Jitter just where do we hear it in our music


#1

As I have two products that improve audio over USB with dacs. I have recently had a little more time then usual to play with toys I own .

I did the following tests. Just how much and what parts does jitter effect our music. I will say what I hear but I would very much like to know others opinions as well. One of the dacs I own is in a hdvd800 by sennheiser . It’s a combo dac headphone amp. The amp is very good with the hd800. But the dac gets ripped apart in reviews at least the honest ones. So I played one of my reference tracks to try and here just what changes. Well the change is jaw dropping to say the least. But I noticed it’s not so much with vocals it’s more with horns and most defiantly cymbals . Big time with them. It also effects the base details but also the amount. One of my songs has the cymbals with the rivets they tend to give them more air around them. Well without the AP1/PP they sounded terrible . Like they were just hissing. When I added the unit the level of clarity was amazing . But as it improved the entire sound of the song it clearly effected some more than others. Does anyone here have any thoughts on this .

Al


#2

I too have been exploring the effects of jitter as it is reduced by various means.

And while I don’t have any of the equipment you list, the effects of improving the ‘resolution’ from the PWD and my entire system is remarkably consistent regardless of how the jitter is reduced.



What I do hear as jitter is reduced, is there is a tighter association between the ‘sound’ of each ‘voice’ and the acoustic energy that is delivered to my ears that makes up all of the ‘voice’.



IOW the delivered spectral energy density of the harmonic and related frequencies are more closely tied to themselves and to their ‘parent’, such that there is more inner detail, more insight into the whole of what each ‘voice’ truly is. And the sound is more compelling, as in it ‘sucks you in’, since there is more there, there.



And it’s not just the increase in inner detail, but the increase in the coupling of the acoustic energy, in both frequency and time, which is also very compelling. So much so that the head starts bobbing and weaving, the feet start tapping, all to the beat. And where the acoustic energy is properly coupled, the bass and leading edge impact of each ‘voice’ makes an audible and obvious improvement to the ‘believability’ of each ‘voice’.



Such aspects as, intelligibility, PRaT, slam, 3d spatiality which surrounds each voice, the ability to hear the acoustic space that is defined by and surrounds the entire song, being able to concentrate on just one ‘voice’ and track with it despite what the rest of the ‘voices’ are doing, all of these and more are improved with reductions in jitter.



And sometimes these reductions are the result of seemingly ‘unrelated’ changes made to other portions of the system. Which seems to reinforce not only the weakest link approach, but the scaleability of various components that make up the system.



But that other changes also make the same or similar improvements to these very same audible aspects, tends to tie them together, at least in my mind. Namely that as the internal resolution of the system improves, as the acoustic presentation ‘tightens up’ and is more cohesive and coherent to itself, there is more there, there.



JJ


#3

Yep. When we got things together enough for Paul to really listen to the current boards in his system he said “There’s a whole lot more than just jitter reduction going on here!” He was right and he was wrong. No matter how good the rest of your system/DAC is, going from more jitter to less has the effects listed above. Now they can be masked by any number of other system problems, but when everything is leaving the music alone there’s something special that happens. I think I’m doing pretty good on lots of fronts in the DirectStream, but lowering jitter is the thing I think that gives us back the magic in music.


#4

Good morning and thanks for the responses. I did this same post in several forums to see reactions and ideas. And although I have no instruments to measure with other than my ears and reference music I know very well. The conclusion I came too is it’s not just jitter but a combination of other types of electrical noise and jitter.

Even though I was using a surface pro on battery and SS drive inside , directly attached to the USB input using a audio quest diamond USB cable . The change was still dramatic when I used the AP1/PP . Keep in mind this is redbook 16/44 PCM .

Now some posters say it is more apparent in higher resolution or dsd . And although I agree to a point

, it does prove the fact there is just more there period to . This makes me feel that the overall improvements

That ted and company has made to the new direct stream DAC must be amazing and something I await to heart very egerly .

Ps update

I did the same test using a P5 as well and although the changes with it were obvious As well.

Thanks



Al


#5
Yep.
snip
No matter how good the rest of your system/DAC is, going from more jitter to less has the effects listed above. Now they can be masked by any number of other system problems, but when everything is leaving the music alone there's something special that happens. I think I'm doing pretty good on lots of fronts in the DirectStream, but lowering jitter is the thing I think that gives us back the magic in music.


I most emphatically agree about getting the magic back into the music due to jitter reduction!

The old adage of, as more and more of the easy to improve aspects are fixed, the costs to go to the next level start climbing, like crazy. The saying 'sorry about your wallet' reflects this quite well.

But sometimes the simplest of improvements, can break thru, in unsuspected ways, a huge log jam.

And as the overall level of the systems degree of precision goes up, sometimes those simple improvements can have a much bigger impact than it would seem they should have.
Of course it is the accumulation of ALL of the previous improvements already made that allows for the full expression of the net effect of even the simplest of changes to happen in the first place.

IOW as fewer and fewer choke points or limiting aspects remain, the amount that any single change can make can be more and more noticeable.
In my mind this is where scaleability plays a huge role…

JJ

#6

I really like your first post in this thread JJ. I couldn’t have written it better myself.


#7

Al and JJ, these are great descriptions of what the DirectStream DAC sounds like in comparison to our old friend, the PWD. I couldn’t have said it better myself (no, really, I couldn’t/didn’t). Ted clearly agrees. :slight_smile:


#8

I can speak for myself and you can write better than me. LOL.

And I am glad I pointed something out that is useful.

But keep in mind I have said I use the pwd mkii only with my offramp 5

It greatly improves the sound quality. This is one of the reasons I bought the new unit without hearing it. I figured as the new unit does PCM and DSD alone and an improvement over the pwd mkii. It’s a win win for me.

And with the offramp 5 I can do DSD over dop using the I2S input of the new unit

So when am I getting the new one man

Al


#9

Since I’ve been fussing with the knobs of my system for a while now, the net effects of the changes that make a difference seem to be similar, or at least in the same ball park…



My latest experiment was with the WA-Q chips on all of the fuses. And as I mentioned elsewhere these chips resulted in very similar audible improvements which sound like the strictly digital jitter experiments and improvements I made previously.



Which, in my mind anyway, expanded my understanding of just what jitter truly is and brought it into a much clearer focus.



And what follows is mostly preaching to the choir, so please bare with me, but some may find useful bits here and there…



So the tricky part is determining just where (or what) is the ‘weakest link’ in terms of sound degradation, or as I call it a ‘choke point’ in any particular system. Sometimes these choke points can be so limiting that enhancements made elsewhere can’t be ‘fully’ appreciated. And in other circumstances the net effect of a change can be wholly detrimental.



Which can make for some mighty vigorous head scratching which can be quickly followed by an early membership into the ‘hair club for men’. ~X( :-w



This is where the scaleability of each component looms large in the ‘upgradeability’ of any system, and this can (and does) drill down to subsystem components as well. And sometimes it can be the smallest of things, the ones that are all to easily overlooked, that become THE Choke Point for the entire system. This can lead to ‘exaggerated’ claims for a particular component that aren’t repeated, to the same degree, by others.



By way of example…

A friend and I visit a 3rd audiophool and he’s just finishing the setup of a secondary system. Well finishing is a bit ahead of where he really was, but we sit down to listen and it sounds like a dull, lifeless, sonic presentation on about the level of a boombox, but without the boom. Think emasculated AM radio… :-q



Clearly THE Choke Point is in full play, it sounded like crap.

I look over at the turntable and notice that the arm is WAY low, I mean I can see it from 5 feet away, low.

So I sez raise the arm height, a little at a time and lets see what happens.

It all starts to get better, to come into focus.

But then it starts getting all phasey and the sound field is everywhere but where its supposed to be.

Clearly the speakers are out of phase.

And here is the point.

The signal being created by the low arm height was so bad we couldn’t hear the out of phase speakers.

At all.

Then when we continued to raise the arm the sound became glorious.

And then we could pay attention to the absolute polarity, because now we could hear which polarity setting was correct.

Of course it still needed a bunch more adjustments to be sure, but now it was toe tapping time.



My point is a choke point may be obscuring other choke points, but you won’t know that until you find the king pin, the worst of the lot and ‘fix’ it. Then the others will become far easier to find and ‘fix’ as well. Mostly because you’ll be able to hear if the changes made in fact actually did have a effect or not, AND if the effects are a keeper, or not.



And jitter is a much more subtle phenomenon than the gross example above.

It may well be among the last set of choke points to ‘fix’.

Mostly because most of the really ‘bad’ or gross choke points have already been ‘fixed’ just in order to hear the ‘subtle’ effects of reduced jitter.

And while sometimes the choke points can be seemingly insignificant (eg. WA-Q chips), in other cases it can be BIG $$$$$s.



But what is most gratifying is when a major choke point is ‘fixed’ and the net change is HUGE.

That’s when the saying ‘I get to hear my music collection as if it were all new’, takes on real meaning. :thumb



JJ


#10
johnjen said: Mostly because most of the really 'bad' or gross choke points have already been 'fixed' just in order to hear the 'subtle' effects of reduced jitter.

I also strongly suspect that when we concentrate on fixing one specific problem - such as jitter- we often inadvertently change something else. This change can be good or bad.

When reading Ted's posts on the development of the DirectStream, it has repeatedly occurred to me that his combination of focusing on specific goals (jitter reduction, phase coherency, etc.) while maintaining a global wide-eyed view of the impact on the overall sound. His focus seemed to continually change from forest to trees to forest.

I have carefully listened to files with varying amounts of jitter. I find it easiest to pick up when listening to piano (just like wow and flutter, excess sibilance, etc.). It takes greater amounts of jitter to be heard than one would expect from reading audio magazines. My experience is consistent with the rigorous studies on the topic.

My bet is thus in focusing on removing jitter, a designer is accomplishing both this and more. It is the overall beneficial improvement we hear.

#11

To me this is where things start to get ‘hinky’.



When we attempt to quantify a qualitative experience.

And I’m NOT saying we shouldn’t try…



But it seems that until the parameters ARE figured out (when the previous attempts/experiments yield truly meaningful results) then we can successfully apply measurements and related techniques to get repeatable and reliable results.



And again we should push the envelope(s) and learn what does and doesn’t have an influence and in which ways etc., to establish what measurements DO have an impact and by how much etc.



But for the subtle stuff like increases in inner details, or resolution, or the ‘air’ surrounding each ‘voice’, all these parameters, and more, are the most cherished and welcome improvements.

And these are the ‘hardest’ audible aspects to quantify.

I tend to lump them together as the cohesion & coherence & coupling of the reconstructed acoustic energy which is presented to us.



And yes I too see ‘jitter’ is a key aspect or term used to describe where ‘errors’ or inaccuracies are introduced into the reconstruction and/or processing of the original audio signal.

But I also see jitter as existing in the analog realm too.

Or at least I’m hearing the same or similar results with improvements made in the analog realm.

So to me it isn’t just about digits and bits, but the optimized realignment of both the voltage amplitude and time as our cherished signal passes thru the entire system and is delivered to our auditory senses.



By way of example, when I replaced the fuses in my DAC and amp, all of a sudden there was WAY more there, there. Just a fuse let me hear bass that wasn’t audible before, let me hear the air and inner details, etc. (see the lists above) all the more.

And as I later learned, the amp alone accounted for at least 50% of this increase. Probably more like 60-70% of the bass increase was due to the new fuse for the amp during this portion of the ‘experiment’.



This very same list of attributes of improvements ascribed to digital jitter, also applied to the analog power amp, with the new fuse.

IOW the amps ability to resolve and pass more of the signal unhindered was improved by the reduction of that choke point’s limitation (via the replacement fuse).



This was further investigated and confirmed by subsequent experiments, namely that the amp’s performance was improved by reducing the ‘jitter’ just on the ac input into the amp.



This would tend to suggest that a broad definition of what ‘jitter’ truly is would be in order, or at the very least determining what digital ‘jitter’ and analog ‘jitter’ have in common besides similar net results.

I suspect it maybe as simple as, ‘the proper voltage at the proper time’, regardless of what portion of the system is under examination.



JJ


#12
I really like your first post in this thread JJ. I couldn't have written it better myself.

Thanks! :thumb

JJ

#13
Al and JJ, these are great descriptions of what the DirectStream DAC sounds like in comparison to our old friend, the PWD. I couldn't have said it better myself (no, really, I couldn't/didn't). Ted clearly agrees. :)

I suspect as much in terms of the gains we can expect from the DS DAC. But it is my imagination that is being employed instead of actual experience, which I can all to easily vouch for being fooled by this approach before… :-?

But the real 'test' will be when we start getting reports from the field, as this DAC gets used in a wider range of systems.
Idiosyncrasies, foibles, hiccups etc. are to be expected to some degree as is the case with any new design as it travels thru the development and into the production cycle.
I do think that they will be minimal this time around what with the expectations and time spent thus far.

And I was especially encouraged as Ted described his development and test 'environment' and that it didn't seem to make any difference to the audible output.
To me this lends huge credence to the DS DAC's scaleability and lack of susceptibility to less than ideal signal input conditions.
These 'insights' along with much else he has stated, directly points out we may well be peeling back and tossing aside major layer(s) of 'inaccuracies' with this DS DAC.

Exciting times… :thumb

JJ

#14
johnjen said: This would tend to suggest that a broad definition of what 'jitter' truly is would be in order . . .

I understand what you mean, and agree there are many things going on, but I think I would leave the definition of jitter alone. :)

I believe improvements are currently being realized in electronics as inconsistencies of all types are minimized - phase relationships remain intact, amplitude changes are linear, ground plane differences minimized, jitter eliminated.

All of these things can be subjected to "jitter," unintended deviations from an ideal.

johnjen said: To me this lends huge credence to the DS DAC's scaleability . . .

Excellent point. I had not thought of it this way.


#15

@elk nicely summed up DS technical improvements, thanks :slight_smile: