Uprgrading my Metrum Octave DAC

Steve,

I won't debate what your transformers sound like since I have not heard them. Knowing what I do about you, they probably sound awesome. I will say that transformers and extremely low distortion are typically exclusive. And, it's hard to imagine much "new transformer technology" when it's still just coil A energizing coil B. (and all the issues with that) (and there are plenty)

As far as not being able to do true balanced topology with solid state, I'm not sure why you take this stance. Every differential amplifier stage is true-balanced by default on both the input and output of the topology.

Lastly, digital volume attenuation is hardly "new." Yes, there are methods of accomplishing this with technically no distortion. To me it's all unnecessary and yet another way to have your bitstream modified via DSP and I have no interest in that.

I'm not saying any of this sounds bad or is "wrong," I'm just more of a purist when it comes to digital manipulation, as in I want none at all if possible.

RM

"Every differential amplifier stage is true-balanced by default on both the input and output of the topology."

In your dreams. Most good balanced output stages are either tube or discrete, so there are several issues that make them imperfect:

1) DC offset
2) common-mode noise
3) difference in amplitude between + and - signals (impossible to make these identical with active circuits)
4) isolation of ground between DAC and amps

A good transformer solves all of these and makes the amps sound better in the process.

"Lastly, digital volume attenuation is hardly "new." Yes, there are methods of accomplishing this with technically no distortion. To me it's all unnecessary and yet another way to have your bitstream modified via DSP and I have no interest in that."

You are not getting it. This is not DSP. This is not software. This is not even a digital chip somewhere in the signal path. This is something entirely different. This is control of the D/A reference voltage. No modification to the digital stream. Nothing else like it.

Steve N.
Empirical Audio

Steve N., you mentioned that WAV files are superior to FLAC files. Can you explain why ?

Mabonn - I can only speculate why they sound inferior to .wav.

I suppose it must be due to the real-time behavior of the CODEC software in the computer. Statically, the data is not changed, but somehow when these compressed formats are played dynamically, it changes things. I have heard it over and over in different systems.

We can debate this until the cows come home, but that is the experience I have had. Also, all of the partners I have had at RMAF shows had the same experience.

Steve N.
Empirical Audio

1) DC offset - almost none
2) common-mode noise - common mode noise is reduced in a diff amp stage
3) difference in amplitude between + and - signals (impossible to make these identical with active circuits) - hello current source
4) isolation of ground between DAC and amps - entirely different topic

You're right in that it's not perfect. You should also realize that none of this is perfect. Especially a transformer!

Transformers are for isolation and impedance matching. They are inefficient and sloppy. They are not extremely linear.

I've heard good music coming through a xformer, who hasn't. I just would not use one unless I had to.

Let's just agree to disagree. From a design methodology standpoint I prefer to keep nonlinear devices, like transformers and capacitors, out of the signal path, and bias devices in their linear region.

Here's a little review for you:
http://www.ecircuitcenter.com/Circuits/BJT_Diffamp1/BJT_Diffamp1.htm

If you bias this with a current source, it's almost perfect. Certainly as perfect as any transformer.

Balanced in, balanced out, and you set the bias voltages so you bias your next stage, which can also be fully balanced. Nothing new.

RM