The Hugo of streaming?

This morning I removed the LT3045 feeding my ultraRendu (uR). This had the effect I expected, body may have increased but what definitely did increase was bloat in the bass, which also masked detail.

I created two cables and inserted TWO 7.1v 1A LT3045s. This had the effect I HOPED. Body definitely increased, especially worthwhile for rock & large scale classical. Even for other music the dynamics for such thing as percussion, string plucking etc is improved.

M

Hi Brilliant,

If by the Ethernet Bridge you mean using copper - fibre - copper I do == FMC > FMC.

Brilliant/NBFP

I don't use wireless inside my network, security. I do use it on my firewall but it has only one route = outbound.

With respect to the RP3, I have closed down everything apart from that which is needed for Logitech Media Server & the USB connection to the StarTech hosted USB sticks. My suspicion would be that in turning in WiFi that I would introduce noise, but if I did use WiFi in this way I would try it.

WRT my connecting the RP3 direct to the uR: With the StarTech/Hub etc I would need to also move 3 LPSUs and the Isolation Transformer into the location of my main HiFi, and the isn't enough room. I have used two lots of network isolation between the backend and frontend to try and maintain the cleanliness of the signal into the uR. I certainly easily here adjustments to my backend infrastructure so I hopeful that it is working!

M

I was thinking of the so called 'direct' connection originally prescribed by Romaz at CA:

Router <---> Switch<--->FMCs<--->Switch <---> 'Bridged' RP3 (server/player) <---> uR <----> DAC.

My setup is basically the same as the above. I don't know how long an interconnect you would need in your setup, but I am currently running 20 ft  from the computer room / DAC-V1 to the pre-amp in my main system. There is a noticeable difference in 'edge' when I bypass the bridge and connect the uR to the switch.

I find this:

Router   <---> Switch <---> FMCs <---> 'Bridged' PC (server/player) <---> uR <----> DAC.

to be better than:

Router <---> FMCs <---> Switch <---> 'Bridged'  PC (server/player)

                                               ^ ------> uR <----> DAC.

some have noted further improvements by inserting a better 're-clocking' switch after the 'bridged' PC but I never tried that.

Edited.

I have tried to follow this thread from the start, but it has had so many permutations and progressive changes that I must profess to being somewhat lost - no doubt I could go through and pick out some, but for the benefit of readers who might be considering this route and who are similarly bemused, it would be very helpful if those here who believe they have reached nirvana, or at least the end game, could give a simple description of the finalised system.

And if anyone has compared, or is in a position to compare, what they have with one or more of the standard sources others will be familiar with that would be great. (E.g. Combined store/renderers like Melco, Innuos Zenith, Unito Core, Mac Mini with audirvana and RF isolator, or separate NAS with ND5XS or NDX etc as renderer.

I’ve been following this thread with great interest and FWIW over the previous few months my main source has changed:

From: QNAP NAS → Squeezebox Touch (SBT) → DAC-V1 USB

To: Vortexbox Music Server → Uptone Audio Regen USB → Uptone Audio USPCB A>B Adapter → DAC-V1 USB

The main reason for choosing the Vortexbox is I wanted to continue using Logitech Media Server (LMS) which was becoming difficult to maintain on the QNAP NAS.

The Vortexbox and Regen USB are both using the stock SMPS.

This upgrade was achieved at relatively low-cost and my system has never sounded better.

Dave

Hi Brilliant,

Not sure how this would work. Th eRP3 has one RJ45 ....and is hosting LMS for me. If I was to connect the RP3 directly to the uR then how would I interface with LMS?

I DID consider using Romaz's suggestion when I was running LMS on my NAS, but the RP3 powered by LPSU is just SO much better.

In terms of edge: I mainly hit this on the head when I sorted out my earthing/grounding in my min HiFi, since then everything has been good to get, but magnitudes less important.

IB

In my case I have one thing left on my ToDo list: replace my 0.5A 5v LT3045 with 1A LT3045 5v cards. When that is done and settled in I may write a post starting at the end point.

M

Innocent Bystander posted:

I have tried to follow this thread from the start, but it has had so many permutations and progressive changes that I must profess to being somewhat lost - no doubt I could go through and pick out some, but for the benefit of readers who might be considering this route and who are similarly bemused, it would be very helpful if those here who believe they have reached nirvana, or at least the end game, could give a simple description of the finalised system.

And if anyone has compared, or is in a position to compare, what they have with one or more of the standard sources others will be familiar with that would be great. (E.g. Combined store/renderers like Melco, Innuos Zenith, Unito Core, Mac Mini with audirvana and RF isolator, or separate NAS with ND5XS or NDX etc as renderer.

I understand that the thread was originally motivated by the introduction of the Sonore microRendu, rapidly followed by the SOtM SMS-200 and later and by ultraRendu and SMS-200ultra.

Considering the bandwith of the threads dedicated to microRendu, ultraRendu, SMS-200 and SMS-200ultra on CA, I think that it is fair to say that these "network players" or "ethernet to USB bridges" have been extremely successful and have significantly chanced the way we implement LAN streaming based solutions. They also have significantly contributed to the renewed popularity of high quality pure DACs.

Lightweight network streamers with support for UPnP and Roon rendering, HQPlayer, Qobuz, Tidal, etc. can be easily implemented on the basis of open source software solutions. Manufacturers can rely on well established, reliable programs (like, for instance, MPD), do not need to "reinvent" anything from scratch and can focus on dedicated low-power hardware with precise clocks, low electrical noise and outstanding sound quality.

The development of low-power, flexible ethernet to USB bridges has obviously raised the problem of how to integrate such devices in an existing network and of how to connect them to non-USB DACs. The perceived high quality of the USB output of the ultraRendu and of the SMS-200ultra has also raised the question of whether these or similar devices couldn't actually also run a UPnP server. This would indeed make them functionally equivalent to devices like the Melcos, the Innuos Zenith, the Uniti Core, the Mac Mini with isolator, etc.

If you look around on the 'net, you will see that, bluntly put, two schools seem to have emerged on how to set up a source for modern DACs using the technology spun-off by the development of microRendu, ultraRendu, etc. One school seems to be fond of building very long chains of devices (and their costly power supplies) between a noisy powerful PC (often running a Roon server) and a high quality DAC. The other school argues that a "clean", well clocked source is all what is needed. The main problem with the latter approach is that clean, flexible solutions are almost inevitably DIY solutions. Dedicated music servers with clean, outstanding outputs still tend to be proprietary solutions with limited flexibility, support and documentation.

My understanding is that Mr Underhill and Brilliant have taken a more balanced approach and are somewhere between the "longest path" and the "shortest path" schools.

Mr Underhill posted:

Hi Brilliant,

If by the Ethernet Bridge you mean using copper - fibre - copper I do == FMC > FMC.

...

I cannot say what Brilliant means by "Ethernet Bridge" of course but what I mean is the practice of using a device with two network interfaces to act as an access point. In the specific case, a server (e.g., a RP3) would be, on one side, directly connected (not via switch or router) to the uR. On the other side it would be connected (via wireless, switch, whatever) to the router. The uR would obtain its IP address from the server. The server would de facto act as a switch and forward packages coming from the uR to their destinations through its second network interface. I understand that this is how a Melco is usually connected to a streamer (or uR, mR, sms-200, sms-200ultra, etc.). Best, nbpf

Hi M,

Glad you have it under control- you have 'crossed that bridge' then.

IB

Will update my setup in a week or so.

B.

Funnily enough I would love to be able to buy a passive hub. They used to make them, but I haven't seen any for years.

However, IF I could buy (or more probably make)  a passive hub to put in front of the RP3 I would also need to connect my USB hub, which is on the other side if a StarTech USB extender etc. i am pretty happy with where my sound is and think that this re-organisation would be messy in my case.

Thanks for the thoughts and suggestions. I have saved that CA thread and may well re-visit it in the future.

M

Current Diagram:

Holey Moley! Mine seems so simple in comparison, even though when I explain it to the curious layman I can see their eyes start to glaze over with confusion and no way could I ever get there. But my set up sounds so good and more tweaks on the way: ferrites, a Ghent audio JSSG Ethernet cable with right angle plug for that last leg from FMC to microRendu, a Ghent star quad dc cable with right angle plugs for the LPS-1 to rendu, and a few cheapo silver Teflon GX-16’s to dc for the HDPLEX. And some solid silver 24ga wire coming for the ground shunts (one CA  member swears it make a difference - cheap enough and I only need 1-2 foot lengths if I measure and cut carefully. Cable dressing here I come. If nothing else should help clean up the spaghetti at the rear. Just had surgery this morning so all of that will have to wait but I’ll check back in. I will say again that in my system the John Swenson ground trick was easily a black box upgrade. The USPCB off them back of them V1 makes a noticeable difference as well.

Hi Charles,

In fact the diagram is not quiet accurate. However, I think of it as:

Loft = Music Server
Living Room  = Player

Network and Mains would be required whatever I was using, and has been extensively covered on the forum.

With that in mind:

Music Server = £1,300
Player = £1,200

That isn't absolutely true as I have taken a number of mis-steps or upgrades along the path.

So, remaining two steps for the moment will be:

1. replace all 5v 0.5A LY3045s with 1A ones; and
2. Sort out the mains for my main system, just engaging an electrician.

Mind you there is vibration control and racking to further consider!

M

Oh, I realize it didn’t cost that much money  (much like mine) but possibly just seeing something diagramed out makes it more complex:

Mine is as follows:

SERVER CLOSET (in office directly below living room stereo):

Vortexbox (Small Green Computer micro build, 7 years old now - just works) running LMS, powered by Chinese R-Core>

Cinnamon Ethernet cable>

Netgear FS105 switch dedicated to the hifi with John Swenson trick/cheap LPS>

basic cable>

LIVING ROOM:

TP Link 10/100 FMC/iPower/TV circuit>

TP Link 10/100 FMC/HD Plex 9V rail>

Melcord Opal Ethernet cable (soon Ghent JSSG)

microRendu/ Uptone LPS-1/HDPLEX 12v rail>

Uptone USPCB>

DAC V1>

Witchhat DIN>

Naim NAP 110>

NACA5>

Rega RX3’s

DAC and 110 on dedicated circuit with Porter Ports, HDPLEX on it’s own with same outlets

OFFICE:

GS108 switch (with other traffic)>

basic Ethernet cable>

TP Link 10/100 FMC pair - both powered by iPowers into same socket>

Cinnamon Ethernet cable>

Unitiqute V1>

Chord Carnival Classic>

KEF Ls50’s.

Everything that is DC is John Swenson ground shunted, and the modem into the house has a low noise grounded Class II wall wart. 

A few new cables on the way and then the 1.4 board upgrade (maybe in Feb when we’re in Hawaii) for the rendu and I think I’m done.

I know the V1 gets a bit disparaged on here (well at least its preamp) but it sounds absolutely amazing to me now that the network is getting so cleaned up. Listening to Monk’s Music right now as I type and it’s just brilliant. 

Hi Charles,

Yes, it looks complex, as would yours, but I hope it is easier to follow than writing it out?

Have you tried the HD-Plex straight into your mR - how would you describe the difference?

I tried stripping out the FMC media bridge yesterday - after all, you never know. I put it straight back in!.

M

A network can quickly get complicated. Can a simpler setup help SQ? It might, it might not. That CA (Romaz) 'direct' connection of the rendu to the  server/player is an example that did help in my setup:- PC-->rendu-->DAC-V1.

If you are clever at IT (which I am not), you might be able to get this rendu 'direct' connection some other way. The rendu uses the Sonore web GUI for configuration so either way it will need to connect to the internet for that, at least initially.

The method as described at CA is what I (cleverly) followed. It requires my PC to have a second ethernet port for the rendu and a 'bridge' via the OS to connect it to my existing LAN (with the DHCP server/router). That thread now has pointers on the first page for a few different OSs. 

I only have one network interface port so I use a USB/Ethernet adapter to provide the second port (in place of a second NIC). I have one adapter by Anker (on the Win 7 64 bit PC) and another by Apple (on a Win 10 64 bit laptop). The Apple one needs  a Win 10 64bit driver which I downloaded from the asix website (AX88772A) after a little google search.

The effort was worthwhile for me and I am grateful to the OP.

B.

Mr Underhill posted:

Hi Charles,

Yes, it looks complex, as would yours, but I hope it is easier to follow than writing it out?

Have you tried the HD-Plex straight into your mR - how would you describe the difference?

I tried stripping out the FMC media bridge yesterday - after all, you never know. I put it straight back in!.

M

The LPS-1 is better, the HDPLEX better than the iFi. The new model HDPLEX looks even better. A great supply if you need to run several things but really nothing special like the LPS-1. 

Hi Charles,

I asked as I found my cheap R-Core LPSU less transparent than the LPS-1, but had more body. I put this down to having a potential 2.5A on tap. I trplaced the LPS-1 with a slightly more expensive Chinese R-Core LPSU + 2 x LT3045 (7.1 A 1A), which I found combined the transparency of the LPS-1 and the body of the R-Core - for half the price of a LPS-1.

M

CP,  You have a nice system. V1/110 must be a foot tapper! I ran the 110 for a little while in the 80s and loved it.

The uR has me a bit buffled regarding PSs. It sounds different yet good with every one, but the LPS-1 appears to be transparent to it, except for a little more detail. With the LPS-1 it takes on the 'tone' of the feeding PS much more so than the mR did (from memory).

1. TP PS ~  2. TP PS+LPS-1: smooth, rounded, fatter bottom, forgiving, musical!

3. ZeroZone (ZZ) ~ 4. ZZ +LPS-1: sharper, detailed, top to bottom balance, revealing, musical!

I have the JSG shunt at the uR DC input. I am not sure I hear anything on A/B testing, but will evaluate its effect over the long term.

The power supplies 'sound can be fine tuned' by using different AC & DC cables (so we have not got rid of all the 'leakage' yet)!

Soundwise - I could live with any combo 1 through 4. The only problem with 4 is that the ZZ cuts out at some point during the booting of the LPS-1/uR probably due to transient current demand it can't handle.

Edited

Mr Underhill posted:

Hi Charles,

I asked as I found my cheap R-Core LPSU less transparent than the LPS-1, but had more body. I put this down to having a potential 2.5A on tap. I trplaced the LPS-1 with a slightly more expensive Chinese R-Core LPSU + 2 x LT3045 (7.1 A 1A), which I found combined the transparency of the LPS-1 and the body of the R-Core - for half the price of a LPS-1.

M

Hi M,

Which one is that then?

B.

Hi Brilliant,

I followed the thread when it first surfaced, and was looking at bridging the network ports on my NAS. A couple of things moved me away from this:

1. Using the RPi3 with the LPS-1 was just SO much better than the NAS, also reducing me to 1 RJ45; and
2. The RPi3 is a simple device generating less noise than a full blown PC or NUC, I suspect that it could be a trade off to further improve on the feed into the uR.

Still on the radar though.

M

Over at diyAudio a member has pcbs with //'d 3 & 6 LT3045s boards for PSs: 2 x 1.5A & 1 x 3A-w/h-sinks!

Brilliant posted:

 

CP,  You have a nice system. V1/110 must be a foot tapper! I ran the 110 for a little while in the 80s and loved it.

 

Well thank you. I keep thinking I want/need a bigger amp but then I listen to my CB110 and it’s always nah... it was rebuilt about 6 years ago and no can no longer power a pre so the V1 is perfect for it. Sounds esp lovely with jazz; not really a stadium rocker per se. The room is very “live” after a recent rebuild - new sub flooring and oak floors and rock wall insulation, 11x17 feet. Perfect boost for the 110. 

This is the DAC-V1 PC setup updated with the Sonore ultraRendu and the UpTone Audio components.

I have compared the playback of some tracks I know and believe to be of the same mastering on the formats: CD  (CD-rip wav), Vinyl, and SACD.

For reference:

TT = Vinyl  played on the Alphason TT/HR100S arm/DV-20X2 cart/ and Belles phono. (I have a Denon DL-304 which might be better for low level detail retrieval but do not want to change current setups)

SACD = Sony XA5400ES player.

Summary (to my ears):

TT gives the fullest tone/timbre and  scores higher mostly for that reason. The uR/DAC-V1 is pretty even with the Sony IME but perhaps a tad thinner sounding on some tracks. The Sony player adds warmth and seems 'sweeter' in tone  which I like especially when playing the SACD layers. In 'musicality' they are all pretty close and in this order (TT>SACD>uR/DAC-V1). I am thinking: they all stop you in your tracks, to listen or to dance! As usual YMMV.

The Tracks used for the comparisons (all subjective, score out of 10= imaginary reference, no format weighting!):

1.Et misericordia (Arnesen 2L 106 SACD, 384.4/24 wav).  This track was recorded in DXD and is offered by 2L for format comparisons. It is available in different formats for free on their website (hires test bench). I also have their SACD. The DSD file was not used as in IMO, DAC-V1 shines better with PCM.

uR/DAC-V1 8.5 (more spatial detail), Sony/SACD 8.5 (more warmth, inviting tones)

2. I am Old Fashioned (John Coltrane -Blue Train AP45 rpm LP, AP SACD, CD-rip wav). The same mastering  transferred to all three and was done  by Kevin Gray & Steve Hoffman from the original Blue Note master tapes!

TT: 9.0 (fuller tone/timbre), uR/DAC-V1: 8.0  (CD Rip , thinner tone), Sony: 8.0 (CD layer) & 8.5 (SACD, warmer, sweet).

3. You Are Too Beautiful - (Oliver Jones - Live At Biddles 1983 LP, CD, CD-rip wav) I believe the CD to be a straight transfer of the  LP- master (but could be wrong). This is a club live recording with background noise from the crowd.

TT: 9/10) (fuller tone/timbre than CD, nice flow), uR/DAC-V1: 8.5, Sony: 8.5 CD layer ( both reveal more low level detail than TT ,  the DV cart is suspect here). This is a nice CD transfer.

It's all good.

ATB

B.

Bowers posted:

Talking about "Hugo of streaming"; After moving from ND5X  to microrendu; Had to have my Hugo DAC repaired (warranty) to feed my Naim NAC 282 and using as source the microrendu (+LPS-1 +DC4 + CuriousUSB);  Just watched an enthusiastic review of the Sms 200 ultra. Might be the top of network streaming devices at this moment or it might be very individual ?!  Anyone with experience/opinion and preference concerning these sources ?   any feedback appreciated. Best Peter

 

There is a Nov. review of the SOtM sMS-200ultra Network Audio Player and sPS-500 Power at audiobacon dot net. It has a comparisons paragraph where comments vs uR are included.

In short (roughly quoting from sections in the para.):

sMS strengths - Macrodynamics, deeper and punchy bass, full bodied richer tone and timbre, layered holographic soundstage. Humanistic!

uR strengths - Microdynamics, tight clean bass, transparent, more detail resolving, quieter, more focused soundstaged but less depth. Insightful!

Note: Some problems (skips) are mentioned when playing DSD files on Chord Dave with the sMS but none with the uR- something to do with the unix incompatibility?

B.