >>>> Distortion Caused By ADC
- For those who fear the digitization of the LP12 output purely on the basis of introducing digital into the analogue chain then this is largely irrelevant for 90%+ of anything recorded this century as it has already been digital earlier in its life (ie the master tapes).
- My understanding is that Exakt is unlikely to ever reach down to Majik level. For an Akurate EDSM then it will be the ADC stage that will be the major difference, in addition to the casework. I'm hoping we'll get a version of the AEDSM with no analogue inputs too.
I do not regard "source first" as a law of nature, but a rule supported by empirical results from a large number of trials. As such it should be used carefully, because it might well not be generally applicable and has to be re-evaluated for every new not yeat validated principal set-up (I am struggling to find the right words to describe what I mean here). In some cases it is not that obvious which functional chain to follow discerning source- or information-path from interference. When it comes to the source units (vinyl or CD, DS players) the obvious source route is the source material (the disc), but it has also been suggested that the furniture, LAN-cables, NAS used is more important than the DS used. I am not sure to what extent this is valid as there are not easy to see what is happening when the different parts forming the analogue output from a DS actually comes together in the ADC. There are so much that still needs to be tested to give some kind of definable understanding in this filed, especially the digital side of the DS units.
One thing that would bring Linn down is if they started to ignore tune-dem testing of changes, purely because they do not think that it would have any impact on the end result. I think we have seen signs of this lately, but I am not sure to what extent this is due to resource limitations or exagerated belief in basic signal and the current knowledge regarding psycho acoustics (I would imagine that they have to make compromises between how much they validate a design change on the digital side and the functional benefit being gained from the same design changes).
Regarding the statement that only analogue processing adds noise, actually is NOT TRUE. Every processing stage where the signal is tampered with in any way, level changed etc actually adds noise. One could argue that the noise added is negligible and it might (in many cases) well be, but the noise added is not zero. AND, Until we fully know what is happening on the digital side, the perceived effect of every processing stage should be evaluated - using tune-dem. Failing this, I think, would be a big mistake. It is one thing not understanding or not believing in a fenomenon, but ignoring it will not make it go away - if it actually is there.