Thread Rating:
  • 3 Vote(s) - 4.67 Average
  • 1
  • 2
  • 3
  • 4
  • 5
A Confused streaming system - Mutec / SOtM Ultra
A bit of an update, and as a spoiler alert, no jaws have literally hit the ground during the lifting of veils.

I have replaced my aging HP i5 PC for a new Dell.  My home PC gets used for many things, including the odd occasion (work related, not gaming or anything else fun) where I need something reasonably high powered.  Anyway, the old HP was getting a bit sluggish, and at the same time I had the chance to buy a new Dell at a significant discount.

The new machine in an Intel i9 9900k, has liquid cooled PSU and processor, lots of fast memory, and other goodies.  The old HP was an "all in one" design", so little more than laptop spec components with a nice screen.  The new machine arrived late October, and I spent a happy weekend getting the thing set up, transferring software, lots of fun finding software keys, backing up music libraries, getting things to talk to the correct directories etc.  Anyway, everything sorted and working without issue.  During early November there were a couple of occasions when I started to think that the hifi system was sounding a little better with the new PC delivering the files.  Nothing major, it was a simple case of the system sounding just a touch nicer.  That said, I have been reading lots of stuff on the "Novel" thread on AS, with people claiming all sorts of sound quality improvements using high powered processors, low latency memory, this kind of thing.  So were the improvements little more than optimism and expectation bias?  Well, maybe not, after all this is not a change of part of a PC, but a whole new machine.

So last weekend I set up a bit of an experiment.  Both PC's are currently connected to my home network, and both PC's have HQPlayer 4 installed.  (albeit the old machine's HQPlayer is now in demo mode, so it stops after 30 minutes).  So I loaded a short playlist of identical music files into HQ Player on both machines to try an A/B test.  This was not an ideal test, when streaming the PC needs to have the web based Eunhasu software running and "talking" the sMS-200.  When swapping from one PC to the other, it was necessary to fully shut down one PC (or pull out the ethernet cable) to get Eunhasu on the other machine working.  So not an ideal A/B test, with much faffing around when swapping from PC to PC.  After much swapping from one machine to the other, I ultimately come to the conclusion that everything sounded pretty much the same, irrespective of which PC was delivering the files.  Not the most exciting of conclusions I know, but perhaps interesting to note that in my subjective experience, making significant hardware changes upstream of the sMS-200 appears to make little difference.

Moving out of the subjective domain, with the old HP "all in one", I always found Roon to be a little clunky in operation, just a little slow and cumbersome.  With the i9 9900k, Roon is super slick, so that's at least something!
1000 Pro - KEF Blade - iFi Zen Stream - Mutec REF10 - MC3+USB - Pro-Ject Signature 12
Reply
I have been running the Mutec / SOtM kit for a long time now, and it might seem from this thread that I have pretty much A/B tested everything possible.  Well there is one major omission, I have never tried the SOtM kit via USB.  In fact, since having the Pro, I don't think I have tried anything via USB.  Last weekend I was thinking about this and thinking about the pending arrival of an EtherRegen, and figured it was about time I tried USB.  Not much time last weekend, but I did a bit of cable swapping to set up the system as PC - Ethernet - sMS200Ultra - tXUSBultra - USB to Devialet, basically taking the Mutec MC3+USB "out of the chain".

With some listening in the week and yesterday, I started to form the impression that there was something a little nicer about the sound via USB, more organic you might say, but I couldn't help thinking a little something was lost.  Doing the armchair audio designer bit, I could see that the MC3+USB has an advantage, as it is feeding the Devialet a REF10'd super low jitter signal to the AES/EBU input which the Devialet could take full advantage of when extracting the clock signal.  On the other hand, maybe the SOtM kit has an advantage, with the "last component before the DAC" being powered by a lovely Paul Hynes SR4, not the Mutec's nasty SMPS.

I had some quiet time today, and set up a little test.  I could run USB using the SOtM kit, and at the same time I could run music from a USB drive via my Arcam Bluray player, feeding the Mutec MC3+USB.  This made for a good test, I have found previously that the MC3+USB sounds not far off identical fed by either the SOtM kit or the file playing Arcam.  I could start identical music on both front ends, and switch between the two using the Devialet remote app.  This worked well, a lot easier than some tests which involve lots of cable swapping or re-starting software, and so on.

An interesting result.  With some tracks things did indeed seem a little nicer and more organic via USB, but also softer.  It was clear that via the MC3+USB things sounded more dynamic, with more PRAT if you like.  Switching the same track from USB to AES/EBU in some cases there was a noticeable increase in bass punch, things sounded much sharper and more dynamic.  Yet, there remains something that with some music sounded a little nicer via USB.  So, a slightly mixed result, but on balance I much preferred the results via AES/EBU.

Of course this does beg the question, is it a consequence of the influence of the MC3+USB, is it the difference between the Devialet's USB implementation versus AEA/EBU, or simply the difference between the asynchronous USB protocol and synchronous AES/EBU.  It's probably a combination of things, there is no way of telling I guess.  What I would say is that both the AES/EBU and USB sources could be classified as "reasonably good", so I am thinking it does say a little about the inputs / protocol themselves.  A degree of speculation here, of course.

To be clear, these are fairly marginal differences, it's not a case of "wow, where have the dynamics gone!" when switching to USB, but it is the kind of subtle stuff that adds up to make a difference.  Plus, I remain a little intrigued by the areas where I seemed to prefer USB.  I shall be leaving the system set up tXUSBultra to USB for a while, it is sometimes interesting to see how small changes like this manifest themselves over time.

It has also got me thinking about the SMPS in the Mutec, maybe changing this to a LPS might help close the gap in areas where USB is ahead?  A little optimistic perhaps, but maybe.  Either way, it has got me thinking....
1000 Pro - KEF Blade - iFi Zen Stream - Mutec REF10 - MC3+USB - Pro-Ject Signature 12
Reply
Quote:Doing the armchair audio designer bit, I could see that the MC3+USB has an advantage, as it is feeding the Devialet a REF10'd super low jitter signal to the AES/EBU input which the Devialet could take full advantage of when extracting the clock signal.  


@Confused are you shure the Devialet uses the clock signal that comes with the AES signal? With all the DSP things going on in there I think they buffer/re-clock the signal with the internal clock but this is pure speculation from my side Smile
Speakers:TAD CE-1. Amplifier: TAD M2500mk2. Digital: TAD DA1000-TX, Innuos Statement Next-gen, Innuos PhoenixNET.

Miscellaneous: Qobuz Studio, Ansuz Mainz 8 D2, Ansuz Darkz DTC, Tubulus Argentus ethernet cable, Tubulus Concentus USB cable, Tubulus Argentus V2 XLR cable, Tubulus Argentus V3 + V3 bass, iFi Nova powercables. 

Second system
Qobuz Studio -> Devialet Silver Phantom, Devialet Tree









Reply
octaviars Wrote:
Quote:Doing the armchair audio designer bit, I could see that the MC3+USB has an advantage, as it is feeding the Devialet a REF10'd super low jitter signal to the AES/EBU input which the Devialet could take full advantage of when extracting the clock signal.  


@Confused are you sure the Devialet uses the clock signal that comes with the AES signal? With all the DSP things going on in there I think they buffer/re-clock the signal with the internal clock but this is pure speculation from my side Smile
No, I am not sure at all!  It would be normal with S/PDIF for the receiving device use a phased locked loop to match the clock signal of the incoming stream, but I have no idea exactly how this is implemented in a Devialet.  I asked Mathieu Pernot about this at a Devialet event once, I got a very detailed explanation about the differences between synchronous and asynchronous protocols, but I did not glean any secrets of the inner workings of a Devialet.  Thinking about buffering / "reclocking", yes, the Devialet upsaples to 24/192, and the ADH runs at 32 bit I think, so there must be some clever stuff going on, but if there is a buffer, this needs to be emptied at precisely the rate that the source fills it, so the clocks do need to match.

What I did gain with the little test yesterday is an understanding of how the differences between synchronous and asynchronous protocols actually sound like, with my system and my ears.  So on a subjective level I know all I need to know, objectively, I am guessing.
1000 Pro - KEF Blade - iFi Zen Stream - Mutec REF10 - MC3+USB - Pro-Ject Signature 12
Reply
In a way this picks up from my post #327 in this thread. Yesterday I had some free time to try some listening tests with the EtherRegen. Referring back to #327, prior to receiving the EtherRegen, my preferred set up was running a standard TP-Link switch, which then feeds my SOtM modified D-Link switch. I only had about an hour spare yesterday, at first I tried comparing the streaming system, PC, EtherRegen, SOtM kit, Mutec kit against feeding the MC3+USB from my USB stick playing Arcam. This allows almost a direct A/B switch, but not ideal as a small volume adjustment is required to equalise, and it does require getting up to change the input on the MC3+USB. This was a little inconclusive. I had a feeling that the streaming set-up was best, but it was marginal. I decided that this was not the best way to test the EtherRegen, so I thought I would try the quick Ethernet cable swap method, to take the EtherRegen in and out of the chain. This method works surprisingly well, no need to worry about volume levelling, and the streaming set-up remained stable even swapping cables between the EtherRegen and the SOtM modified switch. Again, this test was a little inconclusive. I did have a slight feeling that things sounded better with the EtherRegen in place, but at the same time, it was not a big enough difference for me to be sure that I was not kidding myself. At this point I ran out of time yesterday, and other seasonal related tasks go in the way of further experimenting.

Reflecting on the above, it occurred to me that in a way the above test was not entirely fair. Consider that without the EtherRegen, I was running a standard switch, then a SOtM switch, which is effectively running off the Mutec Ref 10. So even without the EtherRegen I have a degree of noise isolation from the upstream kit, plus I am running switch with a decent clock signal.

This takes me to this morning. I thought I would try the following:

A) TP-Link switch, SOtM modified switch versus TP-Link switch, SOtM modified, EtherRegen
B) TP-Link switch versus TP-Link switch, EtherRegen

I chose the above over the more obvious comparison of simply substituting the TP-Link switch for the EtherRegen, because A and B above can be done with a single Ethernet cable swap, and furthermore, the EtherRegen is supposed to make everything before the switch irrelevant, so why should it care if the TP-Link switch is there or not.

Anyway, the conclusion is not far off the one I reached yesterday. Yes, I do feel that there is a very slight improvement from having either the SOtM modified switch or the EtherRegen in the chain, but the difference is very subtle indeed. Could it be imagined? I really am not sure, it is in that strange zone where you go from thinking yes, I can here something, then after a few swaps either way, you become less sure if it is real or imagined. If I have to come off the fence and commit, I would go with an improvement, but very small. What do I think I am hearing as an improvement? It is a sharpening of the sound stage, a touch more focus and resolution to the details.

For now, I will be leaving the system as TP-Link switch, SOtM modified switch, EtherRegen. This seemed best to me. My overcomplicated system gets yet more complicated. One thing to try in the future will be to use the REF10 to provide the clock reference to the EtherRegen.

Whilst in the mood for confusing A/B swaps, I thought I would try the EtherRegen feeding the Devialet direct in CI board Roon RAAT mode, and compare to feeding the Devialet direct from the standard TP-Link switch. Curiously, or perhaps not curiously, the results were pretty much the same as when feeding my SOtM / Mutec kit. I could detect a small difference, but perhaps at the margins of what might be imagined. Same thing as before, the sound seemed a little more focused, tidier if you like, a fraction sharper with more resolution. I think ….

So no jaws literally hitting the floor. To provide a little context to all this, I actually preferred my system with the standard TP-Link switch to the old SOtM modified switch on its own, and the TP-Link switch is one that used in comparison to an AQVOX switch, which I ultimately sent back for a refund, I just could not find any difference between the two. So this is very marginal stuff, at the edge of what might be messing with your mind, but I would take the EtherRegen over the AQVOX, and sit down and enjoy the music happy that it might just be adding that touch of polish to the sound.

The next step for me is to see how I find the system in normal use over time. Sometimes these first impressions hold, sometimes other things become apparent in normal use with a range of music. Time will tell.
1000 Pro - KEF Blade - iFi Zen Stream - Mutec REF10 - MC3+USB - Pro-Ject Signature 12
Reply
@Confused ,

Interesting, especially as I've been thinking about how I'm going to make comparisons when I get my EtherRegen next month. Simply replacing my Cisco Switch with the EtherRegen, what I intend to do, entails swapping 3 cable connections so it can't be done easily or quickly and that makes simple A/B comparisons very problematic, whether you do them sighted on your own or get someone else to do the cable swapping for you.

You may find the following item from Jim Austin of Stereophile interesting: https://www.stereophile.com/content/slow-listening It's a very different take on the way we hear things and on the process of making comparisons of gear. Prior to reading it my intention was to listen to music with the Cisco switch in place for a while, pause and make the swap, then continue with the EtherRegen in place, noting my initial impressions, then to continue with the EtherRegen in place for several days to a week in order to become familiar with the sound, noting any changes, then swap back to the Cisco and see what my reaction to the swap back is. I've found that approach useful on occasion in the past and I think it actually sits well with some of the ideas Jim Austin refers to in that article in the link.

I don't see that article or the research he cites changing the beliefs of those who are wedded to blind and double blind testing protocols but I think there's a lot to be said for the ideas Austin is referring to.
Roon Nucleus+, Devilalet Expert 140 Pro CI, Focal Sopra 2, PS Audio P12, Keces P8 LPS, Uptone Audio EtherREGEN with optical fibre link to my router, Shunyata Alpha NR and Sigma NR power cables, Shunyata Sigma ethernet cables, Shunyata Alpha V2 speaker cables, Grand Prix Audio Monaco rack, RealTRAPS acoustic treatment.

Brisbane, Qld, Australia
Reply
I posted a similar post to post #335 on Audiophile Style. (basically the same post, but some changes for relevant context)

The post received an interesting response from UpTone's Alex Crespi, where he pointed out that if I am using my SOtM modified switch prior to the EtherRegen, this effectively puts the Mutec REF 10 on "both sides of the moat". He has a point, the SOtM switch takes its reference clock signal from the clock board in the sMS-200Ultra, so that places the clock board on both sides of the moat. As to how big a difference this might make, I have no idea, but it is easy enough for me to take the SOtM modified clock out of the system to try. Indeed, I could run router direct to the EtherRegen, easy to do. I will try this when I get time.

@David A The Jim Austin article was interesting, many thanks for the link. In the past I can think of times when I have done a very quick A/B test, or just simply listened to a new piece of kit, come to a very quick conclusion, and that conclusion has held as correct during subsequent long term listening. But not always! I can think of other times when the first impression ultimately turned out to be wrong or missing something. So I see no conflicts here, A/B testing has its uses, as does blind testing, and long term listening, and measurements of course. I think having a variety of methods for evaluation of equipment is useful, there is no need to be in one camp or the other. With subtle and complex matters, why not use all the tools that are available? As an aside, I am a huge fan of the long term listening approach, as this involves nothing more than listening to music and not worrying about anything. Highly recommended!
1000 Pro - KEF Blade - iFi Zen Stream - Mutec REF10 - MC3+USB - Pro-Ject Signature 12
Reply
(23-Dec-2019, 14:35)Confused Wrote: I posted a similar post to post #335 on Audiophile Style.  (basically the same post, but some changes for relevant context)

The post received an interesting response from UpTone's Alex Crespi, where he pointed out that if I am using my SOtM modified switch prior to the EtherRegen, this effectively puts the Mutec REF 10 on "both sides of the moat".  He has a point, the SOtM switch takes its reference clock signal from the clock board in the sMS-200Ultra, so that places the clock board on both sides of the moat.  As to how big a difference this might make, I have no idea, but it is easy enough for me to take the SOtM modified clock out of the system to try.  Indeed, I could run router direct to the EtherRegen, easy to do.  I will try this when I get time.

Could you explain the idea of a „leakage loop“ and how that could apply to your current setup?
Kii Three Speakers | JL Audio Fathom Subwoofer  | KEF LS50 Speakers | Samsung 850 Soundbar
Innuos Statement | Trinnov Altitude 32 | PS Audio Stellar S300 Amp | T+A HA 200 Headphone Amplifier | Meze Empyrean Headphones
Sean Jacobs DC-3 Custom Build LPS | Roon Core DIY Server | SOtM sNH-10G Modded Switch | Oppo UDP-205 Blu-Ray/SACD Player
T+A G 2000 R Turntable | DS Audio E1 Optical Stylus | Gordian Lab 12 Power Conditioner | Artesania Audio Rack
Reply
(23-Dec-2019, 21:48)baconbrain Wrote: Could you explain the idea of a „leakage loop“ and how that could apply to your current setup?

Where did you see the phrase "leakage loop"? It's not anywhere on this page in @Confused's comments and it doesn't appear on Uptone Audio's page for the EtherRegen which instead makes reference to "blockage of all external leakage currents".

I can't recall seeing the phrase "leakage loop" anywhere in relation to the EtherRegen.
Roon Nucleus+, Devilalet Expert 140 Pro CI, Focal Sopra 2, PS Audio P12, Keces P8 LPS, Uptone Audio EtherREGEN with optical fibre link to my router, Shunyata Alpha NR and Sigma NR power cables, Shunyata Sigma ethernet cables, Shunyata Alpha V2 speaker cables, Grand Prix Audio Monaco rack, RealTRAPS acoustic treatment.

Brisbane, Qld, Australia
Reply
David A Wrote:
baconbrain Wrote:Could you explain the idea of a „leakage loop“ and how that could apply to your current setup?

Where did you see the phrase "leakage loop"? It's not anywhere on this page in @Confused's comments and it doesn't appear on Uptone Audio's page for the EtherRegen which instead makes reference to "blockage of all external leakage currents".

I can't recall seeing the phrase "leakage loop" anywhere in relation to the EtherRegen.

The term leakage loop is not specific to the EtherRegen, and I think Mr Bacon may have mentioned this because although I did not use the specific term, it was in fact what I was describing in terms of my SOtM clock board being both before the EtherRegen (clock signal to SOtM switch) and after the Ether Regen (it is in my sMS200Ultra "end point"), hence possibly creating a leakage loop.

Some links to posts from John Swenson might help, the first one explains my situation far better than I could, the rest are for reference:

https://audiophilestyle.com/forums/topic...nt-1007488

https://audiophilestyle.com/forums/topic...ent-800481

https://audiophilestyle.com/forums/topic...ent-865860

https://audiophilestyle.com/forums/topic...ent-592269

https://audiophilestyle.com/forums/topic...ent-668000

EDIT.  I notice that the first link is a response to a question from this forums very own, and very knowledgeable, @octaviars
1000 Pro - KEF Blade - iFi Zen Stream - Mutec REF10 - MC3+USB - Pro-Ject Signature 12
Reply


Forum Jump:


Users browsing this thread: 2 Guest(s)