[Leo-]

@pinnahertz very good words, RF is definitely the issue that gets transmitted through digital inputs. It's bad with USB, typically not that much with Ethernet (although badly constructed cables with shorted shields can ruin them) and optical inputs are typically immune to it (but one needs to watch out for bad transmitters problems though). I don't know much about Coax though, seems to be quite a mixed bag.

Regarding noise interference, there are basically two types of noise. I've already hinted at this before in the thread, which got lost in the noise (sic) of some other forum members who don't grasp the basic concepts of signal transmission. There is signal-independent noise - like mains hum that is basically AC signal that "leaks" due to magnetic effects on transformers and is of constant level and is simply amplified by the system. If you play jazz, classical or rock it will always be there with same intensity. This type of noise is not a big real issue since our brains are quite efficient in filtering out this sound, and the signal is not affected at all.

The bad noise is signal-correlated noise, which means that the analog signal gets affected by the noise injected in the system. This is typical of RF noise, which interfere with components like capacitors which get discharged at very high frequencies and this does indeed affect the signal. The RF noise itself is inaudible, however in the presence of RF noise the analog components will produce an altered signal which is then perceived as altered sound.

When you put a choke on a USB cable feeding a Dave, you reduce some RF frequencies that reach either Dave's PS or other analog parts. Dave should be insensitive to timing changes in the digital signal, however in theory the chokes affect the raise times of the digital signal (remember that the choke affects any current changes in the cable) which can adversely affect the timings of the signal, so folks avoid chokes if you can.

 

[castleofargh]

the Odac measures slightly better with ferrite(and it's not limited to mine). the cable provided had some, so is it considered part of the design, or does it need to be inside the box to count? ^_^
the fact that it's a cheap old usb powered device makes it more forgivable to me than when talking about a 10k$ DAC. I guess I'm a subjectivist in that way.

warning, crazy person ranting!

I trust evidence over people and find discussing noise without any measurement completely ludicrous. just like assuming a causation from a sighted test about fooling around with cables. could be right, could be wrong, it doesn't matter because we have no control to help checking if anything is legit. pure waste of time. even if we were to admit that some impression is real(not yet established), and somehow accept that the correlation is in this specific case, causation(not established and often false). how do we know that it's relevant for another DAC or even the same DAC in a different house with a different computer? how do we know when what the dude heard as better, was really better or personal taste and assumptions?
I'm amazed that anybody can be satisfied after reaching conclusions that way. we're discussing stuff like digital cables and how they somehow should filter a noise which apparently we don't want measured in any way for some reason. because you know, fidelity, that stuff defined entirely by measurements compared to a reference, it's not as good as subjective impression from a super flawed listening "test".
I can't help but being sarcastic, the all thing make no sense.

a few years back all the rage was jitter, everybody was hearing it and how much better the sound was when the anti jitter whatever was stamped on the box. turns out humans are probably not able to notice jitter at the levels often found in consumer gears(while playing music content). in the end, most people don't have a clue what jitter sounds like or how different sorts of jitter might sound different. it's all made up, the marketing guy told us to fear jitter and everybody started "hearing" it.
next comes ringing and dac filters, the time smearing jokes. so now I buy the DAC with the anti time smearing written on the box and anything I notice as a difference (or anything I make up in my head) is thanks to the reduced ringing. and no matter how controlled experiments suggest we don't hear the ringing at 20-something khz from the band limiting. "hermaggerd! ringinrgg! it sounds so bad on those DACs that don't market anti time smearing tech. "
when did technical issues become fashionable? "oh dear you still focus on jitter? poor thing, it's been 3 years since anybody famous cared about jitter. were you living in a cave? now it's all about noise from the computer. it's completely different, but also the same."

I have no doubt that some gears are better than others and that some DACs are better at blocking particularly bad noises from unreasonably bad computers, and power grids, and leaky microwave we keep turned on for some reason. but paranoia and taking all the drugs in advance in case we might have one sickness, well it's not clever. proper diagnostic is what leads to a good treatment, if needed. and proper diagnostic, that's not a sighted test. never was, never will be.

 

[Leo-]

I trust evidence over people and find discussing noise without any measurement completely ludicrous. just like assuming a causation from a sighted test about fooling around with cables. could be right, could be wrong, it doesn't matter because we have no control to help checking if anything is legit. pure waste of time. even if we were to admit that some impression is real(not yet established), and somehow accept that the correlation is in this specific case, causation(not established and often false). how do we know that it's relevant for another DAC or even the same DAC in a different house with a different computer? how do we know when what the dude heard as better, was really better or personal taste and assumptions?
I'm amazed that anybody can be satisfied after reaching conclusions that way. we're discussing stuff like digital cables and how they somehow should filter a noise which apparently we don't want measured in any way for some reason. because you know, fidelity, that stuff defined entirely by measurements compared to a reference, it's not as good as subjective impression from a super flawed listening "test".
I can't help but being sarcastic, the all thing make no sense..

This is true, and one hint that typically can be seen is when "detail" and "brightness" are noticed not only when assessing digital cables. Some sound artifacts can be created e.g. due to excessive capacitance in analog path that may induce people to perceive sound as brighter or more detailed when in reality the signal is actually being distorted on the way. We have that sometimes what may sound more pleasant to the ears may actually be due to a degradation in the sound quality.

 

[theorist]

What's an "$80 Pro DAC" look like?

If an $80 Pro DAC, regardless of how it looked, sounded as good as a mega-dollar DAC, like a Dave, don't you think that market rationality would prevail, at least for most of us, and we would all be using these $80 Pro DACs! But of course we don't because the (likely) accurate analytical sound the $80 Pro DAC produces, while fine for commercial studio mixing, etc, does not sound anywhere near as 'good' to the consumer's ears for enjoyable musicality as does the Dave. This similar logic applies to generic and dedicated audio USB cables as well. They do sound different and this is readily apparent in blind listening testing as my post #205 in this thread confirmed.

 

[Clive101]

What? No placebo comments? But....but....I....trying....to....hold....back....ARRGG!
It's hard for me not to be analytical in casual listening, but if the music is good I tend to relax and push that stuff to the background. If there's something wrong that should not be there, I'll hear it.

What's an "$80 Pro DAC" look like?

Ok you made me smile about the placebo.
Below is the $80.00 in answer to Triode but looking at the product it may not be a DAC...?

A typical DAC used in commercial studios would be the AVID HD I/O or the Apogee Symphony. However, these are typical big ADC/DAC units with 16/32 ins and outs, all the bells and whistles, price tags well into the 4 figures and not really applicable to audiophiles. However, there is a large range of pro audio ADC/DACs which are designed for musicians and smaller project studios, fewer ins and outs and great value for money. A typical one would be the (apparently) best selling USB audio interface in the world, the Focurite Scarlett but surprisingly competently designed for the price is the Behringer UMC204HD, actually it's performance is almost shocking considering the quality of conversion and the amount of functionality you get for just $80.

G

 

[ev13wt]

21 pages, eh?

Wow.

 

[pinnahertz]

@pinnahertz very good words, RF is definitely the issue that gets transmitted through digital inputs. It's bad with USB, typically not that much with Ethernet (although badly constructed cables with shorted shields can ruin them) and optical inputs are typically immune to it (but one needs to watch out for bad transmitters problems though). I don't know much about Coax though, seems to be quite a mixed bag.

Regarding noise interference, there are basically two types of noise. I've already hinted at this before in the thread, which got lost in the noise (sic) of some other forum members who don't grasp the basic concepts of signal transmission. There is signal-independent noise - like mains hum that is basically AC signal that "leaks" due to magnetic effects on transformers and is of constant level and is simply amplified by the system. If you play jazz, classical or rock it will always be there with same intensity. This type of noise is not a big real issue since our brains are quite efficient in filtering out this sound, and the signal is not affected at all.

The bad noise is signal-correlated noise, which means that the analog signal gets affected by the noise injected in the system. This is typical of RF noise, which interfere with components like capacitors which get discharged at very high frequencies and this does indeed affect the signal. The RF noise itself is inaudible, however in the presence of RF noise the analog components will produce an altered signal which is then perceived as altered sound.

When you put a choke on a USB cable feeding a Dave, you reduce some RF frequencies that reach either Dave's PS or other analog parts. Dave should be insensitive to timing changes in the digital signal, however in theory the chokes affect the raise times of the digital signal (remember that the choke affects any current changes in the cable) which can adversely affect the timings of the signal, so folks avoid chokes if you can.

I'd add a few to your two types. Demodulated RFI can be audible on its own. I live 5 miles from a 50kW directional AM station in the main lobe, for example. That RF gets demodulated everywhere that's not had some form of RFI protection, and you get to hear the station in full fidelity. FM signals (broadcast or 2-way)can also be demodulated when the coupling forms a slope detector. RFI can also intermod with internal clocks forming new audible signals that are not related to audio. So add that in with your description of noise.

I think by "choke" you're referring to the ferrite bead/chunk on some USB cables. A few points on that. It's not actually a simple "choke", but rather it contains R, L and C, and is a resonant circuit. Every bead has a peak impedance frequency, in and of itself. However, it also interacts with the rest of the circuit it's installed on, the result being a particular impedance response curve that may or may not help with a particular RFI situation. In fact, there are some conditions where the bead actually makes things worse. It wouldn't be correct to say that the bead affects any current change in the cable, though, and it doesn't necessarily adversely affect timing or signals. It entirely depends on how the entire circuit is tuned. It's not that ferrites should be avoided, it's that they should be properly applied to the total design. Proper application requires full knowledge of the entire circuit. I'm not sure how carefully that's done on a USB cable.

You mentioned Ethernet...an Ethernet interface is essentially a set of balanced line receivers with common-mode filter. It has to be, most Ethernet cables are unshielded twisted pair, so for that sort of interconnect to transmit uncorrupted data it must have extensive CMRR. Using shielded Cat5/6 may help, but may also create a ground loop if the shield is grounded at both ends. Typically the switch should be grounded, the devices not. It's still a point of RFI entry, but probably minimal. Coax would be a ground loop possibility too because the shield grounded at both ends and the signal is carried unbalanced.

 

[gregorio]

Perhaps I should had been more detailed with my question, in a domestic situation when you listen with your trained ear and expertise you would notice noise to a greater extent than for example me. Buy what you say it's a yes !

Probably but you're missing the point! Both of us would hear the noise on the recording and/or from the audio system far more clearly in a well isolated studio. The higher the noise of the listening environment, the less you will notice and be able to identify the noise from your system, hence why we spend so much time and money isolating our studios in the first place! This is the exact opposite of what you have suggested, that somehow system noise would be more noticeable in an environment with a higher noise floor.

I think part of the confusion here is that galvanic isolation is all that is required for perfect RF immunity. It's one method, a good one, you just can't use it for everything, and there are many possibly entries for RF.

Your knowledge of EE is obviously superior to mine but my point throughout much of this thread has been the marketed implication and understanding of some audiophiles that within a DAC, RF/EM and system/source noise is effectively impossible to isolate against to levels below audibility. My knowledge of EE and the specific designs of audiophile DACs is insufficient to counter this understanding with technical arguments and probably wouldn't help any way, as those I've been communicating with would not appreciate it, so I've attempted to use simple logic instead. Namely, how come pro audio DACs, even cheap ones, routinely manage a level of isolation which is apparently impossible for audiophile DACs costing far more? In many/most cases, I believe the differences heard by some audiophiles between USB cables is purely the result of a perception bias but there are some audiophile devices out there which are poorly enough designed to allow interference through to audible levels. In fact, I myself have heard an audiophile DAP which produced audible noise when positioned within a few inches of a laptop, noise which was rendered inaudible by an audiophile USB cable (or by moving it further away from the laptop). My response to this scenario is not that audiophile USB cables are therefore valid but that some audiophile devices are effectively faulty, designed with inappropriate isolation. If an audiophile USB cable really does make an audible difference, it's because it's providing a level of isolation which the DAC itself should be providing.

G

 

[gregorio]

If an $80 Pro DAC, regardless of how it looked, sounded as good as a mega-dollar DAC, like a Dave, don't you think that market rationality would prevail ... But of course we don't because the (likely) accurate analytical sound the $80 Pro DAC produces, while fine for commercial studio mixing, etc, does not sound anywhere near as 'good' to the consumer's ears for enjoyable musicality as does the Dave ... This similar logic applies to generic and dedicated audio USB cables as well.

An unfortunately typical argument. If an $80 pro DAC produces accurate sound then it is high fidelity, if you are looking for a system which produces a different sound then you're looking for a less accurate system with lower fidelity. If we apply "this similar logic" to USB cables, you have or are going to spend a large sum of money on an audiophile USB cable in order to lower the fidelity of your system?! Regardless of any technical details/issues, the logic or "market rationality" you mention is neither logical nor rational! And, if we do consider the technical details/issues, the situation becomes even more nonsensical because to achieve this more "enjoyable musicality" would require the USB cable to contain either: A more "enjoyable musicality" algorithm (which doesn't exist) and the processing power to execute it. OR, A "musicality" pixie or some other magical/inexplicable property.

G

 

[Leo-]

I'd add a few to your two types. Demodulated RFI can be audible on its own. I live 5 miles from a 50kW directional AM station in the main lobe, for example. That RF gets demodulated everywhere that's not had some form of RFI protection, and you get to hear the station in full fidelity. FM signals (broadcast or 2-way)can also be demodulated when the coupling forms a slope detector. RFI can also intermod with internal clocks forming new audible signals that are not related to audio. So add that in with your description of noise.

I think by "choke" you're referring to the ferrite bead/chunk on some USB cables. A few points on that. It's not actually a simple "choke", but rather it contains R, L and C, and is a resonant circuit. Every bead has a peak impedance frequency, in and of itself. However, it also interacts with the rest of the circuit it's installed on, the result being a particular impedance response curve that may or may not help with a particular RFI situation. In fact, there are some conditions where the bead actually makes things worse. It wouldn't be correct to say that the bead affects any current change in the cable, though, and it doesn't necessarily adversely affect timing or signals. It entirely depends on how the entire circuit is tuned. It's not that ferrites should be avoided, it's that they should be properly applied to the total design. Proper application requires full knowledge of the entire circuit. I'm not sure how carefully that's done on a USB cable.

You mentioned Ethernet...an Ethernet interface is essentially a set of balanced line receivers with common-mode filter. It has to be, most Ethernet cables are unshielded twisted pair, so for that sort of interconnect to transmit uncorrupted data it must have extensive CMRR. Using shielded Cat5/6 may help, but may also create a ground loop if the shield is grounded at both ends. Typically the switch should be grounded, the devices not. It's still a point of RFI entry, but probably minimal. Coax would be a ground loop possibility too because the shield grounded at both ends and the signal is carried unbalanced.

I mostly agree with you indeed. Audible RF noise to me could also fall into these two categories. It may be audible in itself and additionally it may change the signal by affecting the circuits, but also may be signal independent. Example is a guy who lived near a powerful radio antenna and his vinyl tonearm was a perfect antenna which let him listen to the radio while listening to vynil. Neat, isn't it?

Yes I meant ferrites and I also meant it affects all current changes. It is correct that they have optimal frequencies, however it doesn't mean that there are no induced currents in other cases (they're just not efficient filters for these conditions). You're absolutely right about ethernet cables, Im using a 5e and haven't noticed differences with respect to unshielded cable but some have noticed the shields shorted and generating ground loops.

 

[Arpiben]

Shielded cables have typically both ends connected.
In some circumstances it may create ground loops.
If the shield is removed at one end only ( either at cable level or at device level ) you solution the ground loop but on the other hand you create an antenna.
When dealing with cables in the 1m length range you basically have an FM radio receiver.Then it may help increasing the length, etc....
Therefore there is no perfect solution until one understands which interferer is audible....

I found the following Jim Brown's document quite helpful: http://audiosystemsgroup.com/RFI-Ham.pdf

 

[JaeYoon]

An unfortunately typical argument. If an $80 pro DAC produces accurate sound then it is high fidelity, if you are looking for a system which produces a different sound then you're looking for a less accurate system with lower fidelity. If we apply "this similar logic" to USB cables, you have or are going to spend a large sum of money on an audiophile USB cable in order to lower the fidelity of your system?! Regardless of any technical details/issues, the logic or "market rationality" you mention is neither logical nor rational! And, if we do consider the technical details/issues, the situation becomes even more nonsensical because to achieve this more "enjoyable musicality" would require the USB cable to contain either: A more "enjoyable musicality" algorithm (which doesn't exist) and the processing power to execute it. OR, A "musicality" pixie or some other magical/inexplicable property.

G

So from what Theorist is saying does this mean if I buy a "Audiophile USB cable" and use it for connecting my laptop for company presentations, the sound will be musical! And the entire powerpoint will glitter and glow with musicality and warmth!!!



Maybe if I also transfer a word document through these usb cables for work there might be some magic that will get the entire management to give me a raise on salary.

 

[Niouke]

I found this thread funny at first....at first.

Living In the current information age how is it possible to believe that bit perfect transmission is not possible over a pair of iron cords, or a cheap USB cable?
Just to put in perspective if you still use regular DSL line to access the internet, the HiRez FLAC files you download probably go througt a cabling that looks like this:

And yet they are bit perfect from the original, and you can probably also stream them on the fly, and they will still be bit perfect.

Also I ticked at the "noisy" computers assertion I've read here. If you think the noise inside a computer is big enough to corrupt a mere 5mb/s PCM stream, please consider that the typical modern volatile memory (RAM) in a computer is able to transfer ten Gibabyte per second, and bit perfect of course. If it's not bit perfect your get a nice blue screen (in windows at least). For reminder 10Gb represents > a single blueray layer.

 

[Leo-]

That's because you despite being in the information age you still don't get it mate

There is no data corruption involved, and no audible noise as some members have suggested. In simple terms, changes due to radio frequency interference and current leakages that are transmitted through cables are the ones which affect the analog circuits in your system, the "bit perfectness of the information" is not an issue however the timings of the information packet transmissions do affect the sound too, since in digital audio time is also information, not only the bits!

I found this thread funny at first....at first.

Living In the current information age how is it possible to believe that bit perfect transmission is not possible over a pair of iron cords, or a cheap USB cable?
Just to put in perspective if you still use regular DSL line to access the internet, the HiRez FLAC files you download probably go througt a cabling that looks like this:

And yet they are bit perfect from the original, and you can probably also stream them on the fly, and they will still be bit perfect.

Also I ticked at the "noisy" computers assertion I've read here. If you think the noise inside a computer is big enough to corrupt a mere 5mb/s PCM stream, please consider that the typical modern volatile memory (RAM) in a computer is able to transfer ten Gibabyte per second, and bit perfect of course. If it's not bit perfect your get a nice blue screen (in windows at least). For reminder 10Gb represents > a single blueray layer.

 

[bigshot]

Like grounding issues, RF interference usually isn't subtle. If you don't experience any obvious problems, there isn't any reason to worry about it. If you do have these sorts of problems, you're unlikely to solve it by just going out and buying an audiophile cable. You've got to go further than that.