MSB Select II arrival

[microstrip]

Well, it could be questioned that without active device there will be lack of oomph and thinness ...kind of like the common criticism leveled at resistor based passive "preamp"...afterall the dac is nothing more or less than a very precise resistor ladder. It is often the reason that people put up with "colored" active preamps because going straight from the DAC through a passive volume control loses something. Not saying this is the case but i could see how that might be argued.

Sorry, but a DAC is is a lot more than a resistor ladder. You need a current source - that is not ideal, and the commutators. Beside it as a logic system feeding the commutators. Except for the DACs that use well known ICs that have great datasheets there is a lot of unknowns concerning most R2R DACs. Trinity also used the DAC chips without further amplification just driving a resistor. DIY people including me did that mod decades ago - suppressing the current converter and the filter/buffer of the DAC. And yes, colorations is the main reason behind this bobby - designers create them , we choose and combine equipment to get those we prefer.

 

[microstrip]

the best continuousness and flow I've heard with digital is the MSB Select II at show demos with the 2 power supplies and the 33 Femto clock. these attributes are not everything, but they are what brought me back multiple times and caused me to choose it. and in that sense it's in analog territory, maybe not quite at the highest analog level. I hear it in other digital too, just not to that degree. in the GG I heard it, but likely the Nagra HD (through XLR and with added output transformers and the best SGM HQP settings) at it's best was a little bit better at that......that was what I heard in my room.

tubes, more tubes, DHT tubes, etc. etc. do not in themselves produce this attribute. it's in the data and how it's presented. just my view of how things seem to work. the stylus tells us something and the rest of the vinyl playback links do other things, but the limitation is the data.

Mike,

Do you feel that all LPs digitally processed lack continuousness and flow?

 

[Al M.]

I could not have said it better myself.

the MSB Select II......has.....nothing.....after the dac chip. that's not the whole story. but like you say.....it matters a lot. there is no analog output/gain stage of any kind. the dacs themselves have enough gain to not only be line level but to drive headphones or amplifiers directly.

is 'nothing' better than lots of something?

to my ears of what I've heard, yes.

Not quite the same, but from RH's review of the Schiit Yggdrasil DAC:

"The analog output stage is fully balanced and built from JFETs rather than op-amps. It is simply a unity-gain buffer. No gain is required in the output stage because the Analog Devices DAC has a built-in current-to-voltage converter, and outputs a voltage at the appropriate level."

***

Of course, the output is fixed, so you still need a preamp (I have a buffered passive one).

 

[microstrip]

Brad, obviously we experience things very differently.

That's o.k., diversity is the spice of human existence.

You said it all in your two short sentences.

 

[Mike Lavigne]

Mike,

Do you feel that all LPs digitally processed lack continuousness and flow?

not at all, they are all over the board in their nuance. but they fall short of the top level of analog sourced Lps. so they don't lack, just come up a little short of whole.

and there are quite a few where a digital recording is mastered to tape, where it's part of the way there.

as always; the whole equation of the recording trumps other factors in total musical satisfaction, including our emotional connection to the music.

like comparing direct to disc to tape based. there are generally differences, but the best DTD are another level.

 

[microstrip]

Not quite the same, but from RH's review of the Schiit Yggdrasil DAC:

"The analog output stage is fully balanced and built from JFETs rather than op-amps. It is simply a unity-gain buffer. No gain is required in the output stage because the Analog Devices DAC has a built-in current-to-voltage converter, and outputs a voltage at the appropriate level."

***

Of course, the output is fixed, so you still need a preamp (I have a buffered passive one).

Just to clarify that it is not related at all with what MSB or others claim. Discrete buffers with JFETs have been used since long in many DACs.

 

[Mike Lavigne]

where digital verses analog based vinyl really jumps out and bites you is with the current crop of Reference Recordings 45rpm vinyl.

choose one that has an analog master and previous Lp or tape where they were analog based. it's almost criminal what they've done in these cases. 'Nojima Plays Liszt' makes me cry. it's a great CD, and a fantastic Lp (the tape is astonishing). the 45rpm, which has the potential to be stunning, is a turd. the CD is better. with 45rpm there is no place to hide....good or bad.

 

[microstrip]

(...) and there are quite a few where a digital recording is mastered to tape, where it's part of the way there. (...)

Some one should produce tapes from top DXD /DSD recordings using the MSB and DCS top DACs and the we would comfortably listen in our machines ...

as always; the whole equation of the recording trumps other factors in total musical satisfaction, including our emotional connection to the music.

like comparing direct to disc to tape based. there are generally differences, but the best DTD are another level.

Here I agree - analog wins. Nothing could "beat" a few great DTD in my system. But the DTSs I have listened are much less complex and easily enjoyable that many recordings I have in digital.

 

[853guy]

I could not have said it better myself.

the MSB Select II......has.....nothing.....after the dac chip. that's not the whole story. but like you say.....it matters a lot. there is no analog output/gain stage of any kind. the dacs themselves have enough gain to not only be line level but to drive headphones or amplifiers directly.

is 'nothing' better than lots of something?

to my ears of what I've heard, yes.

Hello Mike, hello gents...

It appears we’re veering closely toward familiar territory ala SUTs, output stages, transformers and pre-amps. I’ve read many reports over the years how all the aforementioned devices serve to decrease verisimilitude to the original recording (or some such parlance), and result in measurements in which noise, distortion and frequency response are objectively inferior.

That I’ve had too many experiences in which my subjective perception has preferred their inclusion - not always for reasons of sonics, but almost always for reasons of psychological and emotional engagement - leads me to believe ‘nothing’ is not necessarily better than ‘a lot of something’ not only because it seems prudent to deal with these things on a component-by-component and system-by-system basis, but because one person’s transparency is just as often another man’s sterility, (and one man’s flow and ease is another man’s stupor and lack of vitality, etc, etc), and it will depend explicitly on what that ’something’ is.

That eschewing the inclusion of a SUT, an output stage, a transformer and/or a pre-amp may confer an audible benefit to one or more aspects of reproduction does not necessarily mean benefits will be experienced across all aspects of reproduction, and in many cases will come with a corresponding down-side which is masked to a degree by the former (but sometimes discovered only later once contextualized by the subject in their own system, leading to revisions of previous absolute truths, and in some cases, another component upgrade). That many will still be happy with the gain they’ve made in one area at the expense of losses in another only highlights the reality we have preferences weighted toward certain variables, but nevertheless, that those losses exist at all will be a compromise too far for some.

Of course, that ‘a lot of something’ can be in-and-of-itself no better than ‘nothing’ will again be best evaluated on a system-by-system basis relative to the preferences of the subject and their willingness to experience any large gain at the potential expense of a lot of little losses. In that regard, we’re likely to diverge on what we’re willing to sacrifice in order to gain that which we do not yet already possess, as the purchase history of many of us attests to.

Be well,

853guy

 

[Narayan]

Fair enough. I will try to hear the Select DAC sometime soon and give my feedback. I have heard a game changer recently but it was not the Lampi...nor the Select (but as i said...haven't heard it).

Can you tell us a bit more about this DAC? Thanks.

 

[DaveC]

I disagree, especially if it is transformer coupled and care has been taken to have a very stable power. Also, the SRPP circuit, which works with two triodes (or a single double triode tube) will give low impedance, gain and low distortion. It is possible to generate several volts of output with < 0.1% distortion this way (especially transformer coupled) and that will be primarily inaudible (at that level) 2nd harmonic.

As you mentioned SRPP isn't a single triode anymore. Other circuits that produce better results are white follower, aikido, mu follower and others that load the single triode with an identical triode to achieve lower distortion in the form of error correction... the amplifying triode is loaded with the cathode of the 2nd triode. It also helps to buffer the voltage gain with a similar type of follower circuit to lower output impedance as a voltage gain section doesn't do as well directly driving anything requiring much current. This isn't as practical with DHT power tubes, and it explains why I think some feel a tube output section in a DAC isn't the best solution. I do understand it's advantages and that it can sound wonderful, but uncolored it is not. In a system that's otherwise close to perfect, i.e. uses speakers and electronics that are extremely accurate and low distortion, like Mike's system, these things are going to be more noticeable.

I've experimented a lot with the driver section in my SET amp, and I don't feel like a single triode, even if it's supplemented with a SS constant current source to linearize it further, can produce excellent results as a driver section. It takes a circuit like I describe above to make the amp sound neutral, fast, and not "tubey". I've heard SS amps that are subjectively slower and mushier... "tubier"... vs my SET amp.

 

[DaveC]

I could not have said it better myself.

the MSB Select II......has.....nothing.....after the dac chip. that's not the whole story. but like you say.....it matters a lot. there is no analog output/gain stage of any kind. the dacs themselves have enough gain to not only be line level but to drive headphones or amplifiers directly.

is 'nothing' better than lots of something?

to my ears of what I've heard, yes.

It has to, an output section is simply built into the DAC chip. Not saying this is bad, it's probably just the opposite, but there's no way an amp of some sort isn't in the picture here.

 

[microstrip]

I could not have said it better myself.

the MSB Select II......has.....nothing.....after the dac chip. that's not the whole story. but like you say.....it matters a lot. there is no analog output/gain stage of any kind. the dacs themselves have enough gain to not only be line level but to drive headphones or amplifiers directly.

is 'nothing' better than lots of something?

to my ears of what I've heard, yes.

(...)

Of course, that ‘a lot of something’ can be in-and-of-itself no better than ‘nothing’ will again be best evaluated on a system-by-system basis relative to the preferences of the subject and their willingness to experience any large gain at the potential expense of a lot of little losses. In that regard, we’re likely to diverge on what we’re willing to sacrifice in order to gain that which we do not yet already possess, as the purchase history of many of us attests to.

Be well,

853guy

After reading these two posts, I feel we are underestimating digital. Who can say doing 'nothing' when addressing modern digital? There is a lot that is done before entering the DAC chip. IMHO what is happening is that as engineers and designers are better understanding the high-end they are using the resources of high power digital processing to manipulate the subjective sound of their products, creating the sound we prefer - in some aspects moving the part that was carried in the analogue domain in great DACs of the past and current ones to the digital reign. It seems that sometimes they need precision chips, usually custom developed or developed in house, with technical requirements beyond the commonly available chips.

DCS significantly improved the sound of their DACs with a firmware upgrade (v2), Devialet also did it during many years, to the point that they have now a product that most listeners, including me, consider excellent sounding.

Times are changing very fast - as digital designers are managing to merge knowledge coming from psychoacoustics with their powerful algorithms and experience in high-end design we can wait for great products and continuous evolution.

Audiophiles are prone to be victims of the school of minimalism. We have read so many times that the less the better, that sometimes our actions and thoughts become unconsciously driven by the "nothing is better" concepts.

BTW, "nothing after the DAC chip" adds very little to the story, unless we know the details of the DAC chip.

 

[wisnon]

I think its ANALOG that is being underestimated here. Digital is a fallacy...it does not exist...only analog does. What we call digital is electrical representation of sound waves... a code to instruct a device to produce the actual sound waves.

The digital part of the processing is just better encoders and decoders, but all done in the physical realm. Therefore for me it stands to reason that the ANALOG part of the Dac will be of extreme importance, and power quality and circuit simplicity and elegance should assume primacy. To me its more of a do no harm venture to ensure true fidelity.

 

[FrantzM]

Very interesting debate.

What we should not forget is that circuits became more complicated because of the limitations of the simpler circuits or topologies. We can go on forever glorifying the virtues of Triodes and especially DHT. They have severe limitations which called for more complicated ( yes, that word) devices and circuitry. It is another dear fallacy to think that a Digital DAC chip is simplicity itself. It is not,these are for the most part intrically and bewilderingly complex assembly of miniaturized components. Sometimes numbering over thousands of parts.. That is what chips mostly are. They're not "simple" in most instances.
It seems to dawn on many that Digital is capable of conveying music in all its nuances. Well IMO it has been the case for many years. I tend to believe from all the Kudos that the MSB Select is great and for those people who've heard one of the top DACs on the market...
However great that the MSB might be there are many others, some of these not as dear ... as microstrip brushed on, a serious and unbiased audition (is that possible? ) of something like the Devialet could open many ears and perhaps hearts... The Devialet BTW is far from simple it is complex , complicated and extremely good sounding.
Now back to the MSB which in spite of all the resistance seems to have bridged many (all ? ) gaps and becoming the goto listening apparatus for more than one ...
Yes people! Digital can be that kind of good.. has been for a while

 

[microstrip]

I think its ANALOG that is being underestimated here. Digital is a fallacy...it does not exist...only analog does. What we call digital is electrical representation of sound waves... a code to instruct a device to produce the actual sound waves.

The digital part of the processing is just better encoders and decoders, but all done in the physical realm. Therefore for me it stands to reason that the ANALOG part of the Dac will be of extreme importance, and power quality and circuit simplicity and elegance should assume primacy. To me its more of a do no harm venture to ensure true fidelity.

We are not debating elementary digital electronics teaching - then the first thing to say is that all logic signals are represented by an analog signal and give some examples, usually starting with TTL!

Anyway modern encoders have lots of digital processing - and sorry not simple at all! There is not such think as a 24 bit flash ADC's! And most DACs also have a lot of processing. It is not a fallacy, it exists!

 

[morricab]

As you mentioned SRPP isn't a single triode anymore. Other circuits that produce better results are white follower, aikido, mu follower and others that load the single triode with an identical triode to achieve lower distortion in the form of error correction... the amplifying triode is loaded with the cathode of the 2nd triode. It also helps to buffer the voltage gain with a similar type of follower circuit to lower output impedance as a voltage gain section doesn't do as well directly driving anything requiring much current. This isn't as practical with DHT power tubes, and it explains why I think some feel a tube output section in a DAC isn't the best solution. I do understand it's advantages and that it can sound wonderful, but uncolored it is not. In a system that's otherwise close to perfect, i.e. uses speakers and electronics that are extremely accurate and low distortion, like Mike's system, these things are going to be more noticeable.

I've experimented a lot with the driver section in my SET amp, and I don't feel like a single triode, even if it's supplemented with a SS constant current source to linearize it further, can produce excellent results as a driver section. It takes a circuit like I describe above to make the amp sound neutral, fast, and not "tubey". I've heard SS amps that are subjectively slower and mushier... "tubier"... vs my SET amp.

Again, I did not say a single triode that is not at least loaded into an output transformer. I would not take a plate loaded triode out through a cap for example...the output impedance is normally too high. Load it with a good output transformer though and then this works like a charm with extremely low distortion. If you don't need gain then a zero gain buffer might work but with gain then you need something else or transformer loading.

 

[morricab]

After reading these two posts, I feel we are underestimating digital. Who can say doing 'nothing' when addressing modern digital? There is a lot that is done before entering the DAC chip. IMHO what is happening is that as engineers and designers are better understanding the high-end they are using the resources of high power digital processing to manipulate the subjective sound of their products, creating the sound we prefer - in some aspects moving the part that was carried in the analogue domain in great DACs of the past and current ones to the digital reign. It seems that sometimes they need precision chips, usually custom developed or developed in house, with technical requirements beyond the commonly available chips.

DCS significantly improved the sound of their DACs with a firmware upgrade (v2), Devialet also did it during many years, to the point that they have now a product that most listeners, including me, consider excellent sounding.

Times are changing very fast - as digital designers are managing to merge knowledge coming from psychoacoustics with their powerful algorithms and experience in high-end design we can wait for great products and continuous evolution.

Audiophiles are prone to be victims of the school of minimalism. We have read so many times that the less the better, that sometimes our actions and thoughts become unconsciously driven by the "nothing is better" concepts.

BTW, "nothing after the DAC chip" adds very little to the story, unless we know the details of the DAC chip.

Sorry, despite the so called "upgrades" the Devialet is not a really good sounding device...nope...its an OK sounding device...nothing more...oh and the phonostage built-in sucks.

 

[wisnon]

We are not debating elementary digital electronics teaching - then the first thing to say is that all logic signals are represented by an analog signal and give some examples, usually starting with TTL!

Anyway modern encoders have lots of digital processing - and sorry not simple at all! There is not such think as a 24 bit flash ADC's! And most DACs also have a lot of processing. It is not a fallacy, it exists!

Some times to move fwd, we have to go back to basics.
Sometimes we cant see the forest for the trees.

 

[wisnon]

Sorry, despite the so called "upgrades" the Devialet is not a really good sounding device...nope...its an OK sounding device...nothing more...oh and the phonostage built-in sucks.

Hahahahaha

I agree.