How can different CAT5/6 cables affect sound.
While is is beyond doubt that analog cables affect sound quality and SPDIF, TOSlink and AES/EBU can effect SQ, depending on the buffering and clocking of the DAC, I am at a loss to find an explanation for how different CAT5 cables can affect the sound.
The signals over cat5 are transmitted using the TCP protocol. This protocol is error correcting, each packet contains a header with a checksum. If the receiver gets the same checksum then it acknowledges the packet. If no acknowledgement is received in the timeout interval the sender resends the packet. Packets may be received out of order and the receiver must correctly sequence the packets.
Thus, unless the cable is hopeless (in which case nothing works) the receiver has an exact copy of the data sent from the sender, AND there is NO timing information associated with TCP. The receiver must then be dependent on its internal clock for timing.
That is different with SPDIF, clocking data is included in the stream, that is why sources (e.g. high end Aurenders) have very accurate and low jitter OCXO clocks and can sound better then USB connections into DACs with less precise clocks.
Am I missing something as many people hear differences with different patch cords?
I believe Aurender will output DSD via SPDIF as DOP. Some other streamers might as well. Native DSD probably only via USB I believe there are SACD transports that can output DOP via coax. Zenith is one. |
IMHO it should be obsolete. Replace it with something like WAV data over TCP with error correction, or an error correcting USB link, to completely isolate the DAC processing from jitter in the source. The DAC then buffers the data stream and oversamples, or whatever, the data words; it knows the frequency from the WAV data, and feeds the words into the DAC circuitry (chips or discrete) based on an OCXO, or better, clock. An OCXO chip costs from $25 to $250. For cheaper units a TCXO ($3) could be used. The phase noise of a TCXO can be about the same as an OCXO but the long term frequency stability is not as good. Or, go all in and use a Rubidium clock, at $2,000 for the oscillator, although, interestingly, the phase noise of the best OCXO parts is just about the same as a Rubidium clock. Esoteric's latest external clocks are now OCXO based not "atomic" Rubidium. They reduced the price from $25k to $15k. Still crazy packaging a $250 part and charging $15k, rather than packaging a $2k part for $25k. Their internal Master Discrete Clocks are also OCXO based. Google "History of OCXO" it is rather interesting, they were first developed in 1929 (!) but low cost versions have only been available this century. BTW, the SRS FS725 uses the same physics package as the prior ($25k) Esoteric unit and costs $3,995, it has both 10MHz and 5MHz outputs, nominally 50 Ω but doesn't care, the rms voltage is higher with a higher impedance load. |
I am NOT saying that people do not hear differences, I am trying to understand the physics of the cause of those aural difference. I did not study physics beyond 1st year university level, but my mind wants to understand, and experiment if possible. I think I AM lucky in in that: 1) Presto intends that any differences in the download and streaming experiences should be imperceptible. Both are sourced from the highest resolution digital master recording files available. 2) Esoteric, recognizing the deleterious effects of jitter, go to extreme lengths to minimize the effects of jitter in the input. 3) My choice of the Elgar was probably not the most sensitive selection of test material, it is lush and romantic rather than delicate and subtle. 4) I will grant that, at 81, my hearing is not what it was, Treble extension is low on my list of priorities, e.g.! In the '90s TOSLink was regarded as the digital connection to avoid. ATT Glass was the medium of choice but appears to have disappeared, my second phase of early gear had glass fiber digital connections, as sound quality was severely affected by digital cables - timing was driven by a PLL reconstructing a clock from the source, with no real buffering, passing the jitter along to the output waveform. Incidentally I just did another experiment. I have subscriptions to both Presto and Qobuz. My wife's musical choices are broader than my limited one. A favorite of mine is the Fauré Requiem, especially the de Los Angeles/Fischer-Dieskau recording (originally on Angel but later rereleased by EMI). I played selections from both (through the Bluesound Icon), level matching at the preamp as there was a 1.5 dB difference of playback levels. On my primary system the sounds of the Offertory and the Sanctus were, in quieter passages, very subtly different. Qobuz a hair "warmer and laid back", the Presto a fraction more "airy and ethereal". On my less resolving 2nd system (DMP A8, upgraded Elekit 8600, P3ESRs) any difference was imperceptible. I don't know if the two vendors have the same source, their album information appears to be the same, the EMI rerelease. I think I'll try some other works to see if this is a consistent difference. Apologies if this is off topic but, as OP, perhaps I am allowed to ramble a bit to stir up discussion.
|
@retiredaudioguy to your point my Meitner MA3i DAC has a pretty high quality built in network card. I’m using it as a Roon endpoint so the only connection into the Meitner is Ethernet cable. This DAC replaced Aurender N200 and Bricasti M3 combo where I was using a USB cable from N200 into Bricasti DAC. |
S/PDIF can carry DSD when reencoded as DoP, as you pointed out. However it can only support DSD64, which is carried within a 24/176.4kHz PCM signal, essentially equivalent to 4 times Redbook (since DSD64 is 1-bit resolution at 2.8 MHz, v 16 bit at 44.1kHz, or 705.6 kHz, for Redbook). By that token, DSD128 and DSD256 over DoP require 384 and 768 kHz PCM signals, respectively, both of which are beyond S/PDIF capabilities. Additionally, not all DACs are DoP-compatible, so all in all there is little to recommend DoP unless specific limitations rule out native DSD. That's why I didn't mention it. It's probably fair to say that DSD is best enjoyed with digital chains that natively support it end-to-end.
|
