Measurements vs listening

May 1, 2021
 by Paul McGowan

16 comments on “Measurements vs listening”

  1. Walking where angels fear to tread! 🙂 Never shying away from walking the tightrope without a safety net Paul, I think this time around you’ve done a great job. Your food analogy, this time at least, worked for me. Then followed up with the Darren Myer’s phono stage story reinforced your argument. Well, reinforced my view at least. Well done mate! (‘Thumbs up’ emoji.)

  2. I just hope that ‘CtA’ can get his/her head around the concept
    of listening to audio instead of just measuring it.
    Hopefully this short video from Paul McGowan will go some way
    to help said ‘CtA’ to achieve this goal…fingers crossed.

  3. I’m not against listening, but I think we should discuss the importance of using blind testing for listening evaluation. It’s almost impossible to resist other factors (appearance, cost, reputation etc.) in a listening test, even when we’re aware of the hazards. About 1996 I was in a shop listening to Marantz CD63 and CD63SE, priced at £200 and £350 respectively. The SE model sounded significantly better, but I couldn’t decide if it was worth the extra money. The audition was interrupted, and when I resumed listening, I’d forgotten which selector button corresponded with which player. From that point onwards the two players sounded indistinguishable.

    1. Well said, Mark.
      Too many people think their ears are infallible and it’s their eyes doing the listening. Utter nonsense.
      We have to remember Paul stating that anything under 0.1% is inaudible and then this. Sure!
      There also astrology and homeopathy.

    2. Good afternoon Mark!
      For what it’s worth, I agree with you, whole heartedly!
      You can test the crap out of a Pease of audio equipment.
      But you’ll never know what it’s gonna sound like until you set down and listen to it.
      A quite a lot of people miss the point on that when it comes to audio.
      And this not just Audiophiles doing this, it’s also everyday joe people that aren’t Audiophiles doing this too.
      They’ll take a pare of speakers and look at the specs on them.
      But when they read the part that says, “20HZ to 20KHZ”, they’ll say, “these speakers have just as much bass, as they do mid range, and high end.”
      But the problem with that is, they’ll never know what they sound like until they hook them up and listen to them.

    3. You miss a few things: 1) the interruption changed your focus, if not a lot more; 2) that one can infer that you started with a bias toward the more expensive; and 3) that your sample was of one person, one time. Yet, from that one experience you generalize to the rest of us.

      If your ears, aptitudes and personality don’t allow you to tell the difference(s) or no difference between components, let alone better or worse to your tastes, then you’re right to buy the cheap stuff and enjoy. Blind tests are almost invariably artificial because they’re difficult to run in a valid and reliable, i.e., meaningful, way, and in part because they add one or more factors that divert your attention from what’s important in listening — degree of enjoyment and thinking about why the difference in that reaction (if any).

      1. Well, I completely understand that I can’t extend an anecdote from 1996 to draw conclusions about all listeners and all subjective test procedures, but I am convinced that, if we’re going to carry out subjective assessment of hi-fi products, we need to use blind testing. It’s impossible to give an accurate opinion otherwise.

        The important conclusion I draw from my anecdote is that I thought I could hear a difference when I couldn’t. The accuracy (or otherwise) of my hearing isn’t the issue.

  4. We need to measure AND we need to listen. But measurements can tell us more than many believe. They just aren’t the measurements usually cited such as harmonic and IM distortion or even frequency response.

    David Berning once demoed for me an amp with variable negative feedback and the sound improved as the feedback was reduced. The harmonic and IM distortion increased and the frequency response from the speaker got worse and the sound was better. I asked David if any measure reflected the improved sound. He said linearity was improved with reduced feedback.

    My friend Murray Zeligman once picked a cartridge from a 1 kHz square wave and a frequency response curve. And I’ve seen him design a speaker on a computer in half a day and it came out sounding 95% what he expected. He looked at the predicted transfer function of each driver with its crossover and box. The transfer from one driver to another matters a ton.

    As time goes on if we look we will find better and better predictors. But at the end even if you believe your prediction true you still have to listen to prove it. We design space rockets on computers but we still test them without astronauts first.

    1. I’m an engineer, so I’m naturally inclined to want to design something on paper/computer and measure using test equipment. It’s what engineers do. Having said that, there’s normally a poor correlation between the simple standard measurements (THD, IMD etc.) and subjective opinion scores, so we have to listen to our designs as well as measure them. I would like to get to the point where we have better objective measurements that accurately predict the results of listening tests.

      1. Mark,
        Well since everyone’s hearing is different I’d say
        that that is going to be pretty much impossible.
        And I’d like to have world peace.

  5. Clap your hands in a concert hall and do it in a more or less acoustically dampened listening room. The sounds you perceive will be most different. Thus before trusting your ears optimize your listening room and stereo-set up using test tracks with defined and well recorded single (!) instruments and voices and a precise (!) studio monitor under nearfield conditions. Then you can play endlessly with different speakers designs or amps and DACs designed for a specific brand or tube sound. In the result will be that you need different speakers and amps for different recordings, recordings where you are not knowing the specific recording conditions. 🙂

  6. As a budding audiophile with old ears and damaged hearing plus limited listening skills, I’m not going to make any claims on being able to tell differences in electronics, wiring, and even speaker brands. But as a retired, and obsolete, engineer I can still read a frequency graph so I tend to side with the objectivity side of the argument about measurement vs listening. I very doubt those who claim “oh my gawd, …” or “jaw dropping …” differences in cables or brands of amplifiers, etc UNLESS they are very experienced listeners who are aware of specific design differences and are focused on those specific audio characteristics. To me, working on improving the source material however makes a lot of sense.

  7. This argument has been going on since I started an interest in hifi 60 years ago, and it’s amusing to see how intransigent people can become once they commit to a point of view. Therefore, the debate will never be settled. What is unarguable, however, is that, while advances in technology have facilitated much better equipment, the art of recording music has largely gone downhill big time.

    1. Good morning Mr. Kannry!
      What you said about the recording industry, is true.
      But then again, it isn’t true.
      Most of them, are still producing PCM.
      But at the same time, a very select few, are producing DSD.
      Paul and Gus are the only two men that I know today, that are doing that.
      I ordered four SACD’s from PS Audio, and they sound like the people are right in front of me!
      If the recordings are done right the first time, you’ll always get that every time.
      Both Paul and Gus has pulled that off without a plum.

  8. If you can hear a difference in a blind test, then there is a measurement instrument that discerns the difference: our ears. We should then in principle be able to create electronics that can do the same.

    But that doesn’t mean that if our current electronic measurements don’t discern a difference, that the measurement devices we call our ears can’t. So measurements are ok, but they have to be able to measure the right thing! Just appealing to “this measurement” does not cut it.

Leave a Reply

Stop by for a tour:
Mon-Fri, 8:30am-5pm MST

4865 Sterling Dr.
Boulder, CO 80301

Join the hi-fi family

Stop by for a tour:
4865 Sterling Dr.
Boulder, CO 80301

Join the hi-fi family

linkedin facebook pinterest youtube rss twitter instagram facebook-blank rss-blank linkedin-blank pinterest youtube twitter instagram