Server vs. DAC - How you split your funds between the two?

[matthias]

98% to the DAC and 2% to the component that sends music to it over my home network. I should qualify this by saying that if I didn’t have the galvanic isolation of the MSB Reference Digital Director, I would allocate more financial resources towards minimising noise from the digital “transport”.

Would you consider the Digital Director as part of the DAC?
IMO, I would clearly consider it to be part of the server, it "serves" the DAC to make a better conversion.
Sorry, so your 98% to the DAC do not make sense at all

Matt

 

[Young Skywalker]

Would you consider the Digital Director as part of the DAC?
IMO, I would clearly consider it to be part of the server, it "serves" the DAC to make a better conversion.
Sorry, so your 98% to the DAC do not make sense at all

Matt

That’s a really good question, Matt. I am happy to concede that it is part of the server in this instance. Certainly I justified the purchase of the Digital Director on the basis that it would improve any digital source for similar money to a fancy music server.

Thus, my revised ratio of DAC to server is closer to 70:30.

 

[matthias]

That’s a really good question, Matt. I am happy to concede that it is part of the server in this instance. Certainly I justified the purchase of the Digital Director on the basis that it would improve any digital source for similar money to a fancy music server.
Thus, my revised ratio of DAC to server is closer to 70:30.

I am just curious if someone made this comparison:
Select Digital Director vs Taiko Extreme into the same MSB DAC

Matt

 

[Young Skywalker]

I am just curious if someone made this comparison:
Select Digital Director vs Taiko Extreme into the same MSB DAC

Matt

Another interesting question, Matt.

The Reference Digital Director and the Taiko Extreme are similar money here in Australia. If money was no object I would have both and call it a day, as many here on WBF have. More power to them.

If there is a perceptible/significant improvement to be had by running the Taiko into the Digital Director (presumably over USB) vs my Roon Nucleus Plus over my home network into the MSB Network Renderer module, then I would be interested. Having said that, I find nothing objectionable about my current digital setup which is as enjoyable as my vinyl playback rig. I couldn’t say that prior to getting the Digital Director.

 

[exupgh12]

I don’t have any funds for more hifi upgrades right now so it wouldn’t be appropriate to waste a dealer’s time on a hypothetical. However, if there is someone in Sydney, Australia who wants to educate me with a no obligation demonstration as to the sonic improvements that can be had from a lower noise digital transport in the context of an MSB Digital Director and Network Renderer input module, then I am willing.

For those who have already conducted this experiment, what percentage increase would you ascribe to the lower noise digital transport? 1-2%,? 2-5%? 5-10%? More than 10%? Is this improvement evident on every single recording?

I think it won't be responsible predicting how much different you will hear on your setup once the using dedicated music server, my guess is the different will be noticeable that you wouldn't go back.
for me, the difference was notable enough to jump in the water and order my server.

 

[exupgh12]

I am just curious if someone made this comparison:
Select Digital Director vs Taiko Extreme into the same MSB DAC

Matt

That's interesting question, since the Digital Director is still early in the market I doubt if any one done such test.

 

[matthias]

That's interesting question, since the Digital Director is still early in the market I doubt if any one done such test.

It shouldn't be so difficult as there should be a few members here on WBF who have both a Taiko Extreme and a Select DD.

Matt

 

[GSOphile]

Would you consider the Digital Director as part of the DAC?
IMO, I would clearly consider it to be part of the server, it "serves" the DAC to make a better conversion.
Sorry, so your 98% to the DAC do not make sense at all

Matt

Interesting question. But how (other than noise reduction aspects from separation) does this implementation differ from DACs which incorporate filters?

 

[matthias]

Interesting question. But how (other than noise reduction aspects from separation) does this implementation differ from DACs which incorporate filters?

A server has a digital input and a digital output, a DAC has a digital input and an analog output.
As a stand alone unit the Digital Director doesn't fit into the category of a DAC, there is no DA conversion involved, its output is pure digital.
So in the context of the OP question it makes much more sense to put the DD into the server category, but YMMV.

Matt

 

[F208Frank]

Anyone put the muon by network acoustics prior to server (usually MUON with the ethernet switch) going to their msb digital director?

For playback users, anyone use a switch + muon prior to server and then going to their mps-x (renderer) or mpt-8 (renderer) going to their mpd8 via PLINK connection?

I wonder if these optimized connections such as PLINK will lessen the potency of the ethernet filter + muon prior to server because so far the results are significant.

 

[godofwealth]

I have found as a general principle never to invest in an all-in-one solution (speakers, amp, preamp, DAC, server, CD player etc.). I keep multiple references so I can alternative as and when I please. If you use a tubed DAC, e.g., any Lampi, you get a tube coloration. Nothing wrong with that. I love my Pacific! But, it‘s a coloration, nonetheless. All tubes impart a sort of rosy cheeked full bloomed sound, one we all love. My ARC Ref 6SE has that in spades. So, the only way to guard against that is to keep an alternate set of references.

For example, I use a Mola Mola preamp (with built in DAC, which is just the Tambaqui), with their solid state Kaluga monoblocks. The sound is about as far as you can get from the warm bloomy syrupy Lampi sound, so it’s always an interesting contrast. But, that’s the fun part of keeping multiple references, You learn to understand the colorations induced by various design choices.

It’s the same with speakers. I use horns (Klipsch La Scala), electrostatics (Quads, 3 pairs!), and dynamic speakers (Harbeth Monitor 40.1s). Each has a particular sound, some unique strengths and some obvious limitations.

So, for a server, I would do the same, Keep multiple, if you can afford to do that. Generally, I’d recommend investing in a cheap Intel NUC fan less chassis (which I do), along with something more expensive. Compare, and see if you really get your money’s worth with the fancier solution. It takes a while to do long term comparisons, but if you take the time, you learn a lot!

 

[abeidrov]

I have found as a general principle never to invest in an all-in-one solution (speakers, amp, preamp, DAC, server, CD player etc.). I keep multiple references so I can alternative as and when I please. If you use a tubed DAC, e.g., any Lampi, you get a tube coloration. Nothing wrong with that. I love my Pacific! But, it‘s a coloration, nonetheless. All tubes impart a sort of rosy cheeked full bloomed sound, one we all love. My ARC Ref 6SE has that in spades. So, the only way to guard against that is to keep an alternate set of references.

For example, I use a Mola Mola preamp (with built in DAC, which is just the Tambaqui), with their solid state Kaluga monoblocks. The sound is about as far as you can get from the warm bloomy syrupy Lampi sound, so it’s always an interesting contrast. But, that’s the fun part of keeping multiple references, You learn to understand the colorations induced by various design choices.

It’s the same with speakers. I use horns (Klipsch La Scala), electrostatics (Quads, 3 pairs!), and dynamic speakers (Harbeth Monitor 40.1s). Each has a particular sound, some unique strengths and some obvious limitations.

So, for a server, I would do the same, Keep multiple, if you can afford to do that. Generally, I’d recommend investing in a cheap Intel NUC fan less chassis (which I do), along with something more expensive. Compare, and see if you really get your money’s worth with the fancier solution. It takes a while to do long term comparisons, but if you take the time, you learn a lot!

Syrupy Lampi sound? Really? I’ve compared my Lampi to many dacs including MSB Reference. No syrupy sound in my system, unless you use 300B tubes.

 

[jbrrp1]

As a previous Pacific owner, I had the same reaction: what crazy tube combo are you using that renders a syrupy tube sound?!

 

[Mike Lavigne]

wadax has a same expensive server ready for you.

true that.

and turns out that the interface between the dac and server is a severe limitation to how far digital can go. which then causes the synergy between the dac and server to be very significant for that interface to be optimized. so a single brand dac and server might be ideal.

Wadax uses a proprietary optical interface called the Akasa to eliminate noise and optimize synergy. when i compared using a USB interface to the optical Akasa the difference in the realism of the music was significant. i also compared the Taiko Extreme server directly to the Wadax server, both into the Wadax dac, both with USB, and the Wadax server was superior sounding by a good margin. to be fair this was now 18 months ago in early 2022, and the Extreme has advanced since then. the Wadax has had upgrades too.

but ultimately the interface + one brand synergy is hard to beat. at the top of the food chain.

generally i agree that server quality is significant + interface quality and network quality. all three limit what the dac has to work with. maybe like pressing quality matters much with vinyl.

but my opinion is also that a good server, good USB cable, and good network through put gets you most of the way there. not sure how much better the tip top level of these things are than 'good' with digital.

 

[exupgh12]

I have found as a general principle never to invest in an all-in-one solution (speakers, amp, preamp, DAC, server, CD player etc.). I keep multiple references so I can alternative as and when I please. If you use a tubed DAC, e.g., any Lampi, you get a tube coloration. Nothing wrong with that. I love my Pacific! But, it‘s a coloration, nonetheless. All tubes impart a sort of rosy cheeked full bloomed sound, one we all love. My ARC Ref 6SE has that in spades. So, the only way to guard against that is to keep an alternate set of references.

For example, I use a Mola Mola preamp (with built in DAC, which is just the Tambaqui), with their solid state Kaluga monoblocks. The sound is about as far as you can get from the warm bloomy syrupy Lampi sound, so it’s always an interesting contrast. But, that’s the fun part of keeping multiple references, You learn to understand the colorations induced by various design choices.

It’s the same with speakers. I use horns (Klipsch La Scala), electrostatics (Quads, 3 pairs!), and dynamic speakers (Harbeth Monitor 40.1s). Each has a particular sound, some unique strengths and some obvious limitations.

So, for a server, I would do the same, Keep multiple, if you can afford to do that. Generally, I’d recommend investing in a cheap Intel NUC fan less chassis (which I do), along with something more expensive. Compare, and see if you really get your money’s worth with the fancier solution. It takes a while to do long term comparisons, but if you take the time, you learn a lot!

I used to be like that, but over time I realized I’m compromising, why investing resources in more then one of the same product when you can invest in one, and better product then the alternative when splitting resources.

 

[exupgh12]

true that.

and turns out that the interface between the dac and server is a severe limitation to how far digital can go. which then causes the synergy between the dac and server to be very significant for that interface to be optimized. so a single brand dac and server might be ideal.

Wadax uses a proprietary optical interface called the Akasa to eliminate noise and optimize synergy. when i compared using a USB interface to the optical Akasa the difference in the realism of the music was significant. i also compared the Taiko Extreme server directly to the Wadax server, both into the Wadax dac, both with USB, and the Wadax server was superior sounding by a good margin. to be fair this was now 18 months ago in early 2022, and the Extreme has advanced since then. the Wadax has had upgrades too.

but ultimately the interface + one brand synergy is hard to beat. at the top of the food chain.

generally i agree that server quality is significant + interface quality and network quality. all three limit what the dac has to work with. maybe like pressing quality matters much with vinyl.

but my opinion is also that a good server, good USB cable, and good network through put gets you most of the way there. not sure how much better the tip top level of these things are than 'good' with digital.

Mike, I agree with your approach, synergy between components can lead to better results then “each on his own” components (Wadex is great example of that approach), unfortunately my DAC manufacturer doesn’t offer server source Like Wadex.

 

[tomas329]

I don’t have any funds for more hifi upgrades right now so it wouldn’t be appropriate to waste a dealer’s time on a hypothetical. However, if there is someone in Sydney, Australia who wants to educate me with a no obligation demonstration as to the sonic improvements that can be had from a lower noise digital transport in the context of an MSB Digital Director and Network Renderer input module, then I am willing.

For those who have already conducted this experiment, what percentage increase would you ascribe to the lower noise digital transport? 1-2%,? 2-5%? 5-10%? More than 10%? Is this improvement evident on every single recording?

I recently went through the process of comparing servers to add to my setup and, for me, i find that even with the digital director the server did make a perceivable difference over just a roon nucleus. i have the select 2 and select director and i ended up going with the taiko after hearing a demo of their xdms. i would say it added 10%+ to the experience for me. i was unable to test it without my director as that's a whole process as apparently switching the filter option on and off doesn't remove the benefits of the galvanic isolation. FWIW the director also adds another level of performance so given an option between the two i'm not sure which i would chose especially given the price difference.

 

[Macattack]

The OP’s question is something I never considered. Simply been developing my system from point to point. After an upgrade to my server to be completed this month, I will be almost at 1.5 : 1 Server cost to : DC cost.

But I can live with that.

 

[Young Skywalker]

I recently went through the process of comparing servers to add to my setup and, for me, i find that even with the digital director the server did make a perceivable difference over just a roon nucleus. i have the select 2 and select director and i ended up going with the taiko after hearing a demo of their xdms. i would say it added 10%+ to the experience for me. i was unable to test it without my director as that's a whole process as apparently switching the filter option on and off doesn't remove the benefits of the galvanic isolation. FWIW the director also adds another level of performance so given an option between the two i'm not sure which i would chose especially given the price difference.

Thanks for sharing your experiences, Tomas.

I guess the question then becomes, does the Taiko Extreme computer provide a 10% sonic improvement compared to the Roon Nucleus Plus because:

a) it has a significantly lower electrical noise footprint and this pays dividends despite the claimed total electrical isolation of the MSB Technology Digital Director?

or

b) because it is performing digital signal processing that is sonically pleasing, i.e., the data stream being sent to the DAC is no longer bit-perfect? [I think this one has been debunked]

or

c) confirmation bias is playing a role in sonic perception?

My background is in science and I am assured by double blind testing (DBT) regimes in drug trials. Normally, I wouldn't feel the need for a DBT in a field as subjective as music playback, except when the cost of any claimed improvement is very high (46,000 Australian dollars in my market) AND there is no rational explanation for it.

 

[tomas329]

I wish i was able to do the DBT during my home demo but time was limited. i agree there's a high likelihood for confirmation bias especially after reading through these forums and seeing the praise the takio receives. i'm generally a skeptical person and with the director i assumed every server would be equal but i didn't feel that way after demoing a few servers and then finally the taiko. i know you mentioned funds not being available at the moment but if they do become available and you get the itch it could be a fun experiment. Once the taiko is delivered i'll be able to do the DBT and hopefully still be able to confirm my preference for the taiko. I also want to try with and without the director to quantify where money is better spent but with small children its never so simple. i just wanted to pass along my personal experience and i'll come back once the DBT is done.