[xdanny]

John, thank you for the kind words and for making this possible!   And your work is much appreciated!!

Quote:

Danny,
Seeing as nobody else seems to want to acknowledge your post, let me thank you for doing the test & providing some more data about the clocks issue. I'm not going to say anything about the results, they speak for themselves.

 

[shamu144]

Thank you for your work and those comprehensive impressions Xdanny. You seem to confirm indeed - and I have no reason to doubt it - that the Jkeny modded hifaces are somehow far less dependant on the small vs large clock issue, maybe requiring a very resolving system for those differences to show up. That is good news for modded Hiface owners.
 
Now, that doesn't settle the current issue with stock units. When I say a "night and day" differences, please bear in mind we are talking transports here.
 
Quote:

It's not impossible at all.  If there is a real "night and day" difference between the two clocks, with a large enough sampling a "statistical significance" will emerge.  
 

Unfortunately, a proper statistical analysis as suggested is not going to happen. Very few people actually have the opportunity to own at the same time 2 different versions of the stock units. However, what suprises me is that despite the existence of this thread, few people are complaining about the sound quality of their Hiface, when plenty of small clock units have obviously been sold in the last months. Some voices where raised in disagreement with Slim.a review of the original Hiface, but nothing more.
 
I'll have to admit that if everyone is happy with their unit, there is little chance M2Tech will take this issue seriously. Only if there is a strong conviction among users that their device is not performing up to specs, then we will be able to pressure distributors and manufacturer to offer a solution.
 
At this moment, I believe the most convenient method to discover a "faulty" unit without opening the case is the procedure described by Xdanny, listening while using the SoX resampler. As soon as I have some time, I will try it with the small clock unit I have at home and let you know the findings. If differences are easy enough to spot, then it will be interesting to hear other Hiface owners reports.

 

[grokit]

shamu144, a very reasonable post.

 

[xdanny]

+1, I agree... Easy test to do, preferably before opening up a unit.  As blind as it could get...

 

[sleepy dan]

Your interpretation of xdanny's results is only one possible, the other being of course that there is no significant difference between the clocks.  Either is possible, but I believe the second is more likely.  From what I know of oscillator circuits, PSU sensitivity would show up as higher phase noise - it's all related to the Q of the crystal and the impedance of the circuit.  M2Tech have already said that the small clock has better phase noise. 
 
Quote:

Thank you for your work and those comprehensive impressions Xdanny. You seem to confirm indeed - and I have no reason to doubt it - that the Jkeny modded hifaces are somehow far less dependant on the small vs large clock issue, maybe requiring a very resolving system for those differences to show up. That is good news for modded Hiface owners.
 

 

[sleepy dan]

If everyone is happy with their units, what does this suggest?
 
Quote:

Originally Posted by shamu144 /img/forum/go_quote.gif
 However, what suprises me is that despite the existence of this thread, few people are complaining about the sound quality of their Hiface
 
........
 
I'll have to admit that if everyone is happy with their unit, there is little chance M2Tech will take this issue seriously

 

[shamu144]

Now you are not beeing fair, in that you are suggesting that all of us 3 (Regal, myself, my friend) have invented those differences in A/B comparative listening... It is of course not the case. So despite not beeing likely according to your experience, we will have to accept so far the fact that stock units and modded units respond differently to the small clock implementation.
 
Quote:

Your interpretation of xdanny's results is only one possible, the other being of course that there is no significant difference between the clocks.  Either is possible, but I believe the second is more likely.  From what I know of oscillator circuits, PSU sensitivity would show up as higher phase noise - it's all related to the Q of the crystal and the impedance of the circuit.  M2Tech have already said that the small clock has better phase noise. 
 

 

That they don't know what they are missing. And frankly, I never heard such a good transport before the Hiface, be it the Musiland 01, my sony SACD player, the Lavry through USB, etc.... So I would probably have been unable to tell as well what I was missing, should I have received a small clock unit.
 
Quote:

If everyone is happy with their units, what does this suggest?
 

 

[sleepy dan]

Quote:

Now you are not beeing fair, in that you are suggesting that all of us 3 (Regal, myself, my friend) have invented those differences in A/B comparative listening... It is of course not the case. So despite not beeing likely according to your experience, we will have to accept so far the fact that stock units and modded units respond differently to the small clock implementation.
 

 

Yes I do suspect you, regal and your friend are imagining differences.  It's not a personal attack, it stems from an acknowledgement that subjective impressions can be unreliable, and the technical reasons I've given.  How is that unfair?  Being human our sense deceive us all the time.
 

Quote:

That they don't know what they are missing. And frankly, I never heard such a good transport before the Hiface, be it the Musiland 01, my sony SACD player, the Lavry through USB, etc.... So I would probably have been unable to tell as well what I was missing, should I have received a small clock unit.
 

 
 
Again that's one interpretation.  Do you not think that if, after spending $150 on a Hiface, it sounded worse than a $50 sound card, all these people wouldn't be complaining?  Or is it possible that as people are happy with their Hifaces, there is no issue?

 

[mmerrill99]

Quote:

Now you are not beeing fair, in that you are suggesting that all of us 3 (Regal, myself, my friend) have invented those differences in A/B comparative listening... It is of course not the case. So despite not beeing likely according to your experience, we will have to accept so far the fact that stock units and modded units respond differently to the small clock implementation.
 


That they don't know what they are missing. And frankly, I never heard such a good transport before the Hiface, be it the Musiland 01, my sony SACD player, the Lavry through USB, etc.... So I would probably have been unable to tell as well what I was missing, should I have received a small clock unit.
 

 

Shamu, thank you for making reasonable posts in an obviously calm manner!
 
One thing is at odds with your last statement "that they don't know what they are missing" - you are forgetting that a lot of Hiface units have one small & one MEC clock so the differences between these two speed families would be very noticeable according to the claims in this thread & so there should be a lot of people saying this but I don't hear the clamour, do you? This was said before a number of times but was ignored. I completely agree that if you want to be taken seriously (by M2tech or anyone) then reasonable evidence needs to be amassed to back up the claims - simply jumping up & down & making accusations will certainly not get anyone anywhere. That has been the tone on this thread for a while!
 
It's hard to reconcile the claims made here with both the technical data, technical knowledge & also with the lack of further evidence or others reporting the same findings. Again, nobody is accusing you of making things up but maybe you could revisit your tests & look deeper into them for other possible reasons. It may be that there are other differences within the Hiface units besides the clocks for instance? I don't know how you will identify them unless they are visually obvious. It may be other external factors are effecting the outcomes, I don't know, only you can answer that. For instance let me cite something - I have had a couple of people report that the modified Hiface did not live up to expectations & when I delved further into their systems, I found either that they were using iTunes to rip the CD to disk or that they were using upsampling in their playback system  ( as a separate box or in their DAC)
 

 

[sleepy dan]

With regard to the notion that the small clock is much more sensitive to power supply quality ..... Op-amp manufacturers always specify this sensitivity - it's known as PSRR (power supply rejection ratio).  It's given in datasheets, usually along with several graphs showing how it varies with frequency.
 
Have you ever looked at the datasheet for a clock?  I've looked at plenty, and never seen PSRR mentioned.  Here's what the Crystek 950 (the clock used in the latest Buffalo DACs) gives ....
 
 
 
• Frequency Range: 10-330MHz
• Input Voltage: 3.3V±0.3V
• Input Current: <15mA max.
Output:
• TRUE SINEWAVE
• Load: 50Ω
• Power: 0 dBm min. into 50Ω
5 10
Price Each
For quantities of 25 and up, call for quote.
Operating Frequency
(MHz) 1
Crystek
Part No.
MOUSER
STOCK NO.
5 10 25
Price Each
For quantities of 50 and up, call for quote.
Operating
Frequency
(MHz)
Operating
Temperature
Range
Package
Pullability
APR
1
Crystek
Part No.
MOUSER
STOCK NO.
Package
(Other frequencies, stabilities, voltages, outputs and temperature ranges available. Please contact your Mouser representative for more information.)
Note: This Product is also available as a VCXO (Voltage Controlled Crystal Oscillator) CVXO-083/983.
(Other frequencies, stabilities, voltages, outputs and temperature ranges available. Please contact your Mouser representative for more information.)
CVPD 920/940/970 HFF SERIES, LVPECL, VCXO, 9 X 14MM SMD
Specifications:
• Frequency Range: 50~670MHz
• Input Voltage: 3.3V±0.3V
• Control Voltage: 1.65V±1.65V
Output:
• Differential LVPECL
• Symmetry: 49/51% typ., 45/55% max.
• Rise/Fall Time: 550pS max. @ 20%
5 10 25
Price Each
For quantities of 50 and up, call for quote.
Operating
Frequency (MHz)
Operating
Temperature Range
Package
1
Crystek
Part No.
MOUSER
STOCK NO.
(Other frequencies, stabilities, voltages, outputs and temperature ranges available. Please contact your Mouser representative for more information.)
CVHD-950 SERIES, CMOS ULTRA LOW PHASE NOISE , VCXO, 9 X 14MM SMD
Specifications:
• Frequency Range: 50-125MHz
• Frequency Pulling: ±20ppm APR min.
• Input Voltage: 3.3V±0.3V
• Control Voltage: 1.65V±1.65V
Output:
• CMOS
• Symmetry: 45/55% max. @ 50% Vdd
• Rise/Fall Time: 3nsec max. @ 20%-80% Vdd
• Input Current: 15mA typ., 25mA max.
• Low Phase Noise Floor @ -162dBc/Hz
(See Plot Above)
• Jitter: 0.5psec typ., 1psec RMS max.
• Linearity: ±10% max.
• No Sub-Harmonics
5 10 25
Price Each
For quantities of 50 and up, call for quote.
Frequency
1
Crystek
Part No.
MOUSER
STOCK NO.
(Other frequencies, stabilities, voltages, outputs and temperature ranges available. Please contact your Mouser representative for more information.)
CCHD-950 SERIES, CMOS ULTRA LOW PHASE NOISE CLOCK OSCILLATORS, 9 X 14MM SMD
-162 dBc/Hz Phase Noise Floor
-162 dBc/Hz Phase Noise Floor
High Frequency Fundamental Design
Specifications:
• Frequency Range: 50-125MHz
• Input Voltage: ±3.3V±0.3V
• Input Current: 15mA typ., 25mA max.
Output:
• Symmetry: 45/55% max. @ 50% Vdd
• No Sub-Harmonics
• CMOS
• Jitter: 0.5psec typ., 1psec RMS max.
• Stability: ±50ppm @ 0°C to 70°CC
• Low Phase Noise Floor @ -162dBc/Hz (See Plot Above)
549-CVHD950-54 CVHD-950-54.000 9x14mm SMD 54.000 0°C to 70°C 26.80 25.46 24.12 21.44
549-CVHD950-74.1758 CVHD-950-74.1758 9x14mm SMD 74.1758 0°C to 70°C 27.40 26.03 24.66 21.92
549-CVHD950-74.25 CVHD-950-74.250 9x14mm SMD 74.250 0°C to 70°C 27.40 26.03 24.66 21.92
549-CVHD-950-80 CVHD-950-80.000 9x14mm SMD 80.000 0°C to 70°C 27.70 26.32 24.93 22.16
549-CVHD-950-100 CVHD-950-100.000 9x14mm SMD 100.000 0°C to 70°C 27.70 26.32 24.93 22.16
549-CVHD-950-125 CVHD-950-125.000 9x14mm SMD 125.000 0°C to 70°C 41.66 37.50 32.19 31.60
• Low Phase Noise Floor @ -150dBc/Hz
(See Plot Above)
• Stability: ±25ppm @ 0 to 70°C
• Harmonics: -20dB max.
• Frequency Range: 50-125MHz
• Input Voltage: ±3.3V±0.3V
• Input Current: 15mA typ., 25mA max.
 
• Frequency Range: 50-125MHz
• Input Voltage: ±3.3V±0.3V
• Input Current: 15mA typ., 25mA max.
No mention of PSRR. Now, do you not think that if PSRR was at all significant, it wouldn't be mentioned?  Do you think clock manufacturers are really unaware of this parameter?  

 

[sleepy dan]

......

 

[sleepy dan]

..........

 

[digger945]

.... .._ ....

 

[shamu144]

The answer to that question is pretty obvious I think. Most people listen 90% of their music collection - if not more - in RBCD format, which means only at 44.1khz multiples. Do you have lot's of recordings at 48/96/192... I personnally only have 4 or 5 DVD-Audio ripped and some High Res downloads from 2L and HTracks, but nothing more. Regal stated he only had one file at 48khz ripped from a DVD. How do you want people to spot differences with material that only represent 5% of their listening time. Additionnally, chances are that those High Res files will sound better indeed than the RBCD, because of both higher sample rates and better clock handling at those resolutions.
 
So yes, I do see it very likely indeed that people do not realize what they are missing.
 
Quote:

 
One thing is at odds with your last statement "that they don't know what they are missing" - you are forgetting that a lot of Hiface units have one small & one MEC clock so the differences between these two speed families would be very noticeable according to the claims in this thread & so there should be a lot of people saying this but I don't hear the clamour, do you?

 

[upstateguy]

 
Quote:

 I see now that I'm next up for mob lynching!

 
 
As far as I'm concerned, I don't see anything wrong with jkeny or his modifications.  He figured out a way to improve the HiFace, told everyone how to do it and then offered to make modded units for those who couldn't do it themselves. 
 
That he makes a couple of bucks doing the mods for anyone who wants them.... Big Deal!  I don't see anyone offering to do it for free.
 
The units m2tech supplied to jkeny were 'factory units' and from what I've read here, all of them improved with the battery mod.  I don't see how the decision to change components by m2tech has any bearing on jkeny?
 
But it is the stock HiFace that I'm concerned about.
 
 
Quote:

If there is a real "night and day" difference between the two clocks....

Regarding the stock HiFace with a small clock and a large clock, there is not a night and day difference between the small 22.5 clock and the  large 24.5 clock, but the difference becomes noticeable with the SOX resampler.  And once you know what to listen for, it becomes easier to tell them apart.

 
Quote:

If everyone is happy with their units, what does this suggest?
 

It might suggest that they had nothing to compare them to ........... 
 
And, many people were not happy with what they received during the time period in question and sold them off.
 
 
Quote:

I have had a couple of people report that the modified Hiface did not live up to expectations & when I delved further into their systems, I found either that they were using iTunes to rip the CD to disk or that they were using upsampling in their playback system  ( as a separate box or in their DAC)
 

 
 
This is very disconcerting.  I don't understand what you mean? 
 
I see on your web site that you say, "This upsampling will mask the benefits of the low jitter output from the Hiface & will prevent it's true sonic benefits from being realised. "
 
My North Star MKII DAC upsamples.  There are many DACs that upsample. 
 
Does this mean that the sonic benefits from a HiFace are not realized when using an upsampling DAC?
 
Could this the reason why m2tech designed the stock HiFace to be brighter and louder than other transports?
 
On a side note, I don't see how upsampling  would defeat the benefits of the cleaner, battery power in your modification?  Cleaner in cleaner out, right?
 
USG