Mutec REF 10 Masterclock

[Confused]

40 minutes ago, zoltan said:

It seems that MUTEC will not exhibit any new product in Munich in two weeks. I think some of us expected something in the Empyreal class.  

Indeed, I was very much looking forward a Empyreal Class version of the MC3+USB.  It appears to be little more than a groundless rumour at the moment though.....

[ElviaCaprice]

On 4/23/2018 at 2:14 AM, Confused said:

I had been thinking about getting the REF10 for a while, but the sCLK-OCX10 had been announced before I ordered the REF10, so yes, I did consider it.  At that time there were no comparative reviews of the two items, so all I had was the many positive reviews and reports for the REF10, and the published specifications for the sCLK-OCX10.  When reviewing the specifications of the two, I could see that they were both very good, with the sCLK-OCX10 having maybe fractionally better specs in some areas, but the biggest differentiator was that the REF10 was square wave and the sCLK-OCX10 sine wave.  Which of these is best seams to depend on who you ask!  In my case, the primary reason to buy a reference clock was for the Mutec MC3+USB, the last clock before the DAC if you like, so it made sense to buy the product designed to feed the MC3+USB.  One other point I would make here is that I bought the SOtM bundle and the Mutec REF10 at the same time.  The REF10 is a little utilitarian, but it seems to have very good build quality, it looks like it will last forever.  The SOtM kit, by comparison, does seem a bit cheap and cheerful.  In a way, you could argue that this is a positive for the SOtM kit, in that you are paying for the electronics rather than fancy casework, but for something costing £3K, you would expect something reasonably decent.

 

The REF10 does have slightly better functionality also, with 4 off 75ohm outputs, 2 off 50 ohm outputs, which are all switchable, with a clear display to show which outputs are enabled.  The REF10 also includes it's own LPSU, so that is one less box to worry about.

 

I can also see my REF10 going together very nicely with whatever Mutec next bring out in their "Empyreal" class of products, but that remains to be seen.

 

I believe that a REF10 versus sCLK-OCX10 shootout has been planned, so maybe once that one is completed I will discover that I have made the wrong choice!  Meanwhile though, yes, I am very happy with the REF10.  I added it to the system and it just simply made everything sound better, more realistic, dynamic, just all good.  

I think you made the right choice.  Being able to power that SOtM component with a good LPS is a bonus for better SQ.  I like that SOtM makes it's components for separate PS.   If your going to spend that much, might as well do it right and go all the way.  Nothing less than a SR7.

[Confused]

50 minutes ago, ElviaCaprice said:

I think you made the right choice.  Being able to power that SOtM component with a good LPS is a bonus for better SQ.  I like that SOtM makes it's components for separate PS.   If your going to spend that much, might as well do it right and go all the way.  Nothing less than a SR7.

You make it sound like I have made the wrong choice!    Although to be clear, I am more than happy with the choice I made.

 

I agree with your point regarding flexibility offered by SOtM allowing you to use any power supply.  Although another way to look at it is that the REF10 approach removes this as an option, so it is one less thing to worry about or to be tempted to spend money for a minor upgrade.  You can be happy that you have an LPSU optimised for the specific use in the unit, and worry about other things instead.

 

So different approaches, pros and cons each way.  Choice is good!  You pay your money and take your choice.

[Johnseye]

I question whether a MR SR7 would be better than an LPSU designed specifically for the REF 10.  It would be nice if the supply was in a separate enclosure from the REF 10.

 

I have also heard from a bird who has an SR7, compared with OCX10 side by side and preferred the REF 10.  Apologies as I can not share who the individual is.

[ElviaCaprice]

51 minutes ago, Confused said:

You make it sound like I have made the wrong choice!    Although to be clear, I am more than happy with the choice I made.

 

I agree with your point regarding flexibility offered by SOtM allowing you to use any power supply.  Although another way to look at it is that the REF10 approach removes this as an option, so it is one less thing to worry about or to be tempted to spend money for a minor upgrade.  You can be happy that you have an LPSU optimised for the specific use in the unit, and worry about other things instead.

 

So different approaches, pros and cons each way.  Choice is good!  You pay your money and take your choice.

Let me put it this way.  Does Mutec make a reference power supply that audiophiles desire?  So what makes you think that their power supply built into their component is going to do any better?

[ElviaCaprice]

28 minutes ago, Johnseye said:

I question whether a MR SR7 would be better than an LPSU designed specifically for the REF 10.  It would be nice if the supply was in a separate enclosure from the REF 10.

 

I have also heard from a bird who has an SR7, compared with OCX10 side by side and preferred the REF 10.  Apologies as I can not share who the individual is.

Power supplies aside.  If so,  it sounds like the REF 10 is just better designed than the OCX10. 

[Johnseye]

Just now, ElviaCaprice said:

Power supplies aside.  If so,  it sounds like the REF 10 is just better designed than the OCX10. 

 

Yes.  I don't think we'll ever know about the supply. Not even worth the thought considering it's an impossibility.

[Confused]

9 minutes ago, ElviaCaprice said:

Let me put it this way.  Does Mutec make a reference power supply that audiophiles desire?  So what makes you think that their power supply built into their component is going to do any better?

I don't actually think that, I am quite sure that it would be possible to get a power supply that is better than the Mutec supplied item.  The key point is that I am happy that whatever Mutec supply is at least correctly specified  and is of decent quality, and it is one less thing to obsess over.

[zoltan]

11 minutes ago, ElviaCaprice said:

Let me put it this way.  Does Mutec make a reference power supply that audiophiles desire?  So what makes you think that their power supply built into their component is going to do any better?

The vast majority of high-end audio components have a built-in power supply despite the fact that almost none of them manufacture separately sold power supplies that 'audiophiles desire'. In fact, the Ref 10 having a built-in power supply is a plus for me.

[auricgoldfinger]

14 minutes ago, ElviaCaprice said:

Power supplies aside.  If so,  it sounds like the REF 10 is just better designed than the OCX10. 

 

The difference could be a result of the REF 10 square wave vs. the OCX10 sine wave.

[Johnseye]

Just now, auricgoldfinger said:

 

The difference could be a result of the REF 10 square wave vs. the OCX10 sine wave.

 

Do you know what this means exactly, what the implications are, pros and cons?

[ElviaCaprice]

6 minutes ago, zoltan said:

The vast majority of high-end audio components have a built-in power supply despite the fact that almost none of them manufacture separately sold power supplies that 'audiophiles desire'. In fact, the Ref 10 having a built-in power supply is a plus for me.

Which is a negative in my opinion.  Such as the Chord DAVE.  But I live with it.

[auricgoldfinger]

2 minutes ago, Johnseye said:

 

Do you know what this means exactly, what the implications are, pros and cons?

 

I have no idea.  IIRC, others have said the square wave theoretically has greater dynamic range.  I have never seen any detailed explanation or documentation to back up that assertion.

 

In my simplistic, non-technical world, I always saw it as an analog vs. digital question.  Many people are always trying to get the analog sound from their digital equipment.

[barrows]

7 minutes ago, ElviaCaprice said:

Which is a negative in my opinion.  Such as the Chord DAVE.  But I live with it.

An internal power supply is better than an external supply, given that the supplies in question are of the same quality.  DC does not like to travel distances, and the added L, C, & R of sending DC power over a cable between boxes reduces performance.

This is especially true for digital or high speed circuits of any kind, a clock source running at 10 MHz is certainly a high speed circuit.  The problem is that power supply impedance increases over distance, and digital circuits perform best with the lowest possible power supply impedance.  

Less boxes is often better.

[Confused]

11 minutes ago, Johnseye said:

 

Do you know what this means exactly, what the implications are, pros and cons?

According to Mutec:

 

Generates a very high slew-rate square wave signal for superior lock precision compared to the sine wave signals used by competing brands

 

Based on our knowledge of the importance of a clock signal’s rise time (steepness of the edge) for the audio performance, we chose a very high slew-rate, lowest ripple square wave signal for the REF 10. In contrast to the low slew-rate sine wave signals used by competing brands, this leads to superior lock precision and therefor minimized jitter at the receiving end. As a result, connected devices like DACs (D/A converters) and audio clocks will sound even better.

[auricgoldfinger]

16 minutes ago, Confused said:

According to Mutec:

 

Generates a very high slew-rate square wave signal for superior lock precision compared to the sine wave signals used by competing brands

 

Based on our knowledge of the importance of a clock signal’s rise time (steepness of the edge) for the audio performance, we chose a very high slew-rate, lowest ripple square wave signal for the REF 10. In contrast to the low slew-rate sine wave signals used by competing brands, this leads to superior lock precision and therefor minimized jitter at the receiving end. As a result, connected devices like DACs (D/A converters) and audio clocks will sound even better.

 

What is the competing rationale from SOtM for the sine wave?  If the slew rate is really the determining factor, why wouldn't SOtM have also followed that path.  They must have different ideas.

[ElviaCaprice]

36 minutes ago, barrows said:

An internal power supply is better than an external supply, given that the supplies in question are of the same quality.

 

I don't think anyone would dispute this.  But given that it would be difficult to measure or compare, we don't know.  And I for one like to know.  So I prefer external.

 

Additional, external power supplies are modular, allowing their use to continue.

[auricgoldfinger]

From the SOtM site:

 

As many of the audiophiles know, the clock signal is an angled square digital signal, but the sCLK-OCX10 produces a pure analog sine wave of 10 MHz, and this sinusoidal clock signal is the method adopted for the best clock product because of the very less noise than the angled square digital clock signal. In order to implement this, the sCLK-OCX10 consists of only analog circuit.

 

Now we can have a proper discussion. 

[ElviaCaprice]

1 minute ago, auricgoldfinger said:

From the SOtM site:

 

As many of the audiophiles know, the clock signal is an angled square digital signal, but the sCLK-OCX10 produces a pure analog sine wave of 10 MHz, and this sinusoidal clock signal is the method adopted for the best clock product because of the very less noise than the angled square digital clock signal. In order to implement this, the sCLK-OCX10 consists of only analog circuit.

 

Now we can have a proper discussion. 

Next question, does the sCLK-OCX10 outperform the Mutec on certain components?  Like the sCLK-EX?

[Confused]

9 minutes ago, ElviaCaprice said:

Next question, does the sCLK-OCX10 outperform the Mutec on certain components?  Like the sCLK-EX?

It's an interesting point.  As pure speculation, it could be possible that the REF10 is the best with the MC3+USB but could be out performed by the sCLK-OCXO10 with the sCLK-EX.  Who knows? 

 

I also suspect the noise issue as mentioned by @auricgoldfinger is what makes the REF10 super sensitive to cable length and quality, so as ever, there are many factors to consider.  As an example, the REF10 might come out top if it has very short, high spec cables, the sCLK-OCXO10 might be less sensitive in this area. 

 

It would take a few experiments and A/B tests to definitively sort all this out!
 

[jelt2359]

Since we’re speculating I have a question and apologies if it’s been answered- but why did ref10 take a few years to make it from prototype to final product, in an obviously different form?

[Solstice380]

2 hours ago, Johnseye said:

 

Do you know what this means exactly, what the implications are, pros and cons?

 

From a theoretical standpoint:

 

The receiver is triggered by the transition.  I don’t know if that trigger is a simple threshold level or a delta between 2 levels. 

 

@jabbr do you have knowledge of how a chip senses the clock signal to  trigger?

[Summit]

2 hours ago, barrows said:

An internal power supply is better than an external supply, given that the supplies in question are of the same quality.  DC does not like to travel distances, and the added L, C, & R of sending DC power over a cable between boxes reduces performance.

This is especially true for digital or high speed circuits of any kind, a clock source running at 10 MHz is certainly a high speed circuit.  The problem is that power supply impedance increases over distance, and digital circuits perform best with the lowest possible power supply impedance.  

Less boxes is often better.

 

Not necessarily. Big PSUs near sensitive clocks and other devices can have a negative effect and is why MSB, Naim, DCS and Esoteric either have a separate PSU box or a separate clock-box in their SOTA DACs.

[barrows]

22 minutes ago, Summit said:

Not necessarily. Big PSUs near sensitive clocks and other devices can have a negative effect and is why MSB, Naim, DCS and Esoteric either have a separate PSU box or a separate clock-box in their SOTA DACs.

Well sure, if you do a poor internal layout and put the supply too near to critical components it can be a problem, but this is a part of good engineering and layout practices.  A poor design is a poor design, and using an external supply to overcome a poor design is not the best approach. 

For line level components the power requirements of the supply are low, and such that radiating interference fields are very small, and can be accommodated inside a standard size chassis with appropriate layout, and perhaps a bit of shielding in some cases.

Separate power supply boxes offered by some manufacturers for line level components are usually an unnecessary added expense (and allow for much additional revenue to the manufacturer).

 

And it is ALWAYS (necessarily) true that distributing DC power over a cable is a poor compromise.

[jabbr]

5 hours ago, Solstice380 said:

 

From a theoretical standpoint:

 

The receiver is triggered by the transition.  I don’t know if that trigger is a simple threshold level or a delta between 2 levels. 

 

@jabbr do you have knowledge of how a chip senses the clock signal to  trigger?

 

Depends on the chip and logic family. http://bwrcs.eecs.berkeley.edu/Classes/icdesign/ee141_f01/Notes/chapter6.pdf

 

In the simplest situation the gates transition from "low" to "high" based on voltage thresholds. Transistors are fashioned into latches and flip flops. Data is latched into a chip via a clock:

 

In this example "ETL" or edge triggered logic:

http://web.ece.ucdavis.edu/~halasaad/Data/mwscs06.pdf

 

There are also differential logic families e.g. ECL.

 

https://patents.google.com/patent/US5900760

 

Again "edge triggered"