Digital Drive

Upsamplers, DACs, jitter, shakes and analogue withdrawals, this is it.

Return to Digital Drive

Jitter created by various transports vs. jitter rejection on various DACs?

71.146.146.70
Posted on November 4, 2009 at 20:46:15
audiohound
Audiophile

Posts: 255
Location: San Francisco
Joined: December 11, 2004
 Please help me understand jitter as I am not a techie. One explanation that I read described it as being like a bus that is due at the same bus stop every day at exactly 5PM. However, jitter causes the bus to come at 4:59PM one day, 5:01PM the next day, etc.

And here's the issue with my components:

I have an older but very 'expensive' (original MSRP close to $2K) Toshiba SD9200 DVD-A/V player that I formerly used as a transport very successfully with a Lavry DA10 DAC. I never tried my other transports because the Toshiba is so elegant, ergonomic, and has a 26 lb. copper chassis.

I currently have a modded Music Hall DAC25.2, however, it didn't sound good with the Toshiba; it sounds better with my 'inexpensive' $500 Sony DVP NS500V SACD player. (hmmmm...and maybe my former Lavry would have sounded even better if I had used my cheaper transport?). In any case, I'm very disappointed that I can't get the best sound by using my more elegant transport.

What's going on here? Is the expensive player actually introducing more jitter than the Sony? If I want to problem-solve and use the expensive transport, would something like a Monarchy DIP work to reduce jitter?

Or....if I buy a better DAC, perhaps something like the Lavry again, will it "reject" the jitter? How is jitter "rejected" or "corrected" if the "bus isn't coming at the same time each day" from the source component(DAc)?

Thanks for bearing with me on the metaphors. Any insight would greatly be appreciated. Bottom-line: If possible, I want to use my Toshiba as a transport again...but still get the best sound.

Thanks,
Audiohound

RE: Jitter created by various transports vs. jitter rejection on various DACs?, posted on November 8, 2009 at 21:42:06
niws
Audiophile

Posts: 355
Location: Northeast
Joined: September 4, 2003
 My experience with the Sony DVP-NS500V indicates that it is a decent, for the money, transport. It has a laid back, somewhat reticent sound that is slightly rolled off at the frequency extremes, which may work synergisically with some dacs. It doesn't have the resolution and clarity of some of the dedicated transports or cd player/transports I have owned, but it doesn't sound thin and lean like some cheap dvd players that I have tried as transports. To my surprise, the Sony did not benefit from the jitter reduction devices that I tried with it (Audio Alchemy DTI 2.0 and Behringer SRC2496). Usually most transports I have tried, benefit somewhat from a jitter reduction device. However, whether a jitter reduction device will clean up the sound of your Toshiba enough so you will like it, is anybody's guess. If you're going to buy a jitter reduction device, suggest you only buy with a right of return or buy a used one that you can resell at the same price.

RE: Jitter created by various transports vs. jitter rejection on various DACs?, posted on November 9, 2009 at 18:46:36
audiohound
Audiophile

Posts: 255
Location: San Francisco
Joined: December 11, 2004
 Thanks, great info...however, isn't the Sony's ***sound**** only related to its internal DAC, so that its sonic signature wouldn't be carried forward to the DAC? In other words, if it sounds good as a transport, isn't that because its laser/transport unit is somewhat decent without having a sound of its own? It just reads the digital info correctly enough without adding too much jitter and sends it to the DAC, right?

RE: Jitter created by various transports vs. jitter rejection on various DACs?, posted on November 10, 2009 at 09:40:40
audioengr
Manufacturer

Posts: 4088
Location: Oregon
Joined: April 12, 2001
 There is a lot more to this than the laser/Transport. The clock quality and the digital path design and implementation are all critical as is the power delivery system. All of these impact jitter.

RE: Jitter created by various transports vs. jitter rejection on various DACs?, posted on November 9, 2009 at 20:41:18
niws
Audiophile

Posts: 355
Location: Northeast
Joined: September 4, 2003
 I'm not technically oriented so I can't explain in technical terms why I hear what I hear. I rely soley on my ears and what I think sounds good. I have 2 dacs in 2 systems and have had them for a long time. This has given me the opportunity to own or try a number of different transports (over a dozen). While some have sounded similar, most of them have sounded different in one way or another. The best sounding transports I have heard are the Parasound CBD-2000 and Theta Data Basic. Unfortunately, neither one is currently made. I used to own the Parasound but made a mistake in selling it (long story).
When using CD players or DVD players as transports, I have found some correlation in the sound of their analog outputs to how they sound as transports, but not always so. I don't know why the Sony DVP-NS500V sounds the way it does. However, I also use a Sony SCD-C555ES changer in another system as a transport and I find that this player sounds better as a transport (and a player) than the Sony DVP-NS500V. It resolves fine details that the DVD player obscures, presents a bigger soundstage and is more dynamic sounding. Even though the transport is just supposed to read the digital info and send it to the dac, something seems to happen to that data before it gets to the dac which affects the way the dac sounds. I should add that my dacs are older models which do not have any jitter reduction circuitry built into them so they are more sensitive to different transports than newer dacs.

RE: Jitter created by various transports vs. jitter rejection on various DACs?, posted on November 10, 2009 at 10:44:39
audiohound
Audiophile

Posts: 255
Location: San Francisco
Joined: December 11, 2004
 Thanks for taking the time to explain this. Indeed, after posting the original message and reading all the posts, seems like jitter issues around transport/dac combos is one of the most challenging 'phenomena' to explain. And it probably helps to make this hobby fun!

Another factor to consider, posted on November 8, 2009 at 17:01:36
TimB5881
Audiophile

Posts: 559
Joined: September 5, 2001
 It is not just jitter, ie timming errors that vary. Vibration, error correction, and lens misreads can also have a good/bad effect on playback. The more error correction is used, usually the worse the sound. Also how well the error correction is implimented is another factor. Do not always blame jitter when other factors of a transport should be considered.

Surprisingly, posted on November 5, 2009 at 13:57:47
audioengr
Manufacturer

Posts: 4088
Location: Oregon
Joined: April 12, 2001
 What's going on here? Is the expensive player actually introducing more jitter than the Sony?

Likely. The cost or packaging of a component says little about the design quality. This difference can be because of internal jitter due to clock, poor design of the data path, return path issues, crosstalk or impedance mismatch. Just because its an expensive, heavy transport does not mean that the designer knows digital design or transmission-line treatments. These areas are often "copy-cat" designed in the high-end audio industry IME. I've modded a lot of high-end transports, so I have seen crazy things. The well-designed source is definitely the exception.


If I want to problem-solve and use the expensive transport, would something like a Monarchy DIP work to reduce jitter?

It might, a little.

Or....if I buy a better DAC, perhaps something like the Lavry again, will it "reject" the jitter?

It might, a little.

How is jitter "rejected" or "corrected" if the "bus isn't coming at the same time each day" from the source component(DAc)?

DACs that reduce jitter usually use one of two methods:

1) ASRC - Asynchronous Sample-Rate Conversion, where the incoming signal is tracked, but upsampled to usually a higher rate with a DAC clock. The hardware upsampler and the clock jitter are critical. This will usually not make the input immune from the effects of jitter in the signal.

2) multiple PLL stages - the internal clock of the DAC changes frequency to "lock-onto" the incoming stream rate. This has the disadvantage of being somewhat sensitive to incoming jitter depending on the PLL design and of course the clock jitter is critical.
RE: Surprisingly, posted on November 5, 2009 at 16:59:12
audiohound
Audiophile

Posts: 255
Location: San Francisco
Joined: December 11, 2004
 Thanks for the very comprehensive answer. I appreciate that you took the high road and didn't bash my equipment, you just elucidated possible issues. Sounds like I need to be really careful any time I'm pairing transports to DACs.

What is PLL?

If a transport has jitter issues, does it tend to be the same amount on SPDIF vs. Toslink. I'm of the belief that SPDIF is usually better, so this is what I was using when I happened to notice better sound from the less expensive transport. Is it possible that there would be better performance around jitter if I try Toslink on my more expensive Toshiba...or do both signals tend to perform the same?

Lastly, as another person had posted...do you think it's possible that other systems matching issues (not having to do with jitter) could cause one transport to sound better than another with a particular DAC? I think something was said about signal level, not sure.

Thanks!

PS: ...one more thought, I'm seriously thinking about purchasing a Lavry. I think my decision to not keep it after auditioning it last year may have been short-sighted. It sounded great, yet I was so busy around the holidays and wasn't sure I had confirmed ALL aspects of the sound were top-notch. Do you happen to know if the Lavry DA10 has circuitry that would help in my situation...such as you had mentioned...ASRC, PLL, etc.?

Lavry, posted on November 6, 2009 at 19:39:28
mls-stl
Audiophile

Posts: 1621
Location: St. Louis
Joined: April 11, 2006
 >> PS: ...one more thought, I'm seriously thinking about purchasing a Lavry.

I've had my DA-10 for almost 2 years and have been extremely pleased with it. I've had no urge to look elsewhere.

RE: Surprisingly, posted on November 6, 2009 at 12:16:43
audioengr
Manufacturer

Posts: 4088
Location: Oregon
Joined: April 12, 2001
 "What is PLL?"

Phase-locked-loop - its an oscillator with an adaptive feedback loop

"If a transport has jitter issues, does it tend to be the same amount on SPDIF vs. Toslink."

Yes, but the Toslink will be a bit worse due to the additional conversions. If you system has a bad ground-loop, the Toslink will eliminate this, making the Toslink appear better.

"I'm of the belief that SPDIF is usually better, so this is what I was using when I happened to notice better sound from the less expensive transport. Is it possible that there would be better performance around jitter if I try Toslink on my more expensive Toshiba...or do both signals tend to perform the same?"

It may be a combination ground-loop and jitter difference.

"Lastly, as another person had posted...do you think it's possible that other systems matching issues (not having to do with jitter) could cause one transport to sound better than another with a particular DAC?"

It is unlikely but possible. If you system has other sibilance problems, then the lower jitter source may actually sound more harsh by unmasking this other sibilance in your system. It happens.

Steve N.

RE: Surprisingly, posted on November 6, 2009 at 13:18:12
Todd Krieger
Audiophile

Posts: 22046
Location: SW United States
Joined: November 2, 2000
 "Yes, but the Toslink will be a bit worse due to the additional conversions. If you system has a bad ground-loop, the Toslink will eliminate this, making the Toslink appear better."

IMO, the disadvantage of additional conversions is more than offset by the advantage of RF isolation. The RFI generated at the transport stays at the transport. It doesn't add to the RFI produced at the DAC. This is the main reason why Toslink is my choice for digital interconnect type. (I find such isolation almost requisite if your "transport" happens to be a desktop PC.)


No RF through TOSLINK?, posted on November 7, 2009 at 11:17:19
Slider
Audiophile

Posts: 445
Location: NoCal
Joined: September 24, 2006
 Are you saying that the bandwidth of the TOSLINK convertors is so much lower than a normal driver and cable connection that the high frequencies (RF) are completely attenuated, or that the optical cable is immune to outside airborne RFI that may be picked up on a conventional cable? The latter is true, though that won't stop RF from the transport, which will be on the signal line and passed through either wire or TOSLINK connections, though many TOSLINK implementations have pretty severe rise time filtering to avoid ringing on the transmission side. That filtering is one of the reasons that the TOSLINK connection usually results in much higher jitter in the recovered clock in the DAC.

One of the main benefits of TOSLINK or glass fibre connections in audio systems is to prevent the radiated RFI from the digital cable that happens in some installations.

Electrically-Conducted RFI............., posted on November 7, 2009 at 18:50:36
Todd Krieger
Audiophile

Posts: 22046
Location: SW United States
Joined: November 2, 2000
 "Are you saying that the bandwidth of the TOSLINK convertors is so much lower than a normal driver and cable connection that the high frequencies (RF) are completely attenuated, or that the optical cable is immune to outside airborne RFI that may be picked up on a conventional cable?"

Neither actually....... Unlike the electrical type cable, the optical Toslink does not conduct *electrical* borne RFI that's present in the transport to the DAC. Whatever RFI that is generated that can affect the audio is solely in the DAC, as opposed to both the transport and the DAC.

Now you are correct that there might be some airborne RFI from the transport, but its effect should be minimal compared to the conducted RFI via an electrical digital cable.

"The latter is true, though that won't stop RF from the transport, which will be on the signal line and passed through either wire or TOSLINK connections,"

I think a lot of the RFI is transferred via the common side of a coax cable. With the Toslink, there is nothing electrically bringing the transport's RFI to the DAC.

"though many TOSLINK implementations have pretty severe rise time filtering to avoid ringing on the transmission side."

If that were true, Toslink would have major jitter issues. (Caused by slow rise time in the presence of noise. And the noise is unavoidable.) It was purported to have slow rise times simply due to less bandwidth (no actual filtering), but I personally hear virtually no negative effects from a jitter standpoint.

"That filtering is one of the reasons that the TOSLINK connection usually results in much higher jitter in the recovered clock in the DAC."

I don't know if the filtering is really taking place in most cases- If it is, it's a mistake..... I personally haven't encountered a Toslink system that to me had any more audible artifacts of jitter than other means of transmission.

"One of the main benefits of TOSLINK or glass fibre connections in audio systems is to prevent the radiated RFI from the digital cable that happens in some installations."

It might be radiated RFI, but it's mostly the RFI that originates at the transport, and is simply conducted electrically to the DAC.


Hmmm, posted on November 7, 2009 at 23:09:15
Slider
Audiophile

Posts: 445
Location: NoCal
Joined: September 24, 2006
 Not sure I fully understand your electrical RFI argument. The TOSLINK does use an electrical device at each end, on the transmit side there is a IR LED driver, and at the receive side there is a photo transistor, the two ends coupled with a plastic fiber optic cable. If these electrical devices have sufficient bandwidth, the signal at the DAC end recovered from the photo transistor should be the same as the signal that drove the LED at the transport end, RFI and all. Typically it isn't of course, with the addition of noise and other components, along with bandwidth limitations and built-in filtering. In addition, due to the very high amplification required in the receiver, they are very sensitive to ground and power supply issues, and need very good screening, things that don't happen with the integrated package normally used.

In any case, TOSLINK can help remove the very large 120 Hz jitter component seen in many DACs. And if your DAC has very good jitter immunity, could very well be a good choice. I have seen measurement results that indicate the commonly used TORX transmitter and receiver do add large amounts of jitter to the recovered clock compared to a well implemented wire connection, but a lot of the additional jitter is high frequency phase noise related in the tests I've seen. One of the tests I recall was just a clock test, measuring a large increase in jitter when TORX xmtr/rcvr and fiber was added to path, so much of that phase noise jitter would convert to data correlated jitter in the presence of audio data.

Todd: Toslink being better...., posted on November 6, 2009 at 20:41:03
audiohound
Audiophile

Posts: 255
Location: San Francisco
Joined: December 11, 2004
 Todd,

I think you're one of the few that say Toslink is better. I've used it in my system a few times and was surprised at how good it sounds with the right DAC. Do you feel it pays to buy an expensive Toslink cable, or are they all created equal once basic QC parameters are met? I think mine is a cheapie from Blue Jeans.

And let's say one uses digital coax, does it make sense to spend a lot on this type of cable? I ask because, unlike an analog sound signal, I wonder if a digital signal either succeeds in being transmitted...or it doesn't...without a lot of room for fine-tuning?

RE: Todd: Toslink being better...., posted on November 7, 2009 at 10:42:53
audioengr
Manufacturer

Posts: 4088
Location: Oregon
Joined: April 12, 2001
 The issue with S/PDIF is Price. To get a high quality Toslink, it's $50. A high-quality S/PDIF is over $500. S/PDIF is better though if you can afford it.

RE: Todd: Toslink being better...., posted on November 7, 2009 at 21:18:25
roscoeiii
Audiophile

Posts: 54
Joined: April 28, 2009
 Any specific Toslink cables that you recommend at around $50?

RE: Todd: Toslink being better...., posted on November 10, 2009 at 12:25:00
Todd Krieger
Audiophile

Posts: 22046
Location: SW United States
Joined: November 2, 2000
 "Any specific Toslink cables that you recommend at around $50?"

The QED One 1.5m cable goes for about $50..... The problem is they can only be sourced from the UK.

Email me..... I can send you a couple to try.


RE: Todd: Toslink being better...., posted on November 8, 2009 at 11:07:17
audioengr
Manufacturer

Posts: 4088
Location: Oregon
Joined: April 12, 2001
 I have one I really like, but they dont offer it anymore. It's glass with rotating connectors. Very flexible and sounds great. I use it for my home theater.

RE: Todd: Toslink being better...., posted on November 7, 2009 at 18:53:06
Todd Krieger
Audiophile

Posts: 22046
Location: SW United States
Joined: November 2, 2000
 "A high-quality S/PDIF is over $500."

I never realized S/PDIF devices even existed that would cost that much........ Can you maybe recommend a good one that might be used to upgrade a transport? Thanks.


Transparent, posted on November 8, 2009 at 13:39:18
sbrians
Audiophile

Posts: 292
Joined: March 4, 2002
 I have an old Transparent digital cable that I thought was expensive but worth it in my system at the time.

Cable, posted on November 8, 2009 at 11:09:30
audioengr
Manufacturer

Posts: 4088
Location: Oregon
Joined: April 12, 2001
 I'm talking about a 1.5m RCA terminated cable. email me and I'll give you some recommendations.

Not that the S/PDIF output on the transport does not need modding, because it usually does, as well as the clock.....

RE: Todd: Toslink being better...., posted on November 6, 2009 at 23:57:24
Todd Krieger
Audiophile

Posts: 22046
Location: SW United States
Joined: November 2, 2000
 Like anything else, expense with Toslink cables does not mean better performance. I have the expensive AudioQuest Optilink 5 that I don't care for.

My two favorite Tos cables right now are the QED One and the Japanese Saec. The latter is my reference, but it is expensive and can only be sourced from Japan. The Nordost White Light Glass is good to, as is the Van den Hul Optocoupler and for a low-priced cable, the Apogee Wyde Eye.


Why not both?, posted on November 6, 2009 at 14:48:33
audioengr
Manufacturer

Posts: 4088
Location: Oregon
Joined: April 12, 2001
 My choice is to have both isolation and S/PDIF coax. The best of both worlds.

Good Question............., posted on November 6, 2009 at 18:14:49
Todd Krieger
Audiophile

Posts: 22046
Location: SW United States
Joined: November 2, 2000
 "My choice is to have both isolation and S/PDIF coax. The best of both worlds."

I guess you mean the isolation being internal...... That's a nice idea, except for one little problem- Much of the RFI goes through on the common side......... So if you use a coax with isolation inside the DAC, for example, much of the RFI from the transport will still get through via the S/PDIF shield and then the DAC chassis. (This is why I don't like plugging analog devices into the same outlets as the digital devices.)

Now if you can somehow do this with a balanced line (AES/EBU), it should work..........


RE: Good Question............., posted on November 7, 2009 at 10:44:26
audioengr
Manufacturer

Posts: 4088
Location: Oregon
Joined: April 12, 2001
 All you need is transformer coupling. Treat S/PDIF just like AES.

One more bubble to burst, posted on November 5, 2009 at 07:35:44
Charles Hansen
Manufacturer

Posts: 4320
Joined: August 1, 2001
 >> the Toshiba is so elegant, ergonomic, and has a 26 lb. copper chassis <<

Not really. First of all, the chassis is copper plated steel, not copper. Second of all, if you look inside you will see that at least half of the weight comes from a large steel plate screwed to the bottom of the chassis. This mostly gives the illusion of quality and only has a minor effect on the actual performance.

The actual transport mechanism was the same one used in their entry level models.

It's all about the system, not the component., posted on November 5, 2009 at 06:27:46
Gordon Rankin
Manufacturer

Posts: 2247
Joined: June 9, 2000
 AH,

Jitter may have something to do with it. It also can be the effect the digital cable and the two components react. It could be the interface levels that the two components run on.

See with SPDIF you not only have the data but the clock to derive. The better the data and the clock recovery the better the sound and tpically the less the jitter will effect the sound.

Really this is one of the pains of SPDIF and the reason why everyone should try this stuff out before they buy it. There are too many variables to rely on a review in a magazine and take a chance on Audiogon.

In looking at the components above I would think the Sony probably has better clocking on the output circuitry and the Toshiba relies on probably a PLL derived clock. Fixed oscillators will always sound better than derived PLL clocks.

Thanks
Gordon
J. Gordon Rankin
RE: It's all about the system, not the component., posted on November 5, 2009 at 13:39:30
audiohound
Audiophile

Posts: 255
Location: San Francisco
Joined: December 11, 2004
 Thanks, guys!

Page processed in 0.117 seconds.

[ Top Top of Page ] [ Contact Us ] [ Support/Wish List ] [ Copyright © 2009, Audio Asylum, all rights reserved ] [ Go Green ] [ Take5 ]
[ General ] [ Speakers ] [ Tubes ] [ Vinyl ] [ Digital ] [ Hi-Rez ] [ Video Asylum ] [ Cables ] [ Tweaks/DIY ] [ Music ] [ Films ]