Tascam DA-3000

Al - your theory would be easy to test. All he has to do is turn off the Spectral pre and record from the Tascam, using the headphone jack as the monitor. Then, turn on the pre and listen to what was recorded. My guess, is he will get the same result. There is other evidence that the input section of the Tascam is not as good as that on other higher priced ADCs. But it would be an interesting and quick little experiment to do.

Dtc, if I'm not mistaken the phono stage Dhl used is built into his Spectral DMC-10 preamp. And he had indicated earlier in the thread that he was using the DMC-10's main outputs, not its tape outputs, to supply the signals to the Tascam. So the experiments that would seem to suggest themselves would be to either record from a different source altogether, with the Spectral disconnected and turned off, or perhaps to record from the Spectral's tape outputs.

Regards,
-- Al

Yes, my mistake. I was thinking the phono preamp was separate, but it is part of the preamp. The tape outputs sound like the best test.

Al:

I do have oscilliscopes and RF spectrum analyzers and I can assure you there is no RF leaving my Spectral preamp. The Spectral also has a special protection monitoring circuit that is designed to pick up RF leaving the amp. That circuit is not or has ever been active. Plus, I am pretty certain that RF feeding a JC2 would not result in good sound through the rest of my system.

Read my recent post on using the Benchmark ADC USB1. I won't repeat that other post here, but basically the Spectral was coupled to the Benchmark which fed digital AES/EBU to the Tascam. If Al is right, the same RF would have been transmitted to the Benchmark and the sonic results would have been the same.

But they were not. There was a dramatic improvement in sound quality (and yes, even through the headphones), so I am not buying the mystery signal theory.

Clearly, the Benchmark has a better designed and implemented analog front end than the Tascam, and one would expect it given the components used by Benchmark.

Components DO matter, and there is a general consensus that some opamps are better than others, and that copious capacitor coupling with electrolytic caps does impact the sound. Why else do people pay $300 for a Mundorf coupling film capacitor? It is also reasonably agreed that an AD797 will sound better than a NE5532.

There is no mystery here, and we don't need to make up theories of mystery RF signals to explain the observed differences.

Fair enough, Dhl93449. Your post just above solidifies your case persuasively, as far as I'm concerned. My theory was an attempt to reconcile your sonic assessment of the Tascam with Jeff's (whose opinions command great respect in my book), while taking into account the nearly unique ultra-wide bandwidths of the Spectral products.

One minor nit, though:

I won't repeat that other post here, but basically the Spectral was coupled to the Benchmark which fed digital AES/EBU to the Tascam. If Al is right, the same RF would have been transmitted to the Benchmark and the sonic results would have been the same.

My theory revolved around the assumption that the anti-aliasing filter in the Tascam was less than fully effective at RF frequencies, perhaps because the design is such that some RF frequencies might bypass the filter altogether via unintended/parasitic circuit paths. If so, the results would very possibly not have been the same with the Benchmark.

Regards,
-- Al