That SATA cable saga - the aftermath

Summary: Yesterday I pointed to a piece written by HiFi journalist Malcolm Steward (original piece now removed, but it still exists in the Google cache) on something that he dubbed "Super SATA" cables and how replacing the ordinary SATA cables in his NAS box resulted in "easily perceptible improvements." Yesterday Steward pulled the original post and and published a new post which I think makes it worth revisiting this issue.

Yesterday I pointed to a piece written by HiFi journalist Malcolm Steward (original piece now removed, but it still exists in the Google cache) on something that he dubbed "Super SATA" cables and how replacing the ordinary SATA cables in his NAS box resulted in "easily perceptible improvements." Yesterday Steward pulled the original post and and published a new post which I think makes it worth revisiting this issue.

Let's take a look:

I have withdrawn the article that appears to have upset so many computer enthusiasts.

Yeah, well, it would. Any "computer enthusiast" worthy of that title would know that if a SATA cable wasn't faithfully delivering the 0s and 1s from the hard disk to the motherboard, that PC or NAS box isn't going to be getting as far as booting up, let along delivering sub-standard audio.

I realise that the opinion I expressed in it was contentious but the reaction from some individuals was way too extreme. I think that wishing death upon someone because they wrote how they witnessed a change in the way their hi-fi sounded when they swapped a cable in a NAS is a bit of an over-reaction. Anyone in my office, including my wife and children, can read my email and they were not impressed by this and the volume of similarly aggressive correspondance.

This kind of behavior is inexcusable.

I know full well that it is ‘scientifically' not possible for a data cable to exert such influence but I know what we heard and hoped that maybe someone might be able to throw some light on what might be going on.

Problem is, the original article didn't come at the problem from this angle. Instead, it presented these cables as some sort of "wonder SATA cable" that delivered better 0s and 1s to the motherboard. The article fundamentally confused analog and digital data and applied old analog thinking to a digital device.

While a couple of people kindly wrote and did just that most people simply said "It's just ones and noughts, you stupid (expletive)," which wasn't especially helpful.

I've given this some thought, and to be honest I really can't see how these cables make a different. The problem here is that it's easy to fall into the trap of thinking that more expensive is better, which results in confirmation bias. Remove the element of a double-blind test and you have a recipe for wacky conclusions. Maybe the old cables vibrated in some audible way, or maybe one was touching a fan or something, but beyond some mechanical mechanism I can't figure out any reason why the new cables would be better than the old ones (and I don't really believe the whole mechanical mechanism line of thinking either). I certainly can't think of anything that would result in the sorts of celestial improvements that Steward mentioned in his initial post:

The most marked and worthwhile difference, I felt, was in the increased naturalness in both the sound of instruments and voices, which seemed more organic, human and less ‘electronic', and in the music's rhythmical progression, which was also more natural and had the realistic ebb and flow that musicians exhibit when playing live. In short, recordings sounded more like musical performances then recordings.

Anyone got any thoughts?

Topics: Telcos, Hardware, Mobility, Networking, Storage

Kick off your day with ZDNet's daily email newsletter. It's the freshest tech news and opinion, served hot. Get it.

Talkback

23 comments
Log in or register to join the discussion
  • Agreed...

    A keyword to look for when determining whether more expensive cabling is called for is "measurable difference". As I pointed out in your first article, there are most certainly scenarios where signal degradation is both perceptable and measurable. The people who run into these scenarios the most are professional installers, of which I know several. As I also pointed out in my reply in your first article, there is a huge difference between digital medium with built-in error detection/correction and traditional audio/video medium. A SATA cable in your computer isn't likely to have the same issues as say an analog speaker cable carrying a thousand plus watts of signal from the amp or a coaxial cable carrying sound and video from your local cable provider.
    jasonp@...
  • RE: That SATA cable saga - the aftermath

    Fundamentally, the "audiophile" is saying that the "Super SATA" cables solve a bit error problem. If the bit error rate was so high that these "Super SATA" cables made a noticable "audio improvement", I would say the system probably was unusable prior to installation of the cables (ie wouldn't even boot).
    7mgte
  • I don't know, but I'll throw out a thought

    Maybe it was just the fact that he had to reboot the NAS to install the cables that caused the ?improvement.? Maybe there was some sort of transmission issue over the network that was corrected by restarting the NAS.
    pmcgrath@...
  • Feh.. There may have been some issue with the old cable..

    Think about it. We have several grades of ethernet today - CAT 5, CAT 5e, CAT 6. All of them are basically the same - 4 pairs of wires running along the length. There may have been something going on with the old cable. Is it likely? Maybe not. But there's still a possibility, however remote.
    Wolfie2K3
    • Not a valid comparison

      @Wolfie2K3
      The different categories of Ethernet cabling support different run lengths and signaling rates. If you try to use a lower-rated cable for a long run, you would generally get an intermittent link or no link at all. Both SATA and modern Ethernet protocols are sensitive to packet errors, these errors will usually cause a link reset (interruption) as opposed to continually retrying packets which would be perceived as "slower" performance.
      terry flores
    • RE: That SATA cable saga - the aftermath

      @Wolfie2K3
      Cat5 is not capable of 1000Mbit as Cat5e or Cat6 is.
      ryanstrassburg
      • RE: That SATA cable saga - the aftermath

        @ryanstrassburg <br>I will have to technically disagree with you on that statement. Cat5 and Cat5e are mostly the same quality cable. The e on the end of Cat5e refres to the fact it contains 4 pairs of wires compared to only two pairs in a Cat5 cable. I have seen a 1 metre 1000Mbps patch cable made using Cat5 when the Cat5e ran out. The patch cable was made using two 1 metre lengths running between the lan plugs to supply the required 4 pairs of wires needed for a 1000Mbps patch cable. It aparntly worked very well for months till a boss saw it and made the IT guys replace it. I was also told of a situation where a set of LAN splitter cables where used in reverse to create a 1000Mbps connection upto an office that had two LAN plugs in it that only had Cat5 cable to each plug.
        NZJester
  • I tend to think...

    that the blogger fell into the common trap of expecting a difference so his brain provided one. (The trap is "I think, therefore I THINK I am" rather than "I think, therefore I am.") If you enter a test expecting a particular result. All things being equal your brain will try to provide that result.
    Scubajrr
  • Complete and utter cobblers

    The quantity zero was represented on my 1980's 8-bit computer as:
    00000000
    The quantity zero was respresented on my 1990's 16-bit computer as:
    0000000000000000
    ...
    Do you want to have a guess at how the quantity zero is represented on my 64-bit computer?
    0000000000000000000000000000000000000000000000000000000000000000
    You must be a genius!

    This quantity is greater than the value of the pseudo-hi-fi 'enthusiasts' who don't understand the concept of digital. In an ideal world they would be driven out of business accompanied by fines for their spread of misinformation.

    After the invention of the music compact disk in 1982 the same charlatans postulated that the 'natural' sound of analog and vinyl remained superior to digital.

    By 'superior' they meant that we should occasionally represent zero by:
    0000000000000000000000000000000000000000000000000000000000000001
    You see 1 sounds more natural: it includes the 'warmth' added to the original by:

    - obsolete recording, transmission and reproduction equipment

    - faulty placement of microphones (i.e. inside the lid of a concert piano instead of in the middle of nice concert hall which would showcase the venues acoustic)

    - the impossibility of a stylus on vinyl encompassing the dynamic range of live sound

    There are some differences between analog and digital ... but cables isn't one of them.

    Put these people in jail unless they can scientifically demonstrate otherwise.
    johnfenjackson@...
  • RE: That SATA cable saga - the aftermath

    Reminds me of when people would take a green highliter and color the back of an audio CD to improve the audio quality.
    ussafs3
  • From a professional

    I've been a professional in the pro audio business for decades. The people I work with make major films and all sorts of recordings from hip hop to classical. I rely on my ears and take tremendous care with my hearing. It's important that I can tell the difference between things and not fall prey to confirmation bias. The original blogger whom you mocked is an idiot and you let him off easy. Other than the failure to get the bits through the cable (you correctly pointed out the consequences of that) the only other source might be electrical noise passing from a cable to some analog component in the computer chassis. Having D/A components inside a noisy chassis would be a dumb idea and the source of many more problems than that single cable.

    Audio writers have long cultivated an attitude of having superior hearing. In the few instances that they submit to blind testing, they haven't done well. Your clothing, what's on your bookshelf, how you're sitting in your chair, minute placement of your speakers: all of these have more effect than the nonsense this character was spouting. Writers of this sort have long been the object of great mirth among those who really know what they're hearing.
    MC_z
  • RE: That SATA cable saga - the aftermath

    Long ago I had the Hi Fi bug. Vinyl had it's limits but a few artists recorded using a process called direct disk recording. I could hear an improvement. Sony came out with a consumer level reel to reel tape deck with the professional quality heads. To save my vinyl I recorded them to tape for normal use. By chance I met an excellent musician and went to a club and recorded his set. Of course I gave him a copy. This recording was noticably better than any record. When CD's came out I bought a few and they were awful. Because of the 44khz bandwidth limit (Is that the right term?) any soft passages had little detail of the performance. Have there been any improvments in the quality of CD music? If the same standard for CD music recording is in place today then no one knows what music can sound like in a better medium.
    When I came home from work one day and my Duntechs were gone and a Bose speaker system was in their place first I listened and then I felt like something importent had been removed from my body. CD's and toy speakers do not make Hi Fi.
    Call me the lost Hi Fi nut,
    Bill
    Bill1William
  • @Bill1William

    CDs actually sound considerably better than they did in their first few years, for a number of reasons. The converters--both A/D and D/A--are much, much improved. The recording equipment is better. Noise-shaping technology is now the standard for mastering. This moves any remaining quantization noise to bands where you just won't hear it. The disk mastering technology and engineers have the benefits of two decades of improvement. The "limited bandwidth" still gives you better than 20 kilohertz. If you're much older than 25, I guarantee you can't hear that high.

    Any possible (and unlikely) superior frequency response in vinyl is short-lived. Those features are tiny in the grooves and get mashed down after only a few plays. Most vinyl wasn't that good in the first place. You also have tiny pitch variations because the hole isn't absolutely dead-center in the disk. And let's not forget tone-arm resonance. Direct-to-disk WAS a little better, but you sacrificed that as soon as you copied to tape.

    Vinyl has many pleasant associations for people, primarily tied in with their youth. True high-fidelity is not one of those things.
    MC_z
  • RE: That SATA cable saga - the aftermath

    My rather more constructive comments were published for sadly no more that four-to-five hours on Malcolm's blog entry before all comments were pulled.

    I was keen to explore the situation more thoroughly. Maybe the original cable presented a poor impedance match, causing a degree of mismatch reflections across the cable that **continually** kicks in error correction and continually raises the amount of processing, current drawn, noise and electromagnetic interference produced my the microprocessor. Heck, I can even hear my PC's microprocessor grind away when listening through its headphone socket.

    The end result of that constant increase in EMI, noise and current draw would be very much dependant on the system and its configuration - but I wonder whether the following functional blocks would be susceptible: PCM transmitter (where a "digital" output is never a perfect square wave), PCM receiver and analogue output of the DAC.

    Interesting when the areas of computing/information theory collide with quality hi-fi and you see people assume that a DAC and a pair of ears work to the same thresholds as a digital receiver circuit...

    Andrew Randle (www.hifiwatcher.com)
    The Hi-Fi Watcher
    • RE: That SATA cable saga - the aftermath

      @The Hi-Fi Watcher, I read your contribution on Malcolm's blog before it was pulled and have also read the fuller version on your blog. However the original blog by Malcolm said that the SATA cable was inside a NAS device separate from a very well designed and expensive network player. The separation of disk and playback should ensure that no interference between the two (unless the player is badly designed -God forbid!).
      Your original speculation about processor load when errors occur also does not reflect the actual way that SATA errors are handled.
      Modern 24bit Audio hardware and best practice amplifier design give much lower noise floors than was common 20 years ago and thus it is possible that marginal effects that would have previously been buried in noise can now be detected.
      However their is a healthy (and most would say justified) suspicion that many of the low-level improvements that some high-end audio journalists claim to hear are more likely to be perception than concrete differences.
      It is however important to maintain any discussion within the bounds of reasonable behaviour. Nobody is pedalling drugs, just selling expensive components to those willing to pay for them.
      paul_t_milligan@...
  • RE: That SATA cable saga - the aftermath

    Having had the misfortune to listen to mp3s with errors in them (due to transfer problems), I can report that these errors are often audible with extremely loud and unpleasant pops and cracks and what not. The exact way errors are made audible is in this case based on the mp3 algorithm. But even with raw audio samples, random bit errors (which is what this chap must have been having if a new cable fixed it), there is no reason to assume that the errors were among the least significant bits. Meaning that those incorrect samples just as well would have been WAY out of line. I am quite confident that these sort of errors would more likely result in an extremely unpleasant listen, than simply a slightly reduced listening experience.

    Thinking any different is, as already mentioned in other comments, mistaking analog with digital. Either it works. Or it doesn't. And if it doesn't, you WILL notice!

    I rank this among the similarly weird discussions that digital audio sent via coaxial cables sounds different than via optical cables. Why? Why would that be so?? Are those ones and zeros different??

    And while I'm ranting, a comment on what I've seen in some stores: Why on EARTH would you want gold plated optical cables? :)
    funkymoz
    • RE: That SATA cable saga - the aftermath

      @funkymoz while entirely agreeing with your comment on gold plated optical cables I would point out that the major difference between a self-contained digital player (mp3 or lossless) and audio based on S/PDIF is that in the latter case the two devices communicating do not share a clock signal, and it is phase-jitter in recovering the clock that is seen as the pre-eminent source of audible artefacts in an otherwise "perfect" digital system.
      Exactly how great these artefacts are, and their audibility is the basis for most of the debates (some entirely spurious) about digital sound playback quality.
      paul_t_milligan@...
  • Pops and crackles.

    I often read from so called 'audio experts' that they often hear noise and pops and crackles etc. from MP3 or digital sources. Often the kit they are using is exotic and highly specialised.

    I've been using middle of the road gear for computer audio for years and have never heard such bad audio from any of mine. All I hear is smooth music, no pops, no crackles, just tunes. Yes I listen carefully and my main audio gear is all Meridian (a company that embraced digital before many others).

    The question I ask is - Is this high end audio/DAC gear made by folks that havent a clue what they are doing and just poor quality or are these so called 'experts' just hopeless at configuring computer kit properly?

    I reckon a bit of both.
    daglesjunk@...
  • RE: That SATA cable saga - the aftermath

    There are none so deaf as those who will not hear.

    Having been an audiophile for over 40 years I am constantly amazed at the differences I can hear between different pieces of equipment. At my age I am supposed to be half deaf.

    The originaal CD standards, still in use, were set based on technological capabilities in 1974 (between Philips and Sony). Anybody that cannot hear the difference between digital and analogue using those standards IS either half deaf, or has never listened to live, unamplified music.

    While 0's and 1's are undoubtedly constant, they way different pieces of equipment deal with those 0's and 1's varies considerably. Whether it is due to impedance effects, different draws on power supplies, I cannot say. But those who cannot hear those different, or cannot measure them, either do not know how to listen, or what to measure. This last is important as what measures the psycho-acoustic effect of sound on the human ears is still far from defined.
    jorjitop
    • RE: That SATA cable saga - the aftermath

      @jorjitop Having read for years audiophile's opinions lacking any support beyond confirmation bias, I'm amazed anyone still refers to themselves as an audiophile.

      A group of professional magazine writers for a major audiophile magazine were tricked into a blind test pitting monster speaker cables and plain old wire coat hangers straightened out. The majority picked the coat hanger as the superior "cabling".

      Psycho-acoustics is by it's definition unscientific and therefore basically unmeasurable.
      rtk