What and how do we know what IS ‘Better’?
or
It’s all in our heads, or is it?
Part 7 ‘Jitter’, it’s not just the coffee anymore.
The last post was a short one, this however is another long one, so we’re diving into the deep end of the pool again.
‘Jitter’
This is not a technical analysis of Jitter plots nor a ‘scientific’ examination of what Jitter is, nor what it isn’t.
Instead I’m going to delve into what ‘Jitter’ is and what I have come to understand its effects and influences are,
AS I HEAR THEM!
Jitter, we all know it’s ‘bad’ to have too much (any?) Jitter in our digital playback system. But is ‘real’ Jitter limited to just the digital domain exclusively?
I submit that ‘Jitter’ is more than just a digital timing ‘error’ during the sequential re-construction process, indeed I find this to be only 1/2 of the ‘total’ picture.
I submit, at least for this ‘Better” series, that a definition that what we call ‘Jitter’ is ANY resultant error (deviation) during the re-construction of (from) the ‘original’ acoustic waveform signal. This definition encompasses both ‘Domains’ (analog & digital) and ‘Dimensions’ (frequency/time & voltage/amplitude) needed to re-construct and deliver the original ‘target’ signal.
I come to this conclusion based upon lots of observation and fussing with the knobs (experimentation).
But I also submit that ‘Jitter’, is in a sense, a displacement ‘error’ away from the perfect re-construction of the original wave form.
Perhaps another term could be introduced such as smear, or re-construction deviation, etc. But for now I’m going to stick to ‘Jitter’, mostly because the audible effects of reducing any of these errors results in near identical audible improvements, regardless of which ‘Dimension’ or which ‘Domain’ is being improved. And it’s these same audible consequences which I have come to identify and define as to what ‘Jitter’ is and what we can expect as any amount of ‘Jitter’ is reduced.
All of these various sub-sets of ‘Jitter’ have a unique variation for how ‘Jitter’ is introduced into the signal we hear.
And really we as audiophools can only affect Jitter reduction in rather limited ways. Sure we can buy new pieces of equipment, but other than that all we can change is the setup of the system, which includes the cables, and the ac power we feed the equipment, and more. And fortunately these sorts of changes can result in significant sonic improvements which have sonic characteristics that match other improvements made to improve Jitter directly, which is why I have come to equate all of these improvements to ‘standard’ digital Jitter.
So Domains, Dimensions?
What?
Let me explain a bit further…
And for the sake of this "Better" series I categorize our audio systems thusly…
DOMAINS
Currently there are 2 ‘domains’ of signal re-creation methodologies in our playback systems,
analog (tape, vinyl, radio)
and digital (CD’s, DVD’s, digital files from a computer or server).
And as indicated above most don’t associate analog with‘Jitter’, since the term was introduced with DACs, but I hear the affects ‘Jitter’ has in both of these types of 'Domains'.
DIMENSIONS
The 2 ‘dimensions’ of (frequency/time) & (voltage/amplitude) (think of an FFT display and a signal trace on an ‘O’scope) are both necessary aspects of any musical signal.
The frequency & voltage pair applies more to the analog ‘Domain’, while the time & amplitude applies more to the digital ‘Domain’. But these aren’t hard and fast associations, by any means.
In the analog ‘Domain’ the signal is based upon a real time complex dynamic analog signal (voltage), with what we usually term as phase shift (frequency) anomalies due to the nature of the circuits and the physics and the properties of the materials used, and of course the setup of the equipment etc.
In the analog ‘Domain’ we have seen a significant increase in the development of power supply sophistication with multiple layers and levels of regulation and control. And this technique is ‘generously’ applied to the digital circuitry used in dacs and other signal processing equipment.
But not exclusively.
And in the digital ‘Domain’ we have been hearing of and seeing developments of femto-second clocks in the push or attempts to obtain the optimal degree of precision, repeatability, and reproducibility in the time axis ‘Dimension’ of the equation.
I have heard the results of the reduction of digital Jitter made to the incoming digital stream and have ‘categorized’ these improvements in a descriptive narrative.
In the digital ‘Domain’, where timing is it’s ‘thing’, we find that as the degree of exact precision of the timing increases, the results, in complex ways, can be quite audible, but again these sonic changes/improvements are very similar to those where the power supplies ability to delivery precisely the correct voltage is also improved.
So these timing/voltage/frequency/amplitude improvements all share similar sonic characteristics when they are improved or where ‘choke points’ that are associated with them are improved.
And these very same audible changes/improvements are also brought about via, albeit more indirectly, ac power system delivery changes/improvements. And some of these changes are from ‘non-ordinary’ sources as most would view them.
IOW as ‘choke points’ that affect both the domains and dimensions are improved/removed, the acoustical net results have very similar sonic consequences.
So next we will delve a bit deeper in the signal re-creation modes AND the signal delivery modes, and look at how they overlap.
Analog signal creation
Is your phono cartridge diamond tip in THE groove? or just sorta close?
And can you tell when it ‘drops into’ THE groove?
When the diamond tip doesn’t read the groove exactly as it was created, time smearing is the result due to not being in phase and ‘correctly’ reading the original cut in the groove in ‘real time’.
A ‘proper’ read of the groove, that matches the line that was cut by the cutting head, can result in a near ‘Jitter’ free analog signal with little to no timing (phase) nor amplitude (voltage) ‘errors’, assuming the rest of the system is up to the task.
This is a tall order, to be able to ‘nail’ reading the one and only correct portion of the groove, in real time, despite the physical variables of LP thickness, or warpage, or lack of ‘flatness’ or eccentricity, etc. of every record. Not to mention the additional variables of cartridge tip to body alignment, as well as ‘ideal’ cantilever ‘sag’ and the aging of the cantilever mounting viscoelastic material, among other factors.
This difficulty, in (mis)reading the groove, translates into timing and amplitude errors with respect to transferring the physical waveform from the groove wall into an identical mirror image electrical signal. Any deviation of this waveform from the ‘original’ is ‘Jitter’ as I have described it above.
Digital signal creation
Traditional Jitter (the strictly digital kind) exists as a frequency or time dimension ‘error’.
(Yes it’s more complicated than this simplified metric, but it will suffice for this write up.)
But I submit that ‘Jitter’ exists in 2 'Dimensions' (time/frequency) & (voltage/amplitude) and extends into a 3rd element, (‘real time’) as an additional factor.
So really, what is the usual understanding of Jitter? and where does it come from?
But first lets take a quick look at digital circuits.
Digital circuits are just optimized analog circuits that get triggered to ‘go off’ (or on) as ideally as possible.
And since they are operated in a sequential fashion, the sequencing itself becomes yet another issue, at least for dac’s anyway.
So they ‘switch’ to the ‘next’ state as quickly, precisely, and repeatably as possible.
But they also need to ’stabilize’ at that ‘new’ state with equal (or better) quickness, precision, and repeatability.
So we need ‘good’ fast, clean, noise free switches (the time domain) and we need a power supply that mirrors with equal quickness, precision, and repeatability as it delivers the exact voltage (the voltage domain) thru the switch(es) which then becomes the analog signal.
Put another way, the timing variability of the master clock as it is used to trigger the switch(es) to ‘dump’ (add) the precise voltage supplied by the power supply distribution system which is used to re-create the exact amplitude at that exact time, is what is being improved to more precisely mirror the original analog signal.
To me this is why we have Jitter in the first place, as it’s a means of describing how to ‘measure’ the time/frequency deviations in this re-creation process which along with the other variable(s) play a critical role in the re-creation process.
And the 3rd element is to perform this re-construction process, without deviation, with this same degree of precision, repeatability and reproducibility, dynamically, in real time, continually.
And then once we have re-created this signal we need to ‘move it along’, and then ‘present’ it to our ears. This is where the wires and other ac power supply systems come into play, along with the rest of the setup of the entire system.
Analog signal delivery
Many are VERY familiar with the possibility and results of changing the sound by changing both the power and signal cables. A great many theories abound and are bandied about, with relish by some.
And in my experience cables can alter the analog signal voltages that pass the musical signal along, and so can the power cables used by the components themselves. And in many cases these changes reflect the same sonic effects heard as does reducing ‘Jitter’, which can increase the overall quality of the sound we hear.
There are many factors such as slew rate, phase shift, signal propagation down the wires (and in some cases return signal reflections), ampacity of the cables and many more widely known factors all of which can have audible effects.
Digital signal delivery
Strangely enough digital cabling does (or can) make an audible difference. And beyond just the crude, ‘does the music skip or hiccup’, digital signal transfer from one device to another can be effected by ones choice of cabling. And I know according to some, cables, especially digital signal cables shouldn’t make ANY differences.
Except that’s not what I’m hearing, but that isn’t the full extent of all of the improvements made from seemingly unlikely or ‘impossible’ sources.
IOW whether it’s improving the timing of the delivery of the digital signal to its destination, or improving the delivery of the ac power pulses to the power supply, or transferring the analog signal from an output stage to an input stage of the next device, reducing the time and amplitude ‘errors’ of these signals results in audible improvements which have remarkably similar net sonic effects.
And in some cases, the mutual beneficial interaction between associated gear can sound like analog coloration shifts, which was quite a surprise when I first encountered it.
What I am hearing in most cases due to these reductions of ‘Jitter’ are those very attributes I have already brought up and described in my previous missives.
Namely …
Small Signal Dynamics (which most are already familiar with)
C3 (Cohesion – Coherence – Coupling),
T3 (Toe Tapping Time),
HB&W (Head Bobbing & Weaving),
Spooky/Scary,
& more.
As my experiments continued, all of these attributes tightened up, came into sharper focus, produced greater resolution at each level and also influenced each other in a holistic fashion (sympathetically), all the while the musical performance continued to become all the more compelling.
The net result was an ever increasing sense of realism, palpability, and reduction in my sense of disbelief (it became harder to nit pick).
Which is nearly the same description I used after I heard the digital Jitter reduced at the DAC.
Only more so.
It seems as though as further improvements are made, as the ‘choke points’ are eliminated, or their severity reduced such that they no longer act as a primary sound quality limiter, the cumulative benefits add up to more than just the sum of the constituent parts.
New intonations and harmonically related aspects of each ‘voice’ ’snap’ into place, and the complete tonality of the ‘voice’ becomes much more evident.
In short there,
IS…
MOAR,
there, there.
It’s like tweaking any system.
As the roughest edge’s are smoothed, the greatest amounts of change/improvements are noted first.
As additional improvements are made, as the remaining ‘choke points’ are further reduced/eliminated, the degree of change may not be as great but the overall level of SQ keeps increasing with greater degrees of resolution and efficiency, all with reduced ‘error’.
This can have unexpected results in terms of hearing additional sonic nuances that have Never been noticed before.
IOW it’s like hearing your music anew, again.
Or put another way, as the total accumulation of ‘error’ continues to be reduced, and more of the original re-created signal is ‘allowed’ to be presented to our ears, the greater the degree of satisfaction is experienced with more of the creativity of the musical intent being heard.
Or think of it like looking thru a telescope, or microscope, or binoculars.
As the view comes more and more into focus ALL of the details become obvious and immediately apparent as to what they are (shadow?, reflection?, etc) what they belong to (what IS the parent object?), how they belong within the view of what is seen (the nature of the relationships), and more precise 3d spatial relationships come into better focus (in front or behind? bigger or smaller, by how much?), how sharp is the ‘outline’ of each object (how precise are its optical limits?)
And more…
And further, as the optics are improved even more, (as the overall system’s resolution increases) and the distortions (error’s) in the field of view (presentation) are reduced or eliminated still further, the changes to and additional available information that result from these improvements, means there is more there, there.
And what there is there, is all the more apparent and obvious.
JJ
End Part 7
Next up,
Getting Calibrated and CNST
so maybe it's not perfect since it's not made in the USA and some will turn their nose up at chi-fi. But I know kingwa takes care of his employees considering it's a Chinese company. I own two audio-gd products. master11 and nfb28(2015 edition) and used to own the dac19 10th aniv.(sweet Dac before the big upgrade to m11) my next Dac may be a Rockna Wavedream Signature (dreaming 
