[HermannS]

Recently, I conducted an experiment together with a friend. He built a -20dB damping XLR adapter with two of the very best Vishay resistors (1.2kOhm). No passive pre-amp could do that better. We tried it with the 005 direct to the active speakers. So instead of -25dB we drove the 005 with -5dB. Didn't cut the cheese. The music was damped and lost dynamics compared to the HPA4. We also put it between the HPA4 and the speakers. Similar result. It didn't do any good. Passive attenuation is not the way to go, I think. John Siau of Benchmark explained his reasoning in an application note. It's quite interesting.
https://benchmarkmedia.com/blogs/application_notes/benchmarks-256-step-relay-controlled-attenuator

 

[DBB1]

Ok, got it. Still, you are essentially saying you prefer 005 direct over the two active preamps you mentioned, this means you prefer digital and it's attendant bit stripping. This should not be the case. The only thing I can think is the coloration of these two preamps is not sympathetic with the KEF's. Both are ss, perhaps a tube pre would be a better match. The Schitt Saga does give you option of tube buffer or straight through solid state. I'm sure there are other situations in which dac sans pre would seem preferable to active pre, non-sympathetic colorations can make a system unbearable. I'm quite sure all active preamps have some sound signature, one has to find sympathetic match. I'd also suggest looking closely at your power cord and interconnect choices, both are possibilities for attaining preferred colorations. Those are both quite nice preamps, they should outperform dac straight in, I'd give them more time and try some different cording, otherwise the active tube could be worth looking at. I doubt a passive pre is going to outperform what you already have. you'd probably lose some measure of drive, dynamics.

That's it dynamics. As I said earlier over time the preamps in the system were too dynamic with the Kefs. I found it fatiguing over hours of listening. Removing it reduced that feel and added a touch of transparency. You should also be pleased that your credibility caused me to dig deep into the recesses of my closet to find the no-remote Goldpoint stepped attenuator (passive preamp). I set it up so I could listen between 0 and -5 db. Yes it does sound better: more solid. Next I will listen for a week or so then try the Hegel H30 again to be sure.

 

[sns109]

Recently, I conducted an experiment together with a friend. He built a -20dB damping XLR adapter with two of the very best Vishay resistors (1.2kOhm). No passive pre-amp could do that better. We tried it with the 005 direct to the active speakers. So instead of -25dB we drove the 005 with -5dB. Didn't cut the cheese. The music was damped and lost dynamics compared to the HPA4. We also put it between the HPA4 and the speakers. Similar result. It didn't do any good. Passive attenuation is not the way to go, I think. John Siau of Benchmark explained his reasoning in an application note. It's quite interesting.
https://benchmarkmedia.com/blogs/application_notes/benchmarks-256-step-relay-controlled-attenuator

Those notes reiterate many of the negatives I've read by previous authors. I see passives only use as a possible option to dac direct, and for those with limited funds. A cheap active preamp is probably worse than dac direct or passive. I tried a number of passives way back in the day, since the Joule Electra never looked back. Having said that, the Schitt Saga wasn't that bad, not really close to the Coincident, but taking into account price differential couldn't complain. It was also superior to built in transformer based volume control in my 845 Turbo, that is real junk, really just an afterthought in that amp.

 

[TriodeMan]

Could it be there is a sweet spot for dB magnification of the signal? I have noticed that of the twenty or so systems i have put together over the last 40 years they all sounded their best when the dB magnification was 130 or so. I add preamp gain to power amp gain to speaker sensitivity so 15 + 25 + 90 for example. It never really sounded right more than 10dB above or below this figure (lacking dynamics on the low end and an aggressive 'pushed' sound that irritates on the high end of this scale). Could be real or just a coincidence I suppose. If true it would mean that a passive would work when your power amps are 30dB gain and your speakers are 100dB sensitivity and the input/output impedances match and you use a short run of low capacitance signal cable, all of which together is not a common happenstance.

Back to the thread, if anyone wants to gift me a Coincident Statement linestage i'm happy to conduct some confirmatory experiments!

 

[sns109]

Could it be there is a sweet spot for dB magnification of the signal? I have noticed that of the twenty or so systems i have put together over the last 40 years they all sounded their best when the dB magnification was 130 or so. I add preamp gain to power amp gain to speaker sensitivity so 15 + 25 + 90 for example. It never really sounded right more than 10dB above or below this figure (lacking dynamics on the low end and an aggressive 'pushed' sound that irritates on the high end of this scale). Could be real or just a coincidence I suppose. If true it would mean that a passive would work when your power amps are 30dB gain and your speakers are 100dB sensitivity and the input/output impedances match and you use a short run of low capacitance signal cable, all of which together is not a common happenstance.

Back to the thread, if anyone wants to gift me a Coincident Statement linestage i'm happy to conduct some confirmatory experiments!

I'd agree there is likely a narrow sweet spot for a passive, my narrow sweet spot with Coincident is due to using 005 as the fine volume attenuator. When I run 005 in bypass mode, using Coincident volume control the sweet spot is much wider. There is also the problem of human hearing more efficiently at higher volume levels. Those of us with loudspeakers also have to contend with ambient listening room noise, I seem to recall measuring steady state room noise levels of around 30db, and that late at night. Add up internal home sound plus sometime even louder outdoor environment sounds and I might be in the 50db range. This ambient listening room noise floor narrows volume sweet spot, probably somewhere in 70db range minimum for satisfactory listening. At the high end of volume we have to worry about overloading room and proper amp gain to loudspeaker efficiency, thinking about clipping here. And so if my calculations are correct, we may have 30db volume width at best (thinking high 60's db- 90's db). Haven't measured my usual listening levels in a long time, seem to recall high 70's - mid 80's as being my usual volume level. I doubt most of us during intensive listening sessions vary much more than 10db or so. If you could match up passive pre for this exact sweet spot you'd have a greater chance for satisfactory listening experience. Headphone listening, especially closed back type don't have to contend with anything near this level of ambient noise floor.

 

[JecklinStax]

I like buying new hifi gear as much as the next person, but I really hate getting new hifi that does not sound better than what I already had. So, before getting a preamp with a possibly better sounding volume control than the one in the da005, I gave myself a little assignment. I played tracks of 10 SACD's (various genres of music played by real people on real instruments) on a Sony UBP X800Mk2 via a HDMI de-embedder into a da005 coax input at 176,4 kHz PCM. I listened to Stax SR-007 headphones, driven by a Mjölnir KGSSHV amp for electrostatic phones, which has a manual volume control (pot) at the input. I have that pot usually maxed out, because I am used to the remote control on the dac.
I played each track twice, once with the variable volume control of the da005 at -20 dB, and once with the digital control at 'fixed volume'. I chose -20 dB because that is my average day-to-day listening level. During the test I corrected for the enormous difference in loudness between 0 dB and -20 dB by turning the pot on the HPA up or down. Because a bit louder always sounds a bit better, I took care (as far as possible) to get the same sound levels in the phones at 0 and at -20dB.
As a first test I turned up the variable digital control on the da005 up to '0dB', to check if that sounded the same as the digital control on 'fixed volume' (or 'bypass'). It was indeed: the one was exactly as loud and as good as the other. There wasn't even a click or gap when switching between them.
Then I compared the sound of the SACD-tracks via the fixed volume digital control with their sound via the variable digital control at -20 dB. I listened for differences in all the usual aspects of sound quality, such as stereo width and depth, dynamics, openness and transparency, location ambiance, pinpoint location of voices and instruments, delicate treble, weighty and yet taut bass, and so on.
Long story short: I heard no difference in quality between 0 dB fixed and -20 dB variable. None.
This means that for me at this point the purchase of a preamp with a possibly better volume control would make no sense.
Of course there is more to a good preamp than just its volume control. In a (very positive) review on 'soundnews.net' of the Topping Pre90 significant audible differences are described between the interfaces between this preamp and several different dacs, possibly because some dacs have better output configurations than others. But I suspect this is of no concern to da005 owners, because I can hardly believe that Musetec designed a brilliant dac with a sloppy output.

 

[DBB1]

I like buying new hifi gear as much as the next person, but I really hate getting new hifi that does not sound better than what I already had. So, before getting a preamp with a possibly better sounding volume control than the one in the da005, I gave myself a little assignment. I played tracks of 10 SACD's (various genres of music played by real people on real instruments) on a Sony UBP X800Mk2 via a HDMI de-embedder into a da005 coax input at 176,4 kHz PCM. I listened to Stax SR-007 headphones, driven by a Mjölnir KGSSHV amp for electrostatic phones, which has a manual volume control (pot) at the input. I have that pot usually maxed out, because I am used to the remote control on the dac.
I played each track twice, once with the variable volume control of the da005 at -20 dB, and once with the digital control at 'fixed volume'. I chose -20 dB because that is my average day-to-day listening level. During the test I corrected for the enormous difference in loudness between 0 dB and -20 dB by turning the pot on the HPA up or down. Because a bit louder always sounds a bit better, I took care (as far as possible) to get the same sound levels in the phones at 0 and at -20dB.
As a first test I turned up the variable digital control on the da005 up to '0dB', to check if that sounded the same as the digital control on 'fixed volume' (or 'bypass'). It was indeed: the one was exactly as loud and as good as the other. There wasn't even a click or gap when switching between them.
Then I compared the sound of the SACD-tracks via the fixed volume digital control with their sound via the variable digital control at -20 dB. I listened for differences in all the usual aspects of sound quality, such as stereo width and depth, dynamics, openness and transparency, location ambiance, pinpoint location of voices and instruments, delicate treble, weighty and yet taut bass, and so on.
Long story short: I heard no difference in quality between 0 dB fixed and -20 dB variable. None.
This means that for me at this point the purchase of a preamp with a possibly better volume control would make no sense.
Of course there is more to a good preamp than just its volume control. In a (very positive) review on 'soundnews.net' of the Topping Pre90 significant audible differences are described between the interfaces between this preamp and several different dacs, possibly because some dacs have better output configurations than others. But I suspect this is of no concern to da005 owners, because I can hardly believe that Musetec designed a brilliant dac with a sloppy output.

I suspect that the loss of bits when using a digital volume control is audible in a very transparent system, but usually not dramatically so. I swapped in a Goldpoint resister based volume control and felt I did hear some more solidity. To me the more interesting question is under what circumstances can you dispense with a preamp entirely and still benefit in spite of the loss of bits resulting from use of the digital volume control. The answer might be: 1. when the analog output stage of the dac resembles a good preamp in terms of output voltage and output impedance. Or, as I think I am finding, when other components like the voicing of the speakers and preamp were not ideal. This later situation may prevail above all else.

If I understand the above mentioned video (link below) correctly a preamp is likely to benefit when the dac has: 1. low output voltage (less than 6 volts xlr),; and 2. higher than 200 ohms output impedance. I don't know what the output impedance of the 005 is, but the voltage output level is 2Vrms for single-ended RCA and 4Vrms for balanced XLR output. So, other things being equal, the 005 should benefit from a good system compatible preamp regardless of the digital volume control. Or, maybe Musetec did design a brilliant dac, with a less than brilliant (or typical or industry standard) output.

 

[sns109]

The volume control on 005 is not any different from any other Sabre chip or other chip for that matter, they are not designed separately. The volume control is on the Sabre chip itself, this is why you lose bits when adjusting volume down. The only way they could do volume control any other way is to offer analog volume control, Aqua, Waversa are at least two dacs I know of that offer this option. Then what you have is a dac/preamp, they may even have analog inputs on the dac. All I can offer you is the absolute truth that digital volume control does strip bits, this absolutely will affect sound quality of dac.

Now, as to whether you hear the diminished sound quality, I'm not here to argue that. It may even be possible I wouldn't hear the difference in your system. It's also likely the 005 may suffer less from the bit stripping than other dacs because of it's extremely high resolving abilities.

 

[TriodeMan]

I think I see a consensus emerging here between the recent posters. To summarise: a digital volume control will strip bits which at deep attenuation (in the case of the 005 beyond -20dB attenuation) may be audible but wont be at slight attenuation (bypass to -20dB). Makes sense, it is a 32 bit Sabre DAC and you lose 1 bit per 6-12dB attenuation depending on the chip implementation. An active preamp therefore , if it brings something to the party, is doing so by increasing the dynamic gain of the whole system (very useful for where amp gain + speaker sensitivity is <130dB) and/or providing a useful complementary 'flavour" such as can be had with various tubes (the 101D DHT in the Coincident for example). Depending on the resolution quality of the said introduced preamp the net effect can be beneficial or ( in cases of much lower resolution preamp than the 005) detrimental to sound quality.
Such nuances explains the differing experiences and views. I will use this thread as a useful roadmap to stake out my system building.

 

[JecklinStax]

I think I see a consensus emerging here between the recent posters. To summarise: a digital volume control will strip bits which at deep attenuation (in the case of the 005 beyond -20dB attenuation) may be audible but wont be at slight attenuation (bypass to -20dB). Makes sense, it is a 32 bit Sabre DAC and you lose 1 bit per 6-12dB attenuation depending on the chip implementation. An active preamp therefore , if it brings something to the party, is doing so by increasing the dynamic gain of the whole system (very useful for where amp gain + speaker sensitivity is <130dB) and/or providing a useful complementary 'flavour" such as can be had with various tubes (the 101D DHT in the Coincident for example). Depending on the resolution quality of the said introduced preamp the net effect can be beneficial or ( in cases of much lower resolution preamp than the 005) detrimental to sound quality.
Such nuances explains the differing experiences and views. I will use this thread as a useful roadmap to stake out my system building.

Input impedance of my headphone amp is 50k, and my speaker amp is 47k, so they both present a very light load to the dac. I switch between them with a passive balanced selector switch. I have no need of a different 'flavour'. So why then invest in a preamp?

 

[sns109]

I think I see a consensus emerging here between the recent posters. To summarise: a digital volume control will strip bits which at deep attenuation (in the case of the 005 beyond -20dB attenuation) may be audible but wont be at slight attenuation (bypass to -20dB). Makes sense, it is a 32 bit Sabre DAC and you lose 1 bit per 6-12dB attenuation depending on the chip implementation. An active preamp therefore , if it brings something to the party, is doing so by increasing the dynamic gain of the whole system (very useful for where amp gain + speaker sensitivity is <130dB) and/or providing a useful complementary 'flavour" such as can be had with various tubes (the 101D DHT in the Coincident for example). Depending on the resolution quality of the said introduced preamp the net effect can be beneficial or ( in cases of much lower resolution preamp than the 005) detrimental to sound quality.
Such nuances explains the differing experiences and views. I will use this thread as a useful roadmap to stake out my system building.

Yep, I believe the most likely explanations for dac direct preference will be an equal or greater loss of resolution from one's active or passive preamp. The other would be unwanted colorations added by preamp. So, in order to get maximum performance from one's system: run dac full open/bypass, no loss of bits, active preamp that suffers no resolution/transparency loss, and that preamp adds no coloration or a pleasing overall coloration. Passives do have too many inherent limitation (previously spoken of) to offer ultimate solution.

 

[JecklinStax]

I just use a passive switch. No attenuation involved. Taking it out of the chain or putting it back in is inaudible.

 

[sns109]

Input impedance of my headphone amp is 50k, and my speaker amp is 47k, so they both present a very light load to the dac. I switch between them with a passive balanced selector switch. I have no need of a different 'flavour'. So why then invest in a preamp?

Only to not strip bits, and if you don't hear the detrimental effects you have no reason. My problem is I'm an audiophile, eventually, I always get to a point where I detect some weak link in my system. Always seeking improvements, and knowing I'm stripping bits, I'd eventually evaluate this bit stripping as the weakest link in my system. At that point, I'd have to address the issue. Perhaps you'll find long term enjoyment with status of your present system, count yourself lucky if true. For the rest of us, this perfect pleasure is not permanent, there goes some more money out the door! The one thing I do have to say, is after near thirty years building to present system, I do get actual permanent listening pleasure from my system. My system is never less that totally involving, never fatiguing, performers in room every listening session. Every album/cd I put on with intention of listening to only one song ends up in playing entire or near entire album/cd. My summer listening sessions are going till 2 or 3am, its only sleepiness and the knowledge I'm going to end up sleeping in next day that ends the listening session. And with all this pleasure I'm still upgrading, latest purchase audiophile network switch. As an aside, optimizing streaming solution is most complex undertaking yet in audio, every single little thing matters, zeros and ones get converted to analog. For the objectivists, this is not simply a digital network.

 

[sns109]

I just use a passive switch. No attenuation involved. Taking it out of the chain or putting it back in is inaudible.

It may be true that you don't hear it, but that doesn't mean the inherent limitations previously mentioned don't exist. Bottom line on everything in audio, we hear what we hear, who can argue with that! Still, there are very real defects in audio equipment scientific method has proven to exist, the question is, do we hear these defects? I can think of one particular defect that's inherent to my system and can never be solved. That is the ambient and/or steady state noise floor of my dedicated listening room, in other words the noise generated from my internal house, add to that nose from external environment, lowest measured db in my room is typically high 20's, can easily go up to 50's and above. So we spend all this money and effort to lower noise floors, and the most imposing noise floor is the listening room itself! So the point being, there is a major defect in my system (the listening room itself), yet I derive all this pleasure. Goes to show one can look past limitations, our minds control our pleasure. This is what makes subjective evaluation of one's own audio system the absolute truth.

 

[DBB1]

The last few comments sum up our audiophile struggles well. Probably all of us have analytic personalities and at times can be obsessive. I know that only a fraction of my enjoyment of music comes from perceived sound quality. Most is mood. We would all benefit from zen efforts to turn off the critical listening part of our brain. Yet without question sound quality enhances our experience of music. I try to be a critical audiophile only occasionally.