Basic Question on Jitter on PC audio

Oct 18, 2007 at 5:02 PM Post #46 of 401
USB is the best way of getting audio out of a pc seriously read the specs. It suffers from clock jitter but as the data is packetised it dosen't matter. But the only DAC's that convert USB audio directly are very low end and so the usb audio has to be converted to I2S or simlar before the DAC.
 
Oct 18, 2007 at 5:20 PM Post #47 of 401
Quote:

Originally Posted by Maniac /img/forum/go_quote.gif
If jitter does not matter, then I suppose you won't need a motor with stable 33RPM for vinyl records either, heck jitter matters not, it's bogus, so why should mechanical jitter on record be any different?


That's like comparing an ant hill to Mount Everest. Timing issues at a microscopic fraction of a second are nothing like pitch shifts and wow.

In reasonably good equipment, jitter is a non-issue. The reason it gets talked about so much is because it's good for salesmen to use as an excuse to upsell.

See ya
Steve
 
Oct 18, 2007 at 11:16 PM Post #48 of 401
Quote:

Originally Posted by bigshot /img/forum/go_quote.gif
That's like comparing an ant hill to Mount Everest. Timing issues at a microscopic fraction of a second are nothing like pitch shifts and wow.

In reasonably good equipment, jitter is a non-issue. The reason it gets talked about so much is because it's good for salesmen to use as an excuse to upsell.

See ya
Steve



Well, jitter is always an issue, I suppose you haven't seen jitter does its magic in equipments. When your DAC gets costly enough, you will hear the difference between transports and other digital sources. You will be able to hear the difference between different type of transports, cables and such. A lot of it is in some degree, related to jitter.

Maybe you should take a more technical look into jitter's effect instead of refuting everything you do not quite understand as sales excuses. Granted, a lot of those are sales tricks, some dejitter stuff actually makes things worse (Yes, some popular dejitter technology actually lowers the performance, and instead of reducing jitter, it hides the jitter). The matter of fact is that you do not discredit the entire medical industry because of a few snake oil makers, and same goes with audio industry.

other posters of this thread had already posted several links to some very informative articles, I really would encourage anyone talking about this subject to read it carefully. I'd say it is a treasure trove of information.
smily_headphones1.gif
 
Oct 19, 2007 at 12:34 AM Post #49 of 401
Quote:

Originally Posted by Maniac /img/forum/go_quote.gif
Well, jitter is always an issue, I suppose you haven't seen jitter does its magic in equipments. When your DAC gets costly enough, you will hear the difference between transports and other digital sources. snip Maybe you should take a more technical look into jitter's effect instead of refuting everything you do not quite understand as sales excuses.


I spent the better part of a week schlogging through all those technical papers on jitter on the web. I puzzled out what they were talking about and translated the numbers into the concepts they represented. Jitter is tiny. Not just tiny... inaudibly tiny... in most reasonably well designed equipment. I understand perfectly what jitter is. It's something to worry about instead of the things that really matter.

As for "jitter working its magics in equipments" and "when your DAC gets costly enough, you will hear the difference"... I also understand well enough to know that in most electronics, cost is not the determining factor to quality; and there is no magic- just practical application of principles.

You don't solve problems by throwing money at them and hoping for magical solutions. You do it by thinking things out and realizing the limitations of your perception. When you have an understanding of the scale of things, you'll know what matters and what doesn't. Jitter on the grand scale of things is near the bottom of the list.

See ya
Steve
 
Oct 19, 2007 at 4:07 AM Post #50 of 401
Quote:

Originally Posted by bigshot /img/forum/go_quote.gif
I spent the better part of a week schlogging through all those technical papers on jitter on the web. I puzzled out what they were talking about and translated the numbers into the concepts they represented. Jitter is tiny. Not just tiny... inaudibly tiny... in most reasonably well designed equipment. I understand perfectly what jitter is. It's something to worry about instead of the things that really matter.

As for "jitter working its magics in equipments" and "when your DAC gets costly enough, you will hear the difference"... I also understand well enough to know that in most electronics, cost is not the determining factor to quality; and there is no magic- just practical application of principles.

You don't solve problems by throwing money at them and hoping for magical solutions. You do it by thinking things out and realizing the limitations of your perception. When you have an understanding of the scale of things, you'll know what matters and what doesn't. Jitter on the grand scale of things is near the bottom of the list.

See ya
Steve



Actually, I have to say that jitter does indeed matter. I've spent about 1.5 years working on TV chips and I've found that simply by going from a SMT to a DIP socketted oscillator, the jitter makes enough of a difference for it to be visible on the TV after the data is put through the TV DAC. Same with feeding that SMT clock into a PLL. The difference is actually visible on the TV, and I am not talking about subtle differences. On a solid color, the color actually shifts at random places in the image, the noise increases by about 20dBs as well. This was measured on a Tek VM700T.

Sure, the TV is a faster signal, but if a 1~2% increase in peak to peak jitter can make a difference that is THAT visible. It will matter for a slower audio DAC as well.

-Addition-
Also, throwing money at the problem no doubt will not fix the problem most of the time. You have to use the money where it counts.
I've seen the difference between a Toshiba fabbed TV chip and a chip fabbed by another manufacturer. The DAC is the same in both.
The difference in the analog process of Toshiba over the other manufacturer is clear in terms of reduction of noise. This improvement comes from millions of dollars of R&D. This cost will no doubt get passed onto the consumer.
 
Oct 19, 2007 at 5:15 AM Post #51 of 401
I can't speak for other peoples' perceptions, but for me, jitter is a non-issue. I've never heard it, and I doubt I ever will.

Too me it's like infrared, yes it exists, yes we can observe it with special equipment, but otherwise I would be totally oblivious to the fact.
 
Oct 19, 2007 at 6:21 AM Post #52 of 401
Reasons to worry about jitter later:

1) you have electrolytic capacitors in the signal path of the input and or output stage of one or more pieces of your equipment rather than films.

2) you have under-rated electrolytics on one or more pieces of your equipment, including, but not limited to, your PC's Power Supply, your Motherboard, your Audio Board, your Playstation.

3) you haven't invested in power conditioning, yet. Instead you run all your equipment off a power-strip you bought for $10. All equipment plugged into this strip is done so in an order that introduces horrible ground-loops.
 
Oct 19, 2007 at 7:34 AM Post #53 of 401
... I don't buy this jitter stuff. I've heard lots of differences between sources, amps and stuff, but jitter is just this magical attribute that some people use to explain differences.

So what does jitter sound like? How much jitter is needed before it's audible? Why some people claim that even a benchmark dac1 is susceptible to jitter? To sell their snake oil?
 
Oct 19, 2007 at 11:28 AM Post #54 of 401
Quote:

Originally Posted by Logistics /img/forum/go_quote.gif
Reasons to worry about jitter later:

1) you have electrolytic capacitors in the signal path of the input and or output stage of one or more pieces of your equipment rather than films.

2) you have under-rated electrolytics on one or more pieces of your equipment, including, but not limited to, your PC's Power Supply, your Motherboard, your Audio Board, your Playstation.

3) you haven't invested in power conditioning, yet. Instead you run all your equipment off a power-strip you bought for $10. All equipment plugged into this strip is done so in an order that introduces horrible ground-loops.



Actually, none of the stuff you said have much influence in terms of jitter. Not to mention jitter can only be dealt with when it is still original digital data, once it's gone through D/A or some other calculations, it is messed.

And no one I know ever tried to put a electrolytic cap in digital signal path, they use isolator chips, film/ceramic/mica caps or pulse transformers.

Like redwire said above, just changing clocks into different packaging can have a measurable and visible effect. Careful signal path planning and parts selection is just one of the few things required to deal with jitter properly.
 
Oct 19, 2007 at 11:35 AM Post #55 of 401
Quote:

Originally Posted by LeChuck /img/forum/go_quote.gif
I can't speak for other peoples' perceptions, but for me, jitter is a non-issue. I've never heard it, and I doubt I ever will.

Too me it's like infrared, yes it exists, yes we can observe it with special equipment, but otherwise I would be totally oblivious to the fact.



Jitter is not exactly something you hear, it is not a tone, it is an influence, not an effect. There are many ways to induce jitter, having something that is not quite able to produce clean square wave at 2.8MHz (CD's 44.1KHz sample rate needs this speed) would be one way to induce jitter in the circuit that receives it.

Anyone ever wonders why Chord DAC64 and Rega's new line of CD player gone through so much trouble just to deal with jitter?
 
Oct 19, 2007 at 11:40 AM Post #56 of 401
Quote:

Originally Posted by maarek99 /img/forum/go_quote.gif
... I don't buy this jitter stuff. I've heard lots of differences between sources, amps and stuff, but jitter is just this magical attribute that some people use to explain differences.

So what does jitter sound like? How much jitter is needed before it's audible? Why some people claim that even a benchmark dac1 is susceptible to jitter? To sell their snake oil?



Well, benchmark DAC1 uses Analog's ASRC chip (Benchmark calls it ultra lock), to see how that chip work and what exactly it does, I suggest a read of its datasheet and AD's wealth of app note would be very enlightening (Well, more so than brochures that speaks only marketing slogans). I did that a while back and I can feel that they are not just selling their chips but also trying to pass on the knowledge and concepts that they know. Very cool stuff.
 
Oct 19, 2007 at 2:34 PM Post #57 of 401
I think you hit the nail on the head. People get tripped up in this jitter stuff because the prevailing advice is "source first". Why not get a decent, well-regarded source, then headphones, amp, and cables, then power supply. Then go for the $1K+ source.

Quote:

Originally Posted by Logistics /img/forum/go_quote.gif
Reasons to worry about jitter later:

1) you have electrolytic capacitors in the signal path of the input and or output stage of one or more pieces of your equipment rather than films.

2) you have under-rated electrolytics on one or more pieces of your equipment, including, but not limited to, your PC's Power Supply, your Motherboard, your Audio Board, your Playstation.

3) you haven't invested in power conditioning, yet. Instead you run all your equipment off a power-strip you bought for $10. All equipment plugged into this strip is done so in an order that introduces horrible ground-loops.



 
Oct 19, 2007 at 3:00 PM Post #58 of 401
Quote:

Originally Posted by bigshot /img/forum/go_quote.gif
That's like comparing an ant hill to Mount Everest. Timing issues at a microscopic fraction of a second are nothing like pitch shifts and wow.

In reasonably good equipment, jitter is a non-issue. The reason it gets talked about so much is because it's good for salesmen to use as an excuse to upsell.

See ya
Steve



actually in a very good system, jitter is quite the issue. Ive heard the emm lab DAC clocked and unclocked driving solid state amps that fed marten coltrane speakers. The clocked version (less jitter) definitely has tighter bass, more realism from the instruments where i can start to make out more "texture", start point out some incoherent voices during group vocal. Its not noticeable instantly but does take some time to realize this. After that you just find the clocked version providing that slightly more resolution and realism on very good recordings - as if the air molecules have been given slightly more room to interact with many other air molecules.

On a tube amp, i seriously wont bother. I couldnt tell a heck of a difference between clocked vs unclocked no matter how hard i tried.
 
Oct 19, 2007 at 5:29 PM Post #59 of 401
Quote:

Originally Posted by Konig /img/forum/go_quote.gif
slightly more resolution and realism on very good recordings - as if the air molecules have been given slightly more room to interact with many other air molecules.


That is a great analogy, because if you take the time to figure out what all those numbers mean on the sales pitch "informational technical literature" posted on equipment sales sites, you realize that the fractions of time we are talking about are so minute, hearing jitter would be like seeing a molecule.

See ya
Steve
 
Oct 19, 2007 at 5:42 PM Post #60 of 401
The jitter issue or discussions like this about it are like so many polarized issues on head-fi and elsewhere that are volleys back and forth of exagerration from one pole and then the other.

It's pretty clear if you ignore this that the opposing groups in this thread are talking about different levels of equipment. For most people using equipment at all but the higher, pricier end of equipment jitter is a minor issue. For these people/systems it is either inaudible or of much lesser impact than many other aspects affecting playback. Why bigshot and others feel compelled to insist it has none at all no matter what else is the case escapes me. The practical implication and advice he gives is correct and useful without reliance on that impossible absolute claim--at most levels.

From my experience moving up thru a few levels of equipment I have come to a level at which the dac, amp and headphones are good enough that one can hear quite readily the improvement that comes with significant jitter reduction. It's not night and day surely, maybe more like a half-hour before sunrise and an hour after or late afternoon and dusk. Before I had gotten equipment at this level, it was much as bigshot says. Jitter was little perceived and surely superceded by other ways to improve the playback more and for less.

Why is it any less ridiculous to be night and day about this than other aspects?

BTW, Konig was using a metaphor not an analogy. It is notoriously difficult to put in words subtle aspects of what one hears. Also, you might consider what is known as Brownian movement, which a visible affect of movement at the molecular level. That might be an analogy to what happens with increased jitter. I don't know.
 

Users who are viewing this thread

Back
Top