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

rickca · May 16, 2017

33 minutes ago, Johnseye said:

I am in discussion now to build a custom carrier board to my specifications.

Love the all-out assault! Do you have a cost estimate?

Can you please also explain why you want multiple x1 slots for Adnaco cards.

Question is for @romaz please.

romaz · May 16, 2017

On 5/14/2017 at 0:03 PM, Elberoth said:

What extra flexibility do you have in mind when talking about the 10MHz clock option ?

To me, the external clock means more cables, an extra box, and substantially higher cost.

I also have problems seeing any benefits an external clock could provide vs an internal clock, unless the later is of poor quality (It is important to not that the real world phase noise numbers will be degraded by extra cabling, connectors and additional PPL used to get 12MHz clock needed by the SmS-200. So even if you use the best in the world 10MHz clock with ultra low phase noise, the real world #s will be much, much worse. It is even possible that $25 Crystek clock Alex is using in the ISO Regen will have less phase noise than the 10MHz clock + cabling + PLL combination).

I'm not disagreeing with you. I have no intentions of getting an external clock.

SOtM decided to provide this "flexibility" for those who choose to use their own clock. Obviously, whether someone's external clock is better than SOtM's stock internal clock and whether this is worth the hassle of another PSU and the extra cables is a separate matter. A reputable CA member has communicated with me and has told me SOtM's new clock made his Paul Pang OCXO "like child's toys" which says a lot. Having said that, SOtM is designing their own 10MHz external master clock for this clock board and one would assume it will outperform their internal clock. If it were me, I would just design an upgraded sCLK-EX once the better clock becomes available.

romaz · May 16, 2017

On 5/14/2017 at 4:28 PM, austinpop said:

BTW - if anyone is going to the Munich High End show, please talk to the SOtM folks in person and get as much info as possible.

I will be heading to Munich on Wednesday. Absolutely, I will have many questions to ask.

romaz · May 16, 2017

15 hours ago, austinpop said:

The reason SOtM offer the 12v option is not because it is sonically superior, but because it allows you to match to an existing PSU you may already have.

Rajiv, on the SOtM thread, May has indicated that 12V sounds better. This is my experience as well with the sMS-200. I think if you have the option to go 12V, you should do it. The difference is noticeable.

ElviaCaprice · May 16, 2017

12 minutes ago, romaz said:

Looks like a great way to go! Make sure you test with LPS-1s. On boot up, I suspect it will draw more than 1.1A.

From the Prime 95 tests, if one was to power the hard drive separately it will easily run off less than 1.1A and may even do so powering an HDD. Yes, I plan to test. Another good finding is that it has a SATA II connector. I will add an external PCIe 1X connector off the half mini PCIe.

romaz · May 16, 2017

13 hours ago, guerph said:

http://www.6moons.com/audioreviews2/sotm4/1.html

Not a review, but Srajan seems to have been having a bad day when he wrote this up. Echoes many of my frustrations...

" Where the Ultra goes ultra is price. Competing devices with a single output sell for half or less. How many users will exploit the tx-USBultra's ability to distribute source signal to two devices is questionable. While on the q word, ditto for packaging devices whose purpose is to minimize noise with switch-mode wall warts which inject their own HF noise into the power line, hence partnering gear. Sure, it might motivate diehards to shell out even more for a linear or battery supply. Bravo, you get to sell them an extra box*. 'cept that the battery versions invariably arrive with their own SMPS for charging. Hey, sell them one more box. Why, argh, can't this type of kit arrive with a linear supply built in? How about this: "If you use two USB output ports at the same time, the sound might degrade so we recommend using one port." Why then have two in the first place? To run a reclocked printer? There's more: "If you don't use the product, turn off power by unplugging the power supply." Really? For $990, that's how one turns things off? No comment."

I think Srajan was having a bad day. Incorporating a PSU into the tX-USBultra would be the last thing I would want because it would then cost me more money to have to remove and replace that PSU. It is the integrated PSU that brings down such components as the Mutec, RedNET, the Berkeley Alpha USB and SU-1.

romaz · May 16, 2017

15 hours ago, limniscate said:

Are you sure that the sMS-200 Ultra can be powered by an Uptone LPS-1? The website says that the max current input is 2A.

I believe it can be. It states 2A because the sMS-200Ultra incorporates a USB hub with 2 free USB ports for USB storage drives. The USB 2.0 spec calls for 500mA of power per port but if you don't use these ports, you probably will never draw close to 2A.

limniscate · May 16, 2017

Does anyone know anything about the new W4S LPS?

https://www.wyred4sound.com/products/upgrades-mods-accessories/ps-1-modular-linear-power-supply

rickca · May 16, 2017

Maybe someone in Munich can convince SOtM to source switches themselves for the clock transplant. Beats shipping a switch to them.

SOtM has the potential to run the table (in the billiards sense) on these 'middleware' devices like streamers, converters and reclockers.

romaz · May 16, 2017

11 hours ago, mozes said:

I also would like to know how Vinnie Rossi Mini compares to Paul Hynes SR5 or SR7

Based on output impedance alone, I don't believe so (based on my rough calculations and I could be wrong). Vinnie Rossi's LIO's ultracap PSU has a stated output impedance of 16 milliohms which is superior to many other PSUs but remember, according to Paul Hynes, an exceptional output impedance should be <10 millohms over a broad frequency range and his SR7 has a measured output impedance of <3 milliohms from DC to 100kHz. Anyway, based on photos I've seen, Vinnie's LIO utilizes 9 ultracaps per bank which would be equivalent to about 1.78 milliohms per ultracap. I'm not sure if Vinnie is using the same ultracap in his Mini DC-4EVR PSU and I'm not sure how many ultracaps he's using per bank but almost certainly, it's going to be well over 3 milliohms.

The claim to fame of battery PSUs is that they have no leakage current that can lead to ground loops. Paul claims that his floating rail design similarly results in no electrical potential differential (ground loops).

In the end, the proof is in the listening. I have heard Vinnie's Mini several times at shows but I have never directly compared the 2.

limniscate · May 16, 2017

2 minutes ago, rickca said:

Maybe someone in Munich can convince SOtM to source switches themselves for the clock transplant. Beats shipping a switch to them.

SOtM has the potential to run the table (in the billiards sense) on these 'middleware' devices like streamers, converters and reclockers.

They've agreed to source a switch for me.

romaz · May 16, 2017

1 hour ago, feelingears said:

Maybe you mean as many good clocks all along the way upstream from the DAC? FWIW, Audiostream.com just ran a review on SotM's USB decrapifier by Steven Plaskin suggesting that the best clock closest to the DAC with the best power supply possible yields the best sonic results. (This has been my experience too, although only based on my very limited testing in my systems.)

I'm not suggesting you're wrong @austinpop, I'm pointing out Plaskin's point of view as we all explore these relationships further. We're all waiting for the ISO Regen vs SotM USBultra cage match, if only due to the price difference.

I believe you have misinterpreted Rajiv's comment. We have been saying exactly what Steven Plaskin has just said on this thread for quite a while now, that the best clock should be reserved for the component closest to the DAC but ideally, all bad clocks should be replaced as far upstream as possible (possibly all the way to the router/modem but definitely throughout the server) and so this is what Rajiv was alluding to. Rajiv now has 4 Ultra clocks in his chain between server and DAC and so he is operating at an experience level well beyond Steven's when it comes to clocking.

romaz · May 16, 2017

1 hour ago, limniscate said:

@romaz

Thanks for the reply.

Do you know where I could try a SOtM server? Also, I saw an old post where you said the iSO-CAT6 had very little effect on sound quality; however that post was when you had the dCBL-CAT6. Are you saying that the dCBL-CAT7 has some sort of greater synergy with the iSO-CAT6?

Have you had the dCBL-CAT& and iSO-CAT6 combo between the server and switch and the switch and end point simultaneously, i.e., four cables and two iSO-CAT6? That's what SOtM is recommending to me, but I'm wondering if it's just a money grab, lol.

I'm not sure where you can try one but I would negotiate a return policy with SOtM before you buy one.

Yes, the iSO-CAT6 alone didn't make much of a difference. It is the cables that seem to make a much greater difference. You could go dCBL-CAT7 > reclocking switch > dCBL-CAT7.

I have not tried 4 dCBL-CAT7s as I never had access to that many. I don't know how it would sound. I'm sure SOtM wants to sell you as many cables as possible but I think they're also pretty fair at accepting back something that doesn't work for you.

The fact that these cables result in such a dramatic improvement in SQ despite having 3 Ultra devices in my path is what frustrates me. Despite this triple level of isolation, there still seems to be a lot of noise in the line! Very frustrating.

romaz · May 16, 2017

1 hour ago, Johnseye said:

A new business model? Let me know when you start taking orders

No, I don't intend to make money from this venture. I will happily share my findings when the time comes and anyone is free to benefit.

ElviaCaprice · May 16, 2017

How did the initial Adnaco tests go with the tXUSB Ultra?

romaz · May 16, 2017

1 hour ago, rickca said:

Love the all-out assault! Do you have a cost estimate?

Can you please also explain why you want multiple x1 slots for Adnaco cards.

Question is for @romaz please.

So, my Adnaco S3B has come in and it is indeed very good and the combination of Adnaco + tX-USBUltra is just heavenly. I see great potential with the Adnaco with its clocks replaced.

Where SOtM's sound signature is this wonderful detail clarity, it can also sound a bit thin. If there has been a criticism with their latest Ultra server, this is it and perhaps, this signature is even intentional. This is why I value AO with its sound signatures and digital filters because it allows me to put a bit of meat on the bones of the SOtM. I generally value silver over copper DC cabling but I have intentionally gone with copper with my SOtM gear.

The Adnaco has this wonderful quality of gravitas. Sound has more weight. Compared against my server's stock USB port, SQ is significantly improved. It lacks the depth and air and sparkle of the tX-USBUltra but the tX-USBUltra, just like its siblings can sound a touch thin. The Adnaco + tX = pure bliss because the balance now is perfection but keep in mind, this is only my personal preference.

Could the Adnaco with its 3 clocks replaced provide the equivalent SQ of Adnaco + tX? I'm not sure but that would be wonderful. Is it better than switch + sMS-200Ultra + dX-USB HD Ultra or switch + sMS-200Ultra + tX-USBUltra? I don't know yet, I'll find out tomorrow but thus far, I can't stop listening.

Here's something else that I was not expecting. I tried using the tX-USBUltra as a hub for my music drive while I had just the Adnaco connected to my DAC. I connected a compact flash card filled with music via a USB adapter to the tX and then compared SQ against this same compact flash connected to one of my server's USB port. There was some improvement (larger soundstage) when the CF was connected to the tX but the improvement was small and I believe the tX is limited here by the poor quality of my server's USB port. Of course, this could improve once I replace my motherboard clock but that remains to be seen. I then tried plugging this compact flash setup into the Adnaco and WOW! A very noticeable improvement in terms of detail clarity and separation. I am now targeting a 2nd Adnaco for my music drive.

Regarding cost estimate of the motherboard, that is still being discussed.

romaz · May 16, 2017

1 hour ago, limniscate said:

Does anyone know anything about the new W4S LPS?

https://www.wyred4sound.com/products/upgrades-mods-accessories/ps-1-modular-linear-power-supply

I saw this at RMAF last year and I like its flexibility, very similar to the flexibility of the HDPlex. I can't comment on comparative SQ, however. SOtM's upcoming sPS-500 is another interesting option.

ElviaCaprice · May 16, 2017

I am starting to think, listening to others findings, that the clocking isn't as important as power source and that once the music stream has been contaminated it becomes difficult to correct and that is where the extra clocking/isolation comes into play. If we could just keep the stream clean from the get go, server, then maybe we don't need all these expensive solutions/clocking further downstream. Keep it short, with one good clock before the DAC. As long as the power is clean throughout the path.

Or maybe I'm just wishing it was that simple, if even possible.

austinpop · May 16, 2017

1 hour ago, romaz said:

I will be heading to Munich on Wednesday. Absolutely, I will have many questions to ask.

I'm jealous. I still remember some memorable steins of helles beer, and taking delivery of a car at BMW Welt.

And then there's the High-End show. If you have the bandwidth to give us any feedback, it would be greatly appreciated! I'm sure these questions are in your list too, but from my end:

Details on their SMB clock connectors
details of SOtM switch offering with sCLK-EX
what is their envisioned architecture for sCLK-EX distribution? Is it a separate master clock sCLK-EX box, which can then drive up to 4 slaves using SMB connectors? Or will they continue to co-locate the sCLK-EX in the dX and tX offerings?

1 hour ago, romaz said:

Rajiv, on the SOtM thread, May has indicated that 12V sounds better. This is my experience as well with the sMS-200. I think if you have the option to go 12V, you should do it. The difference is noticeable.

Thanks - I must have missed that. In my email communications with her, she didn't mention that, but that is good to know.

12 minutes ago, romaz said:

Where SOtM's sound signature is this wonderful detail clarity, it can also sound a bit thin. If there has been a criticism with their latest Ultra server, this is it and perhaps, this signature is even intentional. This is why I value AO with its sound signatures and digital filters because it allows me to put a bit of meat on the bones of the SOtM. I generally value silver over copper DC cabling but I have intentionally gone with copper with my SOtM gear.

Funny you mention this, Roy. This slight thinness is something I'm noticing too. I am already using AO to tune it. I can get some more warmth by going from 3C to 4D, although I feel that then loses a slight amount of detail.

While I just cannot justify the dCBL-CAT7 cost, I will certainly try it when @limniscate's goodies arrive.

In the meantime I ordered a Supra Cat8 per your comments to see if I found the same warmth difference with it over the BJC 6a. Will report back.

12 minutes ago, romaz said:

Could the Adnaco with its 3 clocks replaced provide the equivalent SQ of Adnaco + tX? I'm not sure but that would be wonderful. Is it better than switch + sMS-200Ultra + dX-USB HD Ultra or switch + sMS-200Ultra + tX-USBUltra? I don't know yet, I'll find out tomorrow but thus far, I can't stop listening.

Fascinating, and much awaited!

romaz · May 16, 2017

29 minutes ago, ElviaCaprice said:

I am starting to think, listening to others findings, that the clocking isn't as important as power source and that once the music stream has been contaminated it becomes difficult to correct and that is where the extra clocking/isolation comes into play. If we could just keep the stream clean from the get go, server, then maybe we don't need all these expensive solutions/clocking further downstream. Keep it short, with one good clock before the DAC. As long as the power is clean throughout the path.

Or maybe I'm just wishing it was that simple, if even possible.

I compared power sources again with the tX-USBUltra when it arrived because my friend who loaned it to me wasn't as impressed as he thought he would be. As I previously posted, he is using SOtM's mBPS-d2s to power it. With my HDPlex, my experience was a letdown as well. Better than with DAC directly connected to server's USB port but not worth almost $1k. From there, I went to my LPS-1 at 7V and it powers it without problems. SQ is MUCHO improved. With the SR7, just another level altogether.

When factoring the cost of these Ultras, you really need to account for the cost of good power, otherwise, I don't think they're worth it.

romaz · May 16, 2017

33 minutes ago, austinpop said:

While I just cannot justify the dCBL-CAT7 cost, I will certainly try it when @limniscate's goodies arrive.

It's hard to justify the price although the difference is remarkable and shockingly so. However, for the cost of a dCBL-CAT7, I was able to buy the Adnaco + change. For the cost of another dCBL-CAT7 + iSO-CAT6, I was able to buy another sCLK-EX for the Adnaco. I think you can see where I'm going with this...

37 minutes ago, austinpop said:

In the meantime I ordered a Supra Cat8 per your comments to see if I found the same warmth difference with it over the BJC 6a. Will report back.

The Supra CAT8 is wonderful and provides a touch of this emotional quality that the dCBL-CAT7 doesn't provide. If you are noticing listener's fatigue with your system, this cable will help remedy that.

AmusedToD · May 16, 2017

3 hours ago, romaz said:

I will be heading to Munich on Wednesday. Absolutely, I will have many questions to ask.

Will be in Munich as well, leaving tonight with a couple of friends!

seeteeyou · May 16, 2017

2 hours ago, romaz said:

I agree. I believe COM Express holds the key and I am in discussion now to build a custom carrier board to my specifications.

Terrific, it's gotta be exciting time.

Let's see if that were gonna pave the way for a brave soul to explore the possibilities of designing something that's similar to SHB Express in the future

https://www.picmg.org/openstandards/shb-express/

http://www.dfitech.com/boards/backplanes-picmg-1-3/

http://www.dfitech.com/boards/backplanes-picmg-1-3/picmg-1-3-backplane/

Even Ethernet didn't seem to be "bulletproof" since we still have to deal with the noise unless we're going for fiber instead of copper, most likely that's why iso-CAT6 would actually make a difference.

It would be great if someone were able to find a way to build a backplane of some sort and then somehow 2 separate Single Board Computers could somehow communicate through PCI Express directly.

And then we could also mitigate the impact of noise with Adnaco-S5

http://www.adnaco.com/products/s5/

It's yet another way to establish a "Direct Connection" between 2 machines in a somewhat indirect way.

What if that weren't doable? Maybe we could "combine" Adnaco-UF1 with one of those USB 3.0 transfer cables if they weren't limited to transferring actual files from one drive to another

http://www.adnaco.com/doc/Adnaco-SU1-DS.pdf

https://www.adnaco.biz/blogs/adnaco-for-medical/13654565-adnaco-su1-usb-3-0-fiber-optic-extension-system

http://plugable.com/products/usb3-tran/

https://www.startech.com/Networking-IO/USB-PS2/usb-file-transfer-cable-windows-mac~USB3LINK

http://www.bafo.com/index.php/products/adapter/usb-cable-adapter/bf-7330-usb-3-0-easy-transfer-cable.html

http://www.laplink.com/index.php/individuals/disk-utilities/538-pcmover/laplink-easy-transfer-cable-modules/1476-laplink-easy-transfer-cable

R1200CL · May 16, 2017

On 4.5.2017 at 5:58 PM, seeteeyou said:

I dunno where to put this but many of us should be interested in replacing DC cables with even better ones

FYI - recently there's yet another audiophile DC plug available in Taiwan

They should be able to ship internationally and here are their contacts

http://www.atl-newcablestudio.com/Contact.html

DC plugs from Oyaide are only good for bulk wires with 6.3mm diameter or below, however ATL ones are good for anything up to 8mm. Basically that's good for CrystalPower Special and the genuine bulk wires are available in Hong Kong as well as China

https://item.taobao.com/item.htm?id=23058296058

http://www.audiofile.nl/sites/default/files/2014-07-01 R6 Crystal Cable Retail Pricelist.pdf#page=3

Of course we could also find a reputable seller on Audiogon etc. and then get one of those genuine Crystal Cables Power cables for cheap, then we could simply replace the IEC plugs on both ends with DC plugs accordingly.