Friday, 1 April 2016

MUSINGS: "USB Audio Gremlins Exposed: Beyond 1s and 0s, by iFi Audio" - Analysis of Industry Content

BTW: Here's the "Flasher Gremlin" for sale if you're a model collector...

Oh my... Run for the hills folks! Didn't you hear? USB is a horrible interface for audio purposes!

Hmmm... At least that's the kind of interpretation one might be tempted to consider after reading this "truthy scientific" article on AudioStream ("USB Audio Gremlins Exposed: Beyond 1s and 0s, by iFi Audio")! What can I say? Brutal! Nice example of content from Industry to instill some measure of concern among susceptible audiophiles.

Okay, let me first say that I'm not claiming that USB is perfect - nothing is... You could run into issues. But the fact really is that these days, with a decent asynchronous DAC (what decent audiophile USB DAC these days isn't asynchronous?), one should not be concerned with data errors, jitter issues, and I have yet to come across noise pollution from a computer other than atypical situations like what I demonstrated a year back.

A typical $10 USB cable made to appropriate specifications by a reputable brand is fine.

With that said, let's delve into this short "article" and consider truth and fiction (delusion?)...

Based on the concept of "fair use", I will quote this article and add comments in context for the sake of debate and critique.

[April 15, 2016 EDIT: I guess iFi was unhappy with me quoting the article... Fine. I'll just remove the majority of the quoted text but leave the first and last few words to indicate the quote being referred to... Go visit the AudioStream site for the full text. I suggest to just open two windows side-by-side and match up the original article text with the comments. I trust this would not be a problem.]

1. Bits are really ...

1.1 Background
... limitations and problems. 

Guys. Please show us where and how "bits are bits" is actually "debunked" and by what amount does this affect the actual perceptible sound quality with a modern digital audio device. Dragging out a Stereophile article from 1996 using jitter simulations in the multi-nanosecond range which is known to cause data integrity errors (100+ns it seems!) when any reputable DAC interface has jitter measured in picoseconds (one thousandth of a nanosecond) is rather disingenuous and adds to the FUD. As someone who wears eyeglasses, it's like claiming that there's no way any person with myopia would ever be able to achieve 20/20 vision and that's cause for concern because someone in 1350 wrote an article claiming that even the best tradesmen are unable to shape the convexity of the glass to an accurate diopter. That's absurd!

Yes, jitter varies between different interfaces but as I have shown previously, this is a function of the DAC/computer interface. No evidence at all that USB cables change the sonic output. Also, no evidence at all that OS tweaks or different machines make any difference with an asynchronous USB DAC. Who's delusional?

Please, do reveal these "reliably measurable differences".

1.2 Asynchronous USB audio ...
... that SPDIF implementations needed care.
Indeed asynchronous USB interfaces improve the ability to provide accurate timing data to DACs as I demonstrated a few years back and can be summarized in the following 16-bit J-Test FFTs:

Old AUNE X1 v.1 DAC playing the 16-bit J-Test using adaptive isochronous USB 1.0 interface vs. same J-Test with asynchronous USB 2.0 CM6631A interface (through a small length of coaxial S/PDIF).
While I'm sure over the years (like in the last 5 years), successive generations of hardware and software drivers improve signal quality microscopically, where is the evidence that this has been audible?

1.3 What about alternative systems...
... there is no silver bullet/panacea
Okay, let's talk ethernet... Is this interface not extremely reliable for data-free transmission? (You better believe this since the whole of the Internet depends on this fact especially if you make financial transactions!) Is ethernet not asynchronous, buffered, and streaming devices using ethernet tend to be very jitter-free? So what if the interface is so powerful that it can handle "general purpose systems"? Folks, if a gigabit home network is capable of transmitting 4K video without a hitch, what's the big deal with hi-res audio which uses even less bandwidth? As I have shown previously, ethernet works well and can transmit over long distances.

2. A Matter of ...
2.1 Audio vs. Data ...
... ‘Bulk/Burst mode’ transfers.
Yeah. So what? Sometimes there's more data to send, other times there's less. Even at stereo 32/768, we're looking at <50Mbps transfer speed. USB 2.0 is capable of 480Mbps ideal top speed (and it can usually achieve at least 350Mbps in the real world with hard drives I've used). Please elaborate on what the concern is here... And BTW, why is anyone even throwing out 32/768 other than trying to impress with big numbers. Seriously, what educated and reasonable human being believes 32/768 is even a desired PCM format for consumer consumption!?
2.2 What is Isochronous...
... still uses Isochronous mode to transfer audio.
Again, so what? Some transfers are isochronous and some drivers use bulk transfers. Just how error-prone do these guys think a decent typical USB cable is for us to fear the lack of CRC error checking and retransmission? Care to demonstrate what a mild case of "signal distortion" sounds like? As for "worst case" clicks and dropouts, how common is that with a decent DAC and standard laptop that's fit to handle basic decompression tasks (eg. MP3, AAC, FLAC) in 2016? (BTW, for those curious, here's what a bad USB cable "sounds" like and the type of distortion one hears due to data error - it ain't pretty.)
3. Of Mice and Fans...
... not work with the same port or cable.
Isn't it a little useless to be so vague?! Please give us an example of what computer, DAC, and cable which is up to basic specifications but suffers from audio subsystem failure yet both low speed (mouse, printer) and high speed (hard drives, flash drives) devices function without signs of an issue. Surely this would be practically helpful for everyone to learn to avoid this hardware combination.

3.2 Which way did USB audio...
... and a low sample rate.
Yeah, sure. Don't buy out-of-spec cables and ports then. The same can be said of out-of-spec SPDIF cables, low quality CAT-5e cables, cheap Firewire ports/cables, etc... But are in-spec cables and ports that hard to find or expensive? I think not. Funny the adjective "catastrophic" being used in this context of what we're discussing?! So scary.

3.3 Encountering unexpected Cyclic Redundancy...
... audio improvement devices and cables.
Where did 'fmak' make these testimonial claims about CRC errors? I presume it must be on the Audio Asylum somewhere? Is this where manufacturers get their research data; from the "inmates"!? Why is his post/claim not included in the References section below? But seriously, we're looking typically at less than 10' of cable. Surely active USB cables are not necessary!

4. Ports, Cables, Packets...

4.1 Being ‘PC’
... technical performance of different USB ports:

"Native Instruments...
... port is 'good' and the other 'bad'."
Okay, sure. Again, some hardware might be faulty or out-of-spec but that can be said of any device out there. And yes, of course you don't put poorly shielded cables where they can pick up interference. Again, just how common are all these "possible source(s) of USB signal distortion"? IMO it's not common at all with decent equipment and a reasonable length of USB cable.

4.2 USB cables
... USB 2.0 use and follow the formal USB specifications closely.
Yup. Fine. Drag out the ol' eye-diagram again huh? (Some measurements came out years ago around this.) Look, so long as the eye diagram is within tolerance, we won't have any data error. What's more important is to demonstrate examples of anomalies that can result in analogue output issues from the DAC; this is something I have never seen from the magazines or from these manufacturers. After all, the analogue output is what we're actually hearing after going through the amp & speakers. So based on certification and specifications, a <$10 length of USB 2.0 cable like this Belkin is just grand from the looks of it, right?

4.3 USB packet noise
We also find factors...
... less expensive USB DACs.
Sure. I can in a special case demonstrate the 8kHz packet noise. But this is far from an issue demonstrated from the DAC's analogue output. In fact, I have never heard or measured any 1kHz or 8kHz packet noise coming out of any of the DACs I have look at over the years. In fact, I would say that if anyone can measure this packet noise coming out of their DAC, then everybody should know about this and avoid this device since it's obviously poorly engineered!

5. Measure > Test > Develop...
Given the issues and limitations...

Computer audio certainly does not stand still, it develops dynamically and with some speed.
Well, I can agree with the "measure > test > develop" engineering concept. Sadly, they have shown no measurements of their own in this article nor demonstrated that they've developed something that has furthered the computer audio world. If anything as I have suggested in the opening, there's just more FUD than substance constituting the core of this article.

References

Stereophile. "Bits are Bits"
Stereophile. "A Transport of Delight: CD Transport Jitter"
Stereophile.The Jitter Game"
...

About iFi
... market advanced audio products.

Okay. That's nice. A few generic articles in the references. The Stereophile articles are from 1992, 1993, and 1996 by the way. Indeed, as they said "computer audio certainly does not stand still"... So why are they referring to articles from before the dawn of the computer audio era when folks were still almost universally spinning CD's, years before the first commercial SACD or DVD-A even saw the light of day? In 1996, USB 1.0 was introduced with almost no peripherals out there yet. The typical home PC was running around 100MHz rocking the original Pentium I and AMD just started to release the K5 processors. Seriously, consider the other areas of consumer electronics like reviews on computer equipment or assessment of the quality of HDTVs, is it not embarrassing that audiophile companies are still quoting articles from a long-gone era as to why we should still fear "gremlins"?

As for the manufacturer, check out the measurements of their current-model Abbingdon DP-777 DAC (~$5000) in Stereophile from 2012. Enough said.

Again, like last time I wrote about this kind of thing, shame on the audiophile press to continue to perpetuate the FUD with no evidence of commonsense or journalistic critical thinking abilities; parroting whatever the Industry throws at them.

--------------------------

Notice that I'm posting this message on April Fools' Day. While this post is not some kind of April Fools' prank, as far as I'm concerned, that original iFi post might as well be. Beyond the health benefits of a good laugh and not to take things too seriously, I think the spirit of April Fools' Day is a nice reminder once a year that we should maintain a reasonable amount of reality-testing in life... I would like to think that the majority of audiophiles are not of the proverbial "audiophool" variety. I really don't care or mind if anyone spends $500+ on USB cables or tweaks or whatever; as I have said before, placing subjective value in non-utilitarian benefits (like a fancy, impressive, nice-looking cable) is fine. Just be honest and rational about it and let's not perpetuate overblown anxieties...

Forget the FUD folks. Enjoy the music!

37 comments:

  1. Thanks for the sensibility. I too read that article and immediately suspected an April Fools joke. AudioStream is well known as an industry shill > FUD in, FUD out.

    Funny how the video space seems to be free of pan handlers and swindlers. Video is infinitely more taxing on a cable than audio, yet somehow the audiophile "industry" revels in perpetuating the fear and mystery to the unsuspecting - one can only conclude its for the benefit of lining their own pockets. They *never* seem to question anything. They are either extremely gullible, or shills.

    ReplyDelete
    Replies
    1. Thanks Qmax. You're right, at least in video other than some marketers who might want to push stuff like expensive HDMI cables, they usually don't get far with the BS.

      I assume it's related to the fact that in video we can "freeze frame" analyze what's in front of us. One is generally able to focus and analyze images with our cognitive facilities and attentional mechanisms better. Expectation of objective analysis comes naturally.

      In audio, the experience is ephemeral and I suspect it's very difficult to "take a step back" and make objective judgments (because perhaps evolutionarily we're programmed to experience and relate emotionally which is why music is so compelling). Compared to the ability to leisurely gauge visual quality, auditory discrimination I suspect has comparatively limited reliability and easily biased by emotional factors.

      Implicitly I think we can say that vision is also our primary modality in judging reality - "seeing is believing". Certainly many more neurons have been devoted to the task :-). Unless one's livelihood is in the acoustic arts, I suspect most of us would appreciate that the burden of disability with blindness is significantly worse than deafness.

      As for the benefit of lining one's pocket. No doubt the profit motive is important... And there's nothing wrong with that. Honest work for honest pay! The operative word of course is *honest* and how that is defined in light of articles like this.

      Delete
  2. Thing is though, Ifi audio accessories do palpably improve the sound, currently using Ifi Mercury USB Cable, Ifi ipurifier 2 and Ifi DC ipurifier with my DAC and linear power supply, definite substantial improvements made in the audio quality.

    ReplyDelete
    Replies
    1. That's interesting. Can you prove it?

      A set of RMAA measurements or something similar would be a good start.

      Delete
    2. I can hear it, now I'm unsure as to whether this can be 'proven' and with measurements but am up for trying and confident there is a significant improvement. How would I go about this (Running Mac with external DAC/Preamp/Active Speakers)

      Delete
    3. Hi Jimixy.

      As Kozmo suggests you could download the RMAA software and run tests on the gear:
      http://audio.rightmark.org/products/rmaa.shtml

      Focus particularly on whether there have been changes in the dynamic range and noise level of your DAC's output as a start. Remember, you'll need your own ADC to make these measurements.

      You see, it's cool if you as a hobbyist make these measurements and try it yourself. And if you can show that that there is an objective change, that would be really nice and I bet iFi would be very happy and might even use your data in the next article :-).

      However, they claim that in their own testing the devices make "reliably measurable differences". So should the onus not be on them to at least show us a few of these measurements and how it could affect sound quality for the sake of consumers interested in these products? Shouldn't they say a little more about what situations one might find a proven improvement and give us some examples of where the product would help? Even infomercials show examples of where the "Super Duper Wet Mop TM" cleans things up like no other with "before and after" samples...

      Yet none of these companies do this. Not for cables. Not for noise filter devices like the iFi or UpTone REGEN or Jitterbug.

      Delete
    4. Hi, thanks for the reply,

      I am on a Mac and I believe this program is windows only, is there a mac version or something similar? I have an audio interface which I can use as a line level ADC which can use to show/analyse changes with and without the Ifi ipurifier 2 and Jitterbug. I am keen to see what results this shows, alongside what I am hearing.

      I agree these companies should show more, Ifi audio do have vague figures and graphics, which I find is more than the other companies offering similar products generally offer, but I agree here.

      Delete
    5. Would these tests work by running audio through my system without either piece of gear present and outputting back into the software via my audio interface (through the line level input) for benchmark then adding jitterbug alone, then ipurifier 2 alone, and then both and comparing results?

      Delete
    6. Hi Jimixy. Although I use Macs, admittedly I haven't checked on freely available measurement software. Don't know if RightMark can be used in Parallels or how it would work off BootCamp.

      Well, you could always just run an FFT off your ADC input and see if the noise floor improves with the Jitterbug/iPurifier off your DAC with these devices in-line versus without them.

      Any Mac users out there with suggestions?

      Delete
  3. Interesting read.

    Re: "3.3 Encountering unexpected Cyclic Redundancy Check (CRC) Errors", I created WasapiBitmatchChecker app that sends digital data from USB audio device through S/PDIF, record it using another audio device and compare sent data and received data. Tested with the following hardware:

    Playback Device: Creative X-Fi HD USB Audio SB1240 connected to PC with generic mini USB cable.
    Digital audio cable: generic unknown brand TOSLink cable 6ft.
    Recording Device: RME Fireface 400 connected to PC with IEEE1394 cable.
    Audio data format: 96kHz 24bit 2ch PCM

    10 billion bits of PCM data transferred through this chain. Result: sent data and returned data was identical, zero error, i.e. zero CRC error happened.

    ReplyDelete
    Replies
    1. Nice! Thanks for the data :-).

      I'm also looking at my router's *wireless* data transmission error amount. With an uptime of close to 150 days, after transmitting 2.635e8 bytes, there were only 14499 errors... That's 0.00055% error rate through much less robust WiFi transmission. I live in a 3 storey house with the router on the lowest level using everything from laptops to tablets to phones to Squeezebox Boom/Radio music streamers among the levels.

      Needless to say, worrying about CRC errors with a wired USB interface makes no sense unless we see some evidence of this.

      Delete
  4. Hi Archimago!

    We can say that this text is basically an iFi ad. The industry need to say that USB cables are garbage, otherwise they won't sell their USB tweaks (like iFi do).

    Until they prove it, i won't spend my money on this kind of tweak.

    I spend my money buying music!

    Best regards!

    ReplyDelete
    Replies
    1. As usual, nice and practical VK!

      It's interesting that companies aren't "publishing" as much of their "white papers" these days... Instead, we seem to be seeing more quotations of Industry message embedded in equipment "reviews" and an article like this.

      At best, an article like this should really be published in the manufacturer website and stay there. Like you said, it's basically an iFi ad...

      Delete
  5. I don't hear the signal entering my DAC, or coming out of my DAC. I also don't hear the signal entering my amps or coming out of my amps. What I do hear is the motion of my transducers (loudspeaker or headphone drivers) generating an acoustic field.

    Until I see actual measurements (non-contact scanning laser vibrometer) showing exactly how the transducers are behaving in terms of changes in vinyl source, DAC, amplification, cables etc., I'm inclined to be sceptical of any comments or electrical measurements suggesting what I should hear or shouldn't hear.

    Yes, it is difficult to obtain this information, but I feel this important aspect is missing from the discussion. As a hifi consumer, I still feel I'm in a position of "blind faith", in spite of all the existing comments, articles, measurements.

    ReplyDelete
    Replies
    1. "I don't hear the signal entering my DAC, or coming out of my DAC. I also don't hear the signal entering my amps or coming out of my amps. What I do hear is the motion of my transducers (loudspeaker or headphone drivers) generating an acoustic field."

      Yes. Ultimate it's the compression and rarefractions of air that we hear. However, the *cause* of the change in sound pressure of course is due to the electrical signal causing the amp and speakers to produce the change in air waves. If there's no change in the electrical properties, how then do differences in soundwaves get generated? Unless you see another mechanism, lemme know!

      "Until I see actual measurements (non-contact scanning laser vibrometer) showing exactly how the transducers are behaving in terms of changes in vinyl source, DAC, amplification, cables etc., I'm inclined to be sceptical of any comments or electrical measurements suggesting what I should hear or shouldn't hear."

      Sure, let me know when you see these measurements and if expensive cables make a difference.

      "Yes, it is difficult to obtain this information, but I feel this important aspect is missing from the discussion. As a hifi consumer, I still feel I'm in a position of "blind faith", in spite of all the existing comments, articles, measurements."

      Faith needs not be blind! And theories that work need not be absolute; just pragmatically adequate. The problem comes when theories are thrown up all the time in the audiophile world with zero adequate evidence to back up the statements which any of us can dream up. This is fine if we're writing a sci-fi novel. But I missed that part if that was iFi's intent :-).

      Remember, audio equipment is born of science. The mechanisms are created by man based on physics and mathematical principles. The gear is *engineered*. It's not like engineers are in a process of discovering basic science principles when they construct a USB 2.0 cable to specifications laid out by the "USB Implementers Forum"!

      Delete
    2. Hi Archimago,

      Judging by your response to my comment, I realize that I didn't phrase my perspective in the right way. My position is actual one of fundamental engineering and physics.

      What I should have said is this - I don't believe any comments or measurements on the electrical properties of a hifi system because no one is relating this to measurements on the actual mechanical behavior of the drivers in loudspeakers or headphones.

      It is these that are generating the actual sound fields that we are hearing. For example, if differences in jitter in the DAC are measurably changing the dynamic mechanical behavior of the drivers, then this will change the radiated sound field. If it isn't, then maybe a jitter measurement is not so important after all. Why then waste time even speaking about it.

      The same applies to cables, etc. No change in the dynamic mechanical behavior of the drivers, then no change in the radiated sound field. Hence, I shouldn't expect to hear a difference.

      I view the mechanical surface measurements of the drivers as being the most important, since this behavior is the final physical phenomena in the hifi chain influencing our hearing (disregarding room or headphone cavity acoustics).

      Since these driver measurements are not generally carried out or even considered, I remain skeptical of all comments and electrical measurements suggesting what is or isn't important in hifi.

      Delete
    3. Hi! Thanks for the clarification and discussion.

      Realize that there are certainly measurements we can make based on characteristics of the sound waves in the room. The most basic of which comes from using a measurement microphone to look at frequency and time domain characteristics as many of us already do with room correction:
      http://archimago.blogspot.ca/2015/11/measurements-digital-room-correction.html

      Also remember that these measurements can be used to assess for cable changes like when I had a looked at speaker cables or the effects of room treatments or even just removing the speaker grilles!
      http://archimago.blogspot.ca/2015/08/measurements-room-acoustic-absorption.html
      http://archimago.blogspot.ca/2015/06/measurements-effect-of-speaker-grills.html
      http://archimago.blogspot.ca/2015/06/measurements-speaker-cables-wires.html

      I fear though that we're wandering into the territory of philosophy and metaphysics with the argument however. In the context of this iFi article speaking about the USB interface, I fail to see how speaker driver measurements are relevant. In a "hi-fi" system, we basically want the "best" fidelity as in it being as accurate to the recording as possible, right? As such, if we can measure electrically no difference between data transfer quality of ports/cables, then we can leave speaker drivers to another discussion about "which is the best hi-fi speaker", right?

      Also, I must point out that when you say "I remain skeptical of all comments and electrical measurements suggesting what is or isn't important in hifi", you are actually stating a philosophical opinion rather than a statement of fact given that audio equipment (especially modern computer audio) is by nature mostly about how the electrical chain is implemented. From the recording, to mixing, to mastering, to digital decoding, to amplification; all of this is done in the electrical domain and they all add up to feeding the speakers with the best electrical representation of the recording. Of course speakers and room make humongous contributions, they're just the last link in the whole chain of events... To dissociate the importance of the electrical steps from whatever meaning of "high fidelity" you are using to define it seems to be in opposition to the typical audiophile mantra that "everything matters!" (not that I necessarily believe this, but there is some truth to it).

      In your first message, you said:
      "Until I see actual measurements (non-contact scanning laser vibrometer) showing exactly how the transducers are behaving in terms of changes in vinyl source, DAC, amplification, cables etc., I'm inclined to be sceptical of any comments or electrical measurements suggesting what I should hear or shouldn't hear."

      I can challenge that statement and say that it's not good enough - I demand we refine it to see "measurements of sound waves 3m away at tweeter level at sea level in a perfect anechoic chamber otherwise it's not hi-fi". We could even get more ridiculous and claim "until I see measurements of the action potentials coming out of the cochlea in the ears, I don't believe we know anything about hi-fi". Or let's go even more "cortical" and suggest that "until I see measurements of neural network changes in the frontal and limbic brain circuitry, we would not know if $500 USB cables make a difference!"

      You see, this is a bit of an idiosyncratic "goal post" you've brought up. As far as I'm concerned, good hi-fi is that which is *transparent*. Neither adding or subtracting anything all along the signal path. In this case, if the USB data transmission is accurate and there is no evidence of timing anomaly (as in asynchronous DACs being jitter-free in the analogue output), then I see no need to worry.

      Delete
    4. "We could even get more ridiculous and claim "until I see measurements of the action potentials coming out of the cochlea in the ears, I don't believe we know anything about hi-fi."

      We know that the hearing ability of people diminishes significantly as they age, yet it is precisely those people (those with the diminished hearing) whose pronouncements on the sound differences between cables we are expected to believe.

      And to a man, they refuse to participate in ABX testing, choosing to believe their diminished faculties over any scientific or engineering reasoning.

      This debate could be brought to a conclusion by properly conducted testing, but the fear of failure is getting in the way.

      Delete
  6. Hey man,

    I am studying your articles, as I am investigating what the issue with alot of people saying they can hear difference between a stock USB cable to DAC async mode (eg. Hegel HD25, even Hegel themselves say buy a really good usb cable, etc..), and a really expencive USB cable like the 200 dollar Nordost cables. I am not entirely convinced, but I do believe a better shielded cable will make a difference between a stock cable.

    I am no electronician, so I will try to explain some findings, and if you mabye can confirm or test this. not sure if you did test this.. I have alot of blog posts to read on your blog.

    But, on the Norwegian forum, someone measured the curve a digital signal have with Oscillator without plugging it to the USB hub or port on computer. Basically measuring the actual copper used in the cable, and he found that bad cables has a dirty curve from top to bottom, and they deviate in the analog signal strength on the cable itself. Now, He did confirm the DAC played without jitter and did not loose signal sync from DAC to PC. But he can hear difference in details.

    Now this has come up:
    So the data transfer from D+D- and +/- are actually electric power signals. Despite the digital transfer on the analogue cable. So the electric signal is transferred to the DAC, and therefore connected electronically directly to the DAC. The theory is that the DAC recieving electrical circuit is poisened by the bad curve signal due to bad copper or whatever happens inside cable, that this bad curve is forwarded into the analogue part of the DAC. After all, the signal is sent electronically and transformed through a chip and transistors to analogue audio signal.

    Have you ever tried this bit curve test on "bad" usb cables compared to a really good one ?

    Excellent blog, and I really enjoyed the 8khz noice from USB on computer and manage to take it down substantially. That itself is a really good find. I believe the combination of several issues with the USB standard is why we can hear the difference.






    ReplyDelete
    Replies
    1. Hi "Unknown"...

      Did this poster on the Norwegian forum put up a detailed post somewhere describing the technique? (Is it in English?) Not clear what a "bit curve test" is. Are they referring to "bit error rate"?

      You see, we could of course measure cables by passing all kinds of signals through - MHz and GHz signals. And any length of cable will at some point have a threshold where the signal doesn't look right anymore.

      But remember that cables are built to *specification* and that basically defines what oscillatory frequencies are important for the intended signal. That is what the eye diagram is about; making sure that the electrical signal achieves an adequate integrity and "looks right" to the receiving equipment. Sure, an expensive cable may be better shielded for example and perform better with high speed transmission (eg. a Cat7 cable is rated better than Cat5e), but in the context of what we're sending, does it matter? Will a Cat7 cable provide noticeable speed improvement from a 100Mbps ethernet switch over an absolutely adequate Cat5 cable, or somehow the data comes out "better"? Of course not! All I can say is I have never come across a problem with digital cables produced to spec...

      Drop a link to this post! Would be interesting investigating what was done.

      Delete
  7. Thank you for the your response, Archimago.

    My goalposts ultimately are to focus on what really matters in the hifi chain. I want to cut out all possibilities of error due to human perception in assessments, so all blind testing variants are out.

    We know that realistically we are not going to speak of measurements inside the human ear, or of brain patterns within our pleasure centres, so that is out too.

    Microphone measurements in the room are room dependent, so they are out too.

    This then leaves the dynamic mechanical driver measurements as focussing on the last part of the hifi chain that can influence your ears (again disregarding acoustic room or cavity effects). A measurement at this stage represents the cumulation of everything happening within the hifi chain.

    Concentrating on these measurements then allows us the best chance of realistically understanding what matters in the hifi chain for "accurate" and "transparent" sound production, and what doesn't. The door is then closed on subjectivity arguments, and on electrical measurements that are reaching levels so low now that one cannot be sure if changes observed are really there, or one is measuring the instability of the measurement system.

    Thank you once again for being a helpful sounding board.

    ReplyDelete
    Replies
    1. Thanks BB.

      Yup, fair enough. I see where you're coming from...

      I think we're on the same page then in that indeed ultimately the speakers (and room) are in relative terms the most important component(s) to the whole endeavor. And as you say, electrical difference are indeed extremely small these days among high quality gear (and speaker anomalies and distortions can easily overwhelm whatever small imperfections of things like DACs).

      Hey, if you ever find some measurements using a "scanning laser vibrometer", lemme know :-).

      Delete
    2. I've got my eye on a scanning laser microscope for the hobby room... will measure vibrations into the GHz. :-(

      Nice blog you have here BTW, keep it up.

      Delete
    3. Cool. Now that could also yield some interesting "pictures of the week" for a fun blog!

      Cheers... :-)

      Delete
  8. Open and tolerant

    I am an electrical engineer, working in the audio field for over 30 years and working highly with measurement systems to improve and evaluate audio products, so this is always my starting point.

    But I am also a hoppy musician for over 30 years and have also a small recording studio, so technique and measurement aside, I enjoy making and also listening to music a lot.

    And so every time to time, I came across some tweaks and products, that I really do not totally understand from the “engineering” standpoint, but I am open and give it a try.

    And so it happens from time to time, that after some time I do find a suitable measurement for that, that wasn't in my mind, when I first “test” a tweaking product.

    So the case, with some above mentioned USB tweaks, that do not change the total response, but I can measure differences when measuring with probes inside.

    And they are also some products (USB and LAN cables), that I still can not measure, but have no problem to accept their influence over the years, and don't call them snake oil.

    Sure, they are some snake oil products out there, no doubt, but they will fail over the time to impress, when the hype is over.

    But my main point here is: Being open to new ideas, to different ideas, even if they look “strange” in the first sight.

    Enjoy the music
    Juergen

    ReplyDelete
    Replies
    1. Thanks for the note Juergen!

      Wondering what was the most surprising products you've come across that at first seemed impossible but with some exploration, you were able to find/explain how it was beneficial? It would make a great blog post to show a few of these and what they actually did!

      Delete
    2. Just one example

      Hi Archimago

      As being a long time in this business, I could tell a lot of stories, but I would like to share just my first experience concerning what I have heard, but what I couldn't measure at first. This was 32 years ago. So looking back, this seams to be too easy, but it was my first case in that.

      I came relatively fresh from finishing my engineer study and finished my first pre amp development. With that I was at a HiFi show and was demonstrating this pre amp, with a loan power amp and some loudspeakers. I was proud and happy with the sound.

      During this show, a visitor came by and asked: Have you checked the polarity of the mains? I was a bit surprised about this question, but after the end of the show day I rotated the orientation of the mains plug and was “confused”, because there was a clear difference in the sound. I have heard this phenomenon from some “high end freaks” mouth, but coming fresh from my engineering degree, I was skeptical and thought this is just some “snake oil talk”.

      Ok, after the show, I set up the preamp again to the measurement gear (those times, a HP meter bridge) and couldn't see any difference in the measurement with both orientations of the mains plug of the pre amp. I was struggled. Then I repeated the listing of the pre amp (those years with a turntable as a source) and some other loaned power amp and also here, I could hear a difference in sound, when rotating the mains.

      Then I began to read some papers, mainly some older papers from bell laboratories that I loaned from a large library (you must remember, that 32 years ago, there was no internet or google, where you can browse) and read some detailed analysis of power transformers, about capacitive and inductive stray coupling and this led my brain to go into the right direction.

      To cross check my thoughts, I set up the pre amp and the power amp together, connected them via RCA, as I had done during listening and with this “complete” setup I was able to measure a difference in performance, when rotating the mains power cord. Now it was clear to me, that with a hifi setup, out of two devices (or more) that needs voltage from the mains, and are connected with the RCA connection, that it will always be an influence of the performance depending of the orientation of the mains.

      If I check the power amp alone with the measurement setup, then again, I couldn't measure any difference with the orientation of the mains cord. This is because good measurement devices, have a very high common mode rejection at the output and at the input, so they suppress the influence of the stray capacitance of the mains transformer of the device under test.

      Delete
    3. With this knowledge I began to work on a method to test and measure the common mode performance of a DUT between mains and signal ground and also to measure the common mode suppression of a mains transformer. From that experience on, I designed proprietary mains transformers with very high common mode rejections, that the stray signal that tries to enter into the device coming from the mains will be shielded to earth and the stray signal that tries to left the device into the mains, will be shielded to power supply ground.

      I would like to have a pre amp that sounds as it should, no matter what orientation the mains cord has to the mains outlet. If you draw a complete signal flow diagram, with all stray currents, you will realize, that beside the wanted connection between the pre amp and the power amp, there is also a “back door”, where the signal creeps via the stray capacitance from both power transformers as a parallel path. And if you draw this diagram, you will see, that with typical impedance RCA Inputs at the power amp, and so with typical low signal current between pre and power amp, that these back door current can have a much lower impedance at mid and high frequencies, than the wanted main signal path.

      This was my first experience, that there are more / different points on a product, when connecting as a complete setup, that influences the sound, than would be obvious, when measuring just one DUT, that is done with most audio magazines for obvious reasons, to see and judge only the performance of the device under test.

      A bit later, I set up a HiFi system, where every power transformer was driven by its own battery driven mains (having a batterie driven sine wave generator, driving a battery driven power amp, driving a power transformer in inverse mode, so at the output is the mains voltage and frequency, but all batterie driven). And with that experience I began to optimize not only the common mode rejection (due to stray capacitance) but also the differential mode rejection (due to stray inductance). But this would be a different story.

      And similar things are happening with digital inputs and outputs, that are galvanically isolated (meaning no Ohm connection, and so no DC current is flowing), but due to stray capacitance of the digital audio transformers, there is an unwanted leakage path of the digital audio signal. Even this stray capacitance can be lower than 10 pF, but with the Biphase Signal in the MHz area, it can influence the total quality of a DAC.

      Delete
    4. For example, nowadays, nearly every Digital Source, or DAC or Pre Amp, measured alone is fairly enough in performance, but when you setup them together as a complete system, with all sources connected, you would be surprised, that common mode disturbances and stray coupling does have a larger impact to the total performance, than the individual devices. So the complete setup has to be measured and not only one DUT.

      And also, using sine waves does not show the complete picture, because even good codecs, can deliver good data with sine waves. You must use a test signal, that shows you differential and intermodulation distortion, noise floor modulation, aliasing, and jitter and has similar DR ratio as typical music. This would give you a more realistic picture, of what is going on.

      And lastly, using only pure ohm load for the power amp, when measuring the complete setup does also not mirror, what you are hearing with loudspeakers. Comparing power amp output on real loudspeakers (speaking complex loads) will show you much more differences than comparing the power amps on a plain ohm load. I could test every phase angle between voltage and current, so simulating everything between a pure capacitive load to a pure inductive load.

      What I want to say is, that you need measurements, that are much more closer to the real setup when listing, in order to better compare and find, what you are able to identify while listening. And with those, more complete setups during measurement, you will see also much more things, that influence the sound, comparing when you just setup one device under test for measurement.

      So in short words: Measure all devices in that way, how they are connected together during listing, and use test signals, that are more close to the kind and dynamic range, of the music you are listening to and use a more realistic complex load for the power amp, as this is the case, when having the speaker connected to and these three points together, will give you a more realistic picture of what you are hearing, even if you just probe the DAC converter output for you measurement, but still sending the signal to the pre and power amp and to the speaker.

      I am sorry for this lengthy story and also for the not native english writing, but I just want to give one example, out of many, that if you look deeper into a problem, you can answer some questions, that are not so obvious at first sight, even sometimes afters years.

      I mainly wanted to say in my former blog entry, that it is good to stay open to different ideas (as also to different cultures, …) and learn from each other. There is much more going on, that we do not understand at first sight, but it is worth to look behind.

      As I said. Enjoy your life and enjoy the music.
      Juergen

      Delete
    5. Thanks for taking the time and the thoughtful response Juergen!

      Indeed, I know you have a ton of experience in all of this so I very much appreciate the reminder that the "system" as a whole needs to be assessed beyond the individual components as well given potential synergy and/or impediment that any one individual "piece" in the chain might cause.

      Well said:
      "Enjoy your life and enjoy the music." :-)

      Delete
  9. Not with an iUSB device but I have measureably shown that similar devices like the Wyrd, Uptone Regen and isolation devices do work to improve jitter/ signal quality. Do keep an open mind and measure more devices.

    Also to note there are DACs which are more sensitive than others in this area.

    ReplyDelete
    Replies
    1. Hi Josh.

      Please show the measurements! I do have an open mind but have not been able to show these things make a difference in the real world! So if you can demonstrate it with the measurements, then it'd be great for everyone to see... Even more important if the measurements show that they change the DAC analogue output. Especially things like the REGEN which as far as I can tell is a "single-port" USB hub.

      Delete
  10. Great article.Facts have a hard time against beliefs though,as we see from some of the comments.

    ReplyDelete
    Replies
    1. Hey there N/A...

      No worries. I believe that ultimately facts do prevail and this is reflected in the marketplace. Snake oil products ultimately die even if initially there might be some excitement. As John Maynard Keynes noted, from a macroeconomic perspective there's always a "bezel" (as in "embezzle") in the financial system. I don't see it as much different in the audio world... At any one time the proportion between "true quality" products that make a difference and those with "pseudo quality"/hyped products out there as well go up and down based on level of interest in the hobby and general affluence and disposable income for the hobby.

      In times of prosperity, we might see more hype and interest... As things get lean, snake oil products will lose out as hobbyists get more conscious of value and cost-benefit ratio. And so it goes.

      The good thing of course is that overall sound quality has gotten better. The deflationary power of technology has made gear *much* less expensive. The audiophile world bemoans the idea that people "don't care about quality". Nonsense. Quality is important just that for the most part, people are not interested in *what they're selling*. The hobby for high-priced luxury audiophile gear is getting smaller as a result because faith overall is waning. People realize that for many items, so-called "mid-fi" (because it's less expensive) gear is just as good as what the boutique is selling but for a heck of a lot less. Forget mention of high-priced cables and weird dongles hanging off USB ports! Thank goodness there are reasonable companies like ELAC who are smart enough to see the writing on the wall, offering reasonable products and using the "audiophile press" to market them.

      This is good for the consumer but threatens those who want to line their pockets with snake oil products though.

      Delete
  11. I also reacted to that article, as they referr old articles which has no or atleast minimum measurement references.

    I think the best findings about the issues is what you have proven about 8khz noise.

    ReplyDelete
  12. Why do you post other people's articles as if they are your own without giving credit to the writer?

    ReplyDelete
    Replies
    1. Hey Sassy:

      Did you not see the link to the original?

      Was the text in this post not significantly my own? Remember, this is meant as a serious and hopefully reasonably complete rebuttal of the usual advertising nonsense.

      Was iFi not mentioned? (Like in the title.)

      Heck, I even quoted the full "About iFi" portion at the bottom. Remember, this is an "INDUSTRY VOICE" article on AudioStream. How did I not give "credit to the writer"? iFi has been given full credit for this embarrassing article (IMO)...

      Delete