[gregorio]

  8.28V is not = 1. It is = 0.92. The resulting signal will be inaccurate, that is why it is important to feed a signal with good quality into your transistors.

 
There is no 0.92 in binary logic. Zero or one are the ONLY permissible results. 8.28V = 1. Therefore the resulting signal will be absolutely perfectly accurate!!!
 
G

 

[lbbef]

It's a waste of my time and effort trying to reason with you when you just don't understand how digital electronics work. This is physics, not something that I made up.

Like I said, digital electronics is not something that is easily understood and is often misunderstood. To fully understand digital electronics, you need to major in computer engineering or computer science.

A DAC can't tell that 8.7V is one. It will take 8.7V in and output an imperfect output signal. The output will be close to one, but is still inaccurate.

 

[manpowre]

It's a waste of my time and effort trying to reason with you when you just don't understand how digital electronics work. This is physics, not something that I made up.

Like I said, digital electronics is not something that is easily understood and is often misunderstood. To fully understand digital electronics, you need to major in computer engineering or computer science.

A DAC can't tell that 8.7V is one. It will take 8.7V in and output an imperfect output signal. The output will be close to one, but is still inaccurate.

 
Its funny to see what Headmania.com wrote about letting the USB cable stay connected, and sound better after hours and days !
https://headmania.org/2015/08/18/schiit-yggdrasil-dac-review/
 
Let me quote:
"One thing that I have to point out before the comparison is that Analog Dac also sounds considerably better if you leave it continuously on. Actually, it also takes 2-3 days before it stabilizes and sounds at its best. Another annoying thing about it is that if you unplug the usb cable and interconnects it also looses from the sound quality as it becomes flatter and less visceral. You have to wait for another ~2 hours after you plug them back in. Don’t get me wrong, Analog Dac still sounds very good even when cold, but not at its full potential."
 
Digital compnents exist of analog pieces. Sure let them warm up and get its work start and stabilize and allow the components to settle with temperature.
 
When in fact that the "good" cables for 400 dollars are a good sales trick.. why should a guy that invested 2500 dollars in a DAC, 2000 dollars on a laptop, 5000 dollars on a Amp, and another 5000 dollars on loudspeakers, and another 2000 dollars on analogue cables (Im not arguing against analogue cables, thats an entire different discussion). lets sell this guy a 500 dollar USB cable. use some silver and good connectors (which you can find in cables that cost 25 dollars).
 
In fact the issue is not the cable, but the 8khz noise and impedance and how good the cable can transfer a bit-perfect signal on the DAC's clock signal. 
 
The discussion never goes into how is the DAC's affected by the 8khz noise ? Because its hard to measure. I searched this up, and only a few guys ever did that. And you dont see that as a reference in eg. headmania.com.

 

[manpowre]

Take this with a grain of salt since he is reviewing a $400 USB cable here but:

"The signal that moves along a USB cable isn’t digital – it most definitely is NOT ones and zeroes – but an electrical-pulse representation of those ones and zeroes. This analogue signal is therefore prone to disturbance from EMI emanating from the host computer and electrical noise arrive over the air, otherwise known as RFI. Greater vulnerability to noise can degrade a cable’s ability to do its job: transfer data from computer to DAC.

Digital audio transfer from computer host to DAC uses the isochronous transfer method that doesn’t specify the error-correcting data packet resends of the bulk transfer method used for moving files i.e. when data packet arrival timing is inconsequential to the outcome.

In the digital audio world, meeting the USB cable specification is only part of the ‘better sound’ equation. The cable geometry, materials used – the conductor, the dielectric and termination plugs – and shielding all influence the cable’s immunity to electrical noise pollution, the rise times of the electrical pulses being carried and the arrival timing of those pulses."


From: http://www.digitalaudioreview.net/2015/09/curious-for-the-curious-a-knockout-usb-cable-from-australia/

 
The sound is still transferred in Async mode even in Isync version ! The error check is either: Keep packet or drop it. No new packet will arrive to replace it if its dropped, as clock continues from DAC, and new packet of audio. Jitter is the sender device's ability to send the packets along with the clock from the DAC.
 
The 8khz noise has been proven to be able to measure and affect the dac chip. It is very low, max 20-30 desibel from silent floor. If the DAC is affected by 8khz noise, and amplified through a 250w mono block, it can me measured. But only if DAC is affected by it. Most high-end dac's reduce the noise on the 8khz area before sending digital signal to DAC chip. Therefore it can't or barely measured on the analogue output. Atleast so little that it can't be audible. I have a Sabre Oppo DAC, where I can hear a 8khz mono tone appear on the right channel and distortion on left channel when Amp is cold. Once warmed up, it plays out on both channels.

 

[Articnoise]

It's a waste of my time and effort trying to reason with you when you just don't understand how digital electronics work. This is physics, not something that I made up.

Like I said, digital electronics is not something that is easily understood and is often misunderstood. To fully understand digital electronics, you need to major in computer engineering or computer science.

A DAC can't tell that 8.7V is one. It will take 8.7V in and output an imperfect output signal. The output will be close to one, but is still inaccurate.

 

I agree it’s a big waste of time! They will never understand that there is no 1 and 0 in analogue streaming of digital data. The difference between digital cables is not something that is only revealed in audio it’s the same when streaming movies from the internet or a Blu-ray player. 

 

[manpowre]

   

I agree it’s a big waste of time! They will never understand that there is no 1 and 0 in analogue streaming of digital data. The difference between digital cables is not something that is only revealed in audio it’s the same when streaming movies from the internet or a Blu-ray player. 

Actually its not the same as streaming from internet or mvoie from blueray player. 2 different protocolls from async audio stream on usb. Thats the thing. and a 0 and 1 is not just a 0 and 1. It also the impedance on cable, and the wavelength the 0 and 1 is being carried on. this is why long cables doesnt work well. And the clock signal from DAC in async mode is not something you have on HDMI or stream from internet. As if the playing unit is off with the clock, you get jitter, and if a packet is corrupt is not resent as with TCPIP transmission.
 
But packet dropouts is possible to measure on the analogue side atleast.

 

[gregorio]

  and a 0 and 1 is not just a 0 and 1.

 
Either it is just 0s and 1s, in which case it's binary digits (digital) or it's not just 0s and 1s, in which case it's not binary digits (digital). If it's not just 0s and 1s (and therefore not digital), you are obviously going to need something other than a Digital to Analogue Converter because by definition, a DAC only converts digital (binary digits)! I can't see what is so difficult to understand here?
 

  It also the impedance on cable, and the wavelength the 0 and 1 is being carried on. this is why long cables doesnt work well.

 
Either a USB cable has been appropriately designed to carry a USB signal or it hasn't. If it hasn't, then by definition it is either not a USB cable or it is a faulty USB cable! AmazonBasics USB cables (for example) have been appropriately designed to carry USB signals, are rarely faulty and do indeed carry USB signals perfectly adequately.
 
Yes, there are different protocols used to transfer digital audio, some generic digital data protocols and some specific to digital audio data. If a DAC is a USB DAC, then by definition it must be able to resolve binary digits from a USB signal, IE., It must be able to resolve binary digits from imperfect square waves. The same is true whatever other protocols the DAC has been designed to accept; SPDIF, AES/EBU or HDMI for example.
 
G

 

[lbbef]

You are an audio engineer, not a computer engineer.

A DAC in analogue domain does not convert and accept 0(eg. 0V) and 1(eg. 5V) It accepts a wide range of voltages between 0V and 5V.

When my computer is back, I'll draw up a circuit diagram of a binary input DAC and you'll understand what we mean.

And how would you explain the results of the various double blind tests that people have done on USB cables?

 

[Mediahound]

   
Either it is just 0s and 1s, in which case it's binary digits (digital) or it's not just 0s and 1s, in which case it's not binary digits (digital). If it's not just 0s and 1s (and therefore not digital), you are obviously going to need something other than a Digital to Analogue Converter because by definition, a DAC only converts digital (binary digits)! I can't see what is so difficult to understand here?
 
 
Either a USB cable has been appropriately designed to carry a USB signal or it hasn't. If it hasn't, then by definition it is either not a USB cable or it is a faulty USB cable! AmazonBasics USB cables (for example) have been appropriately designed to carry USB signals, are rarely faulty and do indeed carry USB signals perfectly adequately.
 
Yes, there are different protocols used to transfer digital audio, some generic digital data protocols and some specific to digital audio data. If a DAC is a USB DAC, then by definition it must be able to resolve binary digits from a USB signal, IE., It must be able to resolve binary digits from imperfect square waves. The same is true whatever other protocols the DAC has been designed to accept; SPDIF, AES/EBU or HDMI for example.
 
G

Yes, it's a stream of digital bits (1's and 0's) but there is an intermediate state that can occur before the signal is so corrupt it cannot be read and cuts out. There is circuitry in the DAC which tries to repair a broken data stream by buffering and error correction but this is based on best possible estimates and is not necessarily exactly what is sent. You might experience noise or micro noise which would affect the sound while the hardware is working overtime doing a number of re-tries trying to repair the bitstream and doing error correction over and over, which can affect the audio quality. This is why different USB cables (and USB audio clean up devices) can and do affect the audio you end up hearing. 

 

[gregorio]

  Yes, it's a stream of digital bits (1's and 0's) but there is an intermediate state that can occur before the signal is so corrupt it cannot be read and cuts out.

 
I'm not sure what you mean by this? There is always a signal and there is no "intermediate state", digital by definition is only 0s or 1s, there are no other "states".
 

  There is circuitry in the DAC which tries to repair a broken data stream by buffering and error correction but this is based on best possible estimates and is not necessarily exactly what is sent.

 
Actually, error correction is generally excellent and well below audibility, at least until there are so many errors that the error correction is overwhelmed.
 

  You might experience noise or micro noise which would affect the sound while the hardware is working overtime doing a number of re-tries trying to repair the bitstream and doing error correction over and over, which can affect the audio quality.

 
1. Noise (micro or otherwise) does not affect the digital data, that's the point of it being digital in the first place!
2. The hardware isn't "working overtime", it's either working within specification or it's not working correctly.
3. Error correction is not repeated ad infinitum.
 

  This is why different USB cables (and USB audio clean up devices) can and do affect the audio you end up hearing. 

 
No, this is why different USB cables (and USB clean up devices) cannot affect the audio you end up hearing!
 
G

 

[Mediahound]

I'm not sure what you mean by this? There is always a signal and there is no "intermediate state", digital by definition is only 0s or 1s, there are no other "states".


Actually, error correction is generally excellent and well below audibility, at least until there are so many errors that the error correction is overwhelmed.


1. Noise (micro or otherwise) does not affect the digital data, that's the point of it being digital in the first place!
2. The hardware isn't "working overtime", it's either working within specification or it's not working correctly.
3. Error correction is not repeated ad infinitum.


No, this is why different USB cables (and USB clean up devices) cannot affect the audio you end up hearing!

G

Nope. As the noise increases in the digital stream the redundancy must also increase in order to receive an error free signal. This decreases the amount of information you can send per unit time (capacity). The signal can drop out, yes, but also the signal to noise ratio of the analog output can be degraded, changing the sound detrimentally.

Here's a short video which basically outlines how it works:


[VIDEO]http://youtu.be/cBBTWcHkVVY[/VIDEO]

 

[CanadianMaestro]

I agree with gregorio above. Packets of 0's and 1's.
 
The components should have low jitter. Cables cannot correct errors in timing due to poor design of digital/analog processors. So no point in spending lotsa money on USB cables.
 
My USB cable is Furutech's Formula-2 from Parts Connexion in Canada. Inexpensive and reliable at 0.5m length for my floor system.

 

[CanadianMaestro]

 

 
Curious from your signature: Did the Cardas Clear cable make a significant difference on your 800? Costs almost as much as the 800 on some sites!

 

[Mediahound]

Curious from your signature: Did the Cardas Clear cable make a significant difference on your 800? Costs almost as much as the 800 on some sites!

It sure did. But check eBay because you can get it for a lot less.

 

[gregorio]

Nope. As the noise increases in the digital stream the redundancy must also increase in order to receive an error free signal. This decreases the amount of information you can send per unit time (capacity).

 
True but we're still well below 0.5% of the maximum 480mb/s bandwidth of USB 2 (for say uncompressed redbook digital audio)!
 

The signal can drop out, yes, but also the signal to noise ratio of the analog output can be degraded, changing the sound detrimentally.

 
You (and your linked video) demonstrates how even the complete loss of the digital signal has no effect on the data recovery and therefore has no effect on the eventual analogue outputs. Noise which doesn't affect the DACs ability to differentiate between a 0 and a 1 has no effect, data transmission is perfect and there are no errors to correct. Noise which is so great it obliterates the bit change threshold is effectively the same as a complete loss of signal, it has to be error corrected and therefore also has no effect on the eventual analogue output! The exception of course is where the noise obliterates the signal (or the signal is lost) for a long enough duration to overwhelm the error correction, then there will be an affect on the eventual analogue output.
 
G