[kiteki]

 
Quote:

Quote:

My head is really starting to hurt now.  People still don't get it at all, or the thread is wildly off topic,  (As if that never happens...

)

 
The thread did expand out significantly from the OP's comment about USB cables, to incorporate a wide variety of audiophile mythology.

 
 
Well, let's go back to the 'mythology' of USB cables and DAC sound quality then if you wish?
 
 
Here is a comment from an electrical engineer.
 

Spoiler: click

 
"I plan to cover this more in the next article. There is some measurable improvement from using a USB cable with a ferrite "choke" at the DAC end of the cable.  If you don't have a cable that has one, you can buy such a cable for a few bucks or get clamp-on ferrites in the appropriate size from electronics suppliers for a few dollars.

With any USB audio device it never hurts to use a USB port with either nothing else sharing that USB controller (cluster of ports) or only something benign like a mouse. High bandwidth devices like USB WiFi adapters, Bluetooth adapters, video hardware, etc. create more bus contention. While I haven't heard or seen any problems with the DAC on shared controllers (in fact I have routinely shared a controller with the dScope which generates lots of traffic) it's still not a bad idea with any USB audio device.

If your PC has front and rear panel USB connections, try to split any high bandwidth hardware between the two sets. Likewise some laptops have two groups of USB connections. As for other variations, the DAC's extra power filtering helps make it more immune to noise issues which can vary from PC to PC."
 

 
 

 

[liamstrain]

A comment without evidence, regardless of source, is nearly useless. His comment is interesting, but doesn't provide any information we can actually evaluate on its own. I'd be curious to see his evidence, though. 
 
With that said - he's dealing with noise elimination in the USB bus, which makes sense to me, as preventing interference. This wouldn't require any audiophile magic - as he states, a ferrite choke can be added for a few bucks. The cable itself should not a factor in that, provided that it is built to spec. 
 

 

[Headzone]

 
Quote:

 
 
 
Well, let's go back to the 'mythology' of USB cables and DAC sound quality then if you wish?
 
 
Here is a comment from an electrical engineer.
 

Spoiler: click

 
"I plan to cover this more in the next article. There is some measurable improvement from using a USB cable with a ferrite "choke" at the DAC end of the cable.  If you don't have a cable that has one, you can buy such a cable for a few bucks or get clamp-on ferrites in the appropriate size from electronics suppliers for a few dollars.

With any USB audio device it never hurts to use a USB port with either nothing else sharing that USB controller (cluster of ports) or only something benign like a mouse. High bandwidth devices like USB WiFi adapters, Bluetooth adapters, video hardware, etc. create more bus contention. While I haven't heard or seen any problems with the DAC on shared controllers (in fact I have routinely shared a controller with the dScope which generates lots of traffic) it's still not a bad idea with any USB audio device.

If your PC has front and rear panel USB connections, try to split any high bandwidth hardware between the two sets. Likewise some laptops have two groups of USB connections. As for other variations, the DAC's extra power filtering helps make it more immune to noise issues which can vary from PC to PC."
 

 
 

I remember old PC's had those ferrite chokes inside them. They were usually put over those wires which turn your pc on. So I can improve my audio quality with one?
 
Hold on, does the guy suggest using DAC in a different usb bus? So I should plug my DAC into the front usb port of my pc, and mouse at the back?

 

[mikeaj]

That's about filtering the power so devices downstream end up with cleaner power supply rails, which of course is good (usually...check exactly what's going on with regards to any filtering).  Somewhat arguably, that's not really a cable but a cable + ferrite bead.  People were focusing on the cable aspect alone for these discussions.  The context there is about a USB audio DAC that get is getting power from the USB connection.
 
For a digital communications system, if all the data gets to the destination without error and on time: mission accomplished.  There's nothing there to be improved with regards to that functionality.  Enough said.
 
The signaling for USB is in the form of voltage differences across two wires, so yes, that's an analog process like for any communications system, but if as long as the data is recovered without error on the other side, you're good to go.  A little noise and weird bumps in the waveform don't matter, unless they actually cause an error.  You need to look at the probability that there are errors, which should be small enough as to be considered zero in many situations.  On a bad (long mostly) cable, the probability of error should be slightly higher, maybe still more or less zero.
 
If the power supply rails are being used, then that's a different matter than just getting the data from point A to B.  Likewise, if the timing of signal transitions over the cable is being used to clock out data in some sensitive D/A application like for everybody's favorite cheaper S/PDIF implementations, then that becomes a relevant factor to that process; it's no longer just about getting the data across.
 
Differences between ordinary cables in these supplemental roles are usually pretty small though, mostly irrelevant compared to the hardware the cables are connecting together.

 

[kiteki]

 
     Quote:

A comment without evidence, regardless of source, is nearly useless.

 
Quote:

 
With that said - he's dealing with noise elimination in the USB bus, which makes sense to me, as preventing interference.

 
So does it have to 'make sense', or have evidence, which one?
 
I provided evidence to support my assertions, like the Apple earbud versus Sony ZX700, yet you've waved off my assertions as 'mythology' anyway.
 
 
 
Quote:

That's about filtering the power so devices downstream end up with cleaner power supply rails, which of course is good (usually...check exactly what's going on with regards to any filtering). Somewhat arguably, that's not really a cable but a cable + ferrite bead. People were focusing on the cable aspect alone for these discussions. The context there is about a USB audio DAC that get is getting power from the USB connection.

 
A lot of the theories on why X USB cables are better than Y, have to do with power supply.  It's just there are close to zero measurements on this, and zero blind tests, as per usual.
 
In my personal view, the difference is most likely low-level noise with little importance, but people with $2000+ DAC's tend to want peace of mind and get a 'better' USB cable at a fraction of the total cost.
 

 

[liamstrain]

Evidence. My makes sense comment was not to say that I accepted it out of hand. Just that I think I see what he's getting at. But I'd want evidence before I ran around telling everyone to buy up bigger and better ferrite chokes with silver quantum wrappers and unobtanium counter windings. 
 
As for your ZX700 v Earbud post - honestly, I couldn't tell what you were trying to prove by posting that data. You just posted someone's data on two unrelated headphones... headphones which not only sound different, but have different corresponding measurements (just as we would expect). What were you trying to show? Bear in mind - at no point in any of my arguments am I saying that measurements are used to prove which headphone sounds BETTER. The whole point, is identifying DIFFERENCES. Something measurement excels at. 

 

[mikeaj]

It does make sense, but I don't think people would generally call that a matter of "interference".  
 
Is there much to disbelieve about ferrite beads or filtering in general making a measurable difference (and often an improvement) in some products?  I like evidence too, but my angle here is about "how much?" rather than "is it true?" since in this scenario it is a very non-controversial statement.

 

[kiteki]

 
Quote:

 
As for your ZX700 v Earbud post - honestly, I couldn't tell what you were trying to prove by posting that data. You just posted someone's data on two unrelated headphones... headphones which not only sound different, but have different corresponding measurements (just as we would expect). What were you trying to show?

 
You've asked me to "provide evidence" or "support my assertion" several times, so I did so.
 
Recap.
 

 
Similarly, there is nothing to suggest that eye/ear/brain system is more sensitive than well designed measurement instrumentation, far from it.

 

 
Quote:

Carbon-dating is an exact science, audio is an inexact science.

 
What about audio exempts it from being an "exact science?" What do you have to support this assertion?
 
there is nothing about designing audio equipment (or electrical engineering in general) that is magic or alchemy - the principles are known, testable, repeatable and predictable.

 

Originally Posted by liamstrain /img/forum/go_quote.gif
 
1. That it doesn't make sense to you, doesn't make it any less true.
 
2. If you want to call into question the accuracy and sensitivity of measuring equipment, you are free too - provided you can support your assertions. It is not enough to say we are "assuming everything audible is visible"
 
The burden is on you to provide examples of something audible that is not measurable, before you can claim that humans can hear something our technology cannot.

 

 

I have provided those examples. Identifying human voices. Identifying sound-stage, imaging and layering in an IEM or headphone, etc.

 
Human voices: computers can do this, measurements can do this. Voice print for one, and as I stated before, all we are looking for is identifying a difference between them, not the content and meaning of the difference.
 
Identifying sound-stage, imaging and layering in an IEM or headphone, etc. - You have not shown that measurements cannot evaluate this.

 

 

How do I show that measurements can't evaluate something? I don't know how to do that, but I'll work on it. It's an intuitive conclusion, you can hear it, and the data for it isn't there.

 
1. First show conclusively that people hear X difference between two headphones.  2. Then show that there is no measurement data which demonstrates that difference. I believe soundstage, etc. are in fact data we can measure via instruments - as for where, precisely, that hides (e.g. what characteristic of headphone performance is responsible), is another matter. I would expect that attack/decay performance at various frequencies (visible in square wave) would be at least partially responsible for perceived separation and soundstage.

 
I answered all of this in post #130
 
1. I think we know, conclusively, that there are X differences between the Apple earbud and the Sony ZX700 (this was a random example).
 
2. In the measurements (which are the most conclusive/extensive which exist), what is there to support these differences? 
 
 
If you blanked out the headphones on those sheets, and only looked at the data, would an expert of any kind, have any idea how the headphones in question would sound, prior to listening to them?  The answer is no.  That's where my original assertion came from, and this is the evidence, that audio is an inexact science, and that the human perception is more exact at discerning the differences.
 
It's disrespectful that you're diverting the conversation to sarcasm.
 
 

 

[liamstrain]

 
Quote:

 
1. I think we know, conclusively, that there are X differences between the Apple earbud and the Sony ZX700 (this was a random example).
 
2. In the measurements (which are the most conclusive/extensive which exist), what is there to support these differences? 
 
 
If you blanked out the headphones on those sheets, and only looked at the data, would an expert of any kind, have any idea how the headphones in question would sound, prior to listening to them?  The answer is no.  That's where my original assertion came from, and this is the evidence, that audio is an inexact science, and that the human perception is more exact at discerning the differences.
 
 
 
 

 
1. Yes.
2. Quite a lot - the frequency response is very different, as is the square wave and impulse performance. That is all very clear in the data. 
3. If you blanked out the headphones and only looked at the data, just about any expert in this field would be able to tell you what to expect from those headphones. You can get a good picture of the sound signature from the frequency response, and the other data will help you fill out the rest. The answer is Yes - you do have quite a good idea of how the headphone would sound. 
 
You have not backed up your assertions at all. 

 

[kiteki]

 

Originally Posted by liamstrain /img/forum/go_quote.gif
 
3. If you blanked out the headphones and only looked at the data, just about any expert in this field would be able to tell you what to expect from those headphones.

 
Do you have any evidence of that?
 
 

 

[liamstrain]

 
Quote:

 
 
Do you have any evidence of that?
 
 

 
Aside from my personal experience and the testimony of my friends and co-workers in the field? No I do not. It would be interesting to go collect it though. A poll of sound engineers would be interesting - though I don't know if I have the time to follow with a test to see if they could match a headphone to its charts... I'll see if I can put one together and send it around.
 
Do you have any evidence for your assertion that would render my experiment unnecessary? 

 

[kwkarth]

 
Quote:

 
 
I thought it was an analogue signal with voltage variances, as in on and off representing the 1s and 0s. Contained within that is another signal to do with timing to make sure the 1s and 0s are going at the correct speed.

Although the 1's and 0's are represented by voltage thresholds, any inaccuracy of representation of that data is interpreted b the system as either a 1 or a 0.  Nothing else.  If the interpretation is incorrect, it bears no relationship to any analog audio information whatsoever and could never manifest itself as anything but audio gibberish or silence.  There can never be anything even remotely subtle about that.
 

 

[Prog Rock Man]

 
Quote:

 
Although the 1's and 0's are represented by voltage thresholds, any inaccuracy of representation of that data is interpreted b the system as either a 1 or a 0.  Nothing else.  If the interpretation is incorrect, it bears no relationship to any analog audio information whatsoever and could never manifest itself as anything but audio gibberish or silence.  There can never be anything even remotely subtle about that.
 

 
That is what I have thought now for some time. If the DAC stops reading the 1s and 0s then you will get silence. Otherwise it reads the 1s and 0s and so long as the timing is not so far off that jitter impacts, no part of the signal can affect sound quality as described my many an audiophile.

 

[liamstrain]

 
Quote:

 
 
That is what I have thought now for some time. If the DAC stops reading the 1s and 0s then you will get silence. Otherwise it reads the 1s and 0s and so long as the timing is not so far off that jitter impacts, no part of the signal can affect sound quality as described my many an audiophile.

 
Correct - there would be no way for a cable to affect just the 0 and 1's that encode the bass, for instance.

 

[proton007]

I think the digital signal part has already been addressed well. Regarding the power part, I can say this. One, the USB power is considered to be one of the cleaner sources out there.
Secondly, digital chips work within a certain min/max range. If you look at the DAC datasheet, most of them use VDD as reference. So it doesn't matter what VDD is, as long as its between 3.3-5V. Thats a big range, even if you consider HF noise. So unless you USB cable is so bad it cannot even transfer 3.3V reliably, it should not cause an issue.