A novel way to massively improve the SQ of computer audio streaming

[romaz]

Exactly my experience. I reported the other night substituting a Mac Mini/MMK/JS-2 for a 2008 Mac Pro. Subsequent listening had me thinking this was among my most significant upgrades. That Mac Pro had been doing much harm - and holding everything else back.

 

 

 

I had thought I installed Sierra on my SD card, but this morning there was a notification that encouraged upgrading to Sierra. Turns out I inadvertently installed El Capitan. :-)

 

 

 

Apologize - are you kidding? It was an awesome post. Thanks so much for putting all that effort into it.

 

 

Sent from my iPad using Computer Audiophile

Thanks, Kenny. I take back my recent recommendation of installing a PCIE SSD drive on your Mac Mini so that you can try an optimized version of Windows. You would be better off trying Windows off of a Thunderbolt hard drive instead.

[Classical Sound]

Hello Roy,

 

Having only understood half of your new impressive (scientifical) approach, the take home message for me is to install the OSX on a (PCIE ?) SD-card.

 

As I have decided to buy a Mac mini, I would like to know, where do you recommend to have the Roon core installed on? On SD-card as well?

 

Jorg- Michael

 

 

 

 

Sent from my iPad using Computer Audiophile

[gadgetman]

Roy,

 

Thank you very much for your feedback. I really enjoy your every article and am always looking forward to your very thorough insight into each topic.

 

Back to sMS-200, by further investigating into it I realize that the CPU used is A20 from Allwinner. It's a dual core Cortex A7 ARM CPU. It needs an external Ethernet PHY which is Realtek RTL8211E. From Allwinner's linux-sunxi community it mentioned:

 

"For reliable Gigabit networking (1000Mbit operation), several sunxi devices require an important tweak that adjusts the relative timing of the clock and data signals to the PHY, in order to compensate for differing trace lengths on the PCB"

 

I believe that SOtM must have put a lot of effort in tweaking their Ethernet driver to come to it's present state. That may also explains why they prefer 100Mbps over gigabit at the early product stage.

 

I totally agree with your low impedance perspective. Yet I have some different point of view regarding the latency issue. The way I look at latency is a little similar to your view of CPU clock. It doesn't have to be the lowest but the consistent and predictable timing performance is one of the key factor to a good sounding system.

 

Another point that we all agree that could degrade SQ is the EMC interference. As we all know wherever electric current goes there's electric-magnetic field generated. It gets worse if the current fluctuates frequently. And as you said, which I couldn't agree more is, that "it is always better to prevent a problem or to address it at the source than to have to fix it downstream".

 

One way to achieve low current fluctuation is to use a fixed system clock. All that an audiophile music server needs is a steady and consistent operation without power saving nor hyper speed function. A fixed frequency system clock can do it just fine. And it is a unique feature in Daphile which I believe can improve the SQ quite a bit.

 

Based off the heavy lightweight media server belief, I'm very interested to give ROCK a try. I'll report back when I done the test.

[lmitche]

Thanks, Larry. Very helpful and I appreciate the tip on the Adnaco USB unit. I am not familiar with this unit and I wonder how it might compare to the Acousense AFI USB modules from Germany which I haven't tried either. The Adnaco looks simpler and to be able to power it with an LPS-1 would make it a better choice probably. To be honest, I am skeptical that the optical isolation portion would benefit me since my DAC's USB input is already galvanically isolated but you never know what complete optical isolation might provide and it would appear the Adnaco would have 2 clocks that could be replaced for the better and this dual layer of reclocking could allow for a single box solution instead of server + NAA. It would help if you had experience with an mR or sMS-200 and could say that this Adnaco solution was better but as it stands, I am bowled over by my new sMS-200 Ultra. If I can secure a Adnaco unit to try, I will definitely do so.

 

Regarding the 6700K (or 7600K), I don't use HQP and have no interest in overclocking and so I am content to be able to use a low power CPU.

 

Thanks for sharing that you like how your DDR4 sounds. I have no experience with DDR4.

 

I prefer not to use network drives either as in my system, local playback for sure sounds a bit better although I maintain a NAS for comparison testing and for backup but I also stream from Tidal and so I can't shut down my network completely. As I plan to upgrade the "incoming" LAN port's clock on my new server and with a Paul Hynes SR7 powering my modem/router, I am hoping to bring Tidal streaming to nearly the same level of SQ as local playback. As it stands, network streaming doesn't sound horrible at all. If I am successful in elevating network playback close to the SQ level of my current local playback, it may encourage me to pursue building an audiophile class modem/router with better clocks and regulators and attempt to integrate larger capacity music storage into the router itself. I find the idea of an audiophile class modem/router/NAS that I can power with a single lead from my SR7 to be an attractive scenario just so I can provide a high quality stream to other areas of my home.

Anyway, interesting times ahead. Thanks again.

 

Roy

Hi Roy,

 

Well I do have experience with both the predecessor to the SMS-200, the SMS-100 and the microRendu. I owned and used the SMS-100 for over a year, using it with minimserver and soxr upsampling.

 

Despite my initial resistance, about a year ago Eurodriver insisted that I learn about HQplayer upsampling and DSD512 DACs. Well a year later I am convinced that using HQplayer to upsample before the DAC is the only way to go. While this becomes apparent at DSD256, DSD512 takes things to another level, and well supported and inexpensive DACs are available that enable us to access this incredible level of SQ without spending 10s of thousands. I am grateful to Eurodriver for his support.

 

Key to this achievement is the use of so called "raw or native DSD" support on the playback chain. Unfortunately at the moment Linux has only nascent support for raw DSD and this results in frequent pop and ticks that makes Linux an unusable solution for day to day use.

 

Hoping this was fixed with the microRendu, I bought one to use as NAA. It was later returned after realizing that the problem persisted. Expectations are the SMS-200 will suffer the same problem as it appears to run a hacked version of the same Fedora based software. Led by Jesus R, many people are working to solve this problem, and it looks like progress is being made, so I am hopeful that Linux will be back in my system the near future.

 

In the meantime and FWIW, HQplayer upsampling to DSD512 with Windows 10 sounds phenomenal, and better then any other solution heard here to date. I would encourage you to purchase a system with this capability in your new build, even if you don't use it initially. You don't need the k version and related overclocking which is why I recommended a fixed clock version. Given your obvious passion for SQ, it would be a shame to see you build a new machine with a processor that is one model upgrade and $10 to 20 from being a dsd512 capable machine.

 

I'll shut up now.

 

Larry

[tedwoods]

I don't know if anyone has noticed this relatively new "audiophile" switch:

best audiophile Switch, high end switch - AQVOX Audio Devices

[chauphuong]

At present, waiting for the new SMS from Sotm, I use a modded Raspberry Pi3 streaming music from a fanless PC. Applying the bridging trick described in Post #1. There is an significant increase in SQ, but the sound is choked up just like insufficient bandwith.

[Qhwoeprktiyns]

I don't know if anyone has noticed this relatively new "audiophile" switch:

best audiophile Switch, high end switch - AQVOX Audio Devices

 

400 euros for a switch! No thanks... This is getting ridiculous.

[Superdad]

But this was not my best overall SQ, however. I went back to El Capitan on the SD card yesterday and without question, right now, this is where the magic is at. That relaxed, natural and effortless presentation that made the Mac Mini so magical when I first heard it was back and gone is the HF noise that even my hard drive seemed to still have some of. Where OSX lacks in precision and imaging compared against my optimized Windows setup, SOtM's better clocks seem to more than make up for it, especially now that I am able to reintroduce my switch back into my chain without irritation. While I still believe optimized Windows is better than Mac OS, the secret is to be able to run the OS off a flash card of some sort.

 

Great post Roy.

 

Of course your above paragraph is a bit of deja vu from 3 years ago :

http://www.computeraudiophile.com/f27-uptone-audio-sponsored/attention-current-mac-mini-users-boot-mavericks-sd-card-load-ramdisk-dismount-your-internal-sata-drives-and-pour-drink-musicians-walking-out-your-speakers-18159/

 

While my optimized SD card is still running Mavericks (with about 65 processes/280 threads), I'm trying to make time to slim an El Capitan install--to at least get under 100 processes--so I can run the current version of HQ Player (via other than the microRendu/NAA). Staying away from the truly bloated Sierra!

 

It is quite something how sonically competitive a simply-tweaked Mac mini can be. My setup with it is:

2012 i7 2.6GHz with 16GB RAM

Our Mac mini DC-conversion/Linear Fan Controller Kit (MMK)

Our JS-2 choke-filtered, dual-output, 5-7 amp LPS

32GB SD card booting heavily-slimmed OS

Internal SATA HD dismounted (I'd remove it entirely, but really need it, with its multiple partitions, for experiments and OS image building; but I "eject" all partitions when using for playback)

 

Only connections to the computer are:

1) the DC power cable (and the coax/SMA for the MMK>JS-2 pseudo-Kelvin-sense voltage feedback circuit);

2) the USB cable (ports 1 and 3 are the best sounding by far) feeding my pre-production ISO REGEN (itself powered by an LPS-1 of course)

3) 25 foot BlueJeans/Belden Cat6a Ethernet cable direct to the 27" 4GHz iMac on my desk--used for Screen Sharing and sharing of a large drive with my music library.

 

I have the direct Ethernet connection set up for Jumbo Frames (MTU 9000) as that reduces overhead and I thought I heard a difference long ago.

 

--------

 

I might build up a high-power machine for DSD512 someday (Larry bugs me about it all the time ), but life is too short for me stress with Windows. Seems likely that Linux DSD issues will get worked out, and I do think that a split high-horsepower SRC/SDM machine to low power, all linear regulated renderer device will be the most desirable.

 

--------

 

Continue on gents. I watch with pleasure...

 

--Alex C.

[tedwoods]

400 euros for a switch! No thanks... This is getting ridiculous.

 

I feel ya...

At times, it seems digital advancement and afford-ability do not always go hand in hand.

My incentive for mentioning this was mostly because of a certain degree of relativity to what's already been discussed in this thread, one of the threads that literally do keep me on my toes!

[Andromedasuite]

Great post Roy.

 

Of course your above paragraph is a bit of deja vu from 3 years ago :

http://www.computeraudiophile.com/f27-uptone-audio-sponsored/attention-current-mac-mini-users-boot-mavericks-sd-card-load-ramdisk-dismount-your-internal-sata-drives-and-pour-drink-musicians-walking-out-your-speakers-18159/

 

While my optimized SD card is still running Mavericks (with about 65 processes/280 threads), I'm trying to make time to slim an El Capitan install--to at least get under 100 processes--so I can run the current version of HQ Player (via other than the microRendu/NAA). Staying away from the truly bloated Sierra!

 

It is quite something how sonically competitive a simply-tweaked Mac mini can be. My setup with it is:

2012 i7 2.6GHz with 16GB RAM

Our Mac mini DC-conversion/Linear Fan Controller Kit (MMK)

Our JS-2 choke-filtered, dual-output, 5-7 amp LPS

32GB SD card booting heavily-slimmed OS

Internal SATA HD dismounted (I'd remove it entirely, but really need it, with its multiple partitions, for experiments and OS image building; but I "eject" all partitions when using for playback)

 

Only connections to the computer are:

1) the DC power cable (and the coax/SMA for the MMK>JS-2 pseudo-Kelvin-sense voltage feedback circuit);

2) the USB cable (ports 1 and 3 are the best sounding by far) feeding my pre-production ISO REGEN (itself powered by an LPS-1 of course

Hi Alex,

Does it mean no microrendu anymore in the chain[emoji6]

3) 25 foot BlueJeans/Belden Cat6a Ethernet cable direct to the 27" 4GHz iMac on my desk--used for Screen Sharing and sharing of a large drive with my music library.

 

I have the direct Ethernet connection set up for Jumbo Frames (MTU 9000) as that reduces overhead and I thought I heard a difference long ago.

 

--------

 

I might build up a high-power machine for DSD512 someday (Larry bugs me about it all the time ), but life is too short for me stress with Windows. Seems likely that Linux DSD issues will get worked out, and I do think that a split high-horsepower SRC/SDM machine to low power, all linear regulated renderer device will be the most desirable.

 

--------

 

Continue on gents. I watch with pleasure...

 

--Alex C.

 

 

 

Sent from my HTC One using Computer Audiophile mobile app

[Cooler]

Glad to see that your system is now working. Clipper helped me as well.

You may even eject and remove the WiFi adapter. HQP still works. You may check if this sounds better :-)

You may disable the LAN that connected to the router as well. HQP still works for at least two hours (see my earlier post) and a total isolation from the router and the WiFi dongle is possible. You may check if this sound better :-)

This is fun :-)

But if your system restarts, you need to setup the bridge again :-( but it just take 2 minutes. What I did was to enable the LAN before shutting down Win10 and reinsert the WiFi adapter just before starting up Win10. The bridge will be re-built automatically so far.

 

Cheers.

Enjoy.

Thanks for suggestions, greenleo!

 

I tried to remove the wifi adaptor, doesnt work for me, HQPlayer is loosing mR, but its really funny, as i turned off wifi for this adaptor and just kept it in usb plug and it works And as i have raid on that PC i cant isolate it from network, but im still thinking to buy sonicTransporter and NAS, maybe i will get even better SQ.

 

Best wishes.

[kennyb123]

I’ve been experimenting with various ethernet cables for both spans of the bridge that I’ve configured on my Mac Mini. Yesterday afternoon a friend (who’s also a Wireworld and Audioquest dealer) and I compared several different cables on both spans of that bridge. (Potentially dangerous to have a friend who brings audio goodies to your home and also brings with him a credit card reader. LOL)

 

Location 1: Paul Pang switch to microRendu (all 1 m):

 

1A: WW Starlight (~ $210 for 1m)

1B: Nordost Blue Heaven (~ $400 for 1m)

1C: WW Platinum Starlight (~ $850 for 1m)

 

Location 2: wall jack to Mac Mini (both 8m):

 

2A: WW Chroma (~ $200 for 8m, or $80 termination fee + $15 per meter)

2B: Audioquest Cinnamon (~ $200 for 8m)

 

(A custom 18" run of Audioquest Carbon CAT700 was used between Mac Mini and Paul Pang switch the entire time).

 

First up was 1A vs 1C with 2B left in place. It took a number of swaps, but I thought I heard a slight veil removed with 1C. It seemed subtle though.

 

Next was 2A vs 2B with 1C left in place. Much easier to hear important differences. The track "Words of Wonder" from Keith Richards' "Main Offender" album features some really interesting high hat playing. With the 2A everything was in the right place such that the high hats and the way they were struck was very convincing both spatially and tonally. 2B disrupted that somewhat by tossing out some important cues that conveyed realism. It also threw less clear of an imaging picture with a few other tracks. Back in the box it went.

 

Leaving the 2A in place we then swapped 1A for 1C and there was an apparent loss of transparency. The spatial and tonal cues of the high hat were mostly retained, but the high hat just didn't stand out as clearly. (Sigh)

 

The more we listened with 1C/2A in place the more impressed we became with its transparency. It did the best job of just getting the heck out of the way - but there’s definitely diminishing returns when considering the price of 1C over 1A. But if one is looking to squeeze out that very last bit, it's definitely worth giving this cable an audition. No, I didn't buy one.

 

Maybe most impressive was 2A - at least relative to the well regarded 2B. Construction-wise it also feels like you are getting much more from your money with 2A. The same connectors are used on all three levels of WW Ethernet.

 

Last up was 1B. This didn't get the same level of attention in terms of comparisons. It was about on par with the 1A, so it seemed too be priced too high. Also disappointing to see a $400 cable with connectors featuring the typical plastic tabs that seem to too easily break off.

 

The 8m run of WW Chroma remained in place last night after my friend left. I thought I heard some nice improvements over what I had been hearing with the 8m run of Audioquest Cinnamon in place, including a blacker background and improved clarity - particularly in the upper frequencies. I should point out that the Cinnamon had about three weeks to burn-in, while the WW Chroma had only 3-4 nights of burn-in prior to yesterday.

 

Hope this was of interest.

[R1200CL]

Ever tried any of the Blue Jeans cables ?

 

 

Sent from my iPad using Computer Audiophile

[kennyb123]

Ever tried any of the Blue Jeans cables ?

 

No not yet.

 

 

Sent from my iPhone using Computer Audiophile

[amgradmd]

I've been following this thread with keen interest. Thanks for all those who have contributed, especially Romaz! I have been thinking as to whether or not this setup with a direct connection from my Mac Mini to my PS Audio DirectStream DAC with Bridge II. Keep in mind, the Bridge II on the DS DAC connects via ethernet and incorporates a Roon endpoint. So, from what I can gather, there is no obvious reason I couldn't "bridge" my Mac Mini with thunderbolt dual ethernet dongle and run direct to the Bridge II card on my DS DAC. Has anyone tried this?? I would love to ditch my USB connection. Thanks in advance for your input.

[greenleo]

Thanks for suggestions, greenleo!

 

I tried to remove the wifi adaptor, doesnt work for me, HQPlayer is loosing mR, but its really funny, as i turned off wifi for this adaptor and just kept it in usb plug and it works And as i have raid on that PC i cant isolate it from network, but im still thinking to buy sonicTransporter and NAS, maybe i will get even better SQ.

 

Best wishes.

Thanks for the info.

 

Yeap, the adapter may be disabled and the bridge always works.

 

In my case, I can eject the adapter and disable the onboard LAN. I guess it's system dependent then and I'm just lucky.

 

It seems somebody has already tried using sonicTransporter but reverted back to PC ->mR -> USB DAC for better SQ. You may check. Again, I guess it's DAC dependent.

 

Best wishes

[greenleo]

Good value for money. Haven't tried pricer things though.

[romaz]

Hello Roy,

 

Having only understood half of your new impressive (scientifical) approach, the take home message for me is to install the OSX on a (PCIE ?) SD-card.

 

As I have decided to buy a Mac mini, I would like to know, where do you recommend to have the Roon core installed on? On SD-card as well?

 

Jorg- Michael

 

 

 

 

Sent from my iPad using Computer Audiophile

Hello Jörg,

 

Yes, that is correct. OSX on an SD card has been the best sounding approach I have found so far with the Mac Mini. With the Mac Mini, the SD card reader is on the PCIE bus which is a nice bonus as this has a more direct connection (lower latency) to the CPU compared to SATA.

 

Roon core (and its database) can only be installed on the OS drive and so in this case, it will install on your SD card. Roon's database can take up quite a bit of space depending on how large your music collection is (about 2GB per 1000 albums is what Roon is quoting). A 32GB card could work but a 64GB SD card would provide you better future expansion.

[romaz]

Hi Roy,

 

Well I do have experience with both the predecessor to the SMS-200, the SMS-100 and the microRendu. I owned and used the SMS-100 for over a year, using it with minimserver and soxr upsampling.

 

Despite my initial resistance, about a year ago Eurodriver insisted that I learn about HQplayer upsampling and DSD512 DACs. Well a year later I am convinced that using HQplayer to upsample before the DAC is the only way to go. While this becomes apparent at DSD256, DSD512 takes things to another level, and well supported and inexpensive DACs are available that enable us to access this incredible level of SQ without spending 10s of thousands. I am grateful to Eurodriver for his support.

 

Key to this achievement is the use of so called "raw or native DSD" support on the playback chain. Unfortunately at the moment Linux has only nascent support for raw DSD and this results in frequent pop and ticks that makes Linux an unusable solution for day to day use.

 

Hoping this was fixed with the microRendu, I bought one to use as NAA. It was later returned after realizing that the problem persisted. Expectations are the SMS-200 will suffer the same problem as it appears to run a hacked version of the same Fedora based software. Led by Jesus R, many people are working to solve this problem, and it looks like progress is being made, so I am hopeful that Linux will be back in my system the near future.

 

In the meantime and FWIW, HQplayer upsampling to DSD512 with Windows 10 sounds phenomenal, and better then any other solution heard here to date. I would encourage you to purchase a system with this capability in your new build, even if you don't use it initially. You don't need the k version and related overclocking which is why I recommended a fixed clock version. Given your obvious passion for SQ, it would be a shame to see you build a new machine with a processor that is one model upgrade and $10 to 20 from being a dsd512 capable machine.

 

I'll shut up now.

 

Larry

Hi Larry,

 

As always, thank you for your valued input. I am very much in favor of upsampling, it's just I do it in my DAC (Chord DAVE) and not in my PC. DAVE upsamples well beyond DSD512 or 768kHz PCM via its FPGA. Chord's new M-scaler (due out next month) will allow DAVE to upsample even more (6x more) and I have my order placed already for this unit. Having heard it at CES in January, the differences are tremendous in terms of clarity, dynamics and depth.

 

I own 2 HQPlayer licenses (for Windows and Mac) and I have played with it extensively over the past year. I have a robust workstation with dual 8-core Xeons, 64GB of RAM and an nVidia K5000 GPU with 1536 CUDA cores and so upsampling to anything has not been a problem. In other systems based on chip DACs that I have heard, upsampling to DSD 128 and 256 resulted in a very nice improvement. Most chip DACs have considerable substrate noise and the smoothness of DSD nicely glosses over any harshness but DSD has its own compromises that I struggle with. With my DAVE, DSD512 resulted in a step down in SQ even though the DAVE's ASIO driver is capable of native DSD512 and not just DoP. Even 16/44 Redbook sounds better to my ears than anything upsampled to DSD512 with this DAC. I am probably in the minority in that I don't like the softness of DSD and its inherent timing issues and with HQP, I actually preferred upsampling to 768kHz over DSD512 but with my DAC, no HQP oversampling still sounded better.

 

This works out in my favor in that I don't have to concern myself with uber powerful (and electrically noisy) hardware.

[romaz]

I don't know if anyone has noticed this relatively new "audiophile" switch:

best audiophile Switch, high end switch - AQVOX Audio Devices

It's a nice looking switch and they seem to have addressed all the issues one would want addressed except maybe optical isolation but I agree, the asking price is steep given that the quality of the clock is uncertain.

 

Paul Pang has now released an actual audiophile-class router/switch with an OCXO clock and as expected, it isn't cheap:

 

SHOPPING AREA: AUDIO GRADE ROUTER ,SWITCH

[romaz]

Great post Roy.

 

Of course your above paragraph is a bit of deja vu from 3 years ago :

http://www.computeraudiophile.com/f27-uptone-audio-sponsored/attention-current-mac-mini-users-boot-mavericks-sd-card-load-ramdisk-dismount-your-internal-sata-drives-and-pour-drink-musicians-walking-out-your-speakers-18159/

 

While my optimized SD card is still running Mavericks (with about 65 processes/280 threads), I'm trying to make time to slim an El Capitan install--to at least get under 100 processes--so I can run the current version of HQ Player (via other than the microRendu/NAA). Staying away from the truly bloated Sierra!

 

It is quite something how sonically competitive a simply-tweaked Mac mini can be. My setup with it is:

2012 i7 2.6GHz with 16GB RAM

Our Mac mini DC-conversion/Linear Fan Controller Kit (MMK)

Our JS-2 choke-filtered, dual-output, 5-7 amp LPS

32GB SD card booting heavily-slimmed OS

Internal SATA HD dismounted (I'd remove it entirely, but really need it, with its multiple partitions, for experiments and OS image building; but I "eject" all partitions when using for playback)

 

Only connections to the computer are:

1) the DC power cable (and the coax/SMA for the MMK>JS-2 pseudo-Kelvin-sense voltage feedback circuit);

2) the USB cable (ports 1 and 3 are the best sounding by far) feeding my pre-production ISO REGEN (itself powered by an LPS-1 of course)

3) 25 foot BlueJeans/Belden Cat6a Ethernet cable direct to the 27" 4GHz iMac on my desk--used for Screen Sharing and sharing of a large drive with my music library.

 

I have the direct Ethernet connection set up for Jumbo Frames (MTU 9000) as that reduces overhead and I thought I heard a difference long ago.

 

--------

 

I might build up a high-power machine for DSD512 someday (Larry bugs me about it all the time ), but life is too short for me stress with Windows. Seems likely that Linux DSD issues will get worked out, and I do think that a split high-horsepower SRC/SDM machine to low power, all linear regulated renderer device will be the most desirable.

 

--------

 

Continue on gents. I watch with pleasure...

 

--Alex C.

Thanks, Alex. I remember coming across your thread a few years back. The title of your thread was too compelling not to want to read through it. As I stated before, what you had done was a source of inspiration.

 

I agree that a Mac Mini is nice setup for a music server. It's housed in an attractive and fairly resonant-free aluminum chassis. Although Apple had different intentions for doing it, in shielding their RAM and CPU in a stainless steel cage, they also effectively provided a nice RF shield against 2 components that generate quite a bit of it. It also has very short signal paths and putting the SD card reader on a PCIE bus and allowing the OS to boot from it was a genius move. Despite its CPU fan, this thing generates only 11dB of noise and is essentially silent. Obviously, it lends itself to being powered by a 12V source and your MMK is the best way I know to do it. On top of that, their hardware just works which is the benefit of their closed system -- all their components have been vetted for compatibility and reliability. Compared against a NUC, I find it to be a better starting platform and many used Mac Minis can be found cheaply on Ebay.

 

I have not yet tried jumbo frames but it makes sense that playback of large music files this way could sound better.

[romaz]

I’ve been experimenting with various ethernet cables for both spans of the bridge that I’ve configured on my Mac Mini. Yesterday afternoon a friend (who’s also a Wireworld and Audioquest dealer) and I compared several different cables on both spans of that bridge. (Potentially dangerous to have a friend who brings audio goodies to your home and also brings with him a credit card reader. LOL)

 

Location 1: Paul Pang switch to microRendu (all 1 m):

 

1A: WW Starlight (~ $210 for 1m)

1B: Nordost Blue Heaven (~ $400 for 1m)

1C: WW Platinum Starlight (~ $850 for 1m)

 

Location 2: wall jack to Mac Mini (both 8m):

 

2A: WW Chroma (~ $200 for 8m, or $80 termination fee + $15 per meter)

2B: Audioquest Cinnamon (~ $200 for 8m)

 

(A custom 18" run of Audioquest Carbon CAT700 was used between Mac Mini and Paul Pang switch the entire time).

 

First up was 1A vs 1C with 2B left in place. It took a number of swaps, but I thought I heard a slight veil removed with 1C. It seemed subtle though.

 

Next was 2A vs 2B with 1C left in place. Much easier to hear important differences. The track "Words of Wonder" from Keith Richards' "Main Offender" album features some really interesting high hat playing. With the 2A everything was in the right place such that the high hats and the way they were struck was very convincing both spatially and tonally. 2B disrupted that somewhat by tossing out some important cues that conveyed realism. It also threw less clear of an imaging picture with a few other tracks. Back in the box it went.

 

Leaving the 2A in place we then swapped 1A for 1C and there was an apparent loss of transparency. The spatial and tonal cues of the high hat were mostly retained, but the high hat just didn't stand out as clearly. (Sigh)

 

The more we listened with 1C/2A in place the more impressed we became with its transparency. It did the best job of just getting the heck out of the way - but there’s definitely diminishing returns when considering the price of 1C over 1A. But if one is looking to squeeze out that very last bit, it's definitely worth giving this cable an audition. No, I didn't buy one.

 

Maybe most impressive was 2A - at least relative to the well regarded 2B. Construction-wise it also feels like you are getting much more from your money with 2A. The same connectors are used on all three levels of WW Ethernet.

 

Last up was 1B. This didn't get the same level of attention in terms of comparisons. It was about on par with the 1A, so it seemed too be priced too high. Also disappointing to see a $400 cable with connectors featuring the typical plastic tabs that seem to too easily break off.

 

The 8m run of WW Chroma remained in place last night after my friend left. I thought I heard some nice improvements over what I had been hearing with the 8m run of Audioquest Cinnamon in place, including a blacker background and improved clarity - particularly in the upper frequencies. I should point out that the Cinnamon had about three weeks to burn-in, while the WW Chroma had only 3-4 nights of burn-in prior to yesterday.

 

Hope this was of interest.

Thanks, Kenny! Very helpful. My impression is that the differences you heard weren't dramatic, is that right? I used to own the WW Silver Starlight and it made an obvious difference in my system (probably the most obvious of any ethernet cable I've ever tried) but not for the better (too much treble energy) and so I sold it recently. I know you didn't buy the WW Platinum. What did you decide to go with for your location 1? It sounds like you're sticking with the WW Chroma for Location 2?

 

I have also compared the BJC Cat6A and I find that cable to probably be the best value given its low cost. It uses floating shields and solid core copper and have rugged terminations. I found the relatively inexpensive Supra CAT8 to also sound good (slightly better than the BJC) in my setup. I tried the AQ Vodka and Diamond and they sounded fine but I didn't feel they were worth the money. I use SOtM's CAT6 only because I already own it (I bought it 2 years ago) and it seems to be as good as the Supra CAT8 resulting in better smoothness. Thus far, I have not found ethernet cables to make a huge difference but rather a nice finishing touch.

[romaz]

I've been following this thread with keen interest. Thanks for all those who have contributed, especially Romaz! I have been thinking as to whether or not this setup with a direct connection from my Mac Mini to my PS Audio DirectStream DAC with Bridge II. Keep in mind, the Bridge II on the DS DAC connects via ethernet and incorporates a Roon endpoint. So, from what I can gather, there is no obvious reason I couldn't "bridge" my Mac Mini with thunderbolt dual ethernet dongle and run direct to the Bridge II card on my DS DAC. Has anyone tried this?? I would love to ditch my USB connection. Thanks in advance for your input.

I don't see why it wouldn't work. I used to own the DirectStream and I would be interested to know if this direct connection improves it.

[romaz]

Roy,

 

Thank you very much for your feedback. I really enjoy your every article and am always looking forward to your very thorough insight into each topic.

 

Back to sMS-200, by further investigating into it I realize that the CPU used is A20 from Allwinner. It's a dual core Cortex A7 ARM CPU. It needs an external Ethernet PHY which is Realtek RTL8211E. From Allwinner's linux-sunxi community it mentioned:

 

"For reliable Gigabit networking (1000Mbit operation), several sunxi devices require an important tweak that adjusts the relative timing of the clock and data signals to the PHY, in order to compensate for differing trace lengths on the PCB"

 

I believe that SOtM must have put a lot of effort in tweaking their Ethernet driver to come to it's present state. That may also explains why they prefer 100Mbps over gigabit at the early product stage.

 

I totally agree with your low impedance perspective. Yet I have some different point of view regarding the latency issue. The way I look at latency is a little similar to your view of CPU clock. It doesn't have to be the lowest but the consistent and predictable timing performance is one of the key factor to a good sounding system.

 

Another point that we all agree that could degrade SQ is the EMC interference. As we all know wherever electric current goes there's electric-magnetic field generated. It gets worse if the current fluctuates frequently. And as you said, which I couldn't agree more is, that "it is always better to prevent a problem or to address it at the source than to have to fix it downstream".

 

One way to achieve low current fluctuation is to use a fixed system clock. All that an audiophile music server needs is a steady and consistent operation without power saving nor hyper speed function. A fixed frequency system clock can do it just fine. And it is a unique feature in Daphile which I believe can improve the SQ quite a bit.

 

Based off the heavy lightweight media server belief, I'm very interested to give ROCK a try. I'll report back when I done the test.

Thanks, gadgetman. I always find it interesting to know the specific components used within a product and whether they were chosen for quality reasons or for cost-saving reasons. I'm sure SOtM was very intentional in their design of the sMS-200.

 

You've brought about an interesting point with CPU clocking that I hadn't considered. It completely makes sense to have as little current fluctuation as possible and so it would make sense to cap CPU frequency at a specific baseline and prevent turbo mode which is easy enough to do in BIOS. I already disable any power saving.

 

Yes, I'm interested to hear how ROCK sounds. Unfortunately, it doesn't lend itself to end-user tweaking and so you can't bridge your LAN ports within ROCK.

[kennyb123]

My impression is that the differences you heard weren't dramatic, is that right?

 

The improvement from WW Platinum Starlight over WW Starlight fell into "subtle but important, but not important enough that I couldn't live without it". So no, definitely not dramatic.

 

But the improvement from WW Chroma over AQ Cinnamon was more along the lines of "I gotta have that!", but of course the price factored into that. Would I have counted it as dramatic though? Not sure I'd go that far but once I was made aware of the harm Cinnamon was doing, it had to go. That was further reinforced in last night's listening.

 

I used to own the WW Silver Starlight and it made an obvious difference in my system (probably the most obvious of any ethernet cable I've ever tried) but not for the better (too much treble energy) and so I sold it recently.

 

As I mentioned I heard nothing like that here relative to any of the other Ethernet cables I've had here. Maybe you could try their Chroma?

 

I know you didn't buy the WW Platinum. What did you decide to go with for your location 1? It sounds like you're sticking with the WW Chroma for Location 2?

 

WW Starlight at location 1 and WW Chroma at location 2. I am *extremely* pleased with how everything is sounding now. I honestly still can't get over how I got such a massive upgrade from swapping my Mac Pro for the Mac Mini/MMK/JS-2.

 

I have also compared the BJC Cat6A and I find that cable to probably be the best value given its low cost.

 

Given all the positive comments about BJC's Ethernet cables, I know I should have tried some. I'm just having trouble getting over a not very positive experience I had listening to one of their interconnects in my system several years ago. Sure their interconnect cost a fraction of a fraction of a fraction of the cost of other interconnects I had on hand, but it just so terribly smeared things. Given that their technology did such a poor job of preserving the fidelity of an analog signal, how could it possibly do a good job passing Ethernet packets with high fidelity? Obviously I won't know until I try one.

 

Thus far, I have not found ethernet cables to make a huge difference but rather a nice finishing touch.

 

That was exactly my opinion prior to swapping my Mac Pro for the Mac Mini/MMK/JS-2. I think I would now say that proper Ethernet cables are a "necessary finishing touch".

 

 

 

Sent from my iPad using Computer Audiophile