What to expect from a new DAC?

First, the digital source is more important than the DAC itself. Muiltiple reviews in TAS show that a $1K DAC can sound almost identical to a $8K DAC when driven by a low jitter source (not a CD transport BTW).

Second, Most D/A chips have gotten a lot better (not all). They deliver more detail and better focus as well as blacker background due to improved S/N ratio.

That said, there are still a lot of poor newer D/A chips (ala AKM) as well as poor DAC designs that dont take advantage of what the better chips can do. There are also chips and DAC designs that are simply not musical. Some of the older ladder and NOS D/A chips are very musical and natural sounding, and so the market for NOS DACs, which is mostly on ebay and from China. The seoncd most irritating thing about newer DAC chips is the digital filtering that is built-in and cannot be disabled. Wrecks the SQ IME. this is why older NOS chips are popular and why some new ultra-high over sampling DACs are popular.

So, the bottom line is: deal with the jitter from your source first (computer or transport) and then think about a new DAC. Unlike what the advertising BS says, the DAC will not reduce the jitter enough. It still needs a low-jitter input signal.

Many cannot hear differences in DACs and sources due to poor active preamps. This is the bain of analog audio and all it takes to kill a systems performance. Try a passive transformer-based linestage instead. Much more transparent and no compression.

Steve N.
Empirical Audio

"So, my transport is a Levinson 31 Reference DAC.... so it was very expensive (and sounds great)."

I've modded ML transports. Not even close, even with clock upgrades.

I have a fully modded ML 38 preamp (my mods). Also not even close to a decent transformer linestage.

"My preamp is an Audio Research LS15 active preamp (tubes)."

That is not bad.

"I tried the passive linestage attenuator 20 years ago and found it not to be the way to go...(with lesser gear then I now use) If memory serves, the music just seemed dead. Funny how things go round and round... What's different with a passive transformer linestage ..... (at the time... I was told... well... the analog impedance matches were just not quite right and you lost too much with the passive Adcom line stage attenuator."

Myths. What is really happening with resistor passive linestages is 1) you are limiting the current going to the amp and 2) you are putting a low-pass filter in the cable. I dont recommend them.

Transformer linestages have neither of these drawbacks. The only real requirement is that the DAC have decent current drive to drive amps directly. Op-amps dont cut it. The ML will have 5 ohms output impedance. That is good. One must purchase a high-quality one however, or you risk transformer saturation distortion.

If you read the 6-moons review of the Music First TVC, you will discover that they believe this is the best linestage ever. More transparent than the most expensive active preamps. I would not go that far, but it is quite good. I have an $8K pure silver wound version that is cryo-treated.

"What I want to do now is add a USB to digital Low jitter interface for a PC server. What would be a good match for the Sonic Frontiers DAC/LS 15 combination?"

The best USB converters on the market are:

1) Off-Ramp 5
2) Diverter HR
3) Synergistic Research

All of these have low jitter. There is really no "synergy" here. If they are good, they are good with all DACs.

That does not mean that you will not take a step backward, because this is also possible. If you have a sibilant component or cable in your system, reducing jitter may aggravate this. More high-frequency energy is usually a result, but it will be "clean". The object will be to identify this component or cable and eliminate it from the system.

Be aware that you must make some wise choices in order to get stellar results from computer audio, namely the computer, the playback software, the ripping software etc.. It can be a minefield. Here are some tips:

http://www.empiricalaudio.com/computer-audio/

Computer Audio Asylum is another good resource.

Steve N.
Empirical Audio

Tom6897 - I really doubt if the "system" jitter is 10psec. Most clock specs are in this range, and that is when they are driven with very expensive low-noise power supplies. In systems, this number goes up usually by a factor of ten. Also, even with the best chips, the internal PLL is affected by incoming jitter.

I have used the Sabre chip that is purported to eliminate jitter. It doesn,t. I had a Wired 4 Sound DAC2 here about 2 weeks ago, which uses the Sabre chip. Each of the inputs; USB, S/PDIF and I2S sound different. The best of these is the I2S driven by an Off-Ramp 5. World-class sound quality. The others are not interesting compared to the I2S IMO. If the Sabre reduced jitter as much as advertised, none of these differences would be audible.

Virtually all DACs benefit from a low-jitter source. Maybe the only exception is the PWD in NativeX mode.

Steve N.
Empirical Audio

Knghifi - No joke - read this:

http://www.6moons.com/audioreviews/musicfirst/passive.html

This is an old review. New units are even better I understand.

Steve N.
Empirical Audio

"Add the Bel Canto REF LINK to this list. It is every bit as good as the Off-Ramp 5. See Steven Stone's review in the current issue of TAS."

It's certainly a good one, but does not offer I2S. Steven could have used a better S/PDIF cable IMO. Might have made a difference. Both the USB and S/PDIF cables are critical.

Also, the OR5 can be further improved by replacing the wall-wart with a Monolith or Paul Hynes power supply. Can't do this with the AC powered BC.

Steve N.
Empirical Audio

Knghifi - Music First has evidently improved the core material or windings etc. because the latest versions of MF are even better. I dont think you need silver anymore. The copper is really good.

Steve N.
Empirical Audio