I find it fascinating that there is a small but persistent minority voice stating that disc playback beats files, at all sorts of price levels. This is one of the reasons I am still staying on the fence and not committing to streaming, despite the splinters in my arse .
Help me beat my CD Transport
- Thread starter joeling
- Start date
The thing is Steve that Ethernet transformers are not fully isolating noise coming through the connection so you are still facing similar issues as you are with USB - just a different noise profile coming through.
In reading your listening impressions of USB Vs ethernet in your link, I was struck by :
This is exactly the issues that USB isolation (done correctly) solves - a solidity & realism of the sound stage & each individual element in the soundfied, emerges
I would also suggest that the USB TIE plot showing two peaks (binomial distribution of jitter) is showing something wrong with your USB setup - it should be a single peak
They are just reporting what they are hearing - nothing strange about it really?
I find that certain Macs are very prone to the current leakage & CM noise issues that cause such lifeless sound from computer audio
J, but all I’ve ever been told about files is that eliminating the spinning disc eliminates all that jitter that makes digital a “spikier” experience than analog, and makes streaming more musical than discs.
Now this isn’t enough, orceven the right diagnosis, now it’s all about usb v Ethernet, or keeping noise down.
No argument from me in the latter, I’ve put massive efforts via isolated room power, balanced power, and dedicated lines, to get noise right down.
There still isn’t a consensus that files beat disc.
And hence another good reason for me not to spend serious cash when my cdp and tt totally satisfy me.
Now this isn’t enough, orceven the right diagnosis, now it’s all about usb v Ethernet, or keeping noise down.
No argument from me in the latter, I’ve put massive efforts via isolated room power, balanced power, and dedicated lines, to get noise right down.
There still isn’t a consensus that files beat disc.
And hence another good reason for me not to spend serious cash when my cdp and tt totally satisfy me.
I got out of the disc spinning business years ago for convenience reasons, but I have been told the new PS audio transport beats the crap out of all other sources to feed bits to the DirectStream DAC.
My router has a LPS with earth ground connected to DC common. I also have an isolator, the EMO EN-70e in the Ethernet cables. This should mostly eliminate any noise through the PHY.
Perhaps, but the Ethernet is still more live sounding. I think the "boldness" of the USB is partly due to either jitter or other data mucking around that is not real.
I also get two peaks when I use my reclocker with ANY device, so I don't think this has anything to do with USB. I would probably get two peaks with Ethernet too, but for the reduced jitter in the input section compared to the isolated USB input section powered with many switching regulators on-board. No switching regulators in the Ethernet interface.
Steve N.
I've read that too. The PSAudio transport is actually a computer I believe with a CDROM drive. It plays from memory and has HDMI I2S output, which can be a bit better than S/PDIF.
I suspect that my Synchro-Mesh reclocker still beats it. Its really hard to get all of that noisy electronics in one box and get 18psec of jitter on S/PDIF or even I2S.
Steve N.
Sure, if you are totally satisfied with your CDP & TT playback why look for anything else?
I believe many audio designers are in learning phase constantly & the noise issues & as one element, like jitter goes low enough not to mask other issues. In other words these other issues (like CM noise) can be evaluated through experiments which can now reveal the part they play (or otherwise) in sound quality.
So when these types of noise issues are reduced to a level that no further sound quality change can be heard through further reduction, it can then reveal what part, if any, the low levels of jitter now play in sound quality & what articular elements of jitter may be important.
In my experiments it wasn't until I eliminated these noise issues & then acquired audio cocks with low measured close-in noise that I could evaluate the effect of this jitter spectrum.
No. I do have a cap meter. I guess you mean across the isolator, input to output?
I think I would have to do this inside the router and in the receiver PHY. Not sure it would be accurate in-circuit. Might have to remove the PHY to do this. I have in receiver PHY that I can measure on the bench. I'll try to do this today.
Steve N.
Empirical Audio
Marc,
You are listening to the wrong people ...
IMHO all systems have a "spikier" sound signature, some have a subjectively less objectionable. Anyway you can not separate the source from the DAC and from the system when considering this problem - please read MBL's Juergen Reis on this subject. URL="https://www.audiostream.com/content/open-and-tolerant-mbls-juergen-reis-listening-measurements-and-uncertainty"]https://www.audiostream.com/content/open-and-tolerant-mbls-juergen-reis-listening-measurements-and-uncertainty[/URL]
I don't know that ethernet isolator or how it works but an LPS also has current leakage issues - just not as many as SMPS
Sure, I found that the more of this type of noise elimination, the less "boldness" to the sound - the more fluid it becomes
I don't know about your configuration - to the best of my knowledge, two peaks in a TIE plot signifies that the timings are distributed across two main time timings, not one that should be seen with any packet based transport.
J, because I know there is a musical education/universe on Tidal/Roon, and I feel I shouldn’t deny myself.
But I also know I’ll struggle to emulate my current sound from many streamer/dac combinations without spending mucho bucks.
I really don’t want to end up with a source that struggles to match the verve and warmth of my current tt and cdp.
And a small stream of comments claiming streaming can be v prone to insipidness does nothing to answer my doubts.
But I also know I’ll struggle to emulate my current sound from many streamer/dac combinations without spending mucho bucks.
I really don’t want to end up with a source that struggles to match the verve and warmth of my current tt and cdp.
And a small stream of comments claiming streaming can be v prone to insipidness does nothing to answer my doubts.
You have that right, at least with this designer. I remember how excited we all were when the first USB designs came out with master clock in the receiver and not the computer. We though that jitter could be really low and the computer, cables and software would be removed from the equation. Boy were we wrong!!
I cannot speak for other designers, but I'm now learning the same things for Ethernet and making changes and adapting. The difference is that there are less impediments with Ethernet to getting really stellar live sound IME.
Steve N.
Empirical Audio
The best way is cutting a cheap RJ45 cable in half, removing the insulator at the extremes and connect the cables of each side together. Using these probes it is easy to measure jRust the device capacity - but surely it should be removed.
I just posted this link in another thread (MBL's Juergen Reis on measurements and sound quality):
Similar things are happening with digital inputs and outputs that are galvanically isolated (meaning no Ohm connection, and so no DC current is flowing), but due to stray capacitance of the digital audio transformers, there is an unwanted leakage path of the digital audio signal. Even this stray capacitance can be lower than 10 pF, but with the Biphase Signal in the MHz area, it can influence the total quality of a DAC.
Read more at https://www.audiostream.com/content...surements-and-uncertainty#y5O8XgLzzSkTHhWG.99
I spin using an Esoteric K-01X and stream from an Aurender W20. IME, the spinner wins every time if I stream from Tidal. Files stored on the W20 sound better but still not as good as cd. Maybe I prefer duel AES/EBU from the K-01X over the Spdif from the W20? I tried USB with the W20 and never liked what I heard but I did not experiment with many USB cables which I have read make a big difference. Regardless of sound, I love the convenience of Tidal for new music exploration.
Certainly driving S/PDIF from a transport is not packet-based. I still get two peaks with my reclocker. No Switching regulators either, all linear. The two peaks are only about 10 psec apart. I think it's a non-issue.
Steve N.
Maybe that says something about the Tidal files or playback software. Maybe it has more to do with your USB interface or AES interface deficiencies. USB requires a lot of optimizations.
Steve N.
Empirical Audio
Similar threads
 | Steve Williams Site Founder | Site Owner | Administrator |  | Ron Resnick Site Co-Owner | Administrator |  | Julian (The Fixer) Website Build | Marketing Managersing |