Superclocks

[Superdad]

3 hours ago, barrows said:

You have measured differences within the range which would be audible?  That is <20kHz and above -110 dB?  Really?  I mean, I get that an early rolloff filter could be considered a measurable/audible difference (although not theoretically for my hearing)-but what about minimum phase vs. linear, for example...

 

4 hours ago, Miska said:

It is not something that would elude measurements but be audible subjectively. Sometimes the differences are easier to analyze than to hear.

 

Ah, the crux of the biscuit as Mr. Zappa would say!  

 

a) Many of the "objectivists" say that if you can't measure the differences as THD, IMD, or frequency response in the 20-20KHz band then you can't hear it; Thus they claim that all these variations in filters must be inaudible and that we are delusional.

 

b) Anyone who has either designed or played with digital filter parameters will tell you that the differences of steepness, cutoff, length, and phase are readily audible;

 

c) Yet Jussi and DAC designers will mostly look at their filters' ultrasonic FR and impulse response.

 

So who is all wet in this?  

 

[barrows]

My POV is that all "should" be measurable at the analog output of the DAC, and perhaps it is, if we just know exactly what measurements to use..  But I have heard too many very audible differences which do not show up in what is traditionally considered the audible elements of the traditional measurement sets: J-test, THD/noise, FR, IMD, etc.  once you hear these differences they are no longer un-hearable, so to speak.  I also do not believe in "magical" audio, but it does appear that we are in need of better measurements to describe all aspects of audio performance.

Interestingly, to me a least, is that Jussi prefers to see alias products at extremely low levels, where most engineers would be comfortable with filter designs with -120 dB suppression of filter alias artifacts, Jussi appears to prefer them to be much lower although clearly anything at -120 dB should not be audible in room (never mind that these artifacts are above 20 kHz as well).  I know Jussi to be a really, really smart guy, so I intrinsically accept that there is probably something to this, but I am at a loss as to how there would be a difference between an alias at -120 dB and -160 dB?

[Miska]

8 hours ago, barrows said:

You have measured differences within the range which would be audible?  That is <20kHz and above -110 dB?  Really?  I mean, I get that an early rolloff filter could be considered a measurable/audible difference (although not theoretically for my hearing)-but what about minimum phase vs. linear, for example...

 

I don't want to specify what figures are "audible", I don't have clear answer for that. Not for clock jitter/phase noise either. I just know what is measurable or detectable in digital domain analysis. For example filter transition for RedBook begins somewhere below 22.05 kHz when it begins to roll off from 0 dBFS. This is certainly measurable in frequency response. Minimum phase vs linear phase is certainly measurable in the phase response too.

 

[Miska]

8 hours ago, mansr said:

Making a filter/modulator that causes such differences is trivial. Then again, so is making one that doesn't.

 

I don't have problems analyzing differences of the stuff in SoX. Your different modulator each produce different results. If two things produce same output values, then they are usually the same.

 

I just don't put any limits like "20 kHz" or "-110 dB" because I think such limits are artificial. What I look at, only limit is computational accuracy or limits of measurement. But usually the filter roll-off point and roll-off curve steepness and shape ("shape of the knee") is certainly measurable.

 

[barrows]

2 minutes ago, Miska said:

 

I don't want to specify what figures are "audible", I don't have clear answer for that. Not for clock jitter/phase noise either. I just know what is measurable or detectable in digital domain analysis. For example filter transition for RedBook begins somewhere below 22.05 kHz when it begins to roll off from 0 dBFS. This is certainly measurable in frequency response. Minimum phase vs linear phase is certainly measurable in the phase response too.

 

of course I understand that, but is it audible? certainly a measurement showing the rolloff above even at 19 kHz is not audible, But still people hear the difference in filters-I would conclude that something else about the filter's response is what we hear, but we are not measuring that thing.  This is the point which I was trying to make.  Same thing with linear phase vs. minimum phase, the phase shift at a very high frequency which we can measure, is also not considered audible by humans (unlike low frequency phase shifts), but still people report differences in sonics of these two filter types (overly simplified here, as we can have intermediate phase filters as well).  We can measure the pre-ring/post ring distribution with an impulse response test, but is that ringing audible?  Very questionable that it is.

I have seen posts where you advocate for very high attenuation of alias products, to levels many times below an audible threshold, indeed many times below what the analog circuitry is capable of delivering, is that audible: well it cannot be as far as what one measures, but perhaps there is some other aspect to it which is audible, otherwise, why do you advocate for it.

In no way am I suggesting you are wrong, and I have huge respect for your work and expertise with digital filters/oversampling, I am just trying to illustrate that the measurements we currently use are inadequate to describe all aspects of the sound of (in this case) a DAC, and that for similar reasons, perhaps the higher performing clock, not showing any difference in an analog domain J-test measurement, is still affecting some aspect of the sound.  I also am not trying to suggest that we cannot measure these things, just that we do not measure them all with the current set of measurements traditionally done-I think we need better measurements, I am really interested in what those might be.

[Miska]

3 hours ago, barrows said:

of course I understand that, but is it audible? certainly a measurement showing the rolloff above even at 19 kHz is not audible, But still people hear the difference in filters-I would conclude that something else about the filter's response is what we hear, but we are not measuring that thing.  This is the point which I was trying to make.  Same thing with linear phase vs. minimum phase, the phase shift at a very high frequency which we can measure, is also not considered audible by humans (unlike low frequency phase shifts), but still people report differences in sonics of these two filter types (overly simplified here, as we can have intermediate phase filters as well).  We can measure the pre-ring/post ring distribution with an impulse response test, but is that ringing audible?  Very questionable that it is.

 

My take is that it is audible. You know I do endless cycles of adjusting the mathematics, analyzing the result and then listening. I have my own idea which things are audible in which way, just based on my own experience.

 

And I have many cases which I would count as "blind listening", because I get a lot of feedback, and in many cases people simply cannot know what is behind. That has shaped my view also what is audible.

 

I think the scientifically defined limits of human hearing are very average and the research methodology there would need more improvement than measurements.

 

3 hours ago, barrows said:

I am just trying to illustrate that the measurements we currently use are inadequate to describe all aspects of the sound of (in this case) a DAC, and that for similar reasons, perhaps the higher performing clock, not showing any difference in an analog domain J-test measurement, is still affecting some aspect of the sound.

 

I think in such cases you need to improve SNR of the J-test measurement, usually things show up when you keep improving that part. But my take on this clock thing is that after changing a clock you must first measure it to check if it performs at least as well as before changing it. Because in many cases the difference is measurable. If it is not measurable and you still hear differences, that is all fine. But at least you know it didn't get objectively worse. I'd call it "sanity check". Because nice numbers on a datasheet/specs doesn't automatically mean that it will perform nice in the particular use case in real world! IOW, you cannot assume the change is unmeasurable.

 

For example using some fancy 10 MHz "atomic clock" as reference means that you need to have DPLL to generate the actual audio clock frequencies from that. And designing a good DPLL is not so easy, so it can easily go bad at that point. Another aspect is that you usually get best performance when you don't have clock dividers, but the clock is really running straight at the same frequency as the conversion section. But you probably cannot do that with any DAC chip, you need to go for a discrete DAC to do that. So these are some of the clocking related things, but unrelated to the clock module itself.

 

J-test24 is good for measuring clock performance, because you have a synchronous undithered test tone and rest of the spectrum is "black hole".

 

3 hours ago, barrows said:

I also am not trying to suggest that we cannot measure these things, just that we do not measure them all with the current set of measurements traditionally done-I think we need better measurements, I am really interested in what those might be.

 

Traditional set of measurements is not good for example for what I do with filters and modulators. I do also standard set, but in addition I do bunch of other measurements that are focused specifically on filter and modulator performance.

 

[mansr]

6 hours ago, Miska said:

I don't have problems analyzing differences of the stuff in SoX. Your different modulator each produce different results.

Of course they do. If they didn't, there would be no point having the options.

[Jking]

Saw a few years ago people were using atomic clocks for Dacs or timing device that had external clock sources.  Would using one make sense?

[mansr]

2 minutes ago, Jking said:

Saw a few years ago people were using atomic clocks for Dacs or timing device that had external clock sources.  Would using one make sense?

If you have to ask, the answer is no.

[Jking]

The Mutec with Rubidium clock use to be popular on these pages.  

 

 

[barrows]

@miska

 

OK, I get that the measurements you use to develop filters are the necessary and applicable ones too show that the filter design is doing what you want from a technical perspective, and I have no doubt that many of your filters do sound better/different.  I do not have HQPlayer (yet, unfortunately) because I do not have a good machine to run it on.  But I have tried quite a few different approaches, and the differences are usually small, but meaningful in an audiophile sense.

I have a hard time saying that a measurement of something down at -120 dB is audible though, in room, through loudspeakers, given the real world dynamic range of systems, and I use Ncore amps with very low floors, and the DACs here are very good in this regard as well.  And I have hard time saying that a measurement of something above 20 kHz is  audible, given human hearing limitations.  Phase relationships at high frequencies, I do not know enough of the psychoacoustics of to know for sure, but my understanding is that it is not audible, I will do some reading on that to try and come to a better understanding.

Clearly you appear to believe that current measurements are enough, this is where we depart a bit.

So your take away is that humans, apparently, can hear things which most research says we cannot.  I certainly cannot hear a single tone at 20 kHz in room, but you are suggesting that perhaps I can hear something at that frequency and higher?  Hmm, I am aware of some research suggesting our bones conduct these frequencies and we may be able to "sense" them somehow, I ma to sure how that would be related to our perception of reproduced music...  Then there are those folks who claim to be sensitive to things like WiFi, bodily.  Of course a lot of those same people also claim to be able to "see" the invisible aliens walking among us, oops, I digress...

As to clocks, for sure I am no believer in external master clocks at 10 MHz (atomic or not) which them have to be carried on a wire to a component, and then converted by a DDS to the actual needed frequency.  Although the recent improvements in DDS does make this a little better proposition.  Give me an XO with -120 dBc/Hz phase noise at 10 Hz, at the audio frequency, right at the DAC and flip flops please.  And as I said before, i would like to have a phase noise analyzer to be able to measure that performance right at the input pin of the DAC to be sure.  But even without that, with clocks using the same implementation, I do hear the difference if a swap to an XO with a bit better (6 dBc/Hz or more) low frequency phase noise performance.  And, I consider this difference larger than what I hear via most different digital filters (but I need to try HQPlayer as well).

I do not see much difference in my advocacy for a better clock (lower low frequency phase noise) and yours for HQPlayer filters.  Both have measurements to show a difference, and both have the anecdotal evidence (which you rely on from your customers as mentioned).  Maybe I can get the AP here at some point, some XOs to test, and take a really close look at the 24 bit J-tests measurements.

As an aside, I know it takes a super computer to run HQPlayers single pass filters (I need to go to DSD 256 and even 512), and quite a bit less power to use the 2 stage filters, how much of the HQPlayer "goodness" can I get via the 2 stage filters?

[barrows]

1 hour ago, Jking said:

The Mutec with Rubidium clock use to be popular on these pages.  

 

 

The problem with an external clock like this is two fold:  First you have to carry the clock signal on a wire, with at least two connections, and then some internal connection inside the component are virtually unavoidable as well.  So you get clock degradation there.  Then, you have the wrong frequency for an audio DAC, so the DAC then internally has to convert that 10 MHz frequency to one of the audio frequencies (like 22.5792, for example).  This is generally done using a DDS chip, this also adds more phase noise, so by the time your clock signal gets to the DAC ir has been degraded quite a bit.

It is much better to put a really good clock in the DAC, right at the DAC chip, at the correct frequency.

External reference clocks like the Mutec are designed for recording studio use where often multiple components need to be clocked together to a single reference, in a home audio playback system we do not have multiple components to clock together (OK there are some rare exceptions to this), so it is better to not use an external clock.

But do not take my word for this, ask @marce, or at @JohnSwenson.

[Miska]

2 hours ago, Jking said:

The Mutec with Rubidium clock use to be popular on these pages.  

 

 

 

Now you'll only need to check how it performs in real world when used as clock for DAC instead of the built-in crystal...

 

Challenge is still the DPLL needed to generate useful clock from 10 MHz instead of internal crystal that is already running at needed frequency.

 

[Miska]

1 hour ago, barrows said:

Clearly you appear to believe that current measurements are enough, this is where we depart a bit.

 

Current measurements already show differences and you hear differences, you just need to find the correlation...

 

1 hour ago, barrows said:

So your take away is that humans, apparently, can hear things which most research says we cannot.  I certainly cannot hear a single tone at 20 kHz in room, but you are suggesting that perhaps I can hear something at that frequency and higher?

 

You cannot hear sine above 20 kHz, but it is not same as having over 20 kHz component as part of transient rise time. It is too simplistic to assume decomposed spectrum components alone would work the same way as the combination.

 

I personally have problem with metal dome tweeters that have resonance frequency somewhere between 22 and 28 kHz. Listening to such in long term is like listening to dentist ultrasonic teeth cleaning tool. If you know what I mean, that cleaning tool sounds just like telephone line modem handshake from the 80's-90's. Of course the tweeter is not as bad, but you get sort of headache kind of annoying feeling in back of your head. And it is much worse if the DAC has leaky filters. You often notice this at shows/fairs where you suddenly enter listening room and get a feeling of the sound in couple of first seconds before you even register what you are listening to.

 

1 hour ago, barrows said:

I do hear the difference if a swap to an XO with a bit better (6 dBc/Hz or more) low frequency phase noise performance

 

I don't have any doubts about that in general. My doubts are more like swapping clocks to commercial products with messy wires and stuff. That's why I'm after sanity check measurements just like for every other aspect too. Always measure first to sanity check that things didn't at least get worse...

 

Certainly those things matter (given good PCB layout design, materials and everything else), no doubt about that. My message is mostly that there is a lot to improve in terms of PCB layout design and other circuit design, just replacing a clock module is likely doing smaller difference than a change in PCB layout or PSU design. Or even capacitor selection. Unless you are starting with a really poor clock. Not to even forget how much design of the analog LPF section matters!

 

 

 - A person who has pile of Crystek CCHD-957's at 22.5792 & 24.576 MHz in stock and looks into 10 Hz phase noise figure...

 

[barrows]

9 minutes ago, Miska said:

I personally have problem with metal dome tweeters that have resonance frequency somewhere between 22 and 28 kHz. Listening to such in long term is like listening to dentist ultrasonic teeth cleaning tool. If you know what I mean, that cleaning tool sounds just like telephone line modem handshake from the 80's-90's. Of course the tweeter is not as bad, but you get sort of headache kind of annoying feeling in back of your head. And it is much worse if the DAC has leaky filters. You often notice this at shows/fairs where you suddenly enter listening room and get a feeling of the sound in couple of first seconds before you even register what you are listening to.

This is a great observation.  Most metal domes are a problem for me as well, although I have heard some beryllium implementations which are very good, they appear to have their resonance higher up.  I have oldish soft revelator dome here.  I ma actually flirting with building some speakers, with the help of an established speaker designer, looking at the Seas soft domes, Crecendo, etc.

 

And of course I agree entirely on the implementation of the rest of the DAC, analog stage, filter implementation.  I am certainly not advocating for just changing the XO in a DAC with poor design overall!  I am advocating for better clock when the other aspects of the design are (near) "perfect".  I consider DAC analog stage design (including analog filter) the most important aspect, followed by power supply implementation (given we are not talking ancient DAC chips, etc), then digital circuitry and clocking.  I would only optimize the XO after the others are addressed.  I have gravitated away from leaky (digital) filters and minimum phase types lately, as DAC/system performance improves it appears I prefer apodizing filters and linear phase...  I wonder if a lot of folks who prefer slow roll-off filters may be experiencing a kind of masking effect (similar to dither?) which is, perhaps, covering up system "problems".

 

I am almost ready to start listening to my DSC-2 DAC in my main system, it will be an interesting contrast to my ESS 9038 based DAC, I will be feeding both DSD 256, and perhaps 512 later on.

[Superdad]

2 hours ago, barrows said:

Of course a lot of those same people also claim to be able to "see" the invisible aliens walking among us...

 

Don’t you?  There must be at least three taking up residence here on the CA forums.  These aliens know who they are... 

[marce]

2 hours ago, barrows said:

The problem with an external clock like this is two fold:  First you have to carry the clock signal on a wire, with at least two connections, and then some internal connection inside the component are virtually unavoidable as well.  So you get clock degradation there.  Then, you have the wrong frequency for an audio DAC, so the DAC then internally has to convert that 10 MHz frequency to one of the audio frequencies (like 22.5792, for example).  This is generally done using a DDS chip, this also adds more phase noise, so by the time your clock signal gets to the DAC ir has been degraded quite a bit.

It is much better to put a really good clock in the DAC, right at the DAC chip, at the correct frequency.

External reference clocks like the Mutec are designed for recording studio use where often multiple components need to be clocked together to a single reference, in a home audio playback system we do not have multiple components to clock together (OK there are some rare exceptions to this), so it is better to not use an external clock.

But do not take my word for this, ask @marce, or at @JohnSwenson.

I agree totally, been talking about this where I am currently working in sunny Pulborough during the week, doing 35kV+ PSU's. The whole point of a master system clock is to avoid clocks being slightly out of sink where you have multiple clocks (as in a studio etc.) this can cause numerous problems, like having numerous hearts all beating slightly differently, inter-modulation, extra beats etc. even worse when you have a multiple phase array GHz microwave, so in those situations on clock to rule them all, because there are multiple clocks of the same frequency. Where clocks have multiple frequencies as seen in domestic audio, local clocks next to the device are best, each clock can then be optimised for its requirements. I dont like external clocks or clock add ons, the extra cable and signal distance adds more problems than it solves.

[Superdad]

2 hours ago, barrows said:

OK, I get that the measurements you use to develop filters are the necessary and applicable ones too show that the filter design is doing what you want from a technical perspective, and I have no doubt that many of your filters do sound better/different.  I do not have HQPlayer (yet, unfortunately) because I do not have a good machine to run it on.  But I have tried quite a few different approaches, and the differences are usually small, but meaningful in an audiophile sense.

 

Hi Barrows:

I strongly encourage you to try HQ Player and spend some time listening to a wide range of its filter, dither, and modulator options.  

About 3 years ago I spent months with Audirvan+, fine tuning (by ear with an exceptional NOS PCM1704 DAC) a filter with the 5 iZotope Advanced parameter sliders, down to a fine degree.  And while A+ seems to do a better job bypassing OS X Core Audio (comparison made by using both players w/o any SRC), @Miska's poly-sinc family of filters blew right past the best of what I was able to accomplish.

 

For PCM SRC you really don't need much computing horsepower, and even DSD256 is not terribly taxing.  And of course HQP NAA on a Rendu is perfectly matched.

 

Have a great weekend.  Weather is lovely here so I'm heading out on a bike ride now.  Got to huff and puff a bunch for the good of my sedentary heart and ass.  

[barrows]

Hi Alex, yeah, I know, I have to get going with HQPlayer eventually, it will happen-I have a DSC-2 version going here, which is pretty cool, but I still need to add an analog volume control (Muses chip option on the way), then I can listen to it in my set-up vs, the very formidable ESS 9038 DIY DAC I am using now.  I have also played around a lot with A+/Isotope filters, ending up preferring an apodizing, intermediate phase option, which rolled off a bit early, to allow a not too steep filter slope, while still reaching -125 or so at Nyquist.

Do you find the 2 stage filters in HQPlayer to be good enough?  I know the single pass ones require a supercomputer.  I need to be able to do DSD 64 to at least DSD 256, and all PCM rates to DSD 256.  Right now I am using ROON, and am very surprised how good it sounds going to DSD 256, considering the little processing power it uses (it does not break sweat doing this even on my I5 Mini!), but I expect HQPlayer to be significantly better.  My ESS DAC is set up for DSD 256 in terms of all the filter settings and the analog filter, and is running just a single 45.1584 XO in synchronous mode (no ASRC or DPLL active).

 

I expect the DSC-2 to be a bit more able to differentiate between OSF settings, considering it is just a discrete converter with no modulator operating, even with the minimum way I am using the ESS chip it is still running its (SDM) modulator.

 

Good for you getting out pedaling, I have been a bit lax in that area this summer, it is amazing how hard cycling is when your not getting out 4-5 days a week like I usually am.  Oh well, backcountry snowboard season is nearly here and I will be skinning uphill like a maniac soon!

 

Best!