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

Johnseye · April 16, 2018

5 hours ago, ray-dude said:

Thanks @d_elm. My understanding (very loose, since we only have been able to speak briefly) is that he recommends a specific 6AWG cable. I suspect it is solid core (based on posts of others), and the typical 6 AWG (and 8AWG) power cable is stranded. Jim recommends the 10 AWG Romex, which is also solid core. I've been poking around, but I haven't (yet) found any solid core options in 6 AWG or 8 AWG.

I defer to others who've been able to work with Jim more than I have to clarify one way or the other (I hope to chat again with him this week)

5 hours ago, Bruce Orr said:

I need 80+ feet!!

My feeling talking with Jim was that the 6 AWG is not solid, but he does prefer that in the 10. I'm not exactly sure what it is that Jim does to customize it. I asked that question in the email I just sent him. If he is customizing it, I asked what he is doing to it, and why he prefers the AWG 10 solid over a "standard" AWG 6

The 6 awg is stranded. You'd have a very hard time running solid 6 awg, especially with bends.

jacquesr · April 16, 2018

Hi,

I have been following this thread and I am seriously considering getting an Innos Zenith SE or maybe a statement.

I have mainly three questions at this stage:

a) I have 1,000 cds ripped in AIFF... Is that supported?

b) I currently use HQP (see my signature) and I have a hard time imagining parting with it. Should I keep my mac mini setup (and eventually get a new Mac Pro 2019) and use the Zenith in NAA?

c) Or would an antipodes CX + ES be a better fit for me?

I don't do/want to do PCM -> DSD conversion. I just like the flexibility and filters. Or maybe the SQ with SE/Statement stands on it own and I will just simplify my chain....

Feedback welcome.

BigAlMc · April 16, 2018

4 hours ago, jacques_racine said:

Hi,

I have been following this thread and I am seriously considering getting an Innos Zenith SE or maybe a statement.

I have mainly three questions at this stage:

a) I have 1,000 cds ripped in AIFF... Is that supported?

b) I currently use HQP (see my signature) and I have a hard time imagining parting with it. Should I keep my mac mini setup (and eventually get a new Mac Pro 2019) and use the Zenith in NAA?

c) Or would an antipodes CX + ES be a better fit for me?

I don't do/want to do PCM -> DSD conversion. I just like the flexibility and filters. Or maybe the SQ with SE/Statement stands on it own and I will just simplify my chain....

Feedback welcome.

Hi,

a) No. The Zeniths support FLAC & WAV.

b) The SE is primarily a server and not an NAA. There's nothing to stop you using it as an NAA with your music files elsewhere but it's Roon implementation doesn't support HQPlayer.

Personally if you want to keep HQPlayer and want to go the NAA route then the Sonore Signature Rendu seems a better fit.

c) dunno on the antipodes front.

Cheers,

Alan

Bruce Orr · April 16, 2018

5 hours ago, BigAlMc said:

Hi,

a) No. The Zeniths support FLAC & WAV.

b) The SE is primarily a server and not an NAA. There's nothing to stop you using it as an NAA with your music files elsewhere but it's Roon implementation doesn't support HQPlayer.

Personally if you want to keep HQPlayer and want to go the NAA route then the Sonore Signature Rendu seems a better fit.

c) dunno on the antipodes front.

Cheers,

Alan

You are right the Zenith does not have the processing power to manage significant upsampling - even from Roon Core in the Zenith.

I use Roon Core in my iMac and do my upsampling on the Zenith from there.

rickca · April 16, 2018

Does the Innuos Statement still use a J1900 CPU or is it something more modern/powerful?

jaspal kallar · April 16, 2018

6 hours ago, BigAlMc said:

a) No. The Zeniths support FLAC & WAV.

I presume it can play back AIFF files (among others), however ripping only supports FLAC & WAV. Is that correct?

BigAlMc · April 16, 2018

33 minutes ago, jaspal kallar said:

I presume it can play back AIFF files (among others), however ripping only supports FLAC & WAV. Is that correct?

Good point. Yeah the UI shows only FLAC & WAV but that's probably more to do with ripping than play back.

47 minutes ago, Bruce Orr said:

You are right the Zenith does not have the processing power to manage significant upsampling - even from Roon Core in the Zenith.

I've played around with upsampling to DSD128 in Roon DSP and that works fine on the SE or MKII. That's the maximum my Directstream DAC accepts so I never had reason to try more. But I agree that it doesn't have the CPU power needed for heavier upsampling like HQPlayer or DSD512.

austinpop · April 17, 2018

On 4/14/2018 at 10:00 AM, Bruce Orr said:

Rajiv, two questions and thank you in advance for al you contribute to these threads:

Hi Bruce, sorry for the delay. I just got back from AXPONA.

On 4/14/2018 at 10:00 AM, Bruce Orr said:

1). I have been waiting for sometime to get the AWG 6 wire from Jim Weil so I can install two dedicated circuits. Actually, I've been waiting since January and growing impatient (it's the supplier, not Jim, that's the hold up). Are you using this or the AWG 10? He claims a significant SQ enhancement using the larger 6 gauge.

I am using Jim's 6 AWG wire. Luckily I only needed a few feet, and got in under the wire - so to speak.

On 4/14/2018 at 10:00 AM, Bruce Orr said:

2) I presume you have a 2-plug socket. Sounds like you are using one to power your amp directly. Are you using the second for your P5?

Correct on both counts.

austinpop · April 17, 2018

I just got back from AXPONA 2018 in Chicago. I'll talk about the show in another post, but I just wanted to thank @Johnseye for opening his home and man cave to myself, Larry @lmitche, Chris @Forehaven and my non-CA college buddy with whom I stayed. We had a blast listening and doing comparisons, and John and I will be reporting on that soon. But most of all, it was a blast meeting all of these great guys in person. That is really the point of these shows.

I came away thoroughly impressed with how knowledgeable and accomplished each of them is, not just in computer audio, but in their professional lives. I hope I get to meet more of you in person some day!

Johnseye · April 17, 2018

16 minutes ago, austinpop said:

I just got back from AXPONA 2018 in Chicago. I'll talk about the show in another post, but I just wanted to thank @Johnseye for opening his home and man cave to myself, Larry @lmitche, Chris @Forehaven and my non-CA college buddy with whom I stayed. We had a blast listening and doing comparisons, and John and I will be reporting on that soon. But most of all, it was a blast meeting all of these great guys in person. That is really the point of these shows.

I came away thoroughly impressed with how knowledgeable and accomplished each of them is, not just in computer audio, but in their professional lives. I hope I get to meet more of you in person some day!

The time spent at Axpona, meeting those who we have conversations with on a regular basis in these forums, listening to new equipment together, discovering, hanging out and just having a lot of fun, was one of the best times I've ever had at a convention. Everyone was incredibly friendly and engaging.

It was a pleasure to have these guys over and I only wish we could have had more time together. One of my biggest take aways from that experience was that we had 3 different approaches together in the same room. Rajiv with the trifecta, myself with the clock modified low powered server and Larry & Chris with the high powered, upsampling PC. Despite our different approaches we all agreed on everything we heard. More to come as we document our findings.

Larry and Chris are out at the Green Mill tonight seeing Patricia Barber with Lee and May from SOtM. I wish I could be there. I'm sure they're having a great time.

austinpop · April 17, 2018

While at AXPONA, I had several long and illuminating conversations.

First, at the SOtM room, May and Lee showed me their prototype Ethernet switch. The price point is still not decided. Physically the switch is rather large and substantial. See the picture - it is the top component on the stack.

Adding to what has already been reported, I got a better understanding of how they plan to implement clocking. The switch can be ordered either with an SMB input, to be supplied by an existing sCLK-EX board in your system, or with an sCLK-EX board in the box, with an optional reference clock input, just like other Ultra components. I am particularly interested in the cost and performance of the SMB version, without the internal sCLK-EX and am looking forward to compare it with my modded switch.

I also had a good chat with Adrian @allaboutacoustic of The Linear Solution. He is planning to come out with a music server of his own design, using the concepts of low power, clean power, and low phase noise clocking, as we've been discussing here on this thread. He has a prototype of the OS they have developed (a stripped down Linux variant) that he demo'd for me on his laptop driving a Hugo2 and Final headphones. Without a point of reference or comparison, I'll reserve judgement, but it was sounding mighty nice. @Johnseye met us there and picked up a loaner sample of Adrian's OCXO switch and PSU, which we tested the next day (yesterday). Based on this, I plan to try out Adrian's products in my system as soon as I get a chance.

Finally, I had a long discussion yesterday with Nuno @nvitorino of Innuos, regarding not just their new Zenith Statement box, but more generally, their approach to server design. First off, I heard the demo of the Statement vs. the SE in their room. The differences I heard were not surprising at all to me, since they're consistent with what I have heard in my own system with better clocking and PSUs. The Statement prototype at the show implements a new custom USB board, with an onboard OCXO oscillator. In addition, they have gone to an 8-rail outboard PSU, that connects via an umbilical to the Statement, which internally houses the final regulation stage of the power supply.

Nuno was very open with me about the issues I raised with him. I expressed concern at the steep jump in price, and suggested it may be too large a gap even for current SE customers. He is aware of this, and they are considering a standalone USB outboard box, that incorporates their OCXO USB interface with its own Sean Jacobs PSU, that could be offered as an add on to SE customers. I also asked him whether they were considering OCXO for the Ethernet and the system (motherboard). He said they were, but they have to be careful about exploding costs.

What impressed me about Nuno was his openness to the input of our community. He credited our explorations as one of the inputs into their deign process for future products. Our efforts are valuable research for them, although I pointed out that they were in a unique position to explore some things that are hard for us to do - for example, do OCXO clocks matter equally for Ethernet, USB, and system clocks? In that context, he described a research collaboration they are undertaking with Warwick University, where they are co-located, to use machine learning and neural nets to aggregate and navigate the experiments they are doing.

It was a great set of conversations. These face-to-face interactions are really the best part of going to these shows!

tapatrick · April 17, 2018

2 hours ago, austinpop said:

It was a great set of conversations. These face-to-face interactions are really the best part of going to these shows!

Invaluable input, many thanks Rajiv!

auricgoldfinger · April 17, 2018

10 hours ago, austinpop said:

Based on this, I plan to try out Adrian's products in my system as soon as I get a chance.

Adrian is building the following custom power supply for me that should be comparable in performance to a Paul Hynes SR5DRXL (and SR7).

2 x 12v/6A rails with dual regulator per each output (fixed voltage)

Cryogenic treatment of transformer

Internal wiring stranded silver (Neotech 16AWG OCC)

Mundorf caps

Furutech IEC FI-06 NCF

Here are the basic specs:

Noise Voltage <0.5 nanovolts root Hz
Noise Reduction +/- 80db 0 to 100KHz *

> 82dB 100KHz to 2MHz *
Efficiency >85%
Continuous Bandwidth > 600 MHz
Output Impedance < 0.0028 ohms

* Cryogenic treatment of the transformer will improve the noise reduction beyond these levels which refer to the standard transformer

The dual regulator makes this PSU statistically equivalent to the SR5DRXL that I will hopefully receive by late summer. With the cryo treated transformer and Mundorf caps, the Linear Solution PSU could be even better. Cost wise, it is a no-brainer compared to the SR5 and SR7.

Johnseye · April 17, 2018

The Sunningdale Session

Sunday afternoon Rajiv @austinpop and his long time friend came by my place after Axpona. Rajiv and I had loosely planned this out. His equipment was brought over the night before and powered on. We hoped to accomplish much more testing than we had time for, but I think it would have taken days to get through everything we wanted to accomplish. Once we began, we realized we needed to reduce the number of songs used as a benchmark to one or two. These tracks were Freddie Freeloader from Miles Davis’s Kind of Blue 24/192 and Mahler’s 5^th Symphony: 5. Rondo-Finale, Osmo Vänskä, Minnesota Orchestra, BIS 24/96 upsampled to 24/192. We both knew these tracks enough to easily discern any change, and both offered differences with musical dynamics. Roon was used as the player. No other DSP settings were applied, other than upsampling Mahler to 192. Both tracks were sourced from my Synology NAS.

Equipment used consistently for all tests were as follows: Paradigm Persona 3F speakers, Benchmark AHB2 amp, Audio Research LS-28 preamp and SOtM tX-USBultra. All cables were the same. Speaker cables were from DH Labs, balanced cables were Audioquest Colorado, USB was the Phasure Lush and AC power cables were Shunyata Venom.

Equipment changed during the tests were as follows: sMS-200ultra, custom server running Roon ROCK (sCLK-EX providing taps to mobo system, LAN, tX-USBexp and tX-USBultra), Windows 10 running Roon Server, tX-USBexp, SATA Filter 2 amongst other components), Audio Alchemy DDP-1 DAC with PS-5 SMPS, Ayre Codex DAC, Cybershaft OP-14 10 mhz master clock, modified Zyxel switch (sCLK-EX from the sMS-200ultra providing the clock tap), Linear Solutions OCXO modified TP-Link switch and included LPS, Paul Hynes SR7 and SR4, and SOtM sPS-500.

We first cycled through a minute or two of about 5 tracks to get a benchmark of my system.

We then added Rajiv’s sCLK-EX modified Zyxel switch and the sMS200ultra. The tX-USBultra powered by the SR7 remained in line just prior to the Audio Alchemy DAC throughout. The sMS200ultra was powered by the SR4 and the switch was powered by the sPS-500. My system was powered by the SR7 except for the hard drive being powered by an LPS-1. After switching back and forth between my system and the sMS200ultra several times we ultimately concluded that my sCLK modified system was slightly smoother while the sMS-200ultra introduced a slightly hard digital edge, or maybe that it didn’t smooth out the edge that is commonly attributed to digital music. The difference was subtle to the point that we needed to make several swaps back and forth to be sure of what we were hearing.

We then sourced Roon from my Windows 10 PC. This test was to determine whether the source upstream to the sMS-200ultra had an impact. Swapping between the Windows 10 PC and my custom server with the sMS-200ultra as the endpoint revealed subtle differences with the custom PC being preferable, having a smoother less edgy sound commonly attributed to digital audio.

At this point Larry @lmitche and Chris @Forehaven joined us.

Wanting to discern whether there was an audible difference between the SR7 and SR4 we used the 12v DR rail, which had been powering my custom server mobo, to power the sMS-200ultra. Using the Windows 10 PC we swapped between SR4 and SR7 as the power source to the sMS, all other equipment being equal. What we all agreed on was that the SR7 elicited more bass, and that this was actually a negative. Reason being that the additional bass drowned out other frequencies. In this case we preferred the SR4. This test would benefit from additional verification.

It was Larry who thought of the possible reason for this. The SR4 is a single rail supply. All components of the supply are dedicated to that rail, while the SR7 is dual rail sharing some components. Based on this hypothesis we concluded that dedicated power supplies for each component results in superior SQ. A big leap, but a very likely possibility that needs additional testing.

Next we added the Cybershaft OP-14. We went back to using my custom server and the Windows 10 PC was no longer used at all. The OP-14 provided the master clock solely to the sCLK-EX housed in my server. There was an immediate and noticeable improvement in SQ. There was more bass, but the bass was very defined to the point of clarity where plucked bass strings were discerned not just blended in a boomy sound. In fact all instruments were more defined creating an overall richness to the sound.

Keeping the Cybershaft in line, I swapped between the Linear Solutions TP Link OCXO modified switch with custom LPS and the Zyxel sCLK-EX modified switch powered by the sPS-500. I swapped this back and forth several times not informing anyone which was which until after. The conclusion was that the Linear Solutions switch sounded better. The difference wasn’t as significant as the OP-14, but it was noticeable. This test could have been refined more by swapping power supplies between the two. We could have also tried using the master clock on the sCLK-EX supporting the Zyxel’s clock.

Finally we compared the Audio Alchemy DDP-1 and PM-5 SMPS with the Ayre Codex. The AA was used for all previous tests. Thanks to Rajiv we took great care to level match. This was an effort I did not do as well when I reviewed these two DACs two years ago. My previous level matching was by ear. In this effort we used an app on Rajiv’s phone. This made a significant difference and is a lesson I will always keep in mind. What everyone agreed upon was that the Codex had a richer texture while the AA sounded slightly thinner.

We could have spent a lot longer and this was such a fun exercise that I look forward to doing it again. It was a challenge to get back to listening to the tracks because we were having such a good time talking. I want to thank Rajiv for making such a big effort to bring his equipment all the way from Austin. I also want to thank Larry and Chris for driving out to my place. It was a very busy weekend and all too short.

TL;DR

· A server with sCLK-EX modified motherboard and tX-USBexp sounds slightly better than an sMS-200ultra, all other things being equal.

· A Linear Solutions OCXO modified TP Link switch with custom LPS sounds slightly better than an sCLK-EX modified Zyxel switch. The Linear Solutions switch and LPS is an even better improvement over a Netgear GS108 powered by a battery (a test I ran previously).

· An sMS-200ultra using a single rail 12v SR4 could possibly have SQ benefits over a dual rail 12v DR SR7. Rajiv and Eric to conduct further tests.

· A Cybershaft OP-14 master clock without a doubt improves SQ.

· An Ayre Codex sounds better than an Audio Alchemy DDP- with PS-5.

auricgoldfinger · April 17, 2018

10 minutes ago, Johnseye said:

It was Larry who thought of the possible reason for this. The SR4 is a single rail supply. All components of the supply are dedicated to that rail, while the SR7 is dual rail sharing some components. Based on this hypothesis we concluded that dedicated power supplies for each component results in superior SQ. A big leap, but a very likely possibility that needs additional testing.

Below is an excerpt from a conversation that I had with Paul Hynes several months ago. Keep in mind that I was considering a power supply with dual regulators. The sonic penalty will be more audible in a 4-rail SR7 that lacks dual regulators. For this reason (and also cost), I chose the SR5DRXL to avoid any potential sonic penalties.

Moving to the SR7DRMR2XL would provide two galvanically isolated supply rails. There is a slight sonic penalty with having all the transformer secondaries on one transformer core as nothing is absolutely perfect and there will be some small level of interaction due to transient load current swings. This may be audible in a high-resolution system but it would be rather subtle due to the very high supply line rejection of the voltage regulator circuitry in the DR mode.

Johnseye · April 17, 2018

1 minute ago, auricgoldfinger said:

Below is an excerpt from a conversation that I had with Paul Hynes several months ago. Keep in mind that I was considering a power supply with dual regulators. The sonic penalty will be more audible in a 4-rail SR7 that lacks dual regulators. For this reason (and also cost), I chose the SR5DRXL to avoid any potential sonic penalties.

Moving to the SR7DRMR2XL would provide two galvanically isolated supply rails. There is a slight sonic penalty with having all the transformer secondaries on one transformer core as nothing is absolutely perfect and there will be some small level of interaction due to transient load current swings. This may be audible in a high-resolution system but it would be rather subtle due to the very high supply line rejection of the voltage regulator circuitry in the DR mode.

And that is exactly what @lmitche said. It appears we've heard the sonic penalty of a single transformer.

AmusedToD · April 17, 2018

20 minutes ago, Johnseye said:

The Sunningdale Session

Sunday afternoon Rajiv @austinpop and his long time friend came by my place after Axpona. Rajiv and I had loosely planned this out. His equipment was brought over the night before and powered on. We hoped to accomplish much more testing than we had time for, but I think it would have taken days to get through everything we wanted to accomplish. Once we began, we realized we needed to reduce the number of songs used as a benchmark to one or two. These tracks were Freddie Freeloader from Miles Davis’s Kind of Blue 24/192 and Mahler’s 5^th Symphony: 5. Rondo-Finale, Osmo Vänskä, Minnesota Orchestra, BIS 24/96 upsampled to 24/192. We both knew these tracks enough to easily discern any change, and both offered differences with musical dynamics. Roon was used as the player. No other DSP settings were applied, other than upsampling Mahler to 192. Both tracks were sourced from my Synology NAS.

Equipment used consistently for all tests were as follows: Paradigm Persona 3F speakers, Benchmark AHB2 amp, Audio Research LS-28 preamp and SOtM tX-USBultra. All cables were the same. Speaker cables were from DH Labs, balanced cables were Audioquest Colorado, USB was the Phasure Lush and AC power cables were Shunyata Venom.

Equipment changed during the tests were as follows: sMS-200ultra, custom server running Roon ROCK (sCLK-EX providing taps to mobo system, LAN, tX-USBexp and tX-USBultra), Windows 10 running Roon Server, tX-USBexp, SATA Filter 2 amongst other components), Audio Alchemy DDP-1 DAC with PS-5 SMPS, Ayre Codex DAC, Cybershaft OP-14 10 mhz master clock, modified Zyxel switch (sCLK-EX from the sMS-200ultra providing the clock tap), Linear Solutions OCXO modified TP-Link switch and included LPS, Paul Hynes SR7 and SR4, and SOtM sPS-500.

We first cycled through a minute or two of about 5 tracks to get a benchmark of my system.

We then added Rajiv’s sCLK-EX modified Zyxel switch and the sMS200ultra. The tX-USBultra powered by the SR7 remained in line just prior to the Audio Alchemy DAC throughout. The sMS200ultra was powered by the SR4 and the switch was powered by the sPS-500. My system was powered by the SR7 except for the hard drive being powered by an LPS-1. After switching back and forth between my system and the sMS200ultra several times we ultimately concluded that my sCLK modified system was slightly smoother while the sMS-200ultra introduced a slightly hard digital edge, or maybe that it didn’t smooth out the edge that is commonly attributed to digital music. The difference was subtle to the point that we needed to make several swaps back and forth to be sure of what we were hearing.

We then sourced Roon from my Windows 10 PC. This test was to determine whether the source upstream to the sMS-200ultra had an impact. Swapping between the Windows 10 PC and my custom server with the sMS-200ultra as the endpoint revealed subtle differences with the custom PC being preferable, having a smoother less edgy sound commonly attributed to digital audio.

At this point Larry @lmitche and Chris @Forehaven joined us.

Wanting to discern whether there was an audible difference between the SR7 and SR4 we used the 12v DR rail, which had been powering my custom server mobo, to power the sMS-200ultra. Using the Windows 10 PC we swapped between SR4 and SR7 as the power source to the sMS, all other equipment being equal. What we all agreed on was that the SR7 elicited more bass, and that this was actually a negative. Reason being that the additional bass drowned out other frequencies. In this case we preferred the SR4. This test would benefit from additional verification.

It was Larry who thought of the possible reason for this. The SR4 is a single rail supply. All components of the supply are dedicated to that rail, while the SR7 is dual rail sharing some components. Based on this hypothesis we concluded that dedicated power supplies for each component results in superior SQ. A big leap, but a very likely possibility that needs additional testing.

Next we added the Cybershaft OP-14. We went back to using my custom server and the Windows 10 PC was no longer used at all. The OP-14 provided the master clock solely to the sCLK-EX housed in my server. There was an immediate and noticeable improvement in SQ. There was more bass, but the bass was very defined to the point of clarity where plucked bass strings were discerned not just blended in a boomy sound. In fact all instruments were more defined creating an overall richness to the sound.

Keeping the Cybershaft in line, I swapped between the Linear Solutions TP Link OCXO modified switch with custom LPS and the Zyxel sCLK-EX modified switch powered by the sPS-500. I swapped this back and forth several times not informing anyone which was which until after. The conclusion was that the Linear Solutions switch sounded better. The difference wasn’t as significant as the OP-14, but it was noticeable. This test could have been refined more by swapping power supplies between the two. We could have also tried using the master clock on the sCLK-EX supporting the Zyxel’s clock.

Finally we compared the Audio Alchemy DDP-1 and PM-5 SMPS with the Ayre Codex. The AA was used for all previous tests. Thanks to Rajiv we took great care to level match. This was an effort I did not do as well when I reviewed these two DACs two years ago. My previous level matching was by ear. In this effort we used an app on Rajiv’s phone. This made a significant difference and is a lesson I will always keep in mind. What everyone agreed upon was that the Codex had a richer texture while the AA sounded slightly thinner.

We could have spent a lot longer and this was such a fun exercise that I look forward to doing it again. It was a challenge to get back to listening to the tracks because we were having such a good time talking. I want to thank Rajiv for making such a big effort to bring his equipment all the way from Austin. I also want to thank Larry and Chris for driving out to my place. It was a very busy weekend and all too short.

TL;DR

·         A server with sCLK-EX modified motherboard and tX-USBexp sounds slightly better than an sMS-200ultra, all other things being equal.

·         A Linear Solutions OCXO modified TP Link switch with custom LPS sounds slightly better than an sCLK-EX modified Zyxel switch. The Linear Solutions switch and LPS is an even better improvement over a Netgear GS108 powered by a battery (a test I ran previously).

·         An sMS-200ultra using a single rail 12v SR4 could possibly have SQ benefits over a dual rail 12v DR SR7. Rajiv and Eric to conduct further tests.

·         A Cybershaft OP-14 master clock without a doubt improves SQ.

·         An Ayre Codex sounds better than an Audio Alchemy DDP- with PS-5.

Thanks for the detailed analysis. You guys had some nice fun together, great to see an Internet forum bringing like minded people together like that!

Anyways, your collective finding that the Linear Solution OCXO switch is better than the SOtM sCLK modded switch is in line with Roy’s findings. That kind of narrows it down. The Linear Solution gets you the switch plus a linear PSU for only $649, whereas the SOtM switch (with a sCLK board) will probably be close to a grand (speculation based on their previous products pricing), and that’s without a PSU.

The LS switch will probably be the better buy, unless the upcoming Uptone switch kills both for less than $500.

lmitche · April 17, 2018

24 minutes ago, Johnseye said:

The Sunningdale Session

Sunday afternoon Rajiv @austinpop and his long time friend came by my place after Axpona. Rajiv and I had loosely planned this out. His equipment was brought over the night before and powered on. We hoped to accomplish much more testing than we had time for, but I think it would have taken days to get through everything we wanted to accomplish. Once we began, we realized we needed to reduce the number of songs used as a benchmark to one or two. These tracks were Freddie Freeloader from Miles Davis’s Kind of Blue 24/192 and Mahler’s 5^th Symphony: 5. Rondo-Finale, Osmo Vänskä, Minnesota Orchestra, BIS 24/96 upsampled to 24/192. We both knew these tracks enough to easily discern any change, and both offered differences with musical dynamics. Roon was used as the player. No other DSP settings were applied, other than upsampling Mahler to 192. Both tracks were sourced from my Synology NAS.

Equipment used consistently for all tests were as follows: Paradigm Persona 3F speakers, Benchmark AHB2 amp, Audio Research LS-28 preamp and SOtM tX-USBultra. All cables were the same. Speaker cables were from DH Labs, balanced cables were Audioquest Colorado, USB was the Phasure Lush and AC power cables were Shunyata Venom.

Equipment changed during the tests were as follows: sMS-200ultra, custom server running Roon ROCK (sCLK-EX providing taps to mobo system, LAN, tX-USBexp and tX-USBultra), Windows 10 running Roon Server, tX-USBexp, SATA Filter 2 amongst other components), Audio Alchemy DDP-1 DAC with PS-5 SMPS, Ayre Codex DAC, Cybershaft OP-14 10 mhz master clock, modified Zyxel switch (sCLK-EX from the sMS-200ultra providing the clock tap), Linear Solutions OCXO modified TP-Link switch and included LPS, Paul Hynes SR7 and SR4, and SOtM sPS-500.

We first cycled through a minute or two of about 5 tracks to get a benchmark of my system.

We then added Rajiv’s sCLK-EX modified Zyxel switch and the sMS200ultra. The tX-USBultra powered by the SR7 remained in line just prior to the Audio Alchemy DAC throughout. The sMS200ultra was powered by the SR4 and the switch was powered by the sPS-500. My system was powered by the SR7 except for the hard drive being powered by an LPS-1. After switching back and forth between my system and the sMS200ultra several times we ultimately concluded that my sCLK modified system was slightly smoother while the sMS-200ultra introduced a slightly hard digital edge, or maybe that it didn’t smooth out the edge that is commonly attributed to digital music. The difference was subtle to the point that we needed to make several swaps back and forth to be sure of what we were hearing.

We then sourced Roon from my Windows 10 PC. This test was to determine whether the source upstream to the sMS-200ultra had an impact. Swapping between the Windows 10 PC and my custom server with the sMS-200ultra as the endpoint revealed subtle differences with the custom PC being preferable, having a smoother less edgy sound commonly attributed to digital audio.

At this point Larry @lmitche and Chris @Forehaven joined us.

Wanting to discern whether there was an audible difference between the SR7 and SR4 we used the 12v DR rail, which had been powering my custom server mobo, to power the sMS-200ultra. Using the Windows 10 PC we swapped between SR4 and SR7 as the power source to the sMS, all other equipment being equal. What we all agreed on was that the SR7 elicited more bass, and that this was actually a negative. Reason being that the additional bass drowned out other frequencies. In this case we preferred the SR4. This test would benefit from additional verification.

It was Larry who thought of the possible reason for this. The SR4 is a single rail supply. All components of the supply are dedicated to that rail, while the SR7 is dual rail sharing some components. Based on this hypothesis we concluded that dedicated power supplies for each component results in superior SQ. A big leap, but a very likely possibility that needs additional testing.

Next we added the Cybershaft OP-14. We went back to using my custom server and the Windows 10 PC was no longer used at all. The OP-14 provided the master clock solely to the sCLK-EX housed in my server. There was an immediate and noticeable improvement in SQ. There was more bass, but the bass was very defined to the point of clarity where plucked bass strings were discerned not just blended in a boomy sound. In fact all instruments were more defined creating an overall richness to the sound.

Keeping the Cybershaft in line, I swapped between the Linear Solutions TP Link OCXO modified switch with custom LPS and the Zyxel sCLK-EX modified switch powered by the sPS-500. I swapped this back and forth several times not informing anyone which was which until after. The conclusion was that the Linear Solutions switch sounded better. The difference wasn’t as significant as the OP-14, but it was noticeable. This test could have been refined more by swapping power supplies between the two. We could have also tried using the master clock on the sCLK-EX supporting the Zyxel’s clock.

Finally we compared the Audio Alchemy DDP-1 and PM-5 SMPS with the Ayre Codex. The AA was used for all previous tests. Thanks to Rajiv we took great care to level match. This was an effort I did not do as well when I reviewed these two DACs two years ago. My previous level matching was by ear. In this effort we used an app on Rajiv’s phone. This made a significant difference and is a lesson I will always keep in mind. What everyone agreed upon was that the Codex had a richer texture while the AA sounded slightly thinner.

We could have spent a lot longer and this was such a fun exercise that I look forward to doing it again. It was a challenge to get back to listening to the tracks because we were having such a good time talking. I want to thank Rajiv for making such a big effort to bring his equipment all the way from Austin. I also want to thank Larry and Chris for driving out to my place. It was a very busy weekend and all too short.

TL;DR

·         A server with sCLK-EX modified motherboard and tX-USBexp sounds slightly better than an sMS-200ultra, all other things being equal.

·         A Linear Solutions OCXO modified TP Link switch with custom LPS sounds slightly better than an sCLK-EX modified Zyxel switch. The Linear Solutions switch and LPS is an even better improvement over a Netgear GS108 powered by a battery (a test I ran previously).

·         An sMS-200ultra using a single rail 12v SR4 could possibly have SQ benefits over a dual rail 12v DR SR7. Rajiv and Eric to conduct further tests.

·         A Cybershaft OP-14 master clock without a doubt improves SQ.

·         An Ayre Codex sounds better than an Audio Alchemy DDP- with PS-5.

Nice writeup John. It was great to meet Rajiv, Vivek and yourself. Thank you for the invitation to your beautiful home and for including us in the evaluation session. Between us, we have taken several different approaches to improving SQ, and it was great to have a chance to compare. Best of all was the sense of comraderie. What a great team!

Larry

austinpop · April 17, 2018

11 minutes ago, lmitche said:

Nice writeup John. It was great to meet Rajiv, Vivek and yourself. Thank you for the invitation to your beautiful home and for including us in the evaluation session. Between us, we have taken several different approaches to improving SQ, and it was great to have a chance to compare. Best of all was the sense of comraderie. What a great team!

Larry

Couldn't have said it better myself, Larry. It was a pleasure meeting you guys.

Pro tip: Larry in person is a true force of nature! Not to be missed if you have the occasion and the opportunity.

auricgoldfinger · April 17, 2018

It’s time for a mea culpa…

In my 35+ years as an audio enthusiast, I have never added a component of any kind – analog, digital, power, or cable – that made my system sound worse. I have heard harshness that always disappeared within an hour or two, but the basic character of the component was always evident and never in doubt. Such was not the case the the sCLK-OCX10, which made my system almost completely intolerable for listening for several days.

After 24/7 burn-in for almost a week, I am now very relieved to say that I am finally hearing the benefits of a reference master clock described by others here and elsewhere. I have returned the silver DC cable powering the sMS-200ultra to its rightful place after earlier substituting a copper DC cable to reduce the glare introduced by the OCX10.

I will continue the 24/7 burn-in for another week and add the additional ferrites that arrived over the weekend to the dCBL-UF cables. I am also awaiting delivery of a Synergistic Research HD BNC grounding cable for the OCX10. I will be very interested to see if the grounding cable further enhances the sound. (Think ‘grounded’ Habst BNC cables, except the dCBL-UF will be connected to a Synergistic Research grounding block rather than the ground plane of the clock itself as with the Habst cables.)

I would like to offer my apologies to everyone (especially SOtM) for sounding a false alarm. I would also like to thank everyone who has offered their ideas and support here and by PM.

The OCX10 is a very worthwhile addition to my system.

austinpop · April 17, 2018

Thanks for writing this up, John. And thank you again for hosting us!

48 minutes ago, Johnseye said:

We then added Rajiv’s sCLK-EX modified Zyxel switch and the sMS200ultra.

Did you mention the effect of just the reclocked switch to your chain? That was our first experiment. Since your music lives on the NAS, it benefits from the switch even for local playback. I heard what I usually do with this change, but I'll let you describe in your words!

48 minutes ago, Johnseye said:

After switching back and forth between my system and the sMS200ultra several times we ultimately concluded that my sCLK modified system was slightly smoother while the sMS-200ultra introduced a slightly hard digital edge, or maybe that it didn’t smooth out the edge that is commonly attributed to digital music. The difference was subtle to the point that we needed to make several swaps back and forth to be sure of what we were hearing.

This is consistent with my own and Eric @limniscate's experience using the Zenith SE, @mozes's recent experiments, and several others.

One of the most important things I wanted to try in John's system was to power the trifecta with the SR-7. This was the burning question - was the trifecta in our previous experiments being held back by the PSUs we had on hand? And would the SR-7 catapult the trifecta above the custom server. The fact that it didn't suggests that beyond a point, the better PSU does not raise the performance of the trifecta.

However, I would still say the sMS-200ultra has a place. If you're handy, and willing to build a system (not technically hard, but does require attention to detail and a certain facility, especially installing OSes, drivers, etc), then an sCLK-EX modded system like @Johnseye's and @ElviaCaprice's is a very economical way to go. The SE (or other similar server) approach adds cost, but does not require assembly, while the sMS-200ultra is a lower-cost option that is comparable to the DIY server approach. Let's face it, the differences we heard between the 2 setups were quite subtle, so I wouldn't feel comfortable declaring one the winner or loser.

I will reinforce one common theme in all of these - the tX-USBultra at the end of the chain, right before the DAC, ideally disciplined by a 10MHz reference clock. I think this is the glue that binds all these approaches together.

48 minutes ago, Johnseye said:

We then sourced Roon from my Windows 10 PC. This test was to determine whether the source upstream to the sMS-200ultra had an impact. Swapping between the Windows 10 PC and my custom server with the sMS-200ultra as the endpoint revealed subtle differences with the custom PC being preferable, having a smoother less edgy sound commonly attributed to digital audio.

This is a recurring theme, I feel. The further upstream you optimize, the more muted the improvements are. Eric and I also had a hard time distinguishing between the SE and his vanilla Roon Core when comparing the SE chain to the trifecta chain.

48 minutes ago, Johnseye said:

At this point Larry @lmitche and Chris @Forehaven joined us.

Wanting to discern whether there was an audible difference between the SR7 and SR4 we used the 12v DR rail, which had been powering my custom server mobo, to power the sMS-200ultra. Using the Windows 10 PC we swapped between SR4 and SR7 as the power source to the sMS, all other equipment being equal. What we all agreed on was that the SR7 elicited more bass, and that this was actually a negative. Reason being that the additional bass drowned out other frequencies. In this case we preferred the SR4. This test would benefit from additional verification.

It was Larry who thought of the possible reason for this. The SR4 is a single rail supply. All components of the supply are dedicated to that rail, while the SR7 is dual rail sharing some components. Based on this hypothesis we concluded that dedicated power supplies for each component results in superior SQ. A big leap, but a very likely possibility that needs additional testing.

29 minutes ago, auricgoldfinger said:

Below is an excerpt from a conversation that I had with Paul Hynes several months ago. Keep in mind that I was considering a power supply with dual regulators. The sonic penalty will be more audible in a 4-rail SR7 that lacks dual regulators. For this reason (and also cost), I chose the SR5DRXL to avoid any potential sonic penalties.

Moving to the SR7DRMR2XL would provide two galvanically isolated supply rails. There is a slight sonic penalty with having all the transformer secondaries on one transformer core as nothing is absolutely perfect and there will be some small level of interaction due to transient load current swings. This may be audible in a high-resolution system but it would be rather subtle due to the very high supply line rejection of the voltage regulator circuitry in the DR mode.

This one truly surprised me. I also heard this, but I'd like to repeat this comparison with more time. Hopefully Eric's SR-7 will be here real soon now, and we'll have that opportunity. As I've mentioned, PSU differences can vary based on the component being powered, especially based on their current demands. I have also found empirically that the tX-USBultra and the DAC are the most sensitive to power quality. I want to repeat this experiment using the SR-4 and SR-7 powering the tX-USBultra and the Brooklyn DAC+ if I still have it.

I also want to control for the dual rails in this experiment. I want to use just one SR-7 rail active, and compare with when both rails are active.

Still, the good news here is that for low current applications, given the Sr-7 isn't really available anymore, the SR-4 is a really attractive option, especially if you happen to live in UK and Europe. How often does that happen!

48 minutes ago, Johnseye said:

Keeping the Cybershaft in line, I swapped between the Linear Solutions TP Link OCXO modified switch with custom LPS and the Zyxel sCLK-EX modified switch powered by the sPS-500. I swapped this back and forth several times not informing anyone which was which until after. The conclusion was that the Linear Solutions switch sounded better. The difference wasn’t as significant as the OP-14, but it was noticeable. This test could have been refined more by swapping power supplies between the two. We could have also tried using the master clock on the sCLK-EX supporting the Zyxel’s clock.

22 minutes ago, AmusedToD said:

Anyways, your collective finding that the Linear Solution OCXO switch is better than the SOtM sCLK modded switch is in line with Roy’s findings. That kind of narrows it down. The Linear Solution gets you the switch plus a linear PSU for only $649, whereas the SOtM switch (with a sCLK board) will probably be close to a grand (speculation based on their previous products pricing), and that’s without a PSU.

The LS switch will probably be the better buy, unless the upcoming Uptone switch kills both for less than $500.

One caveat I would add is that in our tests at John's, the Zyxel switch, while being driven by the sCLK-EX, was not reference clocked. So it wasn't quite an apples-to-apples comparison. With more time, we could have swapped the Cybershaft input to the sMS-200ultra that was sCLKing the switch. This illustrates the limitation of the single-output OP-14.

Still, the Linear Solutions switch seems like a winner. More importantly, it seems to be a no-brainer for those who do not have sCLK-EX components in their chain, but want to benefit from the same mechanisms.

Regarding the SOtM switch, don't dismiss it yet. Remember, it will have an option to purchase with a 25MHz SMB input, rather than an internal sCLK-EX board. I hope SOtM will price this configuration attractively. In this mode, it would be a replacement for the modded switch people are using.

This switch space is heating up, so I think we're in for an exciting few months as SOtM, Uptone and others come out with new products for us to try.

48 minutes ago, Johnseye said:

Finally we compared the Audio Alchemy DDP-1 and PM-5 SMPS with the Ayre Codex. The AA was used for all previous tests. Thanks to Rajiv we took great care to level match. This was an effort I did not do as well when I reviewed these two DACs two years ago. My previous level matching was by ear. In this effort we used an app on Rajiv’s phone. This made a significant difference and is a lesson I will always keep in mind. What everyone agreed upon was that the Codex had a richer texture while the AA sounded slightly thinner.

Yes, I credit @limniscate for convincing me of the importance of level matching. We did this the first time we compared his Yggy to my Codex. We first listened without careful level matching, and formed a strong preference for one. After level matching, all that went out the window, and we were hard pressed to tell them apart!

austinpop · April 17, 2018

5 minutes ago, auricgoldfinger said:

It’s time for a mea culpa…

In my 35+ years as an audio enthusiast, I have never added a component of any kind – analog, digital, power, or cable – that made my system sound worse. I have heard harshness that always disappeared within an hour or two, but the basic character of the component was always evident and never in doubt. Such was not the case the the sCLK-OCX10, which made my system almost completely intolerable for listening for several days.

After 24/7 burn-in for almost a week, I am now very relieved to say that I am finally hearing the benefits of a reference master clock described by others here and elsewhere. I have returned the silver DC cable powering the sMS-200ultra to its rightful place after earlier substituting a copper DC cable to reduce the glare introduced by the OCX10.

I will continue the 24/7 burn-in for another week and add the additional ferrites that arrived over the weekend to the dCBL-UF cables. I am also awaiting delivery of a Synergistic Research HD BNC grounding cable for the OCX10. I will be very interested to see if the grounding cable further enhances the sound. (Think ‘grounded’ Habst BNC cables, except the dCBL-UF will be connected to a Synergistic Research grounding block rather than the ground plane of the clock itself as with the Habst cables.)

I would like to offer my apologies to everyone (especially SOtM) for sounding a false alarm. I would also like to thank everyone who has offered their ideas and support here and by PM.

The OCX10 is a very worthwhile addition to my system.

Kudos to you. Thanks for your honesty.

FYI, @limniscate and I hope to make it out soon to @auricgoldfinger's in Houston, to pit the Ref-10 and OCX10 head to head, and check out Brian's toys!

elan120 · April 17, 2018

9 minutes ago, auricgoldfinger said:

After 24/7 burn-in for almost a week, I am now very relieved to say that I am finally hearing the benefits of a reference master clock described by others here and elsewhere. I have returned the silver DC cable powering the sMS-200ultra to its rightful place after earlier substituting a copper DC cable to reduce the glare introduced by the OCX10.

I will continue the 24/7 burn-in for another week and add the additional ferrites that arrived over the weekend to the dCBL-UF cables. I am also awaiting delivery of a Synergistic Research HD BNC grounding cable for the OCX10. I will be very interested to see if the grounding cable further enhances the sound. (Think ‘grounded’ Habst BNC cables, except the dCBL-UF will be connected to a Synergistic Research grounding block rather than the ground plane of the clock itself as with the Habst cables.)

I would like to offer my apologies to everyone (especially SOtM) for sounding a false alarm. I would also like to thank everyone who has offered their ideas and support here and by PM.

The OCX10 is a very worthwhile addition to my system.

This is similar to my experience as well that sCLK-OCX10 takes a bit of burn-in to really excel, in my case, past 250 hours.

3 minutes ago, austinpop said:

Kudos to you. Thanks for your honesty.

FYI, @limniscate and I hope to make it out soon to @auricgoldfinger's in Houston, to pit the Ref-10 and OCX10 head to head, and check out Brian's toys!

Very much looking forward to this comparison.

Johnseye · April 17, 2018

25 minutes ago, austinpop said:

Did you mention the effect of just the reclocked switch to your chain? That was our first experiment. Since your music lives on the NAS, it benefits from the switch even for local playback. I heard what I usually do with this change, but I'll let you describe in your words!

I briefly mentioned it in the TL;DR. It does have a positive impact. I swapped it with my battery powered GS-108 a few times and the difference is clearly audible. Not as impactful as the master clock, but still an improvement. I'll be waiting on the Uptone and SOtM switch reviews before I make a decision.

25 minutes ago, austinpop said:

One caveat I would add is that in our tests at John's, the Zyxel switch, while being driven by the sCLK-EX, was not reference clocked. So it wasn't quite an apples-to-apples comparison. With more time, we could have swapped the Cybershaft input to the sMS-200ultra that was sCLKing the switch. This illustrates the limitation of the single-output OP-14.

Because the Linear Solutions switch was not master clocked I don't this would have been a fair comparison. A good one, because of the option to do so with the sCLK, but not apples to apples. Adrian should add a master clock option.

http://thelinearsolution.com/ocxo_switch.html

25 minutes ago, austinpop said:

Still, the Linear Solutions switch seems like a winner. More importantly, it seems to be a no-brainer for those who do not have sCLK-EX components in their chain, but want to benefit from the same mechanisms.

Regarding the SOtM switch, don't dismiss it yet. Remember, it will have an option to purchase with a 25MHz SMB input, rather than an internal sCLK-EX board. I hope SOtM will price this configuration attractively. In this mode, it would be a replacement for the modded switch people are using.

The only draw back to master clocking is that the components need to be in close proximity. The switch is one component, at least in my environment, where I'd prefer it not be with the other equipment as it allows me to eliminate an upstream switch, potentially benefiting Tidal streams.

25 minutes ago, austinpop said:

Yes, I credit @limniscate for convincing me of the importance of level matching. We did this the first time we compared his Yggy to my Codex. We first listened without careful level matching, and formed a strong preference for one. After level matching, all that went out the window, and we were hard pressed to tell them apart!

I recently ordered a Holo Spring Level 3 DAC. I'm looking forward to the many possibilities this brings. It will be a new chapter as I'll also have a new sCLK motherboard and the ability to upsample to DSD512. I'll be sure to compare with level matching this time around.

austinpop · April 17, 2018

5 minutes ago, Johnseye said:

I briefly mentioned it in the TL;DR. It does have a positive impact. I swapped it with my battery powered GS-108 a few times and the difference is clearly audible. Not as impactful as the master clock, but still an improvement. I'll be waiting on the Uptone and SOtM switch reviews before I make a decision.

Because the Linear Solutions switch was not master clocked I don't this would have been a fair comparison. A good one, because of the option to do so with the sCLK, but not apples to apples. Adrian should add a master clock option.

http://thelinearsolution.com/ocxo_switch.html

It helps to think of this comparison as follows: how do I get OCXO-level clocking delivered to the switch? In the Linear Solutions design, it is to engineer an OCXO chip directly in the switch. In the sCLK-EX approach, it is to take an OCXO reference clock, connect it to the sCLK-EX master clock input, and distribute to the switch. Without the reference clock in the modded switch, we're not getting the OCXO-level quality. I realize this is a non-technical description. What does it really mean to use an OCXO reference clock to discipline an sCLK-EX clock generator? I guess in layman's terms, and based on what we are hearing, the "quality" of the OCXO reference clock seems to propagate to the outputs of the sCLK-EX.

A fixed OCXO design like Adrian's does not need, nor can it even accept, a master clock input.

5 minutes ago, Johnseye said:

The only draw back to master clocking is that the components need to be in close proximity. The switch is one component, at least in my environment, where I'd prefer it not be with the other equipment as it allows me to eliminate an upstream switch, potentially benefiting Tidal streams.

Agreed. This is an important point.

5 minutes ago, Johnseye said:

I recently ordered a Holo Spring Level 3 DAC. I'm looking forward to the many possibilities this brings. It will be a new chapter as I'll also have a new sCLK motherboard and the ability to upsample to DSD512. I'll be sure to compare with level matching this time around.

Looking forward to this comparison.

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

Recommended Posts

Link to comment

Link to comment

Link to comment

Link to comment

Link to comment

Link to comment

Link to comment

Link to comment

Link to comment

Link to comment

Link to comment

Link to comment

Link to comment

Link to comment

Link to comment

Link to comment

Link to comment

Link to comment

Link to comment

Link to comment

Link to comment

Link to comment

Link to comment

Link to comment

Link to comment

Create an account or sign in to comment

Create an account

Sign in