Active monitors and Hi quality dacs

[Fancypants]

Thanks for this interesting discussion.

I was considering purchasing the KH150 or KH120 II for desktop PC music listening in my office. However, it's a bit of a bummer that my recently acquired ADI-2 DAC FS (which I use as my headphone AMP) will not garner any benefit to the audio chain, and would in fact add an superfluous double conversion (versus going straight from my PC to the monitors via SPDIF).

I really love the idea of having my ADI-2 DAC FS' volume knob, and the highly regarded DAC in play -- switching seamlessly between headphones and monitors. Also, don't love that all audio is resampled to 48hz.

So it sounds like going for an "all analogue" active monitor, like the Focal Shape 65 (or original KH120) would be a good option to avoid the in-monitor DSP and keep my DAC in play? Be open to other active monitor suggestions as well that would pair nicely with my RME.

 

[voodooless]

So it sounds like going for an "all analogue" active monitor, like the Focal Shape 65 (or original KH120) would be a good option to avoid the in-monitor DSP and keep my DAC in play? Be open to other active monitor suggestions as well that would pair nicely with my RME.

Or just hook the stuff up the analog way via your RME, and don’t worry about it. The active DSP speakers will make up for any potential negatives that this chain may provide. Or save up for the ADI-2 Pro version with digital outputs. It will be able to control the volume digitally as well.

 

[bodhi]

Thanks for this interesting discussion.

I was considering purchasing the KH150 or KH120 II for desktop PC music listening in my office. However, it's a bit of a bummer that my recently acquired ADI-2 DAC FS (which I use as my headphone AMP) will not garner any benefit to the audio chain, and would in fact add an superfluous double conversion (versus going straight from my PC to the monitors via SPDIF).

I really love the idea of having my ADI-2 DAC FS' volume knob, and the highly regarded DAC in play -- switching seamlessly between headphones and monitors. Also, don't love that all audio is resampled to 48hz.

So it sounds like going for an "all analogue" active monitor, like the Focal Shape 65 (or original KH120) would be a good option to avoid the in-monitor DSP and keep my DAC in play? Be open to other active monitor suggestions as well that would pair nicely with my RME.

Sound quality wise one extra conversion AD-conversion will not matter, at all. If you had 20 or 50 conversions then there might be a discussion, but one? No, don't go there.

Nothing wrong with buying stuff based on feelings but if you want to make rational decision then just get whatever monitor you like best, plug in whichever way is supported and never think about it again.

 

[Fancypants]

Thanks all!

 

[b407driver]

I'm glad I found this thread, as I was getting ready to ask the same question (so I won't duplicate with a new post). Have had the Kali IN-UNF + WS-6.2 subwoofer for a couple of weeks, and have been waiting for DAC/headphone amp to be shipped (small office desktop environment, Mac OS). I know these studio monitors haven't garnered much attention around here (they're weird, I know), but they solve some problems (and perhaps create some of their own) for users like myself, admittedly probably a small sample. That being said, after spending a week tweaking the DSP settings (via dip switches) and volume levels, the sound from them is really quite good; remembering all the YT reviews where the 'reviewers' basically chucked them on the desk and called it a day now seems ridiculous, as they almost surely didn't hear what these speakers are capable of (the addition of a sub really, really helps). That being said, I do have a 120Hz hump I need to get rid of, so REW/SoundSource comes next.

My question: I read this thread, and think I understand the general 'take' of experienced members, but want to make sure. Given my situation (active monitors that have built-in DSP), do I net anything by adding a high-quality DAC into the signal chain? Will the signal output by the speakers be any different with a (Topping) DAC feeding an analog signal into the monitors, compared to analog out from the Mac? [Yes, the IN-UNF has USB in, but when using with the WS-6.2 subwoofer, you cannot use it, as the way they designed it (a miss, in my opinion) the subwoofer has to feed (balanced) analog to the IN-UNF bass unit.]

Another way to look at this, if you put the crappiest DAC that exists in 'front' of the speakers, could you tell on the output side that it was being fed a low-quality signal, or would the ADC/DAC conversion in the speakers clean that up? And, if it did, would the sound effectively be the same as (instead) having that Topping in the mix?

Basically trying to figure out if there is any getting away from the 'lowest common denominator' of the built-in ADC/DAC in the active monitors, or whether it even matters from a technical standpoint.

Thanks again for any clarity, you all have been great help over the years!

 

[Vincent Kars]

would the ADC/DAC conversion in the speakers clean that up

Of course not. ADC converts analog to digital. If the analog sound is crappy, so will be its digital equivalent.

 

[JIW]

To what degree the quality of the analog source affects the output after the internal AD/DA cycle does depend on the relative levels of noise and distortion as well as the correlation between the distortion of each.

The higher the noise of the analog source relative to the AD/DA cycle, the higher the noise of the output after the AD/DA cycle. For distortion, it is more complicated. Negatively correlated distortion will lead to lower distortion for a higher relative level, while positively correlated distortion will lead to higher distortion for higher relative levels. This is all assuming the analog source has noise and distortion levels no higher than the AD/DA cycle. All values used in the calculations below are for RMS voltages.

For noise, if independent, meaning it is uncorrelated, with mean zero, if the analog source has the same level of noise as the AD/DA cycle, noise after the AD/DA cycle will be
10*log10(1 + 1) = 3 dB
above the noise of the AD/DA cycle. If the noise of the analog source is 10 dB lower, the increase in noise is
10*log10(1 + 10^(-10/10)) = 0.412 dB
and if the noise of the analog source is 20 dB lower, the increase in noise is
10*log10(1 + 10^(-20/10)) = 0.043 dB.

For distortion, the calculations are more complicated since it cannot as reasonably be assumed that it is uncorrelated. Still, it is straightforward, assuming the coefficient of correlation is known. The level difference between cumulative distortion and the distortion of the AD/DA cycle is
10*log10(1 + 10^(deltaL/10) + 2*rho*10^(deltaL/20))
where deltaL is the level of distortion of the analog source relative to the level of distortion of the AD/DA cycle and rho is the correlation coefficient.
Since the distortion difference is strictly increasing in the correlation coefficient, a range can be calculated even if it is not known. If the correlation coefficient is at its minimum of -1, if the distortion levels are the same, there is no distortion at all, if the distortion of the analog source is 10 dB lower, the cumulative distortion is
10*log10(1 + 10^(-10/10) + 2*(-1)*10^(-10/20)) = -3.302 dB
lower and if the distortion of the analog source is 20 dB lower, the cumulative distortion is
10*log10(1 + 10^(-20/10) + 2*(-1)*10^(-20/20)) = -0.951 dB
lower. If the correlation coefficient is at its maximum value of 1, if the distortion levels are the same, the cumulative distortion is
10*log10(1 + 1 + 2*1*1) = 6.02 dB
higher, if the distortion of the analog source is 10 dB lower, the cumulative distortion is
10*log10(1 + 10^(-10/10) + 2*1*10^(-10/20)) = 2.38 dB
higher and if the distortion of the analog source is 20 dB lower, the cumulative distortion is
10*log10(1 + 10^(-20/10) + 2*1*10^(-20/20)) = 0.828 dB
higher.

Thus, a high quality analog source is not entirely wasted when using speakers with AD/DA cycle. For noise, the noise of the AD/DA cycle will be the lower bound for cumulative noise but for distortion it is unclear.