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

[exupgh12]

In today's digital audio, there are two essential components that affect the sound quality of your music: the source (usually a Music Server) and the DAC.

So, how would you split your funds between your source and DAC to maximize results, is there any rule of thumb the forum members act by?

 

[TDX]

Both are important so maybe half half? Towards DAC a bit more.

 

[Young Skywalker]

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”. I expect that devotees of single-bit DACs would enjoy the same isolation advantages with the Playback Designs DAC and matching transport via their fibre optical connection.

 

[Sablon Audio]

I am an outlier here in that I use a Taiko Extreme to feed a far cheaper Lampizator Baltic. The latter is however a delight and highly transparent (with some special tubes) to changes upstream. I believe the focus should be on maximising signal integrity at the source because it can never be recreated further downstream.

 

[exupgh12]

Both are important so maybe half half? Towards DAC a bit more.

fair enough and logical, but there is a limit to using this formula - if you have DAC such as WADEX Reference, MSB Select 2, or Kassandra Reference MK2, those DACS cost much more than any Server.

 

[exupgh12]

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”. I expect that devotees of single-bit DACs would enjoy the same isolation advantages with the Playback Designs DAC and matching transport via their fibre optical connection.

"garbage in garbage out"

but even if the galvanic isolation protects from noise, don't you think that if your transport is all ready feeding your DAC with nosy packets then the galvanic\fiber network will pass them through, just not adding other noise to the traffic?

 

[Young Skywalker]

"garbage in garbage out"

but even if the galvanic isolation protects from noise, don't you think that if your transport is all ready feeding your DAC with nosy packets then the galvanic\fiber network will pass them through, just not adding other noise to the traffic?

No, I believe that it is 100% effective at eliminating any noise prior to transmission of the synchronous but-perfect datastream into the sensitive DAC chassis. That’s how galvanic isolation works. By extension, I also believe that no other bit-perfect data source would make any perceptible sonic difference under these test conditions.

I expect many to disagree and I am completely comfortable with that. What a boring existence it would be if we all had the same intellectual and emotional response to every external stimuli.

 

[TDX]

fair enough and logical, but there is a limit to using this formula - if you have DAC such as WADEX Reference, MSB Select 2, or Kassandra Reference MK2, those DACS cost much more than any Server.

wadax has a same expansive server ready for you.

 

[audiobomber]

No, I believe that it is 100% effective at eliminating any noise prior to transmission of the synchronous but-perfect datastream into the sensitive DAC chassis. That’s how galvanic isolation works. By extension, I also believe that no other bit-perfect data source would make any perceptible sonic difference under these test conditions.

I expect many to disagree and I am completely comfortable with that. What a boring existence it would be if we all had the same intellectual and emotional response to every external stimuli.

"Galvanic isolation" is not to do with comfort or emotional response, it is purely technical.

"While you may see discussion of "galvanic isolators,” by definition galvanic isolation means only that DC is blocked. Galvanic isolators—be they transformers or any form of capacitive or optical digital logic isolator—still pass AC (alternating current). The transformers in Ethernet ports are galvanic isolators—they block DC—they also block low-source-impedance AC leakage. However, high-source-impedance leakage sails right through them."
John Swenson

https://cdn.shopify.com/s/files/1/0660/6121/files/UpTone-J.Swenson_EtherREGEN_white_paper.pdf?v=1583429386

 

[Republicoftexas69]

I think that they are both critical as every thing in the chain contributes to the quality of the sound or end product. I have 60-40 split on spend DAC, Streamer NAS.

 

[abeidrov]

It probably depends on the quality of your dac inputs and how it connects to a server. In my case the server is more expensive, than my dac. It may be unusual, but the new server (Aurender) has made a bigger impact on sound than MSB Reference compared to my Lampi GG. So 50/50 would be a good starting point.

 

[audiobomber]

In today's digital audio, there are two essential components that affect the sound quality of your music: the source (usually a Music Server) and the DAC.

So, how would you split your funds between your source and DAC to maximize results, is there any rule of thumb the forum members act by?

I would say there are three components in my system: network, renderer and DAC. All are important, with priority to the DAC, then by proximity to the DAC in the chain.

In other words, DAC first, then renderer, then the network feeding the renderer (NAS, router, switches, cabling, etc.).

 

[exupgh12]

No, I believe that it is 100% effective at eliminating any noise prior to transmission of the synchronous but-perfect datastream into the sensitive DAC chassis. That’s how galvanic isolation works. By extension, I also believe that no other bit-perfect data source would make any perceptible sonic difference under these test conditions.

I expect many to disagree and I am completely comfortable with that. What a boring existence it would be if we all had the same intellectual and emotional response to every external stimuli.

Ron had a video interview with Antipodes owner and CTO, as I remember he explained what are the benefits to taking care of the signal even before it sent to galvanic network or revolved.

I have MSB Premier with Pro USB\ISl, when I connect the usb side to my Roon core computer (Mac mini) the sound inot as good as when i connect the usb cable to my Medco server - all ought in the two cases the pro usb is connected to the dac via the optical side.

have you ever tried connecting your dac using audiophil server?

My home network between the router and the sotm switch that connects to melco server is optical.
The computer and the Melco server are connected to different switches via Cardas ethernet cable.

 

[exupgh12]

wadax has a same expansive server ready for you.

The Wadex Reference server is an exceptional case. No other server that I know of is priced as high as the Atlantis Reference. Not even Wadex's own Atlantis server costs as much as their Atlantis DAC.

 

[exupgh12]

I am an outlier here in that I use a Taiko Extreme to feed a far cheaper Lampizator Baltic. The latter is however a delight and highly transparent (with some special tubes) to changes upstream. I believe the focus should be on maximising signal integrity at the source because it can never be recreated further downstream.

That’s really is unusual, but in honesty I can’t argue with any of the approaches.

 

[exupgh12]

It probably depends on the quality of your dac inputs and how it connects to a server. In my case the server is more expensive, than my dac. It may be unusual, but the new server (Aurender) has made a bigger impact on sound than MSB Reference compared to my Lampi GG. So 50/50 would be a good starting point.

The lampi and the MSB have different sound qualities, did both were connected trough usb, or did you use different connections?

 

[abeidrov]

The lampi and the MSB have different sound qualities, did both were connected trough usb, or did you use different connections?

Both were connected through USB.

 

[jbrrp1]

No, I believe that it is 100% effective at eliminating any noise prior to transmission of the synchronous but-perfect datastream into the sensitive DAC chassis. That’s how galvanic isolation works. By extension, I also believe that no other bit-perfect data source would make any perceptible sonic difference under these test conditions.

I expect many to disagree and I am completely comfortable with that. What a boring existence it would be if we all had the same intellectual and emotional response to every external stimuli.

It might be interesting for you to challenge those beliefs with a trial of a high quality server. Others have found them to make important improvements in sound quality feeding MSB's best. If you can arrange a "try before you buy" approach, you've got nothing to lose and possibly something to gain.

 

[Young Skywalker]

It might be interesting for you to challenge those beliefs with a trial of a high quality server. Others have found them to make important improvements in sound quality feeding MSB's best. If you can arrange a "try before you buy" approach, you've got nothing to lose and possibly something to gain.

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?

 

[Willgolf]

Great topic and one discussed often on WBF. Both are important. I can not say it is 50/50 either way. I really think it depends on the quality of your DAC and Server.

In my case, I weight it 60% server / 40% DAC. A couple of years ago, I purchased a Lampi Pacific to replace a Lumin X-1. I was using a Synology NAS with a Lumin Mini when I got the Pacific. To be honest, I was disappointed with the sound of the Pacific. I was expecting leaps and bounds from the Lumin X-1. I knew the Lumin U1 mini was decent but not good enough so I began researching servers.

After a long study, I ended up going with a Lucas Audio LDMS. When I hooked it up to the Pacific for the first time, I just said WOW. I could not believe how much better my music sounded.

I have since upgraded to the Horizon and an updated LDMS. The LDMS just feeds the Horizon with so much detail and clarity. I will be getting a AC Kassandra 2 this fall so I will be interested to see how my music changes.

BTW, if anyone in the USA is interested in my other LDMS send me a private email. Lucas D owns it but I still have it. He thought rather than sending it back to him it could be sold to someone in the USA. It is an outstanding unit.