Reality is Cruel : Cybershaft new Ultimate OCXO 10M Clocks Shootout OP20 vs OP17

renan33 said:

Hello since the brand new cybershaft (21A) is a giant killer most of clock user will sell their previous clock to gzet the better one.

So what about the senescence of the clock after 10 years. Is it ok to buy a ten year word clock?
Do some calibration is needed. I saw some ten year old esoteric clock to sell. Is there any pb to buy it??

I have gpsDO on a DCS network bridge and I'm very surprise of the improvement. I thought that a 4000euros gear (DCS NB) could better handle its clock and no a 100 euros clock put some air in the audio...

Click to expand...

It is advised to have the clock tested and recertified after 'X' amount of hours/days of continuous operation. See product documentation for details. If the clock you are using doesn't call for it, it might be a good idea to inquire with the manufacturer as to their guidelines. True these devices have large long-term stability numbers however with such a precise instrument, I can see it reasonable to have it re-tested ever few years as a preventative measure given the multiple other components other than the OCXO or other timing references inside master and word clocks in general. Note this doesn't imply recalibration is always needed; that would depend upon the test result. In the case of my Cybershaft OP21A it's been a few years and the musical playback is rock solid; I do plan to send it in for a pit stop just in case in another year or perhaps later in 2021 if I'm ever on the road for multiple weeks for work (or MUNICH HIGH END 2019 and business on either side!).

Thanks for your answer

Given the imperfect nature of cables, conductors, terminations, and interfaces between outboard master clock crystal oscillators and external components which are supposed to be synchronized to it, the empirical results shown in the article below makes intuitive sense... not to mention that 10.0000 MHz does not divide equally into neither 44.1kHz nor 48.0kHz, requiring the need for an intermediary PLL oscillator to synthesize an approximation. Hence, outboard clocks typically help INCREASE (not reduce) jitter.

Does Your Studio Need A Digital Master Clock? (soundonsound.com)
"Overall, it should be clear from these tests that employing an external master clock cannot and will not improve the sound quality of a digital audio system. It might change it, and subjectively that change might be preferred, but it won't change things for the better in any technical sense. A?D conversion performance will not improve: the best that can be hoped for is that the A?D conversion won't become significantly degraded. In most cases, the technical performance will actually become worse, albeit only marginally so."

Can anyone provide links to tests which demonstrate that outboard clocks actually help to "reduce" jitter?

QuadDiffusor said:

Given the imperfect nature of cables, conductors, terminations, and interfaces between outboard master clock crystal oscillators and external components which are supposed to be synchronized to it, the empirical results shown in the article below makes intuitive sense... not to mention that 10.0000 MHz does not divide equally into neither 44.1kHz nor 48.0kHz, requiring the need for an intermediary PLL oscillator to synthesize an approximation. Hence, outboard clocks typically help INCREASE (not reduce) jitter.

Does Your Studio Need A Digital Master Clock? (soundonsound.com)
"Overall, it should be clear from these tests that employing an external master clock cannot and will not improve the sound quality of a digital audio system. It might change it, and subjectively that change might be preferred, but it won't change things for the better in any technical sense. A?D conversion performance will not improve: the best that can be hoped for is that the A?D conversion won't become significantly degraded. In most cases, the technical performance will actually become worse, albeit only marginally so."

Can anyone provide links to tests which demonstrate that outboard clocks actually help to "reduce" jitter?

Click to expand...

Have you tried a decent external clock in your own system?

The article you coded was posted in 2010, i.e. more than 10 years ago. At that time external reference clocks just started to be introduced. Things continued to develop and improve quite a lot in the past 10 years. I changed and upgraded my reference clocks four, five times in that time and noticed improvement all along.

We had done many A/B tests in our homes and the result was very obvious. The music sounded better, much better with good clocks. This point was echoed again and again in this forum and others. What other evidences do you need?

TLi said:

Have you tried a decent external clock in your own system?

The article you coded was posted in 2010, i.e. more than 10 years ago. At that time external reference clocks just started to be introduced. Things continued to develop and improve quite a lot in the past 10 years. I changed and upgraded my reference clocks four, five times in that time and noticed improvement all along.

We had done many A/B tests in our homes and the result was very obvious. The music sounded better, much better with good clocks. This point was echoed again and again in this forum and others. What other evidences do you need?

Click to expand...

Hi TLi,

Yes, I've been aware of the evolution of audio hardware addressing the need for a solution to jitter - from Esoteric's G-0 (from circa the mid-1990s) which I've seen and heard in Hong Kong and Japan, and the more recent decade's slew of clocking devices, both onboard and outboard - VCOs, OCXOs, TCXOs, 10MHz sine waves, multiples of base clock frequencies of 44.1kHz and 48.0kHz, etc. Despite the passage of time, the reality is that things are still very much as I've described it - there is an unavoidable corruption of the integrity of the reference clock signal between the oscillator and the components (typically located quite a distance away, with incompatible or at best imperfect electrical and signal interfaces) which supposedly benefit from the synchronization. That's exactly why I posted what I wrote in post #343 - my questions are not rhetorical, as there has been a gaping absence of quantitative data showing the benefits of the outboard clock connection. Actually, it's quite understandable because commercial companies have a huge incentive to sell / supply products to meet a demand and generate the requisite revenue and profits.

I presume that since this Forum allows a variety of opinions to be expressed for the mutual benefit of everyone (including lurkers) I was attempting to ask a value-added question, as I'm genuinely curious.

One of the designs / products on the marketplace which does a good job at mitigating the many inherent pitfalls in remedying the fundamental problems of jitter, is to place an ultra-low phase-error reference Master clock just millimeters from the DAC, and utilize that Master clock as the reference from which to Slave all other outboard devices - this is done as well as practically possible, by MSB Technologies. In the case of their Pro-ISL and Pro-USB module is to use the same Master clock to reclock the cadence of the asynchronous data present on the USB cable. Ditto for MSB Technology's BNC, RCA and AES/EBU input modules which conform to the S/PDIF standard whose synchronous data stream marches to the embedded clock data, for which the Master clock also provides a clock sync for, via a BNC word clock output. This design topology which MSB Technology has pursued is NOT the only way, but IMO a near-perfect one. My MSB Select II DAC has the Femto 33 clock module as its Master clock, and slaves a multiple of external data sources; there are three input modules in my unit, each having a dedicated "clock out" connector: BNC-in (synchronous S/PDIF), AES/EBU-in (synchronous S/PDIF), and Pro-ISL-in (asynchronous USB).

My three SOtM components (nH-10G x 2, and USBultra - all fully upgraded, with their own independent onboard clocks) had the potential to benefit from an ultra-precise 10MHz sine-wave outboard clock for synchronization, so about a year ago, I borrowed the OCX10 (together with three of SOtM's expensive clock link BNC cables) for a home audition. The DC power supplies included the SOtM PS-500 for the first nH-10G, but otherwise for the remaining three units, I used the PLiXiR Elite BDC x 2 (12V) and Farad Power3 which are no slouches in delivering stable and noise-free DC current. For over two weeks, I listened very carefully, and gave the OCX10 every opportunity to shine. However, I heard an undesirable uptick in blurring and veiling - my system sounded clearly superior WITHOUT the outboard clock and its associated clock link cables.

I wrote about this on WBF:
https://www.whatsbestforum.com/thre...of-usb-network-gadget-setup.27758/post-614620

The bottom line is that in my limited experience, the SOtM OCX10-linked system, the addition of the outboard clock made the sound worse. Hence, I'm curious to see some objective data demonstrating the effectiveness of these clocks, even highly-evolved more-modern versions. I suspect that it's the case that the "difference" or "improvement" in sound may in fact be an increase in jitter, given the inherent limitations in the vast permutations of variables present in the spaghetti of cables, connectors, boards, power supplies, impedance mis-matches, etc.

I have always read and referred to this forum.
It was very informative.
But I had to read and speak out of your rating now. #345
I am an electronics designer.
It has also been proved by the emulator that the PLL circuit refers to the external clock and improves some phase noise.
Your opinion totally ignores those technical grounds.
What are the criteria you rate as "good sound"?
I think you simply have a low ability to evaluate your ears for sound.

QuadDiffusor said:

One of the designs / products on the marketplace which does a good job at mitigating the many inherent pitfalls in remedying the fundamental problems of jitter, is to place an ultra-low phase-error reference Master clock just millimeters from the DAC, and utilize that Master clock as the reference from which to Slave all other outboard devices - this is done as well as practically possible, by MSB Technologies.

Click to expand...

Thats one possible way, but your inference that it's better does not bear out in actual measurement results; MSB does not measure better than manufacturers whose systems perform best with an external master clocks, like DCS for example. Go look at Stereophile's measurements.

Your argument just fell apart with your example.

QuadDiffusor said:

Hi TLi,

Yes, I've been aware of the evolution of audio hardware addressing the need for a solution to jitter - from Esoteric's G-0 (from circa the mid-1990s) which I've seen and heard in Hong Kong and Japan, and the more recent decade's slew of clocking devices, both onboard and outboard - VCOs, OCXOs, TCXOs, 10MHz sine waves, multiples of base clock frequencies of 44.1kHz and 48.0kHz, etc. Despite the passage of time, the reality is that things are still very much as I've described it - there is an unavoidable corruption of the integrity of the reference clock signal between the oscillator and the components (typically located quite a distance away, with incompatible or at best imperfect electrical and signal interfaces) which supposedly benefit from the synchronization. That's exactly why I posted what I wrote in post #343 - my questions are not rhetorical, as there has been a gaping absence of quantitative data showing the benefits of the outboard clock connection. Actually, it's quite understandable because commercial companies have a huge incentive to sell / supply products to meet a demand and generate the requisite revenue and profits.

I presume that since this Forum allows a variety of opinions to be expressed for the mutual benefit of everyone (including lurkers) I was attempting to ask a value-added question, as I'm genuinely curious.

One of the designs / products on the marketplace which does a good job at mitigating the many inherent pitfalls in remedying the fundamental problems of jitter, is to place an ultra-low phase-error reference Master clock just millimeters from the DAC, and utilize that Master clock as the reference from which to Slave all other outboard devices - this is done as well as practically possible, by MSB Technologies. In the case of their Pro-ISL and Pro-USB module is to use the same Master clock to reclock the cadence of the asynchronous data present on the USB cable. Ditto for MSB Technology's BNC, RCA and AES/EBU input modules which conform to the S/PDIF standard whose synchronous data stream marches to the embedded clock data, for which the Master clock also provides a clock sync for, via a BNC word clock output. This design topology which MSB Technology has pursued is NOT the only way, but IMO a near-perfect one. My MSB Select II DAC has the Femto 33 clock module as its Master clock, and slaves a multiple of external data sources; there are three input modules in my unit, each having a dedicated "clock out" connector: BNC-in (synchronous S/PDIF), AES/EBU-in (synchronous S/PDIF), and Pro-ISL-in (asynchronous USB).

My three SOtM components (nH-10G x 2, and USBultra - all fully upgraded, with their own onboard clocks) had the potential to benefit from an ultra-precise 10MHz sine-wave outboard clock for synchronization, so about a year ago, I borrowed the OCX10 (together with three of SOtM's expensive clock link BNC cables) for a home audition. The DC power supplies included the SOtM PS-500 for the first nH-10G, but otherwise for the remaining three units, I used the PLiXiR Elite BDC x 2 (12V) and Farad Power3 which are no slouches in delivering stable and noise-free DC current. For over two weeks, I listened very carefully, and gave the OCX10 every opportunity to shine. However, I heard an undesirable uptick in blurring and veiling - my system sounded clearly superior WITHOUT the outboard clock and its associated clock link cables.

I wrote about this on WBF:
https://www.whatsbestforum.com/thre...of-usb-network-gadget-setup.27758/post-614620

The bottom line is that in my limited experience, the SOtM OCX10-linked system, the addition of the outboard clock made the sound worse. Hence, I'm curious to see some objective data demonstrating the effectiveness of these clocks, even highly-evolved more-modern versions. I suspect that it's the case that the "difference" or "improvement" in sound may in fact be an increase in jitter, given the inherent limitations in the vast permutations of variables present in the spaghetti of cables, connectors, boards, power supplies, impedance mis-matches, etc.

Click to expand...

Your opinion is much respected and thank you for supplying alternative argument for discussion.

I agree some older clocks actually sounded worse, usually they gave a edgy presentation which initially appeared to sound better but after a while it felt unnatural.

I lent my Cybershaft OP20 to my local dCS dealer. They also sold some reference external clocks such as Antelope before. The manager there was first sceptical, he was not very impressed with the previous clocks on dCS Vivaldi system. He thought they sounded thin and bright. After a week, he called saying he was very shocked with what OP20 can do. There were improvement everyday up to the fifth day. His opinion on external clock has completely changed from then. I told him I still have OP21 in my place, want to try?

I have SOtM network switch and I find it benefited a lot with good external clock as well.

TLi said:

Your opinion is much respected and thank you for supplying alternative argument for discussion.

I agree some older clocks actually sounded worse, usually they gave a edgy presentation which initially sounded better but after a while it felt unnatural.

I borrowed my Cybershaft OP20 to my local dCS dealer. They also sold some reference external clocks such as Antelope before. The manager there was first sceptical, he was not very impressed with the previous clocks on dCS Vivaldi system. He thought they sounded thin and bright. After a week, he called saying he was very shocked with what OP20 can do. There were improvement everyday up to the fifth day. His opinion on external clock has completely changed from then. I told him I still have OP21 in my place, want to try?

I have SOtM network switch and I find it benefited a lot with good external clock as well.

Click to expand...

Likewise, my experience with Cybershaft's OP21 clock linked to my dCS Vivaldi Upsampler and DAC have been very positive. Not at all hard to hear and definitely improves overall sound quality. Visitors to hear my system before and after the addition of the Cybershaft OP21 were also able to hear the improvements.

Steve Z

Same for me ...

Same for me...

oldmustang said:

Likewise, my experience with Cybershaft's OP21 clock linked to my dCS Vivaldi Upsampler and DAC have been very positive. Not at all hard to hear and definitely improves overall sound quality. Visitors to hear my system before and after the addition of the Cybershaft OP21 were also able to hear the improvements.

Steve Z

Click to expand...

same here...I am using Mutec SE120 by the way.....

I must confess, added OP21A with external power supply to Vivaldi Clock, made music sound mo betta!

2ndLiner said:

I must confess, added OP21A with external power supply to Vivaldi Clock, made music sound mo betta!bn

Click to expand...

better clock cables further improve........

Habst from Germany are what I use. Fabulous cables at a reasonable price!

Thanks to fellow HK-based hardcore audiophile TLi who so kindly brought over his Cybershaft OP-21A with two Cybershaft BNC clock cables! I was able to audition it, connecting it to my SOtM-nH10G (one of two) and SOtM-USBultra.

I was able to experiment with the following:
1) 50 ohm cable impedance setting
2) 75 ohm cable impedance setting
3) TIDAL streaming (both clocklinks)
4) Aurender ALAC file playback (one clocklink to the USBultra)

The verdict?

Apart from setting 2), a deliberate impedance mismatch, which introduced jitter and resulted in a flatter, grayer, and hazier soundstage with diminished musicality, there was a delightful and significant improvement in all audiophile performance criteria for both 3) and 4). Having heard it for myself in my own system, I’m now convinced !

Memo to myself - pinning the link to a review of the SotM OCX10:

Inbound from Japan soon are the OP21A-D and its companion outboard power supply, the LTPW-01P - three BNC clock cables will connect the OP21A to three SOtM components, the USBultra, and 2 x nH-10G. As for the fourth BNC clock cable, I'm not sure whether it will be 100% compatible with my Aurender W20 music server.

Does anyone have any experience connecting a 10MHz sine wave Master Clock signal into the Aurender's coaxial BNC word clock input? Will the W20 happily accept it, and will be a "plug it in, and forget it" experience?

For reasons related to full compatibility with all formats at all sampling frequencies including DSD and Tidal's MQA, I've been using ONLY the USB output from my Aurender W20 to my MSB Technology ProISL module (via the ProUSB optical module). As the USB output data signal is asynchronous with no embedded clock, will the OP21A's 10MHz sine wave be (choose the appropriate word): rejected, ignored, accepted (sonically helpful), or corrupting (sonically damaging)?

Aurender's User Manual does not explain the compatibility permutations adequately, and an extensive search on various audiophile forum chat sites did not provide any answers, either.

As far as the word clock signals generated by the MSB Select II DAC's modules go (synchronous data streamed S/PDIF, AES/EBU, etc.), they should not, and can not be connected to the Aurender's BNC word clock input, because the fixed base frequencies of either 44.1kHz or 48kHz will often clash with the asynchronous USB and widely-varying base frequencies of the source signal.

Any insights would be greatly appreciated !

If memory serves, I posed the question to Aurender regarding use of the W20 with a 10 mHz master when it first came out and they did write back that it would auto-sense 10 mHz.

Please check with them to verify...

Yes, it’s likely to auto-sense, with the combination of a toggle in the Aurender Conductor software to turn the clock function on. But beyond that, I’ve been asking Aurender, but they have not been able to explain exactly how the W20 utilizes a 10MHz sine wave master clock signal, particularly when the data output is USB (not S/PDIF or AES/EBU). Will the W20’s front panel display “10MHz Master Clock”? How does the master clock signal override the W20’s internal crystal oscillators, and how does this affect the processing of data files and data packets, particularly during streaming?

Oh well... I went ahead and ordered an extra 50 ohm clock cable from Cybershaft anyway. Guess I’ll just have to experiment!

There has been no definitive reply from Aurender yet regarding the potential value of a “perfect” Master Clock in improving all things digital inside the W20. However, the link I found describing the importance of the “flip flop” function inside digital devices is very pertinent: