Mb9061 Couldn't you go wireless with Squeezebox or something similar and eliminate the cable issue? |
I think I heard Audioengr's suggestion on a similiar post sometime back, about keeping USB cable length to a minimum, i.e. preferably 6' or less.
I'm running a benchmark DAC1 and couldn't agree more. I'm using a Belden gold USB 6' usb cable and when I substitute a 15' cable in it's place (same make), it honestly tames the dynamics. Very easy to tell the difference on acoustic tracks. I changed majors away from electrical engineering, so I can't wax on the technical aspects of why this is the case, but I wholeheartedly stand behind the viewpoint.
That said, it p@sses me off because I can't have the computer more than 6' from the DAC and amplifier it's tethered to, but it's a compromise for now. |
Undertow - here are two good USB cables:
Synergistic
Locus Design Axis
As for DAC's, they are all somewhat sensitive to the incoming jitter, even those with isolated free-running reclocking. In theory they shouldn't be, but they are. I know because I make reclockers and I mod a lot of DAC's. I also read a lot of anecdotal posts. Maybe it's power supply, RFI or ground-plane coupling causing this, I dont know.
I do know for a fact that lesser anti-jitter techniques such as ASRC chips are generally sensitive to incoming jitter. Making ANYTHING that totally rejects all incoming jitter is the holy grail and to my knowledge has never been achieved. Some manufacturers claim this, but when I hear their gear and do some cable swapping, I disprove these claims. They usually just need more resolving systems to hear this. I've been working on this problem for years and thought I had it licked. Closer than anyone has gotten before I believe, but still no cigar.
Steve N.
Empirical Audio |
Audioengr
So who makes the optimal USB cable? And dac in your opinion to handle this for that matter?
Thanks |
"Is USB effected by all this "Jitter" for example a laptop connected to a USB Dac with like a 15 foot USB Cable etc
?"
Yes, and it depends on the DAC and it depends on the cable as to how much. Cables with small gauge silver conductors and air or teflon dielectric add less jitter. Less dielectric absorption and therefore less dispersion of the signal.
Steve N. |
Undertow,
I think you are right on. I use five megachangers to feed a DAC for a total cost that is still way less than that of a high end player. The sound is probably 9/10ths as good but most important is that I get to hear my music instead of fighting with jewel cases and a filing system (now where is that CD?). When you have over a thousand CD's it is nice to be able to cue any one of them up at the push of a button for contiinuous play. So I am with you there and would not recommend going for an extremely expensive high end player. One day I will junk the tranports and connect the DAC straight to a PC where my music will eventually be stored - that, my friend, like it or not, is the future! |
Well then the solution would be a DAC with a USB interface powered off a separate power supply, or maybe a Battery powered supply for the USB charged off the main transformer tap to isolate it with the batteries, and don't batteries supposedly help eliminate jitter on their own anyway or do I have this backwards? Whatever bottom line is investing too heavily into digital at this point is more silly in my opinion cause you can get excellent results today out of many sources without too much money or work compared to 5 or 10 years ago I guess. |
Undertow,
Yes and No. In theory Yes. In practice sometimes NO - as the bursts required for USB operation may feedback through the power supply to make the DAC inside the USB device more jittery than an equivalent AES/EBU or Toslink setup. Unfortunately activities close to the DAC itself can affect jitter. It is the same problem in a CD transport...control signals to the servo motor may end up inducing jitter in the DAC clock.
Remember that Toslink and Digital coax have been around longer than USB and therefore DAC designers have had a bit more time to work out all the possible jitter issues and good ways to attenuate them. In that sense a USB communications protocol running off the same power supply might be an added "unknown". |
Shadorne
Thanks, I figured it was very similar to how an HDMI type device works, almost identical.. Quality will be built depending on the end device. So it makes sense that the quality will be in the USB dac itself doing its job more so than the device storing a file.
This is why I actually put that question out, which is why I can't see any advantage to going the route of a standard Coax digital or Optical cable connection converted at the music serving device in the first place, it seems to me it is much more sensible to put a little more money into a USB DAc these days, Eliminate a 300 dollar Coax digital cable for example and replace with a decent USB then just let a standard computer, or music server device do its job on its end and the DAC take on clocking it and allowing the DAC also the actual responsibility to make it sound good!
Seems to eliminate much of the guesswork in cabling, cost, and basically only one conversion done at the input of the USB Dac once data arrives.. I really don't see putting a cheap USB card on a computer that outputs the Digital coax etc.. at the main device, but of course I can understand people with really good old school DACS wanting to keep their old DAC and this is the only way to convert it price effectively. |
Is USB effected by all this "Jitter" for example a laptop connected to a USB Dac with like a 15 foot USB Cable etc
? No it should not be - USB is sending "data" without a timing signal. Jitter has no affect on data. You can copy a CD a million times - it does not matter. Jitter occurs when A to D or D to A is occuring. Digital works wonderfully provided timing is VERY accurate (or erros are randomized). Unfortunately maintaining a precise clock over interfaces is not easy - it is much better to do it all on one chip with crystal clock close to the converter. However, the reason for passing a clock signal along a digital audio chain is to maintain the relative timing of the devices. This is so nothing gets too far ahead or too far behind....if for example the DAC got too far ahead then it would end up having no data and music would stop. If a DAC gets too far behind then all the "bits" need to be temporarily stored somewhere in a buffer. Also if you are watching a DVD with video then you want the sound to stay in sync with the video - so relative timing is often important. With A USB connection the devices can "handshake" - data is sent or resent upon request so buffers do not overfill and no data is lost and no synchronization is needed. Your only concern with USB is the jitter quality in the DAC device itself. (Note that USB protocols and communications will have periodic bursts and some data packest will be repetive - so there is still a risk that correlated noise from USB communications (and the draw on the shared power supply) reaches the DAC clock... |
Is USB effected by all this "Jitter" for example a laptop connected to a USB Dac with like a 15 foot USB Cable etc
? |
Thanks to you both for the most generous advice.
That jitter stuff is, I guess, only OK if you have a career in pole-dancing!
Regards,
8^( |
So much loss of quality even though it would be a digital signal to the DAC? The benchmark DAC1 claims to be able to work with a 1000 feet of Cat 5 cable but that is the only one I have seen making such claims. Any digital interface seems to suffer from jitter and longer runs are worse. It can vary from barely audible or subtle to plain bad and each situation is different so it is hard to predict. Good power conditioning seems to help. You need a reclocker or get a DAC that you think can handle it....and test it thoroughly by A/B short versus long runs to satisfy yourself it is doing its job. |
Vvrinc - I spend a LOT of time on the forums educating audiophiles about one of the most pervasive problems with digital audio, and that is Jitter.
Digital audio is comprised of two elements, the data words and the timing of these data words. The two cannot be separated because the D/A conversion of the digital datastream must have both elements. The data must be delivered to the D/A without error and the timing must be recovered or generated so that timing variations are minimized (jitter). This attempts to matche the timing of the original A/D that was performed in the studio with a low-jitter clock. These clocks are critically important for both A/D and D/A.
Here are some white papers I wrote for PFonline that tell more about this:
http://www.positive-feedback.com/Issue22/nugent.htm
http://www.positive-feedback.com/Issue14/spdif.htm
Steve N.
Empirical Audio |
Audioengr,
Thanks for your answer.
So much loss of quality even though it would be a digital signal to the DAC?
Regards,
8^0 |
In my system, this would completely crush the quality. Even 2 feet of 1694 does this. Depends on your system.
I would recommend using WiFI instead, such as Squeezebox, duet or AppleTV. The same data is delivered. The only difference is jitter and some of these are pretty good.
Steve N.
Empirical Audio |
