Why do we care about jitter?

Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
Jump to Latest
T

thoriated

Member
Joined 2003
Exactly what amount of jitter do people claim is audible? I just have trouble understanding why this is a problem that would cost more than $20 to fix.
Click to expand...

Which is more than 95% of mass market manufacturers are willing to spend to fix it.

I presume you are already aware that jitter over about 40ps is enough to compromise Redbook full scale linearity at 20khz with the effects becoming progressively more pronounced the more jitter there is (the exact sonic effect depending also upon jitter spectrum, of course). Jitter that is semi-correlated with the musical signal produces some of the most objectionable effects.

Audible jitter can easily also be introduced into a recording where there is no amelioration possible in a playback system.
 
jan.didden

jan.didden

AX tech editor
Joined 2002
linearaudio.net
Paid Member
Guido Tent said:


mmm, not exactly. Practice shows that crosstalk (on chip, power supply) jitter still ripples through the reclocker, but In theory Jan is right.

In addition, one needs to mention that the clock used for reclocking is crucial for final performance.

cheers
Click to expand...

Guido,

I think we agree here. Very often power supply issues 'ripple down' (pun intended) to the DAC and clocking stuff and wreck the jitter performance, you know that better then most here. But that's another kettle of fish, it reinforces the need to improve that final clocking including its supply, grounding, RF immunity etc.

Jan Didden
 
jan.didden

jan.didden

AX tech editor
Joined 2002
linearaudio.net
Paid Member
thoriated said:
In theory, digital is 'perfect':)
Click to expand...


Parts of it are. CD transports and their copying mechanisms, their error correction mechanisms are for all purposes perfect. Try to copy and install some PC program from your CD reader. Make copies of copies. Overwhelming chances are that everytime your software runs perfectly. Your copies and reads are 'bit perfect'. The final jitter performance (which is important for audio) then is determined at the end of the chain.

Jan Didden
 
K

KBK

Member
Joined 2003
yes, it does seem odd that we can 'hear' picoseconds of jitter, but if you analyse the way the ar works and wha those picoseconds of jitter do to the audio signal, it should become more apparent as to why this is. We haar by transinet timing and level. The ear-as a diode. So all of our hearing is tuned to the transient peaks level and placement -in tiime, and their inter-relations. Thre will, of corse, hopefully be a point at which jitter ceases to be an issue, but at this time, it is still not. but we re getting close. It's the jitter spectrum, is my understanding.

I can show you a place where other types of similar jitter levels are visually seen as well. I could prove it to you, right in front of your eyes. But that is another kettle of fish.
 
jan.didden

jan.didden

AX tech editor
Joined 2002
linearaudio.net
Paid Member
KBK said:
yes, it does seem odd that we can 'hear' picoseconds of jitter, but if you analyse the way the ar works and wha those picoseconds of jitter do to the audio signal, it should become more apparent as to why this is. We haar by transinet timing and level. The ear-as a diode. So all of our hearing is tuned to the transient peaks level and placement -in tiime, and their inter-relations. Thre will, of corse, hopefully be a point at which jitter ceases to be an issue, but at this time, it is still not. but we re getting close. It's the jitter spectrum, is my understanding.

I can show you a place where other types of similar jitter levels are visually seen as well. I could prove it to you, right in front of your eyes. But that is another kettle of fish.
Click to expand...

I see or read nothing in this that gives me any clue as to the audibility of a certain level of jitter. Sorry.

Jan Didden
 
preiter

preiter

Member
Joined 2006
Well, at least one mystery is solved for me. I can well believe jitter levels in the hundreds of picoseconds in audio equipment.

I can't quite accept that such levels of jitter are audible. But if people do think that they need their jitter down in the range of 10s of picoseconds, I can well believe that it would cost thousands to achieve that.
 
darkfenriz

darkfenriz

Member
Joined 2004
Re: Re: Re: Re: Re: Re: Re: Why do we care about jitter?

preiter said:

Perhaps I am missing the definition of bandwidth as it applies to a PLL. Could you explain what you mean by it and how it affects jitter?
Click to expand...


Phase comparator's output feeds a VCO through a loop filter (pole or pole-zero). -3dB of filter is PLL bandwidth.

I believe that answer to original question is more philosophical.
SPDIF is very primitive communication interface, no FEC, no impedance matching, no nothing and plastic waveguides. What is 2Mbps compared to synchronous optical networks of today? Who cares about jitter? Who cares about medium?
Cummulation of turning blind eyes made jitter audible.

Just my opinion... might be wrong

regards
Adam
 
preiter

preiter

Member
Joined 2006
BlackCatSound said:
Putting to one side the issue of jitter, what about the fact that most (if not all) DACs are not amplitude perfect?

If you sweep a DAC from -FSD to +FSD you won't get 2^x perfectly spaced steps.

The major downside to the multi-bit DAC.
Click to expand...

That's why they invented single bit DACs, and dacs > 16 bit (so you can use just the more linear upper bits), and stacking multiple DACs to average the response.

But that's an issue I can understand and accept
:cool:
 
Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
Top Bottom