Announcement

Collapse

HUG - here for all audio enthusiasts

The Harbeth User Group is the primary channel for public communication with Harbeth's HQ. If you have a 'scientific mind' and are curious about how the ear works, how it can lead us to make the right - and wrong - audio equipment decisions, and about the technical ins and outs of audio equipment, how it's designed and what choices the designer makes, then the factual Science of Audio sub-forum area of HUG is your place. The objective methods of comparing audio equipment under controlled conditions has been thoroughly examined here on HUG and elsewhere and should be accessible to non-experts and able to be tried-out at home without deep technical knowledge. From a design perspective, today's award winning Harbeths could not have been designed any other way.

Alternatively, if you just like chatting about audio and subjectivity rules for you, then the Subjective Soundings area is you. If you are quite set in your subjectivity, then HUG is likely to be a bit too fact based for you, as many of the contributors have maximised their pleasure in home music reproduction by allowing their head to rule their heart. If upon examination we think that Posts are better suited to one sub-forum than than the other, they will be redirected during Moderation, which is applied throughout the site.

Questions and Posts about, for example, 'does amplifier A sounds better than amplifier B' or 'which speaker stands or cables are best' are suitable for the Subjective Soundings area only, although HUG is really not the best place to have these sort of purely subjective airings.

The Moderators' decision is final in all matters and Harbeth does not necessarily agree with the contents of any member contributions, especially in the Subjective Soundings area, and has no control over external content.

That's it! Enjoy!

{Updated Oct. 2017}
See more
See less

The loudness dynamics of music (examples)

Collapse
This is a sticky topic.
X
X
  • Filter
  • Time
  • Show
Clear All
new posts

  • The loudness dynamics of music (examples)

    This thread is a fork from another thread covering how much amplifier power is likely to be needed at home to approximately deliver classical music life-life dynamics. Original thread here.

    Humans are not good at judging the loudness of sound. In fact we are so bad at it that we allow ourselves to physically destroy our hearing and only when significantly deafened, realise our mistake. It's still happening, right now in your communities where young people are exposing themselves to destructive loudness.

    Since we're not built to accurately assess loudness, we have to use technical equipment, and its convenient to measure in units which correlate well to how we actually perceive sound. So we measure frequency in cycles per second (these days in Hertz, same thing) and we measure matters concerning the quantity of sound, or voltages relating to sound as they pass through audio equipment in volts. Actually we don't quite use volts directly, we convert the audio voltages into decibels just for convenience and to be more appropriate to the way our ears grade sound waves.

    That's really all we need to know. Truly! The pitch of sounds is measured as their frequency, and the amplitude of the sounds are measured in decibels. (Let's not get into issue like sound intensity, let's just concentrate on sound pressure - let's keep it simple).

    That's it then. Hertz (Hz) and dBs.

    Now, frequency is not debatable. It's a matter of measurable fact. But decibels are not quite so universal in their character. We could phone our musician friend half way around the world and ask him to 'play me an A flat on your piano please, right now over the phone...' and he'd do that. His A flat would have (or should have) an identical pitch to your A flat in London.

    But if you asked him to 'play me an A flat on your piano please, right now over the phone... and play it at 74.5dB' he couldn't do that with any accuracy until you and he agreed a frame of reference because decibel units must have a defined frame of reference or they are meaningless, which they really are anyway because they do not exist in isolation. You can't order a box of decibels.

    You could say to him 'play me an A flat on your piano please, right now over the phone... and play it at 74.5dB relative to the sound of dropping a cricket ball 1m onto a tiled floor...' or you could say 'play me an A flat on your piano please, right now over the phone... and play it at 74.5dB relative to the international standard of what audiologists have long agreed is the quietest sound that a normal, health 20 year old human ear can detect in a silent room...'

    Give him either - or make up your own standard which you must both agree to - and you're in the business of describing sonic events in units of decibels. That really is all we need to know. Lock, stock and barrel.

    OK, so imagine that we are promoted from being a humble audiophile home listener to being a recording engineer with a big label. For some, it would be a dream job, although thanks to overcapacity, not a well paid one. the big day has arrived, and we find ourselves sitting at the mighty mixing console - mixing desk in BBC speak. Here you are, at the controls ... picture or here.

    Your job is to balance the sound contribution of the many musicians into a sonically pleasing whole. You may have allocated on microphone to each performer, which gives you maximum technical flexibility in setting the relative level from each, or you may have fewer microphones, but placed strategically by performers playing the same instruments. Either way, your mixing desk will have lots of channel strips, each one connected to a microphone. Nearest to you will be a fader or level control, and according to how far you slide the fader towards or away from you will determine how much sound from that performer's mic will appear in the final mix. You are painting a sound picture using not brushes and paint but tone colour and loudness.

    It's time for the rehearsal. You will decide what sound the most pleasing balance to you across the performers and you will note down (or these days, store in the auto-fader memory) page by page of the score, how you want to fade up or fade down individual performers, or maybe you'll aim for a minimalist recording and once you have the relative balance set, just record 'as is'. It's up to you and your boss, the Producer, entirely.

    Now, aside from the artistic matter of painting your sonic picture - which anyway, is a nebulous sort of business wholly depending upon taste - there is one technical matter which you have to keep an eye on or you're going to be in deep trouble, fast. You have to know with confidence what the peak loudness will be on every microphone channel at every moment from the first bar to the last note. Put another way, you have to know what voltage every microphone will generate as it hears the musician it's pointed at, all the time. You have to be certain that that not even one note in the entire performance is so energetic that the mic will generate a massive audio voltage that will overload the individual microphone input preamp (every mic strip has one mic preamp)*, clipping the preamp, ruining the performance, ruining the conductor's day, ruining your bosses week and jeopardising your career path. That's because clipping cannot be undone; the only solution is to re-record the performance. And that's expensive. It cannot be allowed to happen under any circumstances, or you're out of that comfy chair and back home to being a listener agsin.

    Can we know the minimum sound level in a performance, and do we care about it? The minimum sound level, and hence minimum voltages generated by the microphones, is going to be directly related to the musicians sitting as quietly as they can in readiness to perform, background traffic noise, aircraft and trains, air conditioning and if an audience is present, the inevitable rustling, wheezing and coughing they generate. That minimum noise, and minimum microphone voltage is one of the vital elements in our determination of dynamic range of the performance. Nothing - repeat, nothing - below that noise floor will be heard by anyone in the hall or at home on replay. That sets our sonic obliteration threshold. All sounds, wanted or unwanted, below that threshold are masked by noise - gone, kaput, vamoose.

    What about the other end of the loudness decibel scale? The terrifyingly loud peaks? Those we can and must do something about.
    Alan A. Shaw
    Designer, owner
    Harbeth Audio UK

  • #2
    The loudness dynamic of music - 2

    To determine just how much audio voltage is going to come from the microphones under peak fff conditions we are going to get onto the talkback intercom and ask our friendly conductor if he would mind skipping to a certain page in the score, where you have noted the composer's fff markings and to play a few bars before and through it. This is where you need eagle eyes; you'll have to scan all the mic channels (and there could be 200 microphones) and see if the red 'clipping/overload' light comes on for any of them. If it does, you have to fix that and you also have to keep in mind for all mics, that come the real performance, adrenaline will result in somewhat more decibels of sound generated by the musicians, and therefore more voltage from the microphones. You need to bakc-off the gain of the individual mic preamps to be 100% certain that none of those sources will clip. Or big problems for you.

    So, our minimum background lnoise level is not determinable in advance with any sort of certainty (you can't be sure that come the recording a police siren wont go rushing past) but the conductor has cooperated with you in setting the maximum level. You, being smart, have allowed a safety margin even beyond that for the recording plus an allowance for audience participation, which cannnot be determined in advance. In other words, you're now pretty sure what your peak voltage recording level will be from the microphones but you don't know what your minimum voltage signal will be. But it's the peak your really concerned about, and we only have one relatively fixed yardstick - the peak sound.

    Remember back to phoning your friend and asking him to play at a precise level? That was doable because you both agreed in advance what your decibel frame of reference would be - a very quiet sound. In our situation sittling at the recording desk we have a sort of opposite problem: our quiet sound is vague and indeterminable, no two halls the same background noise, no two cities, no two days, no two audiences, no two rain showers. But there is a finite limit to how much sound a human orchestra can generate, day in, day out, so for our purposes, why don't we think in reverse. Instead of working in dBs setting a mutually agreed working reference as voltage above a vague and uncertain background noise, why not count backwards from our absolute sonic peak? That we know, and we know pretty accurately.

    Why don't we call our orchestral absolute microphone voltage output signal peak, zero level, or zero dB, or full scale deflection or fully modulated - all mean the same.

    Now we're getting somewhere. Providing that we keep an eye on the music signal, scaled in dB relative to the absolute signal peak, we should be able to capture the recording in all its glory as intended by the composer, at least as far as a non-fidgeting audience will allow, and make best use of the dynamic range between the quietest sound we can record and the absolute, maximum, maximum that our technical equipment can hold, undistorted, unclipped.

    What we need now is some signal metering system that shows us at a glance how close our micropone signals are getting to the absolute maximum. There are a number of ways of displaying this career-saving information, and the (BBC invented) Peak Programme Meter is a really good way of doing that.

    The PPM has two defining characteristics:

    1) Even though it was a mechanical meter until recently, it was driven by a circuit card that, in effect, accelerated the needle movement so that it was able to display microphone voltage transients that could catch out the sound engineer

    2) It has a slow fall-back time, so that the display was not flickering wildly as the accelerated needles tried to follow every single (confusing) note

    The PPM was, and is, a brilliant piece of engineering that in the hands of a trained user, tells him all he needs to know about the audio peaks. It doesn't tell him much about perceived loudness; for that we need another solution, but right now, to cling onto our job as junior sound engineer, perceived loudness is the composer's problem. Peak voltage, peak loudness is ours.
    Alan A. Shaw
    Designer, owner
    Harbeth Audio UK

    Comment


    • #3
      The loudness dynamic of music - 3

      To recap:

      We are going to work backwards from our absolute (and absolute means absolute!) peak voltage that we can handle, unclipped, in our recording/replay system. We're going to call that 0dB. We have on our mixing desk a stereo PPM, that is, a mechanical meter with independent red and green, left and right, signal level needles. We can see, at a glance, the peak voltage for both our left and right channels which normally, with classical music, will approximately measure the same voltage, instant by instant.

      What about the scale we print of the face of the PPM meter dial? Ah well, that's another subject, but let's mark in in dBs. No, that would mean we have to define our dB reference level. We may have reasons not to tie ourselves down so rigidly. Ok, let's make the meter in percentages (%). Not a good idea: the human ear is logarithmic and what we'll find is that the scale is crushed and the instrument virtually useless. OK, plan B: lets arrange the electronic accelerator card that drives the meter to present the audio signal in graduated steps - of 4dB - and each one of those steps is clearly marked on the faceplate of the meter. Let's mark the face not in dB or % but in steps, 1 -7, equally spaced. We could mark it in letters if we wish, or pictures of The Seven Drwarfes or anything we like, but numbers seems to be clean and simple.

      All we have to do now is get a screwdriver out, inject a signal into the PPM that is exactly the maximum that our recording system can handle, clean, and tweak L and R gain trimmers on the PPM driver card so that with that peak signal, the L/R needle tips point to some number on the scale. Say, point to the number 7.

      In addition, since we're going to be busy when the recording starts managing the desk, let's build a little red LED into the face of the PPM so that it illuminates to catch our attention if the audio voltage is creeping up towards PPM7, a level that we've agreed is the absolute maximum. Just to give ourselves a fighting chance to turn down a microphone or two when the signal is getting rather 'hot', let's make that warning LED illuminate when the signal is one whole step, 4dB, below our PPM7, absolute maximum signal marking.

      In other words, let's make the red warning LED turn on, using our new lexicon, "at -4dB", by which we mean 'when the signal reaches a threshold that is 4dB below our peak level'.

      Phew! We're ready to make our recording! We're trained and ready to go. Ten years condensed into ten minutes! It's a miracle! I don't know about you, but I have a definite feeling that now we have a proper understanding of the voltage levels here in the studio, we might just be able to blag our way through this recording session and keep our job!
      Alan A. Shaw
      Designer, owner
      Harbeth Audio UK

      Comment


      • #4
        The loudness dynamic of music - 4 - PPM ballistics

        Just to be absolutely sure that everything is really calibrated, we'd be wise to put in a call to the duty maintenance engineer.

        'Would you mind popping up to Studio 3 with a signal generator just to double check for me that the PPM level meters are perfectly calibrated to our studio maximum level - I mean - can you confirm that the PPMs are calibrated to the house standard 0dB, zero level?'

        'Oh, and before you go, would you mind showing me if these mechanical meters really do respond fast to transients? Can you inject just a cycle or two of 100Hz, 1kHz and 10kHz (not that music would contain just a cycle of any note) so I can see for myself the operation of the accelerator circuitry?'.

        He does, and we are truly impressed. We can see that our entire thinking about signal levels must be top down, from our zero level maxima, not ever from the bottom up. Because that quietest, bottom signal is indeterminate.

        And this is what we see when he injects that 0dB zero level at a fixed tone (1kHz), and then attenuates it downwards in steps of 4dB. Note the operation of the red LED.

        Loading the player ...


        Right-o. Now we have in mind the top-down concept of audio signals as they pass along the equipment chain from the microphones to the speakers, we're really well equipped to get to grips with the concept of music dynamics, power and clipping. We have another measuring tool that can give us an objective display of the dynamics of any audio source.

        The next thing I'd like to show you (over the next days, before we slip off to Italy) is what the PPM displays when we play real music. How close to the absolute maximum level does it go? What are the recording safety margins? How does classical music look on the PPM compared to pop, or speech? Can we guess what the music is - Don's harpsichord, for example - just by watching the ballistics of the PPM needle movements with the speakers silenced?

        And what can we see of dynamic range in real music?

        With me so far?

        BTW: can we appreciate why the PPM only displays the 'top' 24dB or so to the sound engineer?

        P.S. PPMs in action here and here. This is particularly interesting as the L & R channel PPMs are side by side. Note how after 30 sec. or so, the energy in the bass beat is clearly reflected on the meters (PPM7). Could you have judged by ear how much louder that beat is than the average music level? How many dBs would you say the beat is above the average? Remember: there is 4dB level difference between each marked step. So from an indicated PPM4 to an indicated PPM7, that represents (3 x 4dB =) 12dB.
        Alan A. Shaw
        Designer, owner
        Harbeth Audio UK

        Comment


        • #5
          After reading these four postings, I feel as if I'm ready to earn my degree. Thanks so much for the enlightening discourse.

          Comment


          • #6
            The loudness dynamic of music - 5 - loudness v. dynamics (don't confuse them)

            The next step may come as a bit of a sonic expectation shock.

            We hear the words "dynamics" and "loudness" associated with both music composition, music playing, music recording and music replay at home. I suspect that there is great confusion as to what these terms actually mean. They have precise and wholly different meanings.

            1) Let's define sound dynamics as, approximately, the range in audio voltage (or when sound, range in sound pressure) between the quietest element of the music/recording and the absolute loudest, quietest being defined by composer's intention, ambient noise in the studio, microphone hiss, tape hiss, electronics hiss and any other factor which sets a practical voltage level below which music cannot be heard properly

            2) Let's define level as the magnitude of the audio voltage, or when converted to sound by the speakers, the sound pressure level. (Roughly, we can say 'loudness' but pressure is measurable with a meter, loudness is a perception issue and rather more difficult to measure.)

            3) We can measure everything in volts (or pressure units, Pa) and because it's more convenient, we can convert all those measurements to dBs.

            4) We can talk about the dynamic range of a piece of music (or audio equipment) by measuring the loudest signal it can pass, undistorted, and the quietest before the background noise swamps the audio as xdBs. For example, if we know that vinyl LPs have a granularity, roughness, to the PVC plastic itself that causes a random motion in the stylus dragged along the groove, the voltage produced at the cartridge terminals in response to that random motion sets our audio signal minima. Conversely, the loudest sound that can be cut into the medium before breaking into the next groove sets the maxima. Subtract one from the other, and we can make a statement like 'the vinyl LP has a dynamic range of about 65dB at best'.

            In the previous post I demonstrated that the Peak Programme Meter has a scale marked-off in dBs (that's not printed on the scale, but it's true), and I showed you that there is a precise 4dB dynamic range between each adjacent pair of markings on the meter scale. That means that the space between marking 1 and 2 covers 4dB of signal range, between 5 and 6 also 4dB and 6 and 7, also 4dB. We could see that as the steady audio test voltage was reduced in 4dB steps, the meter movement tracked the signal.

            Now for what may not be obvious; in fact, you probably won't believe your own ears. Don mentioned on the source thread that he was enjoying the Goldberg Variations, and I suggested that pleasant though that is, it is not a dynamic recording. It can't be, because the harpsichord itself has a very narrow dynamic range (unlike it's successor, the modern piano) - the note is either played, or not played. There are limited shades of level, hence the sound is of a narrow dynamic range. How narrow? Very narrow indeed: as we will see from the PPM, roughly one and a half PPM scale units. Since we know that each step represents 4dB, we can say that the harpsichord recording 'has a dynamic range of about 6dB'. Which is extremely low dynamics. The sound is almost constant in level, and to a modern audience familiar with the wider dynamics of modern instruments, not very interesting or captivating.

            So, let's prove that 6dB dynamic range by playing the recording and watching the PPM movement. Before we do that, I want to pre-warn you that I've taken that 30 sec. Goldberg clip, and reduced it in level by 4dB four times. Taking our definition 1) above, that the dynamic range of the music is the difference in level between the peak and the quietest elements in the music, by reducing the level of the entire 30 secs. in steps of 4dB, I have not altered the dynamics of the piece at all. But that is probably not what you will think when you hear the same clip playing at progressively lower sound pressure levels. You will probably think that as the level diminishes, the dynamics of the music diminishes too, so that it becomes flat and lifeless. That's a trick of your ears. The dynamics are absolutely identical between the source clip, and the same clip played at a sound pressure 16dB lower.

            Listen here and watch how even as the level drops, we still see about one and a half PPM units (6dB) betwen the quietest notes and the loudest notes i.e. the dynamic range is objectively constant regardless of the replay loudness. There are many significances of this observation relating to the subjective assessment of audio equipment and the vital necessity to control replay loudness otherwise the slightly lounder equipment will take on a subjectively more dynamic sonic personality, as we know.

            Loading the player ...
            Alan A. Shaw
            Designer, owner
            Harbeth Audio UK

            Comment


            • #7
              The loudness dynamic of music - 6- surprising dynamics in the real world

              Now we can read and interpret the Peak Programme meter, and we know that each marked division on the face scale represents 4dB, we can use that information to give us some insight into the dynamics of music (or speech) monitored by a PPM. Note that any voltage signal peak that even momentarily exceeds the marked PPM6 will trigger the red LED. Also note that you cannot interpret any meaningful information for needle movements below PPM1, because they represent signals that are very much lower than the -24dB (rel. zero level) that is indicated by PPM1.

              I mentioned that Don's harpsichord example had a very small dynamic range, being the objective measurable difference between the instantaneous maxima and minima sound levels, note by note, bar by bar, throughout the piece. Observing the left & right PPM needles playing the Goldberg variations, we see that from the lowest marked level (the most left point either needle drops to) to the highest marked level (the needle furthest to the right) roughly, the range is about one and a half marked PPM levels, which equates to 6dB.

              In the next video I want to compare that 6dB dynamic range, clip A (which is unsurprisingly low for the harpsichord, an instrument abandoned by musical evolution for the much more dynamic concert piano) with a nicely mastered rock track, clip B. Observing the PPMs again we see that the rock track has also a dynamic range of about one and a half PPM divisions, again 6dB. Surprising?

              Clip C and D are both recordings of full orchestra. The peak levels are about the same in A, B, C and D (PPM6.5 or thereabouts) but look how low the needles drop for classical pieces C and D. The conductor and the orchestra are following the composers markings in the score and they are adjusting the muscle energy they are applying to their instruments to give them control of loudness throughout the piece. It's not the recording engineer that is turning up and down the level according to his personal whim: the development of the music through shades of pitch and loudness is defined in the score, bar by bar. If we are serious about recreating that real live sound at home, our replay equipment must be capable of delivering a similarly wide dynamic range to our ears as we would hear live.

              Otherwise we might as well forget all about fancy hi-fi equipment, save the money and buy a supermarket micro system. There is rather a nice all-in-one micro "hifi" system in my local supermarket this week for under $100.

              Faithful dynamic reproduction is, in my opinion, not only a defining characteristic of a worthwhile investment in audio equipment, it is often terribly lacking.

              If we agree that the dynamic range in the music of clips A (harpsichord) and B (rock) is about the same, what do we estimate the dynamic range of exampled C and D to be?

              As a statement to stimulate debate, having seen what we have and perhaps re-listened to examples in our own music collection, is there any justification at all for using pop/rock music as test material to probe the dynamic potential of our treasured hifi system? Should hifi dealers/audio fair exhibitors only use the most sonically demanding material to showcase audio systems - i.e. classical music?


              Loading the player ...
              Alan A. Shaw
              Designer, owner
              Harbeth Audio UK

              Comment


              • #8
                Justification for low-dynamic music

                Originally posted by A.S. View Post

                As a statement to stimulate debate, having seen what we have and perhaps re-listened to examples in our own music collection, is there any justification at all for using pop/rock music as test material to probe the dynamic potential of our treasured hifi system? Should hifi dealers/audio fair exhibitors only use the most sonically demanding material to showcase audio systems - i.e. classical music?
                In the spirit of debate, I will answer your question with an emphatic "yes", for the following three reasons:

                1. While much rock music may be mastered with relatively "squashed" dynamics, it isn't all like that, and while little or no rock music may have the dynamic range of symphonic classical music, at least some of it has more dynamic range than the worst-case (I'm assuming) example you present.

                2. Dynamic range may be important, but it isn't everything. Some might like harpsichord music, even if it has been superseded by the piano. Some might like rock music for reasons other than its dynamic range. It probably does make some sense to demo a speaker with the music that the prospective purchaser will actually be listening to.

                3. Even program material with a relatively restricted dynamic range might, if otherwise well-recorded, showcase positive attributes of a speaker's design, e.g. freedom from colouration or fatigue, natural frequency balance, etc.

                Having said that, I do accept that well-recorded classical music probably is the ultimate test.

                Comment


                • #9
                  Supermarket audio adequacy?

                  As a statement to stimulate debate, having seen what we have and perhaps re-listened to examples in our own music collection, is there any justification at all for using pop/rock music as test material to probe the dynamic potential of our treasured hifi system? Should hifi dealers/audio fair exhibitors only use the most sonically demanding material to showcase audio systems - i.e. classical music?
                  Excellent question.

                  The obvious answer is that hifi dealers/audio fair exhibitors should use whatever music they consider will give the gullible pubic the quickest, cheapest thrill so that the maximum number of punters buy the product being demonstrated. I presume this is what happens most of the time.

                  I recall AS mentioned a while back (prior to a Bristol show) that a diet of orchestral music, heaven forbid Shostakovich 8 from start to end, is not going to attract many punters at an audio show (although let me know and I'll come along happily). The wonderful except in post 7 is about 40 minutes into that work and by then most punters will have relocated on elsewhere.

                  The main issue I suspect is how much time you have on your hands. There is a lot of classic rock that is reasonably dynamic (as previously mentioned, LZ and Genesis, Stairway to Heaven and The Muscial Box for example), although you are still looking at over 10 minutes per track.

                  Rephrasing the question to "What music should a purchaser choose to demonstrate the dynamic performance of a system?", there is no doubt that it would have to be mostly or entirely classical. The dynamics of a Mahler symphony range from a single note plucked on a violin or blown from a piccolo to over 100 instruments including a double brass section hammering away at full blast. Nothing compares. To keep in context, I note that the final movement of Shostakovich 8 starts with a solo bassoon. So it obvious to look at the masters of orchestration, such as Mahler and Shostakovich, and for dynamic contrast I would plump for the opening movement of Mahler 1 as much for the beauty of the quieter bits, especially the unprecedented opening section.

                  I also have in mind the change in dynamics, perhaps referred to as attack and decay, which to my understanding are features of acoustic music that make it sound real. I understand this issue relates to AS's discussion of transients, above. I recall using Beethoven op59/3 when demo'ing my SHL5+, but the opening movement of the Kreutzer sonata might be equally appropriate.
                  https://www.youtube.com/watch?v=WOIsY7SodBc
                  Whilst there are only two instruments involved, would I be correct in considering this piece to be appropriately dynamic?

                  Outside of classical (OK, she was a classical pianist), Nina Simone's "Who am I?" took my breath away when first heard on the SHL5+, for the naturalness of the attack and decay.
                  https://www.youtube.com/watch?v=WOIsY7SodBc
                  The introduction to this piece uses the upper registers of the piano, which AS noted as being his focus for realistic piano (around the 3khz crossover point).

                  Turning the question on its head, if the customer listens only to music that that has a very limited dynamic range, whether pop music or Bach's keyboard works on harpsichord, will AS's $100 supermarket system suffice?

                  Comment


                  • #10
                    Low dynamic listening

                    In response to Eric, I like lots of relatively undynamic music such as the Bach violin partitas, and try the Gustav Leonhardt box set if you want a real challenge.

                    Such music invites fatigue. We all know Harbeth excel in this regard, and Harbeth literature refers to their lack of fatigue. However, I put this down to excellent loudspeaker design (drivers/crossovers) rather than a test of dynamics.

                    Comment


                    • #11
                      Dynamic truth is vital

                      Originally posted by EricW View Post
                      In the spirit of debate, I will answer your question with an emphatic "yes", for the following three reasons:


                      1. While much rock music may be mastered with relatively "squashed" dynamics, it isn't all like that, and while little or no rock music may have the dynamic range of symphonic classical music, at least some of it has more dynamic range than the worst-case (I'm assuming) example you present.

                      2. Dynamic range may be important, but it isn't everything. Some might like harpsichord music, even if it has been superseded by the piano. Some might like rock music for reasons other than its dynamic range. It probably does make some sense to demo a speaker with the music that the prospective purchaser will actually be listening to.

                      3. Even program material with a relatively restricted dynamic range might, if otherwise well-recorded, showcase positive attributes of a speaker's design, e.g. freedom from colouration or fatigue, natural frequency balance, etc.

                      Having said that, I do accept that well-recorded classical music probably is the ultimate test.
                      Actually, the rock track has a very typical dynamic range. If you look across music published by the pop/rock industry, you can get an idea of real-world DR from here. As for a bit of history of the misuse of the DR potential ...

                      Back in the 1970s, digital audio held remarkable promise. Though the early digital equipment had its share of problems -- and there were vigorous arguments over the way the conversion between analogue and digital would color the sound -- there was little argument about the glorious 90 db or more of dynamic range that digital could provide. A really good analogue tape machine with noise reduction could give you in the high 60s. But once the music got onto an LP for distribution to consumers, only the very finest vinyl formulations, and a virtually virgin pressing could give a signal to noise ratio approaching 60 db. Most commercial pressings, especially those played a few times, averaged in the low 50s, and intermittent pops and scratches could actually be louder than the music itself (a negative signal-to-noise ratio). So even with the best quality manufacturing, the recording that music lovers bought in the store never sounded as good as the master tape.

                      In 1982-83 when compact discs were introduced, it was like an epiphany for us audio folks. For the first time, consumers could purchase a recording in a medium whose dynamic range exceeded that of $20,000 professional {analogue} tape machines. Now I know that there are vinyl-philes who still swear that LPs sound better than CDs. But right now I'm talking about signal-to-noise ratio and dynamic range. Putting aside the arguments about the analogue digital conversion process, I don't think anyone can make a convincing case that an LP (or a cassette for that matter) has a dynamic range that comes within 20 db of that available on a CD.
                      Source here.

                      I find it particularly difficult to present quality audio equipment to the public (e.g. the Bristol show) playing crowd pleasing, familiar, low dynamic music, which by implication will be non-classical. Unfortunately, past experience is that the public are significantly disinterested in classical music (they seem to represent about 15% of our customers, dealers say) which means that 85% of listeners are listening to relatively low dynamic material. Whilst I obviously agree that low coloration, low distortion and adequate frequency response are core values in high fidelity equipment at home, if we are honest with ourselves, a replay system meeting those criteria can be assembled for perhaps $500 or less. A good quality wooden box DAB portable radio meets that basic spec at under $100. I have several.

                      What neither the $100 DAB radio nor $500 micro hifi can do is pretend to reproduce real-life dynamics. Frequency response and low distortion they can do pretty well and they make fine background musak systems. And for the vast majority of the public they are all they need, playing away to accompany whatever other activity is going on in the home.

                      So what is there left to strive for in quality home audio if the basics can be acquired for so little money? It must be dynamics. Above the basics, that surely is the defining difference between 'being there' and being at home. And of course, it demands speakers that can take the power and deliver clean sound which in turn demands adequately generous amp power.

                      If the listener lives exclusively on a diet of low dynamic music he is setting himself a very low expectation (and thrill) threshold, achievable for a very modest outlay in a non-specialist store. Listening on a supermarket micro system I can hear the piano, I can even guess the make of the piano, I can hear the hall acoustics all with low distortion and adequate frequency extension. What I can't kid myself of, no matter how hard I try, is that I'm in the same room as the piano because the dynamic range capabilities of the micro system is absurdly small compared with the real instrument. Ditto my car audio system.
                      Alan A. Shaw
                      Designer, owner
                      Harbeth Audio UK

                      Comment


                      • #12
                        Shocking dynamics

                        Thank you for posting the link to "real-world DR" - I think you've posted it before.

                        My first thought on DR was the 2014 album "Royal Blood", I mentioned it at the time and someone else agreed it was almost unlistenable. (Great live, according to my son.) On the database is ranks an average DR of 5, amongst the very worst, only the second band from the bottom of the list that I'd heard of (the other was Mettalica, unsurprisingly). It did of course win numerous BRIT and NME awards. Heaven help the hifi industry.

                        Nice album cover, mind you.

                        Comment


                        • #13
                          Try

                          The ideal music video? AS's favourite (excluding Thriller)?
                          https://www.youtube.com/watch?v=hQ5dJlQjYZ8

                          {Moderator's comment: Alan has actually passed out. The dynamics were too extreme.}

                          p.s. You have to watch this full-screen or preferably in the cinema to get the full oscilloscope sensation. It hurts my eyes.

                          Comment


                          • #14
                            Loving the punch

                            Originally posted by ssfas View Post
                            My first thought on DR was the 2014 album "Royal Blood", I mentioned it at the time and someone else agreed it was almost unlistenable. (Great live, according to my son.) On the database is ranks an average DR of 5, amongst the very worst, only the second band from the bottom of the list that I'd heard of (the other was Mettalica, unsurprisingly). It did of course win numerous BRIT and NME awards. Heaven help the hifi industry. Nice album cover, mind you.
                            I picked one track and out of curiosity had a look at its dynamics with the PPMs. I actually really like the grinding sound, and it would fill the demo room in a flash. But would it really challenge a quality audio system?

                            As for DR, I'd say that the 5-6dB you mention is about what we see at best, mostly a DR of 2, on our meters, which considering that they are strips of wire thrown about by magnets, correlate adequately well with a full technical deconstruction of the sound waves. I wonder what the similarity is to finding oneself in a boxing ring and being repeatedly punched in the face.

                            Providing one took adequate precautions to protect ones ears, and tanked up beforehand as a mild anesthetic, I can indeed imagine how thrilling the band would be live. 'Respek' to your son.


                            Loading the player ...
                            Alan A. Shaw
                            Designer, owner
                            Harbeth Audio UK

                            Comment


                            • #15
                              A right Royal battering - and why we spend on Quality

                              Originally posted by A.S. View Post
                              I picked one track and out of curiosity had a look at its dynamics with the PPMs. I actually really like the grinding sound, and it would fill the demo room in a flash. But would it really challenge a quality audio system?

                              As for DR, I'd say that the 5-6dB you mention is about what we see at best, mostly a DR of 2, on our meters, which considering that they are strips of wire thrown about by magnets, correlate adequately well with a full technical deconstruction of the sound waves. I wonder what the similarity is to finding oneself in a boxing ring and being repeatedly punched in the face.

                              Providing one took adequate precautions to protect ones ears, and tanked up beforehand as a mild anesthetic, I can indeed imagine how thrilling the band would be live. 'Respek' to your son.
                              Thank you for the analysis. DR of 2 - blimey. Try listening to the whole album and see if you retain a sense of humour. I went to a kick-boxing event a couple of weeks ago. Given the choice of 5 rounds and loud random noise in a 2dB range, it's a tough call.

                              It is worth remembering that we listen to music for pleasure, that derives from (non-random) variation in frequency and sound pressure level, as you have so excellently elucidated in this thread. Take away those extremes and add randomness and the result is used for purposes of torture of prisoners of war.

                              I think the point also has to be made that some reputable speaker manufacturers (and disreputable ones) design/tune their speakers for their target demographic, or at least they have specific ranges tuned and priced for a target demographic. The intention may be to shift as much air as possible and stuff the dynamics. This may not meet Harbeth's criteria for existing, or indeed any monitor manufacturer, but it would seem to be good business.

                              Anecdotally, my father is a music lover and for 20 years has used an Aiwa 100 supermarket audio system. He has a quiet apartment and listens at low level. He never understood why I spent more than 100 on my audio system. I posted "Who am I?" above because when I got my SHL5+ I played him that album, in particular for that track, and he finally understood, at age 75, what HiFi was about. I think it is more than dynamics. It is timing, attack, decay, the general sonority. It is the ability to make a complex sound genuinely lifelike. That is some achievement.

                              Comment

                              Working...
                              X