My current DAC in my Amp/DAC combo upsamples to 192Khz and cleans up jitter as described here:
 

The CS4398 is the best performing DAC chip in Cirrus' line-up with numbers like 120dB dynamic range and -107dB THD+Noise, and the Analog Devices AD1896 Asynchronous Sample Rate Converter up-samples all incoming digital data to 24-bit, 196kHz data with an interpolation algorithm using its own ultra-stable clock to insure a jitter-free output data stream.
 

Click to expand...

 
 
If I were to now put in front of it a USB to S/PDIF converter with it's own upsample and dejitter routines, would they synergise or would the effect of one nullify the other.  
 
Could the DAC stifle the potential dejittering benefits of the USB to S/PDIF converter?

I don't think so Aim.  My dac wasn't cheap, and is also meant to have the dejittering thing going.  But if that is the case - why do I feel there are small differences between transports going into the DAC.  I never felt any difference in the past but recently with a new headphone purchase, this particular headphone tells me jitter counts.  I have never spent a cent on digital cables or anything like that - overall I leaned on the bits-are-bits camp...but recently I cannot say anymore.
 
The dejittering USB convertors would make the job of the dejittering Dac a lot easier - meaning even further reduction in jitter - that's what I am lead to believe.  Yeah they'd synergise for a super duper dejitter job - more than one alone can manage - many hands make light work etc.

The reason I ask is because I am not hearing a difference between my Mac's optical out and my USB to S/PDIF converter that comes showered with praise.  I'm told that I'm failing to appreciate a difference because my upsampling DAC is making all my transports sound the same.

Quote:

aimlink said:

The reason I ask is because I am not hearing a difference between my Mac's optical out and my USB to S/PDIF converter that comes showered with praise.  I'm told that I'm failing to appreciate a difference because my upsampling DAC is making all my transports sound the same.

Click to expand...

Before I heard the LCD2s I would have said that combo dac amps are good enough and any more expenditure will see very little return in investment.  I had Bryston Dac, Dacmagic, Pioneer SACD for source - with 5 head amps at different performance levels.
 
After experimenting with the LCD2s - my mind has changed dramatically.  The differences between sources and amps are much greater than I'd imagined.  I do believe HD800s are as revealing of downstream components as LCD2s - maybe even more so.  My latest experiences are pointing the culprit at your amp/dac combo - I don't think it's revealing enough to expose jitter differences.  This is my genuine opinion, and if you feel you heard differences with the ALO cable and your K702s - I have no problems suggesting further upgrades downstream of your HD800s.

 
Quote:

aimlink said:

The reason I ask is because I am not hearing a difference between my Mac's optical out and my USB to S/PDIF converter that comes showered with praise.  I'm told that I'm failing to appreciate a difference because my upsampling DAC is making all my transports sound the same.

Click to expand...

Jitter or lack thereof is one of those topics where there really is little solid evidence that jitter below absurd amounts really makes a difference. There are many anecdotal accounts but in terms of solid experiments with controlled listeing conditions there is almost zero, the empirical evidence that does exist from reasonable research (Benjamin and Gannon, Ashihara et al.)  places the thresholds of audibility for jitter rather higher than most of the audio industry would like us to believe. There are numerous models of jitter audibility based on theory and higher mathematics but these have never (to date) been supported by solid empirical research.

Quote:

nick_charles said:

Jitter or lack thereof is one of those topics where there really is little solid evidence that jitter below absurd amounts really makes a difference. There are many anecdotal accounts but in terms of solid experiments with controlled listeing conditions there is almost zero, the empirical evidence that does exist from reasonable research (Benjamin and Gannon, Ashihara et al.)  places the thresholds of audibility for jitter rather higher than most of the audio industry would like us to believe. There are numerous models of jitter audibility based on theory and higher mathematics but these have never (to date) been supported by solid empirical research.

Click to expand...

Thanks Nick.  I had seen you mention this in another thread.  However, what of two dejitter routines interfering with each other?  Could my DAC's upsampling and dejitter routine be making the signal it's processing end up with more jitter prior to the D/A conversion process?  This seems possible, though far-fetched to me.  I guess you didn't entertain this part of the question since if this is all below the audible threshold, then in a pragmatic sense, who cares? 

Quote:

aimlink said:

Thanks Nick.  I had seen you mention this in another thread.  However, what of two dejitter routines interfering with each other?  Could my DAC's upsampling and dejitter routine be making the signal it's processing end up with more jitter prior to the D/A conversion process?  This seems possible, though far-fetched to me.  I guess you didn't entertain this part of the question since if this is all below the audible threshold, then in a pragmatic sense, who cares? 

Click to expand...

Problem is how can you measure the jttter without some serious and expensive kit. Jitter will show up also in downstream distortion such as IMD and THD but again casualy measuring it is not easy. Steve Nugent does have the kit to measure jittter but he does not visit this subforum much. You can crudely measure noise or even FR differences by recording the digital or analog streams from two set-ups but jitter tends to normally operate at the below -90db or -100db level of distortion which can easily get lost in random noise from non pro recording kit. Really bad jitter will show up but by really bad I am talking about -60db i.e several 10s of ns...
 
Of course if you are just interested in if the differences are audible a good old DBT will do for that...

Upsampling to remove jitter will only benefit you once. So if the S/PDIF converter already upsamples the digital signal to 192Khz before handing it to the DAC, then the DAC will just use that without changing it further.
 
If all transports sound the same in your particular setup, my interpretation is that there was no jitter problem in that setup to begin with. In other words, your computer's native S/PDIF was already more or less jitter free and/or your DAC's ASRC function already removes what little jitter there is. Once jitter is completely removed, no amount of further jitter reduction will yield any improvement.

Jitter reduction through ASRC is nonsense.
 
ASRC's use is for variable-speed CD Players (for DJ's). Another use is to mix several digital sources that are not slaved to the same master clock in the digital domain.
 
At a given time, it seems that audiophile sort of became convinced that it could remove jitter. In some conditions, it may indeed, but no more than any conventional DAC. Worse : once the jitter removal is done, the residual jitter is forever fixed into the upsampled digital stream.
Send this stream into an S/Pdif output. At the other end, you get the usual jitter caused by the cable (that is the main source of jitter), plus the residual jitter in the upsampled data.
 
Perform this operation twice, and the result is even worse : at the output, you get the jitter of your DAC clock, as in any ordinary DAC, plus two tiny bits of residual jitter coming from the two ASRC operations.
 
In practice, these effects are negligible, but I don't see the necessity to tamper with the data. If one of the two devices is not designed with perfect care, you may have unneccesary dither, possibly badly noise shaped, or not at all, or even aliasing. Especially since asynchronous resampling is much more prone to aliasing than regular resampling.
 
The safe side is not to de-jitter at all. The only improvement would be in the case of a broken DAC (for it to be sensitive to jitter) in an integrated player (for the de-jittered signal not to be re-jittered by the S/Pdif link).

Interesting stuff guys.... thanks.
 
We may be seeing the birth of a new generation of equipment with questionable benefits.  Maybe cables will get a rest then since there's a new victim to beat.

reclocking skeptics should try a WM8804 board: http://www.twistedpearaudio.com/digital/wm8804.aspx
 
whatever you feed it, it'll reclock it to 50ps, there's plenty of white papers there: http://www.wolfsonmicro.com/products/spdif_transceivers/WM8804/
 
I've tried it on a glass toslink input(said to be +800ps jitter) and coax output w/ very transparent phones and a DIR9001(50ps clock recovery) DAC, the difference is shocking.

Quote:

leeperry said:

reclocking skeptics should try a WM8804 board: http://www.twistedpearaudio.com/digital/wm8804.aspx
 
whatever you feed it, it'll reclock it to 50ps, there's plenty of white papers there: http://www.wolfsonmicro.com/products/spdif_transceivers/WM8804/
 
I've tried it on a glass toslink input(said to be +800ps jitter) and coax output w/ very transparent phones and a DIR9001(50ps clock recovery) DAC, the difference is shocking.

Click to expand...

 
That you can lower jitter is not a big controversy. That the difference is in fact audible under controlled conditions is more open to debate. You do know what I am going to ask next right
 

 

nick_charles said:

 
That you can lower jitter is not a big controversy. That the difference is in fact audible under controlled conditions is more open to debate. You do know what I am going to ask next right

Click to expand...

 
yeah yeah, DBT yada yada...this board is quite cheap, try it w/ the cheapest/longest toslink cable you can find as input, the shortest 75Ω coax you can solder(10/15cm) as output, on a DIR9001 DAC w/ good opamps(AD797B/LT1028AC/LT1363/etc) w/ a proper DPS, a very transparent phone(CD3K/K701/HD800/T1/SA5K), and become a believer.
 
The WM8804 sounds amazing in my Firestone Bravo transport. The exact same difference when used as a toslink>coax reclocker as between the STX/ST soundcards. After you've heard the ST, the STX sounds like a mushy mess...and many other ppl have confirmed it.
 
I also tried to connect the Bravo to the Spitfire DAC using a 6ft glass toslink Dayton cable and a 15cm coax, the coax won hands down..the sound was far more focused and tighter. Granted that the cd3k drivers are extremely fast and that you're only as strong as your weakest link, it more than likely wouldn't be audible on a DT770/Pro.
 
All those pointless "debunking" DBT's are always conducted on speakers in a non-acoustically controlled room and w/o room EQ....using high end gear and high grade/fast headphones drivers, reclocking DOES matter.
 
Jitter distorts music because S/PDIF is based on clock recovery to boot: http://www.gearslutz.com/board/so-much-gear-so-little-time/172143-spdif-vs-word-clock-question.html
"S/PDIF is a horrendously poorly designed interface. This is because it combines the clock and audio coding onto the same signal. The receiver is supposed to recover the clock from this signal as well as extract the audio data. This turns out to be a non-trivial task, and one that almost always leaves the recovered clock contaminated with signal correlated jitter artefacts."
 
Being skeptical is great, but nothing beats real world experience.

Also, the faster your music the more reclocking becomes audible....slow music acts as a bottleneck, but for instance on 160BPM DNB, a low jitter is just amazing.
 
And I don't think that upsampling to lower jitter is a good idea at all, as this will inevitably increase THD.

 
Quote:

leeperry said:

reclocking skeptics should try a WM8804 board:

Click to expand...

Yes, but this one doesn't use ASRC. That's all different.
 
Quote:

Originally Posted by leeperry 
 
yeah yeah, DBT yada yada...this board is quite cheap,

Click to expand...

This board is called Sound Science. Of course, if you don't like science, this board looks cheap to your eyes.
 
Quote:

Originally Posted by leeperry 
 
All those pointless "debunking" DBT's are always conducted on speakers in a non-acoustically controlled room and w/o room EQ....
 

Click to expand...

Wrong. A big lot of DBT are conducted with headphones.
 
And though some rare ones were conducted in acoustically controlled rooms, there is always someone to complain that since the room was unknown to the listeners, the test is not valid.
 
Actually, it is necessary to perform the test on the system, and in the room, where the difference is audible. If a difference is audible in an ordinary room, it is a waste of time and money to rend an acoustically controlled auditorium to run a DBT, especially if once in the new room you realize that the difference is not audible anymore and that you need to go back into the original room to start the test...
 
Quote:

Originally Posted by leeperry /img/forum/go_quote.gif
 
Jitter distorts music because S/PDIF is based on clock recovery to boot: http://www.gearslutz.com/board/so-much-gear-so-little-time/172143-spdif-vs-word-clock-question.html
"S/PDIF is a horrendously poorly designed interface. This is because it combines the clock and audio coding onto the same signal. The receiver is supposed to recover the clock from this signal as well as extract the audio data. This turns out to be a non-trivial task, and one that almost always leaves the recovered clock contaminated with signal correlated jitter artefacts."
 
Being skeptical is great, but nothing beats real world experience.

Click to expand...

Your two sentences say two different things. You say that jitter distorts music, then quote a sentence that tells that jitter leaves jitter artifacts in the recovered clock.
 
DBT is real world experience too.
 
And since this is a scientific forum, let's quote jitter measurments performed in analog outputs. And not amplitude modulation ones, that you get looking at the bottom of the spectrum of a 11 kHz sine.
I don't have any, and I would be interested in seeing some.