Go Back   Home > Forums > Blogs > abraxalito

Rate this Entry

Digital that sounds like analog

Posted 10th December 2012 at 04:18 AM by abraxalito
Updated 15th February 2013 at 06:11 AM by abraxalito

For those who missed Frank (fas42)'s link on a thread I started then here's where I'm continuing my minimal oversampling DAC developments for the time being : Digital that sounds like analog
Views 4023 Comments 108
Total Comments 108


  1. Old Comment
    I'm curious whether you've noticed this effect at any time, in your experimenting - and, where you able to listen to that clip at a decent enough quality to pick the transition taking place?

    I've personally found the timescale can be anything from several minutes, to the length of a CD or longer. I've never had the instrumentation to really ascertain what's happening, but in essence it appears, sounds as if the DAC mechanism is gradually tipped, or edged into a state of electrical imbalance where it develops that classic, undesirable, "digital" sound.

    Playing at louder levels can make it happen faster, but I believe this is purely because the amplifiers working harder spawns more interference from stressed power supplies.

    I've long had a simple method of "measuring", and resetting, the effect. Run the system until the sound is well and truly "off", and then listen to a particular track that you use as a test piece. Immediately switch off all the gear; DACs, amps, etc, but leave only switched off long enough until all the smoothing caps have fully discharged. Power up everything again, and immediately listen to that test track again. If the problem exists then this track should sound definitely, or even dramatically, better than on the first listen.

    Posted 8th January 2013 at 10:35 AM by fas42 fas42 is offline
    Updated 8th January 2013 at 10:37 AM by fas42
  2. Old Comment
    abraxalito's Avatar
    I haven't noticed it happening over time with my system no. I was able to listen to the clip just too bored to. Listening to other people's systems via Youtube just doesn't teach me anything, I prefer to be working on my own system.

    My suspicion is that the system must have a CM noise problem - perhaps the transformers are the things changing over time. Certainly iron cores have properties which change over those timescales. Ferrite does too.

    One thing I noticed early on in my travels - I was using small ferrite cored inductors to protect opamp inputs from RF. On first fitting them the sound was fine. But then after a day, it went back to how it had been before - harsh, sibilant. This lasted about another day and we then returned to the clean sound which didn't change subsequently. This effect was quite repeatable on first modding a pair of active speakers with said inductors. Charles Hansen I believe has had uneven experiences from using ferrites.
    Posted 9th January 2013 at 02:42 AM by abraxalito abraxalito is offline
  3. Old Comment
    Very interesting to hear you say firstly that you didn't have such an experience; followed by a story which in fact details you having the very thing!

    My experience has been that the more recent gear is less prone to these problems than older stuff, whether by design or happenstance. It is extremely difficult to pin down precisely the origin of these sorts of issues a lot of the time, trial and error comes into it; if you don't have the artifacts up front it's a good, and a bad thing. Good for obvious reasons, bad because the gremlin can come and strike you at the worst possible moment, such as when you're demonstrating to a 3rd party -- Murphy always wins!! :D

    My suspicion with the ferrites is that the stability of the winding is the culprit: while installing, soldering in place the wire is disturbed with the handling. Now somewhat physically "disturbed" out of its normal "resting" state it manifests the very symptoms I'm speaking of, but with time the wire "settles" down into a "relaxed" orientation again.

    I would suggest as an experiment to deliberately manhandle the ferrite winding in situ, just a bit roughly, to mimic the installation handling, and see whether the rough sound comes back again. And then goes away also ...

    Posted 9th January 2013 at 04:04 AM by fas42 fas42 is offline
    Updated 9th January 2013 at 04:06 AM by fas42
  4. Old Comment
    abraxalito's Avatar
    Its a different timescale my story to the one you were talking about. You noticed degradation over the length of a CD - mine was much slower. Ferrites have what's called 'disacommodation' so rather than it being the winding it could be the thermal stress caused by soldering.
    Posted 10th January 2013 at 12:37 PM by abraxalito abraxalito is offline
  5. Old Comment
    Nearly all the distortion phenomena I worry about have varying timescales, from minutes to many hours. The type which most people are concerned with, which are easily measured, have timescales of the order of seconds. For example, the RF related interference artifacts that I have noted are almost instantaneous: switch on a problem device, sound immediately degrades; switch it off, sound immediately regains quality.

    What is clear is that everything matters, and really the biggest hurdle is determining where and why the next most "problematic" behaviour is ocurring.

    The extremely heartening overall result I've achieved is understanding that all audible bad behaviour [I]can[/I] be eliminated, meaning reduced to a point where it is subjectively irrelevant. Hence my motto, "there is no such thing as bad recording" -- if the playback of a track is subjectively not "right" then it always, always tells me that there exists a fixable issue ...

    Current, highly compressed, pop recordings are excellent candidates: do yourself a favour, as a local music "guru" is famous for saying, and try a recent Foo Fighters album at high volume. If your system can unravel this cleanly then you're travelling well ... ;)

    Posted 10th January 2013 at 11:13 PM by fas42 fas42 is offline
  6. Old Comment
    BTW, you're doing a nice job, working the crowd in the "other place" -- I'm impressed! :cool: ...

    Posted 13th January 2013 at 11:05 PM by fas42 fas42 is offline
  7. Old Comment
    abraxalito's Avatar
    Ah you mean doing a great marketing job for Thorsten's kit?
    Posted 14th January 2013 at 02:28 AM by abraxalito abraxalito is offline
  8. Old Comment
    Still workin' over there ... :)

    I will take issue with this:

    [QUOTE]When I listen to DSD-mastered CDs I notice that the noise seems to be added at HF - bass dynamics remains fairly decent, but the top end gets congested, fuzzy[/QUOTE]

    If you're hearing this then you've got a system problem: congestion in the top end is the giveaway that there is a solvable issue in the playback side of things; dumping the blame on the recording is an easy way out ...

    Why do I say this? Because this is exactly what I've been dealing with for years: fuzzy, congested treble in "poor" recordings. Like most, decades ago I just passed the blame back to the record companies, etc, but I don't use that excuse any more. I [I]know[/I], because I've done it over and over again, that such recordings can be revealed to be pristine clear, but [I]only[/I] if I eliminate every last ounce of added distortion, contributed by my system weaknesses ...

    Posted 2nd February 2013 at 01:37 AM by fas42 fas42 is offline
  9. Old Comment
    abraxalito's Avatar
    So why do only some recordings show the HF congestion and not others? And also why does the presence of HF congestion correlate with S-D (DSD) recordings almost perfectly? Its not that the recordings are 'poor' ones, rather its that they've used particular ADCs.

    If you have answers for these questions then (and only then, no interest at all if you have no suggestions on them) what would be your suggestion for what my system is doing wrong? Incidentally did you see the THD measurement of my chipamp? Any criticism to raise there?
    Posted 2nd February 2013 at 10:29 AM by abraxalito abraxalito is offline
  10. Old Comment
    When my own, I repeat, my own system is not in "good tune" certain recordings, styles of recordings, will show up as being worse off. Typically these will be more modern recordings, that done using more recent studio techniques, which for want of a better term, use more "digital" approaches. Historically, this will probably show to correlate almost perfectly with the wholesale use of S-D converters, and hence the linkage.

    My thinking is still evolving on what underlies all this, but perhaps bizarrely to you and others, it's because these recordings have a greater effective dynamic range that they sound "worse". Older style recordings have an inherent noise floor in the recording which is higher, which is a highly effective masking agent for sounds which can unpleasantly intrusive. It can be likened to the distant water fountain in the city effect, the soothing, random hissing and gurgling of the water distracts the mind from the grinding sounds of the street traffic.

    The street traffic noise correlates with HF congestion obviously, and the latter is a mixture of wanted, recorded, low level sound, and distortion. Which the ear can't separate, hence it becomes a "bad" recording. I have a nice collection of such albums, which are good "stress tests" ...

    Answers? You are very familiar with many of the techniques I use, so your thinking is certainly in the right space. Power supplies, grounding are the obvious ones you've done plenty about already.

    HF congestion is also the giveaway marker of interference effects. How I would address the situation there would be to temporarily shut down every, I repeat, every source of possible interference to as far away as physically possible. Every outlet switched off, power cord in the house, flat, pulled out, all lights off, all wireless devices switched off, mobile, cordless phones, even the circuit breakers in the power box pulled. You want the environment of your system to an electrical desert, apart from obviously the circuit supplying the system. Having thoroughly warmed up, conditioned the system beforehand, put on a particularly "nasty" test track, how does it sound? Give it time to settle in, put it on repeat perhaps. If there is a difference, start putting electrical devices back on line, one by one - you get the idea.

    This is a technique I've used over and over again to isolate where my problems may be coming from. How you "fix" it then obviously depends on everything ....

    Posted 3rd February 2013 at 01:12 AM by fas42 fas42 is offline
  11. Old Comment
    I'm appreciating what a character Lynn Olson is, from his comments "over there", he's been in an excellent position to monitor the amount of crud that actually ripples through real circuits, from his own and acquaintances' monitoring. I have not been fortunate enough so far to have access to the right gear that picks up this sort of goings on everytime, so I stayed with good ol' reliables, the ear/brain. This has always told me that normal audio sound is full of muck, and whether I'm going in the right direction or not ...

    His approach is obviously very different from mine, but it appears he has the same goals. The fact that a totally alternative philosophy can achieve satisfying sound merely emphasises that the end point is very real, and is essentially the same for everyone. The language used is the giveaway, he noted his surprise that "bad" recordings could sound, well, not bad!

    Audio [I]will [/I]get there in the end, there are enough people out there with enough passion, and part-knowledge to make it happen. The fumbling in the dark, pawing the poor elephant in the room, trying to understand the bigger picture, will finally combine to give everyone the "good stuff" ...

    Posted 24th February 2013 at 11:08 PM by fas42 fas42 is offline
  12. Old Comment
    abraxalito's Avatar
    Yep, Lynn's writings have been a major encouragement to me over the past couple of years in terms of maintaining the passion for 'audio nirvana'. Where he differs from me is mainly in the realm of cost - his Karna amp is pretty expensive, even to build for yourself. I'm motivated much more by bringing digital to the mass market - which ultimately means portable. This rules out valves
    Posted 26th February 2013 at 04:14 AM by abraxalito abraxalito is offline
  13. Old Comment
    I'm watching the tussle with xiphmont with great interest ... the man's got a bit more ammo than I did ... ;), :)

    Posted 1st March 2013 at 12:08 PM by fas42 fas42 is offline
    Updated 1st March 2013 at 12:19 PM by fas42
  14. Old Comment
    abraxalito's Avatar
    But now seems he's out. But andy_c has come like the cavalry over the hill...

    <later> Looks like andy_c didn't come equipped with any real ammo after all, he's been blowing smoke the whole time.
    Posted 2nd March 2013 at 03:24 AM by abraxalito abraxalito is offline
    Updated 2nd March 2013 at 06:37 AM by abraxalito
  15. Old Comment
    Taking turns ... I like it! Good viewing from the dress circle ... :D

    Posted 2nd March 2013 at 10:01 AM by fas42 fas42 is offline
  16. Old Comment
    abraxalito's Avatar
    Did you view andy_c's hypocrisy ? I can just feel the disdain for me dripping from his posts It could be he's still carrying a torch after all these years - he banned me from DIYoverengineering which then subsequently sank without trace
    Posted 2nd March 2013 at 10:49 AM by abraxalito abraxalito is offline
  17. Old Comment
    Okaaay ... I was wondering why the conversation turned tetchy in that way at the end there ...

    And, of course, you now have Monty back to keep the party lively ... :)

    Posted 2nd March 2013 at 11:05 AM by fas42 fas42 is offline
  18. Old Comment
    abraxalito's Avatar
    But Monty don't seem to be getting it - perhaps he was over tired, I dunno. Think I'll show a clean pair of heels to that thread unless he evinces a lightbulb moment or two
    Posted 2nd March 2013 at 12:01 PM by abraxalito abraxalito is offline
  19. Old Comment
    Monty [I]is [/I]having a hard time getting through, the latest round demonstrates that the frustration felt by even the most reasonable person can be overwhelming, trying to deal with the twisted thinking of so many, especially those who should know a lot better

    So much of this game is tied up with confusing faults of implementation, "bad" engineering, with supposed inadequacies of the fundamentals. And while this tangled nest of worms continues to writhe, everybody stumbles around in the dark trying to make sense of the elephant -- how that's for a brilliant mixed metaphor ...!!

    Posted 4th March 2013 at 10:42 PM by fas42 fas42 is offline
  20. Old Comment
    abraxalito's Avatar
    The fact that he got angry shows he's not the most reasonable person. I missed that it was a joke, but figured that if someone was daft enough to really believe it then he was too daft to understand being corrected. Hence I left it well alone
    Posted 5th March 2013 at 01:41 AM by abraxalito abraxalito is offline

New To Site? Need Help?
Copyright 1999-2017 diyAudio