New vs. old DACs - opinions?

See the HIFICRITIC web site for info. Some of the ratings are there. They don't publish the articles as that is how the Mag is supported. As to Mapmans point, specific DACs will vary but whether there is a general trend upward or downward is a legitimate question.

You also need to consider whether jitter rejection will be an important consideration for the DAC or no depending on your source(s). If you are not sure, then safer to go with a DAC that is more resistant to jitter from the source, like most upsampling DACS that must re-clock the bits.

Although it doesn't directly address the question at hand, an excellent point was made by Audiogoner Tobias in this thread a couple of months ago:

http://forum.audiogon.com/cgi-bin/fr.pl?fcabl&1233206618

DAC's that don't have re-clocking provisions that minimize jitter, and that don't have precisely controlled input impedances, will be particularly sensitive to cable length, assuming their electrical (not optical) digital input is being used. He has found that 1.5 meter lengths usually work better than 1 meter lengths.

The reason is that cable length will affect the re-arrival timing of signal reflections caused by impedance mismatch at the DAC input, after the reflections travel back to the source and then are partially re-reflected back to the DAC. Depending on the cable length, they may or may not re-arrive at the DAC input coincident with signal edges in the original incoming waveform. If they do arrive coincident, the resulting edge distortion could significantly increase jitter. And that would be pretty much independent of what source component is being used (unless the source component provided an essentially perfect impedance match at its end, preventing the re-reflection).

Something to keep in mind when comparing a jitter-resistant DAC with one that doesn't have that provision.

Regards,
-- Al

Stan,

The general trend has to be upward.

Its a legitimate question but I don't think the answer is very useful for determining which way to go other than if you shoot blind, the odds say you will probably find the right DAC sooner with new versus old.

Better to know what you need and then find the DAC that seems to best fit the bill.

If it doesn't try another.

In the end, you'll get to the right one faster by looking at other factors rather than whether a DAC is old or new.

If you limit it down to several choices, some old and some new, chances are you'll pay less in the end with an older second hand unit than a new one.

This question has been asked before, it comes up from time to time. You may find more information by searching the archives.

IMHO, I'd say go older dac. For one reason, yes, the newer dacs have newer digital chips, but these are relatively inexpense. What made, say for example the Classe DAC-1 so great was that it had a fabulous analog output stage. This was a $4K DAC back in the day, with more money going into analog than the DAC chips. That great analog stage will still beat any analog output stage on a $1K new DAC.

The second reason would be re-sale. Say you buy a used Classe DAC-1 (again, just using this as an example), you get a $4K DAC for $1K. In two years should decide to re-sell and move on, you'll still get about $1K for the unit. Spend $1K on a new unit, in 2 years you'll be lucky to get $300 for it.

Cheers,
John

I agree with Bob, in general, newer DACs more resolving and exposing of warts, older less resolving and more forgiving nature.

I just recently revisited a review from Feb. 1992 Stereophile of the ML No.30 DAC by Robert Harley. This was a state of the art processor priced at 14k in '92. Harley raves about this machine as head and shoulders above anything else available at the time, needed it's own category in recommended products. He describes precisely the sonic attributes I hear in my present digital setup, a newer and lower priced setup. I also hear these attributes in current more upmarket digital from Wadia and Esoteric.

My take is digital, at least at the top of the market is making real strides forward. Now, do these improvements trickle downward into the less expensive spread? I would think this can only be ascertained on a case by case basis.

I do think a lot of newer digital relies solely on processing speed and other parts upgrades without addressing other, perhaps more important aspects of performance. I certainly see a lot of newer digital with very low parts count, weak power supplies, just lots of empty space, selling at serious money. In this case, I might say the older digital with more guts may be better. Still, a case by case comparison would be needed.

My own experience has been that digital made some major strides in the reasonably priced spread somewhere around 2000. Generalizing, I would look limit my purchases to digital after this period. The highest priced earlier digital may also be a viable choice, if I could get a ML No. 30 or 30.5 at perhaps 4 or 5k (I don't know current market prices), that could be very enticing.