Results 1 to 13 of 13

Thread: ABX test(s) for your digital files

  1. #1
    Senior Member
    Join Date
    Feb 2004
    Location
    Central Coast California
    Posts
    9,042

    ABX test(s) for your digital files

    This might have gone into the "iPod playback systems" or "How good is your hearing" threads, but it seems to be unique enough to warrant its own discussion. There's software for both OSX and Windows playback.

    http://techland.time.com/2012/03/02/...nd-lossy-audio

    I'll try this over the weekend, and if others are so inclined, maybe we can have a civilized discussion. For me it's more about discovery than defending a position or particular point of view.
    Out.

  2. #2
    Member
    Join Date
    Aug 2007
    Location
    Amish Paradise, Ohio
    Posts
    44
    Spent the weekend working with the Foobar ABX comparator. I had several tracks I encoded to WAV, FLAC, VBR Mp3 (LAME 3.98 encoder), 64k Mp3, and 8-bit -q7 Mp3. Except for the 8-bit, which had a constant low-level wobbly hiss in quiet passages, I could not even venture a guess in every A/B/X/Y trial. I created the 8-bit preset with the aim of getting the worst possible quality when I could not distinguish between WAV and the highly-compressed 64k Mp3. I was really trying to hear some sort of roughness or distortion in the compressed audio, the "hiss" was all I could find. At times I thought I found a difference between A and B, but in trying to match A with X or Y, I couldn't do it.

    The equipment was all solid mid-range stuff. L100t's properly placed out from the walls in a 14 x 24 room/ Yamaha DSP-A1 (My beloved Mac 2105 developed a hum in both channels so it's down for a couple months), SPDIF from the computer using the Yamaha's internal DAC. With some of the other computer A/V players using HDMI out of the computer I'm able to do an A/B between my TV's $5 DAC and the Yamaha's. There is a definite audible difference between the two but the Yamaha DAC isn't always preferable. I plan on buying a new DAC/preamp later this year, we'll see what changes.

  3. #3
    Member
    Join Date
    Apr 2010
    Location
    North of NYC
    Posts
    48
    Here is an interesting read that touches on this subject and probably explains why I am having trouble finding anything wrong with my Spotify streams when compared to the same tracks in 16/44.1.

    http://people.xiph.org/~xiphmont/demo/neil-young.html

    Also feeling no need to pursue Hires formats. That's a relief!
    Control 25AV on the deck - L1 - L20t & L80t in piano black - 4312A - 4430

  4. #4
    Administrator Mr. Widget's Avatar
    Join Date
    Apr 2003
    Location
    San Francisco
    Posts
    9,720
    Quote Originally Posted by Jonis View Post
    Here is an interesting read that touches on this subject and probably explains why I am having trouble finding anything wrong with my Spotify streams when compared to the same tracks in 16/44.1.

    http://people.xiph.org/~xiphmont/demo/neil-young.html

    Also feeling no need to persue Hires formats. That's a relief!
    Well... Monty at xiph.org should get a job at Consumer's Reports... another case of someone looking at specifications and not giving the real world practical applications a serious listen, or perhaps being so convinced by their understanding of what is going on, that they don't hear what they are hearing...

    Monty, you are not entirely wrong, but you are absolutely far from right... your arguments will hold under certain circumstances and most certainly not in many others!


    Widget

  5. #5
    Member
    Join Date
    Apr 2010
    Location
    North of NYC
    Posts
    48
    Quote Originally Posted by Mr. Widget View Post
    Well... Monty at xiph.org should get a job at Consumer's Reports... another case of someone looking at specifications and not giving the real world practical applications a serious listen, or perhaps being so convinced by their understanding of what is going on, that they don't hear what they are hearing...

    Monty, you are not entirely wrong, but you are absolutely far from right... your arguments will hold under certain circumstances and most certainly not in many others!


    Widget
    Granted, his real world and your real world are far from each other. His pecuniary interest is to make a living developing codecs that push the envelope of getting as high a quality music/audio feed to the ears of the masses. Your pecuniary interest is to make a living selling equipment/high end systems that are targeted at a discriminating few, where money is no object. Definitely different real worlds. I can understand why you would disagree with him.

    If, as you say, he is “absolutely far from right...”, which statement(s)/assertion(s) is/are furthest from right?
    Control 25AV on the deck - L1 - L20t & L80t in piano black - 4312A - 4430

  6. #6
    Senior Member
    Join Date
    Feb 2004
    Location
    Central Coast California
    Posts
    9,042
    Quote Originally Posted by Mr. Widget View Post

    Monty, you are not entirely wrong, but you are absolutely far from right... your arguments will hold under certain circumstances and most certainly not in many others!


    Widget
    Quote Originally Posted by Jonis View Post

    If, as you say, he is “absolutely far from right...”, which statement(s)/assertion(s) is/are furthest from right?

    Yes, I think an enumeration of the "certain circumstances" and the "not in many others" would be both enlightening and, perhaps, entertaining.
    Out.

  7. #7
    Senior Member richluvsound's Avatar
    Join Date
    Dec 2006
    Location
    london england
    Posts
    2,060

    Myth Busting

    Thanks Ti,

    an interesting read ,some of which I understood and some is stuck to the wall above my head . Would be great if Grumpy could lend some insight for the layman !

    Rich

  8. #8
    Administrator Mr. Widget's Avatar
    Join Date
    Apr 2003
    Location
    San Francisco
    Posts
    9,720
    Quote Originally Posted by Jonis View Post
    Granted, his real world and your real world are far from each other. His pecuniary interest is to make a living developing codecs that push the envelope of getting as high a quality music/audio feed to the ears of the masses. Your pecuniary interest is to make a living selling equipment/high end systems that are targeted at a discriminating few, where money is no object.
    I take exception to this statement as I read it to imply my primary interest is in making a profit. While it is true I do make a living from designing A/V systems and selling them to people, less than 1% of my customers are interested in high fidelity. I always try to encourage it, but I understand that for most of them "invisible" sound is their goal even if mine is "transparent" sound. I give them what they want and need to achieve their goals even if I might prefer a much higher end product.

    Quote Originally Posted by Jonis View Post
    If, as you say, he is “absolutely far from right...”, which statement(s)/assertion(s) is/are furthest from right?
    His principal criticism of high rez files is that by having a Nyquist frequency of 96KHz (24/192 files) you may drive your electronics into nonlinearites... true, if you have rather poor electronics. However should we damn high rez just because inexpensive electronics can't handle the frequency spectrum? Most of my electronics are rated flat and low distortion to 100KHz. I'm not sure how my "antique" Marantz Model 7(c) handles ultra high frequencies, but I'd assume it will simply roll them off in a benign fashion. As for the importance of a frequency response out to 96KHz, that is up for a real debate, but having a Nyquist frequency of 22.05KHz (redbook CD) with the necessary "brick wall" filter ~19KHz to avoid harsh digital noise is widely believed to be a very real source of negative sonics.

    My point is simply that making a grand condemnation of high rez files because it might cause problems for some electronics is a wrong headed approach. Now, if he wanted to say, "high rez files are unnecessary for most musical playback systems and may even cause undue distortion in some lesser systems", I'd agree completely.


    Widget

  9. #9
    Senior Member grumpy's Avatar
    Join Date
    Oct 2003
    Location
    SoCal
    Posts
    5,742

    Opinion time...

    OK... slogged through both articles.

    Not too much to argue against... I don't really agree with all of the absolute statements
    and "obvious" conclusions, but the 2nd piece is self-consistent. The truth is, most people
    would be thrilled with the sound quality of a well-produced, well-mastered CD. The
    truth also is: many/most(?) are not. Sometimes (not always) more care is taken to create a
    'premium' product in the form of high-rate PCM or SACD... this, in itself, can produce
    amazing differences (e.g., JVC's XRCD process.. still a 16bit output). The author stated this.

    High-rate and resolution recording is a necessity. The author also stated why.
    What is being recorded -often- has ultrasonic content that is not completely filtered
    by microphone or preamp (ahead of the ADC). If not properly filtered at, or ahead of,
    the ADC, aliasing will occur (crap folds back over into the audio spectrum).

    Assuming that is taken care of with well implemented filters (nothing is recorded
    in excess of 1/2 the sample rate), and we ignore the effect of the typical 5-cent
    oscillator that typically determines the phase noise (timing of each sample
    that is again modulated by another such oscillator on the DAC side ), then the "proper"
    playback of 20KHz-limited material or 96KHz-limited material (192KHz samplerate)
    is indeed a function of the following electronics (that certainly have the -potential-
    to non-linearly mix ultrasonic input with the equipment's designed-for bandwidth ... again, sort
    of folding ultrasonic "music" into the normal audio range... likely in an un-harmonically
    related fashion) and then what comes out is most often passively low-pass filtered with
    the physical reproducer (speaker). There are a lot of assumptions in there, and frankly,
    for the DAC end, 96KHz seems plenty high enough to me (most low-level electronics can
    pass 48KHz... power amplifiers at more than a few watts might be an issue).

    Bit depth? I don't see that this is such a big deal... It's hard enough to produce a DAC that
    can linearly output much more than 21-bits outside of laboratory settings. Internally, I'd
    like to see 24 or 32-bit -processing- (especially if the user can fuss with the bits, e.g., digital
    EQ) but more than 24bits (input from media or output via DAC) seems... unnecessary.
    Well mastered discs (perhaps with a modicum of gain riding) can probably almost all fit
    into the HDCD 20-bit equivalent or less (avoiding fireworks and coal trains).

    Arguing for 16-bits and 44.1KHz as being completely adequate is a valid mass-consumption
    train of thought. To say that we know everything there is to know about human perception,
    playback-chain improvement (including better characterization of each element's full-range
    bandpass such that we aren't operating it outside of it's capabilities), and that anything
    outside of that box is a complete waste, seems ... uninspired and short-sighted.
    Last edited by grumpy; 03-07-2012 at 12:43 PM. Reason: originally typed 2x ... meant 1/2

  10. #10
    Administrator Mr. Widget's Avatar
    Join Date
    Apr 2003
    Location
    San Francisco
    Posts
    9,720
    Quote Originally Posted by grumpy View Post
    Bit depth? I don't see that this is such a big deal...
    To be fair to Marty, he didn't make a big deal about bit depth either... and as you suggest, if we can get a true bit depth of 20 bits at the playback end we're doing good, but even a true 16 bits is "quiet" enough for me.


    Widget

  11. #11
    Member
    Join Date
    Apr 2010
    Location
    North of NYC
    Posts
    48
    Quote Originally Posted by Mr. Widget View Post
    I take exception to this statement as I read it to imply my primary interest is in making a profit.

    Mr. Widget, sorry about my statement that you took exception with. Naturally, it didn’t come out the way I necessarily meant. My apologies, but remember- profits are important for all of us. We all need to make them.

    My take away from Monty’s article—if there is no sonic benefit, why go there? Why pressure consumers to upgrade hardware, repurchase already owned music for no audible benefit? Why force device makers to design flat/low distortion out to 100kHz when no sonic benefit is derived? I know…. no one is forcing anyone to do anything, but let’s face it, market pressures and “keeping up with the Joneses” are real forces that cause companies and individuals to make bad choices.

    As an aside and not meant in a patronizing manner at all, I appreciate the likes of Monty, Mr. Widget, Grumpy and others providing their learned opinions and viewpoints, particularly when they are science based. Helps the rest of us.
    Control 25AV on the deck - L1 - L20t & L80t in piano black - 4312A - 4430

  12. #12
    Senior Member grumpy's Avatar
    Join Date
    Oct 2003
    Location
    SoCal
    Posts
    5,742
    I'm just as happy to read well-grounded tests results as much as listen
    to the 40yr experience of an industry capturer/masterer of live performances
    (having just had the pleasure of the latter experience at AIXrecords in LA).

    When someone can sit and mix at a desk all day and not get tired of it,
    vs. getting burned out after a few hours, that tells me -something- is
    going on, whether we completely understand it or not (the experience
    was a 96K/24bit environment, both 2ch and 5.1).

    I like to keep an open mind; in my view science follows experience to
    help us understand our world and ourselves... sometimes one shoots
    out in front of the other

    This is an area where in my opinion there is work to be done. Other
    areas in audiophilia-land grind on me hard (in a bad way); they
    make anyone with hi-fi aspirations look like a moron by association...
    So I do understand Monty's effort to peel away some of the BS that
    can accompany certain claims.

  13. #13
    Administrator Mr. Widget's Avatar
    Join Date
    Apr 2003
    Location
    San Francisco
    Posts
    9,720
    Quote Originally Posted by Jonis View Post
    My take away from Monty’s article—if there is no sonic benefit, why go there? Why pressure consumers to upgrade hardware, repurchase already owned music for no audible benefit? Why force device makers to design flat/low distortion out to 100kHz when no sonic benefit is derived? I know…. no one is forcing anyone to do anything, but let’s face it, market pressures and “keeping up with the Joneses” are real forces that cause companies and individuals to make bad choices.
    I agree in principal... a good case in point would be the advent of SACD. A noble design but poorly handled. The original "high end" SACD players operating in a high end system showed just how good digital audio could be, but putting the SACD technology in $200 CD players was just stupid. (So why go there.) It offered little to no sonic benefit and basically made people think SACD was a hoax of some kind... "audiophile-dom" being full of them.

    Unfortunately because of this, those of us who do listen and care and are willing to make the financial sacrifice to buy the better gear are not going to be able to buy many SACDs going forward. My gripe with Monty's position is he is so resolute.

    "The number one comment I heard from believers in super high rate audio was [paraphrasing]: "I've listened to high rate audio myself and the improvement is obvious. Are you seriously telling me not to trust my own ears?"Of course you can trust your ears. It's brains that are gullible. I don't mean that flippantly; as human beings, we're all wired that way."


    I even agree with his notion about brain gullibility... I ranted, some would say endlessly, about perception not long ago, but that aside, I think it is easily proved on very good gear that even those of us with only average hearing can and do hear the advantage of high resolution audio. I also know that on a typical AVR playing a 24/192 vs. 256-kbps MP3 it might be difficult to impossible to hear the difference. But that is no excuse to damn high resolution files for the few of us willing to pay for them and the equipment to enjoy the difference.

    BTW: The differences as I subjectively enjoy them are a slightly less bright and glaring top end and a greater sense of space in a stereophonic recording.


    Widget

Thread Information

Users Browsing this Thread

There are currently 1 users browsing this thread. (0 members and 1 guests)

Similar Threads

  1. Replies: 21
    Last Post: 03-13-2008, 12:15 PM
  2. Replies: 7
    Last Post: 05-07-2007, 02:48 PM
  3. TIFF files
    By Earl K in forum Forum Feedback
    Replies: 27
    Last Post: 02-12-2005, 05:27 PM
  4. A.C.A. NEWSLETTER - 18 new files from the last 1 1/2 months!
    By skaloumbakas in forum Lansing Product General Information
    Replies: 0
    Last Post: 08-03-2003, 03:09 AM

Posting Permissions

  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts
  •