Does high quality digital cables matter?
Comments
-
So you found someone who doesn't understand blind testing...bravo!
If you were looking for people who don't understand blind testing, you really didn't need to leave this thread.
In your case, you wouldn't need to leave a mirror.
Audio is full of people who don't understand blind testing. I have written extensively on the misapplication of blind testing methods to stereo audio.
I am continually amused by the reactions I get when I point out the correct applications for certain blind tests and ask blind test cultists to scientifically justify their use for stereo audio.Proud and loyal citizen of the Digital Domain and Solid State Country! -
DarqueKnight wrote: »Audio is full of people who don't understand blind testing. I have written extensively on the misapplication of blind testing methods to stereo audio.
So what? You write a lot of things, doesn't make any of it true. -
DarqueKnight wrote: »Audio is full of people who don't understand blind testing. I have written extensively on the misapplication of blind testing methods to stereo audio.
Audio is also full of people who don't understand the digital domain, and the differences between digital and analog audio. Let me ask you something DQ...does a file on a hard drive contain Jitter?Too many good quotes to list..waiting for some fresh ammo.
-
Audio is also full of people who don't understand the digital domain, and the differences between digital and analog audio. Let me ask you something DQ...does a file on a hard drive contain Jitter?
Not speaking for DK, but nobody ever said the stored file has jitter. Why would you ask that? Jitter can be added as the file is transfered from the drive to the device playing the file. Any link and/or circuitry between the drive and the DAC can add the jitter.Lumin X1 file player, Westminster Labs interconnect cable
Sony XA-5400ES SACD; Pass XP-22 pre; X600.5 amps
Magico S5 MKII Mcast Rose speakers; SPOD spikes
Shunyata Triton v3/Typhon QR on source, Denali 2000 (2) on amps
Shunyata Sigma XLR analog ICs, Sigma speaker cables
Shunyata Sigma HC (2), Sigma Analog, Sigma Digital, Z Anaconda (3) power cables
Mapleshade Samson V.3 four shelf solid maple rack, Micropoint brass footers
Three 20 amp circuits. -
Not speaking for DK, but nobody ever said the stored file has jitter.
Actually you did, in post 700 of this thread. -
Actually you did, in post 700 of this thread.
If I did then I was wrong. It is easy to make mistakes in these conversations.
Wait. I just reread that post, and I was specifically referring to the transfer of the data. So, at least in that post, I did not say that a file on a hard drive has jitter.Lumin X1 file player, Westminster Labs interconnect cable
Sony XA-5400ES SACD; Pass XP-22 pre; X600.5 amps
Magico S5 MKII Mcast Rose speakers; SPOD spikes
Shunyata Triton v3/Typhon QR on source, Denali 2000 (2) on amps
Shunyata Sigma XLR analog ICs, Sigma speaker cables
Shunyata Sigma HC (2), Sigma Analog, Sigma Digital, Z Anaconda (3) power cables
Mapleshade Samson V.3 four shelf solid maple rack, Micropoint brass footers
Three 20 amp circuits. -
Not speaking for DK, but nobody ever said the stored file has jitter. Why would you ask that?
Point in case ^Jitter can be added as the file is transfered from the drive to the device playing the file. Any link and/or circuitry between the drive and the DAC can add the jitter.
Exactly. But don't forget (And this is a fact) that every link or step along the way nullifies any jitter that was added by the previous step as jitter simply does not exist while outside of the real time domain. There is no jitter in a stored digital file. It can only be added *After* being "Unstored" and moved to a real time domain. But, the second it's stored again..poof..all jitter gone, erased, and reset to zero (Or technically to the same clocking as the master recording).
Here's a good article for everyone to read. Nice subjective look at the subject.
http://nwavguy.blogspot.com/2011/02/jitter-does-it-matter.html
So to summarize something important to remember to is that Jitter only has meaning during the real-time ADC procedure. Physically Jitter can ONLY exist once there is some form of clock or timing device...so files on or in any storage medium (Hard disk, cloud storage, RAM memory/buffers) cannot have any jitter. Each time that the file is transferred and taken out of the real time domain, jitter that has been introduced is nullified. However, jitter that is recorded during the original Analog to Digital process (At the recording studio) cannot be fixed, corrected for, or removed. If jitter exists in the playback of an audio stream (Whether it's audible or not) it's either original recording jitter or jitter that was introduced in the very last step of digital/analog conversion before analog playback via the wires going to your speakers. Given that this is the case I would take a serious look at your DAC/AVR/Amps rather than playing the cable blame game...as the DAC/AVR/Amps are the only place that jitter heard during playback of audio could possibly be introduced. Every other step of the process effectively nullifies jitter since you are moving back and forth from the Real Time domain (Where jitter exists) to a domain where jitter doesn't (As a data file or digital data packets in a storage medium). Think about it..if this wasn't true then by the time any type of audio file was delivered to you it would be nothing but jitter. Especially after traveling through thousands of miles of fiber strands (Which are notorious for introducing jitter) in the digital domain before ever reaching you. Also consider that nearly every movie theater in the country now displays moving images and sound delivered via a cloud service. That is still true "Reference Level"...correct?Too many good quotes to list..waiting for some fresh ammo. -
Point in case ^
Well, I am not going to discuss semantics. If your reading comprehension skills are that weak then no wonder you are having such a hard time understanding why it is hypothetically possible for an Ethernet cable to affect the sound of a musical file.Lumin X1 file player, Westminster Labs interconnect cable
Sony XA-5400ES SACD; Pass XP-22 pre; X600.5 amps
Magico S5 MKII Mcast Rose speakers; SPOD spikes
Shunyata Triton v3/Typhon QR on source, Denali 2000 (2) on amps
Shunyata Sigma XLR analog ICs, Sigma speaker cables
Shunyata Sigma HC (2), Sigma Analog, Sigma Digital, Z Anaconda (3) power cables
Mapleshade Samson V.3 four shelf solid maple rack, Micropoint brass footers
Three 20 amp circuits. -
Taken from a recent Audioholics reply regarding "Club Polk" and Polk speakers:
"I'm yet to hear a Polk speaker that merits more than a sentence and 60 seconds discussion."![:\](https://forum.polkaudio.com/resources/emoji/confused.png ":\")
My response is: If you need 60 seconds to respond in one sentence, you probably should't be evaluating Polk speakers.....
"Green leaves reveal the heart spoken Khatru"- Jon Anderson
"Have A Little Faith! And Everything You'll Face, Will Jump From Out Right On Into Place! Yeah! Take A Little Time! And Everything You'll Find, Will Move From Gloom Right On Into Shine!"- Arthur Lee -
Let me ask you something DQ...does a file on a hard drive contain Jitter?
No, but a file on a hard drive can contain noise.The difference is that while every bit of interferrence effects the output quality of an analog signal, interferrence and degradation within a certain range doesn't effect the final product of packetized data being transported on an ethernet cable. You'll either get the signal correctly, or you won't. There is no grey area. Like Habanero said, once the amplitude of the signal being carried on a digital line hit's the demarcation point for a solid 0 or 1 it's done.
It's not quite that simple. "Ones" and "zeroes" are not being transmitted. Voltage pulses that correspond to logical high ("1") and logical low ("0") are being transmitted. The high and low voltage pulses have nominal values, for example low may be 2 V and high may be 12 V. In reality, high and low correspond to a range of values: low may be any value from 0-5 V and high may be any value 10-15 V.
Consider the sequence 1-0-0-1-1-0-1-1-0-1-1-1: Ideally this would be represented by the following pulse amplitudes according to the scheme given above:
12-2-2-12-12-2-12-12-2-12-12-12
A noise-affected version of the sequence is:
10-0-1-11-11-3-12-14-1-15-10-11
Here, the demarcation between logical high and logical low is preserved, but the signal looks a lot different than the ideal case.
What happens when a more seriously noise-affected version occurs?:
14.79-4.38-1.13-9.9-13.8-0.3-11.5-0.45-11.5-13.88-10.65
Here, the demarcation between logical high and logical low is preserved, but the signal now approaches the amplitude characteristics of random noise.
Each of these voltage pulse sequences are within spec with regard to amplitude:
A. 12-2-2-12-12-2-12-12-2-12-12-12
B. 10-0-1-11-11-3-12-14-1-15-10-11
C. 14.79-4.38-1.13-9.9-13.8-0.3-11.5-0.45-11.5-13.88-10.65
However, due to noise contamination, there are now three different versions of the pulse sequence. Consider that, in a typical stereo music file, you will be dealing with millions of bits, therefore, the grossly simplified cases represented by A, B, and C translate to three files with multiple millions of differences between them.
Questions:
1. Will all three files require the same processing effort?
2. When the D/A conversion is done, is one file going to be more prone to error than the others, or will they all convert to identical analog files?
Consider another case of the sequence contaminated by electric fast transient noise, so that the following sequence results:
16.79-4.38-5.13-9.9-15.8-6.3-11.5-0.45-11.5-10.88-15.65
The values in red are outside the ranges for high and low. What happens in this case?Amplitude has no effect on volume, frequency, or anything else that it would normally have in an analog signal...it's simply a signaling method to determine 0's and 1's for perfect reproduction. If the MD5 hash matches the source MD5 hash then there is NO variance in the file being output that wasn't in the source file to begin with. They are 1:1 bit copies. Perfect copies. It's impossible to have additional "Data" within that signal that wasn't there at the get go.
My example showed that multiple versions of the same file can result due to noise contamination. They were not perfect 1:1 bit copies, although each file maintained the proper relationship between logical high and logical low.IF there is anything there that wasn't in the source file it's being introduced along the way by means other than the computer system or network components, and to make matters more interesting it's not actually part of the signal nor is it within it..althought it can creep along with it..around/outside/within the actual cable being used to transport the digital signal but not actually reproduced within it. Things like electrical interferrence, ground interferrence, magnetic interfrrence, etc...and not things like routers, switches, or fibre. But regardless those are just using the cable for a carrier, and are not within the digital signal being decoded. They are creeping from cable to cable to your receiver to your tv, to whatever they can to cause interferrence where they can. Do not confuse those resulting specks in your tv's image as interferrence caused by the cable itself, or the digital stream that is being played back. That noise is an overlay, nothing within.
If there is a high degree of amplitude randomness in a digital signal, why wouldn't this affect processing efficiency, especially with regard to digital to analog conversion? If the randomness affects processing efficiency, why wouldn't this translate to audible differences?Proud and loyal citizen of the Digital Domain and Solid State Country! -
I don't know about others, but I try to get information about a subject from a variety of sources. For example, jitter is a relatively common topic on audio forums. Here is an excellent post, and with good timing (ha ha) for this thread. Monk, I hope your write-up reaches this level of detail.
First paragraph:
"One of the common arguments made against jitter mattering is that: "the data is buffered and clock regenerated in the DAC so jitter won't be there." This makes all the sense in the world. Once we capture the data and then push it out at our will, there shouldn't be a problem. Well, there is a problem. A serious one. Buffering and clock regeneration do not deal with jitter by themselves. I have explained this in words many times but this time I am bringing in some specific data to hopefully put this myth to bed (yeh, wishful thinking )."
http://www.whatsbestforum.com/showthread.php?14957-Yet-another-look-at-Jitter-Clock-Extraction&p=270683&viewfull=1#post270683Lumin X1 file player, Westminster Labs interconnect cable
Sony XA-5400ES SACD; Pass XP-22 pre; X600.5 amps
Magico S5 MKII Mcast Rose speakers; SPOD spikes
Shunyata Triton v3/Typhon QR on source, Denali 2000 (2) on amps
Shunyata Sigma XLR analog ICs, Sigma speaker cables
Shunyata Sigma HC (2), Sigma Analog, Sigma Digital, Z Anaconda (3) power cables
Mapleshade Samson V.3 four shelf solid maple rack, Micropoint brass footers
Three 20 amp circuits. -
DarqueKnight wrote: »No, but a file on a hard drive can contain noise.
Yes it can. Noise that is introduced to the file by everything from capacitors, resistors, transistors, etc. to ethernet cables.DarqueKnight wrote: »It's not quite that simple. "Ones" and "zeroes" are not being transmitted. Voltage pulses that correspond to logical high ("1") and logical low ("0") are being transmitted. The high and low voltage pulses have nominal values, for example low may be 2 V and high may be 12 V. In reality, high and low correspond to a range of values: low may be any value from 0-5 V and high may be any value 10-15 V.
Consider the sequence 1-0-0-1-1-0-1-1-0-1-1-1: Ideally this would be represented by the following pulse amplitudes according to the scheme given above:
12-2-2-12-12-2-12-12-2-12-12-12
A noise-affected version of the sequence is:
10-0-1-11-11-3-12-14-1-15-10-11
Here, the demarcation between logical high and logical low is preserved, but the signal looks a lot different than the ideal case.
What happens when a more seriously noise-affected version occurs?:
14.79-4.38-1.13-9.9-13.8-0.3-11.5-0.45-11.5-13.88-10.65
Here, the demarcation between logical high and logical low is preserved, but the signal now approaches the amplitude characteristics of random noise.
Each of these voltage pulse sequences are within spec with regard to amplitude:
A. 12-2-2-12-12-2-12-12-2-12-12-12
B. 10-0-1-11-11-3-12-14-1-15-10-11
C. 14.79-4.38-1.13-9.9-13.8-0.3-11.5-0.45-11.5-13.88-10.65
However, due to noise contamination, there are now three different versions of the pulse sequence. Consider that, in a typical stereo music file, you will be dealing with millions of bits, therefore, the grossly simplified cases represented by A, B, and C translate to three files with multiple millions of differences between them.
Questions:
1. Will all three files require the same processing effort?
2. When the D/A conversion is done, is one file going to be more prone to error than the others, or will they all convert to identical analog files?
Consider another case of the sequence contaminated by electric fast transient noise, so that the following sequence results:
16.79-4.38-5.13-9.9-15.8-6.3-11.5-0.45-11.5-10.88-15.65
The values in red are outside the ranges for high and low. What happens in this case?
Your example above does a phenominal job of demonstrating exactly why the ideas of "digital transfer is "bit perfect"" and "It's an exact copy of the original file" are falsehoods. There are absolutely no perfect, exact copies of electical signals. there is always degredation to higher or lower degrees.DarqueKnight wrote: »My example showed that multiple versions of the same file can result due to noise contamination. They were not perfect 1:1 bit copies, although each file maintained the proper relationship between logical high and logical low.
Exaclty Ray. Also, I wanted to point out that the more times a file is copied, the more error is introduced to the file (to varyingdegrees as in your example above). After the files is copied enough times (fewer for higher indroduction of error, and more times for lower introduction of error) the more audible the errors become when listening to the file.DarqueKnight wrote: »If there is a high degree of amplitude randomness in a digital signal, why wouldn't this affect processing efficiency, especially with regard to digital to analog conversion? If the randomness affects processing efficiency, why wouldn't this translate to audible differences?
I know it was rhetorical, but it would (translate to audible differences).
Taken from a recent Audioholics reply regarding "Club Polk" and Polk speakers:
"I'm yet to hear a Polk speaker that merits more than a sentence and 60 seconds discussion."
My response is: If you need 60 seconds to respond in one sentence, you probably should't be evaluating Polk speakers.....
"Green leaves reveal the heart spoken Khatru"- Jon Anderson
"Have A Little Faith! And Everything You'll Face, Will Jump From Out Right On Into Place! Yeah! Take A Little Time! And Everything You'll Find, Will Move From Gloom Right On Into Shine!"- Arthur Lee -
Jitter can be present in the original recording and this can never be mitigated. It's a permanent fixture to the recording unless they make a new digital master."Appreciation of audio is a completely subjective human experience. Measurements can provide a measure of insight, but are no substitute for human judgment. Why are we looking to reduce a subjective experience to objective criteria anyway? The subtleties of music and audio reproduction are for those who appreciate it. Differentiation by numbers is for those who do not".--Nelson Pass Pass Labs XA25 | EE Avant Pre | EE Mini Max Supreme DAC | MIT Shotgun S1 | Pangea AC14SE MKII | Legend L600 | BlueSound Node 3 - Tubes add soul!
-
Audio is also full of people who don't understand the digital domain, and the differences between digital and analog audio. Let me ask you something DQ...does a file on a hard drive contain Jitter?
Yes, it absolutely CAN.
H9
P.s. Villian and Habanero you guys have lots to learn about digital audio playback."Appreciation of audio is a completely subjective human experience. Measurements can provide a measure of insight, but are no substitute for human judgment. Why are we looking to reduce a subjective experience to objective criteria anyway? The subtleties of music and audio reproduction are for those who appreciate it. Differentiation by numbers is for those who do not".--Nelson Pass Pass Labs XA25 | EE Avant Pre | EE Mini Max Supreme DAC | MIT Shotgun S1 | Pangea AC14SE MKII | Legend L600 | BlueSound Node 3 - Tubes add soul! -
Yes, it absolutely CAN.
They were talking about packet jitter, which only arises in digital transmission. You are correct that sampling jitter, which is a possible artifact of A/D or D/A conversion, will become a noise characteristic of a digital file.Proud and loyal citizen of the Digital Domain and Solid State Country! -
DarqueKnight wrote: »They were talking about packet jitter, which only arises in digital transmission. You are correct that sampling jitter, which is a possible artifact of A/D or D/A conversion, will become a noise characteristic of a digital file.
To be fair he asked if jitter could be present in a file on hard drive. I answered as asked. Can't help if the person posing the question isn't articulate enough to narrow down a vague question. And yes, I have been following this thread (even though they are all over the board) to know what he might have been getting at. But I like to deal with the specific and not the general. His question was asked in a general way."Appreciation of audio is a completely subjective human experience. Measurements can provide a measure of insight, but are no substitute for human judgment. Why are we looking to reduce a subjective experience to objective criteria anyway? The subtleties of music and audio reproduction are for those who appreciate it. Differentiation by numbers is for those who do not".--Nelson Pass Pass Labs XA25 | EE Avant Pre | EE Mini Max Supreme DAC | MIT Shotgun S1 | Pangea AC14SE MKII | Legend L600 | BlueSound Node 3 - Tubes add soul! -
DarqueKnight wrote: »One of the things I continually stress is the importance of the noise performance of cables and components. With regard to Ethernet cables, which are UTP (unshielded twisted pair), one would think that environmental noise would be of concern...and it is.
You certainly need to apply the correct cable to the correct application.
A recent IEEE paper by Leersum et al, "Ethernet Susceptibility to Electric Fast Transients", which was published in the Proceedings of the 2013 IEEE International Symposium on Electromagnetic Compatibility (EMC Europe 2013), Brugge, Belgium, September 2-6, 2013, provided some insight:
Conclusion
It is shown that already a small EFT pulse is capable of disturbing or disrupting Ethernet communication at protocol level, which makes this not just a matter of signal integrity. It seems that the protocol is not designed to handle EFT phenomena in an efficient way. We focused on the quality of the interface, i.e. cables, connectors and feed-thoughs. While affordable high quality cable is readily available, it is more difficult to find low cost and robust connectors without a large variability in performance for EMC, but to decouple data handling electronics from disturbing transients is of paramount importance for interference free data communication. As expected, screened cables help to mitigate the interference, but only when the screens are properly connected on both sides. Further research will focus on the coexistence of power lines and Ethernet cables in the same conduits and on the use of low cost connectors.
A couple of important points from the paper:
An earlier study* that focused on crosstalk showed that EFT bursts on nearby power lines are very effective in causing interference and loss of data on Ethernet cables. The impact of these disturbances, referred to as EFT, on the performance of an Ethernet connection is further investigated.
and
It is expected that digital circuits are more likely to exhibit non-linear failure characteristics compared to the gradual degradation of analogue signals with increasing noise.
*T. Wijnands, Analysis Report: Crosstalk in Cables", Thales Nederland B.V., Internship Report, July 2008.
Discussion
From what I have read so far, it appears that, as with analog signals, cable quality matters with Ethernet's digital signals. It seems that jitter is not the only gremlin that can affect the integrity of Ethernet signals.
More later...[/QUOTE]
I'll get that paper from the research library in the next day or two. I'm curious as to what connector type they are using. The industry for CAT7 was primed for GG45 or Tera Connectors as a new standard but vendors dug heals in for the standard 8 pin that is common.
I'll also be curious how much UTP vs STP or STP wrapped FTP cabling is used in a Navy ship or those that insist on running your LV cabling next to power. Did all the interconnects pass certification? Did they find any brands or model of connectors that were consistent?
Are ships of such a nature that HV and LV cabling need to share the same conduit?
We don't run Ethernet parallel to any power. 3 foot minimum is our separation when it comes even close (common wall etc). If anyone is really worried then they should run certified STP.
So what does this all mean for audio enthusiast that either don't live on a Navy ship or run their Ethernet on top of their house wiring? -
DarqueKnight wrote: »
However, due to noise contamination, there are now three different versions of the pulse sequence. Consider that, in a typical stereo music file, you will be dealing with millions of bits, therefore, the grossly simplified cases represented by A, B, and C translate to three files with multiple millions of differences between them.
This is completely wrong. The transport mechanism is not the same as the data being carried by that mechanism. That electrical pulse sequence does not alter the bits being transmitted. You're implying that 00000001 somehow can become 00000011 in the transport process. If this were the case then we would all see a different version of polkaudio.com every time we visited.
You're statement about analog conversion is irrelevant in this thread because it's topic was about digital cables which are a part of the transport process, not encoding or decoding.
The engineering and science behind the digital realm ensures that the bits and bytes transmitted are the same on the receiving end. If they are not then there are CLEAR and OBVIOUS problems. In the audio world that would be chirps, skips, etc. Digital error detection and correction prevent that unless there is a serious flaw in the transport process.
Everyone is allowed their own perception. If someone does or doesn't hear a difference between digital cables, cool. That's their perception and nobody can argue with or disprove that. It's impossible, so just stop trying and let people enjoy their own experience. Jeeze. -
DarqueKnight wrote: »No, but a file on a hard drive can contain noise.
No, it can contain data. You will have to expound on the 'contain noise'. It's too broad for interpretation.DarqueKnight wrote: »It's not quite that simple. "Ones" and "zeroes" are not being transmitted. Voltage pulses that correspond to logical high ("1") and logical low ("0") are being transmitted. The high and low voltage pulses have nominal values, for example low may be 2 V and high may be 12 V. In reality, high and low correspond to a range of values: low may be any value from 0-5 V and high may be any value 10-15 V.
Consider the sequence 1-0-0-1-1-0-1-1-0-1-1-1: Ideally this would be represented by the following pulse amplitudes according to the scheme given above:
12-2-2-12-12-2-12-12-2-12-12-12
A noise-affected version of the sequence is:
10-0-1-11-11-3-12-14-1-15-10-11
You have to show this in usage however. This could manifest as packet loss. This could manifest as no packet loss but malformed packet (which various protocols would most likely re-transmit) and re-order for correctness.
Also your optimal would be 15-0-0-15-15-0-15-15-0-15-15-15
Your "noisy versions" would be
12-2-2-12-12-2-12-12-2-12-12-12
10-0-1-11-11-3-12-14-1-15-10-11DarqueKnight wrote: »
Here, the demarcation between logical high and logical low is preserved, but the signal looks a lot different than the ideal case.
What happens when a more seriously noise-affected version occurs?:
14.79-4.38-1.13-9.9-13.8-0.3-11.5-0.45-11.5-13.88-10.65
Here, the demarcation between logical high and logical low is preserved, but the signal now approaches the amplitude characteristics of random noise.
Each of these voltage pulse sequences are within spec with regard to amplitude:
A. 12-2-2-12-12-2-12-12-2-12-12-12
B. 10-0-1-11-11-3-12-14-1-15-10-11
C. 14.79-4.38-1.13-9.9-13.8-0.3-11.5-0.45-11.5-13.88-10.65
While the mechanics are certainly sound, the differences, if they are in the signaling thresh hold, are inconsequential. You have a PHY that is designed to either accept 1-5 or 10-15 and ignore everything else. It's not random noise to the transceiver. It's delta has either been hit or missed. 9.9 isn't a near miss, it's a miss and data will be requested again.DarqueKnight wrote: »However, due to noise contamination, there are now three different versions of the pulse sequence. Consider that, in a typical stereo music file, you will be dealing with millions of bits, therefore, the grossly simplified cases represented by A, B, and C translate to three files with multiple millions of differences between them.
Questions:
1. Will all three files require the same processing effort?
2. When the D/A conversion is done, is one file going to be more prone to error than the others, or will they all convert to identical analog files?
That would be a question for someone that develops silicon for NIC's. Now lets say that if the Amplitude voltages are sent with some variance, but they are within the delta of the PHY and they come out to the exact same binary stored in buffer with some timing variance as they were transmitted on the line:
We can say that they are indeed the same exact file
We can say that the timing variance, that the difference in amplitude response, is non-sequiter at this point. The buffer is filled with exact data that took milliseconds or less to fill for seconds or 10's of seconds for playback.DarqueKnight wrote: »Consider another case of the sequence contaminated by electric fast transient noise, so that the following sequence results:
16.79-4.38-5.13-9.9-15.8-6.3-11.5-0.45-11.5-10.88-15.65
The values in red are outside the ranges for high and low. What happens in this case?
That is protocol dependent. TCP would resend, SMB/CIFS would re-request. UDP you are screwed.
I doubt that a $350 AQ Vodka cable would do any better at insuring amplitude response characteristics than my $13 certified BJC cable as long as I'm not on the USS Nimitz or running my Ethernet cable zip tied to my 12/2 20 amp service that is servicing my listening setup I'll take my chances.DarqueKnight wrote: »My example showed that multiple versions of the same file can result due to noise contamination. They were not perfect 1:1 bit copies, although each file maintained the proper relationship between logical high and logical low.
In your example the amplitude response may not be identical. But the transceiver that gets the stream will code out the same binary file regardless.
I'm sure there is some V drop at the 328 foot of cable vs 3 foot. So the spec is most likely designed to work at the extreme and do so reliably.DarqueKnight wrote: »If there is a high degree of amplitude randomness in a digital signal, why wouldn't this affect processing efficiency, especially with regard to digital to analog conversion? If the randomness affects processing efficiency, why wouldn't this translate to audible differences?
Because it doesn't have to be a hard 15 or 0. It's not randomness to the transceiver. It's either 1/0 or trash and let the higher layer protocols whether network, application, session etc work it out.
That is why at least getting a cable that is certified has merit. -
I don't know about others, but I try to get information about a subject from a variety of sources. For example, jitter is a relatively common topic on audio forums. Here is an excellent post, and with good timing (ha ha) for this thread. Monk, I hope your write-up reaches this level of detail.
First paragraph:
"One of the common arguments made against jitter mattering is that: "the data is buffered and clock regenerated in the DAC so jitter won't be there." This makes all the sense in the world. Once we capture the data and then push it out at our will, there shouldn't be a problem. Well, there is a problem. A serious one. Buffering and clock regeneration do not deal with jitter by themselves. I have explained this in words many times but this time I am bringing in some specific data to hopefully put this myth to bed (yeh, wishful thinking )."
http://www.whatsbestforum.com/showthread.php?14957-Yet-another-look-at-Jitter-Clock-Extraction&p=270683&viewfull=1#post270683
Did you bother to try comprehending the fact that Amir is speaking about the bitstream to the DAC and the buffer on the DAC?
He's not speaking to, nor can he speak to, if the file was read from local or remote storage.
Go ahead and ask him if his measurements of the bitstream can let him intrinsically determine if the data is derived from over at network cable or traditional HD. -
This is completely wrong. The transport mechanism is not the same as the data being carried by that mechanism. That electrical pulse sequence does not alter the bits being transmitted. You're implying that 00000001 somehow can become 00000011 in the transport process. If this were the case then we would all see a different version of polkaudio.com every time we visited.
All you have done is reveal your deficiencies in reading comprehension. My example had nothing to do with bits being changed from high to low or vice versa.You're statement about analog conversion is irrelevant in this thread because it's topic was about digital cables which are a part of the transport process, not encoding or decoding.
Again, reading is fundamental.The engineering and science behind the digital realm ensures that the bits and bytes transmitted are the same on the receiving end. If they are not then there are CLEAR and OBVIOUS problems. In the audio world that would be chirps, skips, etc. Digital error detection and correction prevent that unless there is a serious flaw in the transport process.
Again, reading is fundamental.Everyone is allowed their own perception. If someone does or doesn't hear a difference between digital cables, cool. That's their perception and nobody can argue with or disprove that. It's impossible, so just stop trying and let people enjoy their own experience. Jeeze.
You should be preaching this to Habanero Monk. He is the one taking issue with people claiming that they hear differences in Ethernet cables. He is the one challenging people to prove to him that they can hear differences in Ethernet cables.Habanero Monk wrote: »When you take the dubious step of saying it does X/Y/Z to the sound vs another competent CAT6 cable then I have to take issue since I understand how packet switched networks operate.Habanero Monk wrote: »My position is that it's not possible. I have actively put this out there at some other forums. A position that I would like to find a way to actively exploit people and the money they like to seemingly be fleeced of.
Because if they think there is a night and day difference between two Ethernet cables that pass spec and are willing to pay hundreds of dollars. Well it's immoral to let a sucker keep his or her money.Habanero Monk wrote: »All you need to know is there are some chumps here that are all talk. All bark no bite.
As a matter of fact I offered, while at the PETT GTG in Dayton, last summer to put together a computer with dual Radeon cards (that max out the HQV Blu-Ray benchmark) all going to a high quality display (with 2 HDMI inputs).
The display would be calibrated for both inputs. A $20 Belden HDMI and any HDMI cable they wanted to bring. I offered up a ~$600 AQ.Habanero Monk wrote: »Does anyone want to bet on a certified CAT6 cable vs any AQ Ethernet cable you care to pick?Habanero Monk wrote: »It's certainly not a 'fact'. Also some are claiming differences. I'm not testing cables. I'm testing claims. Discern the difference, and then form your opinion. I'm curious how this would work.
Heck if anyone is 90 minutes or so outside of Cinci it would be neat to test on their equipment.Habanero Monk wrote: »I'm saying I'm willing to bet you the cost of a DAC you can't follow the ball under the cup routine when one Ethernet cable is used vs another.
Let me know. I can get a contract drawn up.Habanero Monk wrote: »You said you can discern differences in Ethernet cables. So I'll pay you off with a $1600 DAC if you can.
You either can or you can not. Based on your conviction of your position it's an easy $1600 for you. I can always send the DAC back and give you $1600.
My vested interest is in showing people, rather publicly, that in this instance you are 100% in error of your understanding about how this works in regards to computer based networked audio.
Even if you decline my offer it still proves my point about your convictions. There is no downside for me other than you hitting it 13 out of 15.
I am the one who said I think it's possible that Ethernet cables might make an audible difference, but I have never heard a difference in any coax or optical digital cable I have used.Proud and loyal citizen of the Digital Domain and Solid State Country! -
This is completely wrong. The transport mechanism is not the same as the data being carried by that mechanism. That electrical pulse sequence does not alter the bits being transmitted. You're implying that 00000001 somehow can become 00000011 in the transport process. If this were the case then we would all see a different version of polkaudio.com every time we visited.
You misunderstood what Ray is saying. The transport mechanism(s) is/are affecting the data being transmitted. Ray is not saying thay the data and transport mechanism(s) is/are the same. The electrical pulse sequence most definately does alter the bits being transmitted. This is because (as shown in Rays example) the transport mechanisms including many many things such as: capacitors, resistors, diodes, etc. as well as ethernet cables all affect the electrical impulses sent through them. What makes you think that electrical impulses (making up the data being sent, which makes up the bits you referred to) are not affected by any of the transport "mechanisms"? What makes you think that there is no affect on these electrical impulses?
Ray was not saing that 00000001 can become 000000011. You are implying that extra data was added to the bits. This is not what Ray was saying at all. Ray is saying that the electrical impulses for the data being sent were to varying degrees of accuracy based upon the amount of noise affecting the data. Not that extra data was added, but that the noise altered the data to make it less than exact, while still maintaing the correct bits being received. Do you undertsand the difference?The engineering and science behind the digital realm ensures that the bits and bytes transmitted are the same on the receiving end. If they are not then there are CLEAR and OBVIOUS problems. In the audio world that would be chirps, skips, etc. Digital error detection and correction prevent that unless there is a serious flaw in the transport process.
They may appear to be the same when you look at them on a computer screen or they are analyzed with a scientific measuring device. When something more sensitive (as in the case of your ears, brain and consciousness) are used, it becomes much more apparent that what is received is not equivalent to the orignial source (to varying degrees of course). That's I believe the problem with "digital know-it-alls" is that they cannot use their brains and get to a fine enough detail (scale) to realise that when you get to a fine enough scale, there are electrical problems that show up in digital signals. Since their brains cannot get a hold of this, they think that digital transmissions are "perfect". Again, there are no perfect transfers of electrical signals, period.
When you are referring to chirps, clicks, etc. showing up in music transmitted digitally, you are referring to extreme cases. What about the more subtle cases of error. They are perceptible, but it does not seem that "digital-know-it-alls" can detect them.Everyone is allowed their own perception. If someone does or doesn't hear a difference between digital cables, cool. That's their perception and nobody can argue with or disprove that. It's impossible, so just stop trying and let people enjoy their own experience. Jeeze.
Yes, they are aren't they. Please see my response in the paragraph above this one.
People who suggest that there are differences in ethernet and other digital cables are trying to prevent others from enjoying their own experience by any means. In fact, people promoting better (higher quality) digital cables are promoting an improved audio experience.
Taken from a recent Audioholics reply regarding "Club Polk" and Polk speakers:
"I'm yet to hear a Polk speaker that merits more than a sentence and 60 seconds discussion."
My response is: If you need 60 seconds to respond in one sentence, you probably should't be evaluating Polk speakers.....
"Green leaves reveal the heart spoken Khatru"- Jon Anderson
"Have A Little Faith! And Everything You'll Face, Will Jump From Out Right On Into Place! Yeah! Take A Little Time! And Everything You'll Find, Will Move From Gloom Right On Into Shine!"- Arthur Lee -
Sorry Ray. I just noticed that you responded before I did. Carry on..... :redface:
Taken from a recent Audioholics reply regarding "Club Polk" and Polk speakers:
"I'm yet to hear a Polk speaker that merits more than a sentence and 60 seconds discussion."
My response is: If you need 60 seconds to respond in one sentence, you probably should't be evaluating Polk speakers.....
"Green leaves reveal the heart spoken Khatru"- Jon Anderson
"Have A Little Faith! And Everything You'll Face, Will Jump From Out Right On Into Place! Yeah! Take A Little Time! And Everything You'll Find, Will Move From Gloom Right On Into Shine!"- Arthur Lee -
I don't know about others, but I try to get information about a subject from a variety of sources. For example, jitter is a relatively common topic on audio forums. Here is an excellent post, and with good timing (ha ha) for this thread. Monk, I hope your write-up reaches this level of detail.
First paragraph:
"One of the common arguments made against jitter mattering is that: "the data is buffered and clock regenerated in the DAC so jitter won't be there." This makes all the sense in the world. Once we capture the data and then push it out at our will, there shouldn't be a problem. Well, there is a problem. A serious one. Buffering and clock regeneration do not deal with jitter by themselves. I have explained this in words many times but this time I am bringing in some specific data to hopefully put this myth to bed (yeh, wishful thinking )."
http://www.whatsbestforum.com/showthread.php?14957-Yet-another-look-at-Jitter-Clock-Extraction&p=270683&viewfull=1#post270683
Since we are quoting Amir:
While his explanation of what the J-test is, his reasoning for why it is used at the end is incorrect. The file itself contains no jitter whatsoever. Nor does it simulate "subtle timing inaccuracies." Remember, a file sitting on your hard disk before being played is just a bunch of digital audio samples. it is "perfect" in that regard. We can copy it 100 times and the last copy will be identical to first. So no timing problem exists or can exist in a digital file on a computer.
The J-test signal is designed to cause more jitter on the S/PDIF link. Due to the way the clock is transmitted on that link, the J-test signal can excite them due to the way the digital values are picked in that test signals.
http://www.avsforum.com/t/1531357/jitter/0_100#post_24730005 -
Apparently Monk's reading comprehension skill level is equal to his networking knowledge, which is negligible. He has just made two posts based on a post in the WBF forum. Neither of Monk's post have anything to do with the WBF post. The WBF post is showing jitter is not eliminated when the digital signal is buffered. Rather, Monk is either intentionally trying to distract from that idea, or he just has no idea what he is reading. I suspect it is the second item since he has already shown he does not understand jitter when he even posted the link to a jitter explanation.
Interestingly, the WBF post destroys the entire premise for both Monk and Villian. If buffering does not eliminate jitter then they have no case. Therefore, case closed. Ethernet cables have the potential to affect audio sound quality.Lumin X1 file player, Westminster Labs interconnect cable
Sony XA-5400ES SACD; Pass XP-22 pre; X600.5 amps
Magico S5 MKII Mcast Rose speakers; SPOD spikes
Shunyata Triton v3/Typhon QR on source, Denali 2000 (2) on amps
Shunyata Sigma XLR analog ICs, Sigma speaker cables
Shunyata Sigma HC (2), Sigma Analog, Sigma Digital, Z Anaconda (3) power cables
Mapleshade Samson V.3 four shelf solid maple rack, Micropoint brass footers
Three 20 amp circuits. -
Apparently Monk's reading comprehension skill level is equal to his networking knowledge, which is negligible. He has just made two posts based on a post in the WBF forum. Neither of Monk's post have anything to do with the WBF post. The WBF post is showing jitter is not eliminated when the digital signal is buffered. Rather, Monk is either intentionally trying to distract from that idea, or he just has no idea what he is reading. I suspect it is the second item since he has already shown he does not understand jitter when he even posted the link to a jitter explanation.
Interestingly, the WBF post destroys the entire premise for both Monk and Villian. If buffering does not eliminate jitter then they have no case. Therefore, case closed. Ethernet cables have the potential to affect audio sound quality.
Again: Ask Amir, in that thread, if his jitter measurements would enable him to deduce of whether the data is retrieved from local HD or pulled over the network.
If you read what he is writing about in PLL recovered clock sync inputted over SP/DIF you would see he is talking about the jitter induced error of the PLL and that of the DAC buffer (not computer buffer).
If you had also read the other post of his from AVSForum:
The file itself contains no jitter whatsoever. Nor does it simulate "subtle timing inaccuracies." Remember, a file sitting on your hard disk before being played is just a bunch of digital audio samples.
The buffer on a computer is the file that contains no jitter what so ever. Also keep in mind that JRiver can pull over an entire file to RAM (up to 1 GB).
You are truly out of your depth here and the sad thing is you don't even realize it.
Also find out where I said buffering destroys Jitter in absolute terms. Go ahead now and find it. What I said is what Amir said: Remember, a file sitting on your hard disk before being played is just a bunch of digital audio samples".
Jitter will be introduced once it is read into RAM, when it is read out of RAM, when it is transferred over bus X/Y/Z. It's the READ and WRITE process that will have some form of jitter. Not the storage state. Whether it's RAM, HD, RAM DISK. -
One of the common arguments made against jitter mattering is that: "the data is buffered and clock regenerated in the DAC so jitter won't be there." This makes all the sense in the world. Once we capture the data and then push it out at our will, there shouldn't be a problem. Well, there is a problem. A serious one. Buffering and clock regeneration do not deal with jitter by themselves. I have explained this in words many times but this time I am bringing in some specific data to hopefully put this myth to bed (yeh, wishful thinking ).
Introduction
The way a clock is "regenerated" is to have a local oscillator (clock) that we can change its frequency to eventually match and track the incoming digital stream. As you may know, S/PDIF is a serial digital connection with clock and data intermixed. By using this circuit which is called a Phase Locked Loop (or PLL for short), we are able to extract a clock that is cleaner than the incoming one. This clean up allows us to capture the digital samples reliably.
I've underlined the items of import that BlueFox is having a failure to understand the ramifications of and also his conflation as to what buffer is being referred to.
The data is buffered in the DAC. The clock is regenerated in the DAC. This is also DAC specific since there are many options for syncing clock.
The data buffered on the computer is being confused with the above by BF. -
Also from the thread Blue Fox mentioned:
There are no "inaccuracies of information" - the actual data gets transferred without errors and without any change. The issue is with the clock - cheaper DACs use a clock that is generated by having a PLL lock to the incoming data clock, and that PLL is not always very good at rejecting jitter. Thus the timing of the DAC conversion can be affected, resulting in jitter-induced distortion/noise on the analog output.
I have to thank him for that linked thread. It completely makes my point that there are some that don't understand what is going on here due to failure to launch in their understanding of what is being spoken about.
If one did understand they would never have attempted it's use in fortifying their already indefensible position.
Now I'll just have to wait and see how it can be tied back into a BJC vs some other high $$ Ethernet cable. -
DarqueKnight wrote: »No, but a file on a hard drive can contain noise.
It's not quite that simple. "Ones" and "zeroes" are not being transmitted. Voltage pulses that correspond to logical high ("1") and logical low ("0") are being transmitted. The high and low voltage pulses have nominal values, for example low may be 2 V and high may be 12 V. In reality, high and low correspond to a range of values: low may be any value from 0-5 V and high may be any value 10-15 V.
Consider the sequence 1-0-0-1-1-0-1-1-0-1-1-1: Ideally this would be represented by the following pulse amplitudes according to the scheme given above:
12-2-2-12-12-2-12-12-2-12-12-12
A noise-affected version of the sequence is:
10-0-1-11-11-3-12-14-1-15-10-11
Here, the demarcation between logical high and logical low is preserved, but the signal looks a lot different than the ideal case.
What happens when a more seriously noise-affected version occurs?:
14.79-4.38-1.13-9.9-13.8-0.3-11.5-0.45-11.5-13.88-10.65
Here, the demarcation between logical high and logical low is preserved, but the signal now approaches the amplitude characteristics of random noise.
Each of these voltage pulse sequences are within spec with regard to amplitude:
A. 12-2-2-12-12-2-12-12-2-12-12-12
B. 10-0-1-11-11-3-12-14-1-15-10-11
C. 14.79-4.38-1.13-9.9-13.8-0.3-11.5-0.45-11.5-13.88-10.65
However, due to noise contamination, there are now three different versions of the pulse sequence. Consider that, in a typical stereo music file, you will be dealing with millions of bits, therefore, the grossly simplified cases represented by A, B, and C translate to three files with multiple millions of differences between them.
Questions:
1. Will all three files require the same processing effort?
2. When the D/A conversion is done, is one file going to be more prone to error than the others, or will they all convert to identical analog files?
Consider another case of the sequence contaminated by electric fast transient noise, so that the following sequence results:
16.79-4.38-5.13-9.9-15.8-6.3-11.5-0.45-11.5-10.88-15.65
The values in red are outside the ranges for high and low. What happens in this case?
My example showed that multiple versions of the same file can result due to noise contamination. They were not perfect 1:1 bit copies, although each file maintained the proper relationship between logical high and logical low.
If there is a high degree of amplitude randomness in a digital signal, why wouldn't this affect processing efficiency, especially with regard to digital to analog conversion? If the randomness affects processing efficiency, why wouldn't this translate to audible differences?
If the logical output is the same as the logical input and the result is a 1:1 bit copy, then what else matters? You should be banned for this post. No joke. You've knowingly attempted to mislead anyone reading this thread, and this post is the worst bit of semantics yet. No wonder people here are stuck in this 1960's train of thought and unable to comprehend what does and doesn't effect digital signals..
Completely disgraceful DK.Too many good quotes to list..waiting for some fresh ammo. -
Your example above does a phenominal job of demonstrating exactly why the ideas of "digital transfer is "bit perfect"" and "It's an exact copy of the original file" are falsehoods. There are absolutely no perfect, exact copies of electical signals. there is always degredation to higher or lower degrees.
Sheeple being completely mislead. Sigh.
Wow. That's about all I can say.
DK himself just explained how there IS "Bit perfect" 1:1 digital copies, and you just dismiss that right out the window in the name of ignorance..Too many good quotes to list..waiting for some fresh ammo.
This discussion has been closed.