The funny thing about dynamics

September 26, 2022
 by Paul McGowan

It’s somewhat of a mystery why vinyl can sound more dynamic when in fact it is more compressed.

Much has to do with the technical measurement versus the perceived measurement.

The meaning of dynamic range is the difference between the loudest and softest sounds. The bigger the gap between loud and soft determines the technical dynamic range number of a recording (but not the perceived dynamic range).

Let’s imagine for a moment two recordings of the same event. One has a dynamic range of 90dB and the other is restricted to 60dB—same live event, only the first version is untouched with full dynamic range while the second has been compressed to fit into the smaller dynamic range space.

We can all agree that from a numeric standpoint the 90dB is far more dynamic than the 60dB.

Let’s now imagine we are playing back these two recordings at the same volume: the loudest notes on both recordings will always be at the same level.

The compressed version has all the info of the entire recording and sounds just right. However, when we switch to the higher dynamic range version something happens: the quietest notes will be 30dB lower than those same notes on the compressed version and likely lost. To our ear/brain, this higher dynamic range version doesn’t sound more dynamic, in fact, it likely sounds less dynamic because much of the information we use to mentally measure dynamics are simply lost.

When we playback at the same volume the compressed recording our ear/brain registers a greater range of loudest and softest and thus we decide it is the dynamic winner.

With me so far?

Now, let’s change the game. Let’s instead adjust our listening level to the softest portions. To do this on the higher dynamic version we turn the level up so we miss nothing. (We leave the volume the same as in our first example on the lower dynamic version because we can already hear all the softest notes).

Now when the two tracks are played back the loudest portions on the higher dynamic range version is much more dynamic than the compressed one and succeeds in scaring the crap out of us.

A great example of this can be found in the San Francisco Symphony Mahler recordings with Tillson Thomas. Most of the music is at a softer level and so we crank up the BHK preamp from its normal listening number of 30 to 45. We hear everything from the pin drop to the crash of the tympanis and the furious bowing of the basses.

The dynamic range on that recording is stunning.

If you set the level correctly.

Subscribe to Paul's Posts

52 comments on “The funny thing about dynamics”

  1. Cranking the loudspeakers up to play SPLs of 120 dB (background noise normally some 30 dB)??? You must have damned good loudspeakers which keep the distortions inaudible when listening to SPL of 100 dB or higher!

    1. You are exactly right! It is already very hard to get a background noise level of 30 dBs. And very few speakers in your room will reach 120 dBs!
      I wonder which recording Paul has with a 90 dB dynamic range. Or even a vinyl with 60!

  2. “…the loudest portions on the higher dynamic range version is much more dynamic than the compressed one and succeeds in scaring the crap out of us.”…as the drivers in our loudspeakers distort & maybe even go open circuit.
    For the protection of our loudspeakers a, little (when necessary) compression is not a bad thing.
    The TELARC – ‘Tchaikovsky’s 1812 Overture’, recorded with digital cannons, is a good (ok, extreme) example of pushing the dynamic envelope…another one is Pink Floyd’s – ‘Get Your Filthy Hands Off My Desert’.

    This is why an important statistic is to let a prospective customer know what a loudspeaker’s maximum SPL in dBs is, as well as it’s unclipped peak power.

  3. Yep; good explanation.
    But of course you can get the benefits of (much of) the higher dynamic range recording without going to the limits of ‘full scale’.

    And – in this vein – some cartridges are said to be ‘more dynamic’ than others. I wonder if this is correlated with stylus cantilever length? Imagine the cartridge body remaining still, along with the cantilever pivot point. Suppose a record displaces the stylus 1mm. Then the long cantilever will move the stylus through fewer degrees than the short cantilever. If the movement is 2mm, the long cantilever might displace through very nearly 2x the degrees it moved for 1mm; the short cantilever somewhat fewer – the stylus is moving tough an arc, remember.

    Thus a short cantilever cartridge will compress the higher levels of a recording, generating a lower dynamic range playback – but may well sound more ‘dynamic’ for the same reasons a Paul’s explanation.

  4. Hi Paul, another perfect example are the two versions of Zuill Bailey’s execution of the Bach cello suites on SACD and LP (long playing record). I prefer the 45 Rpm LP and know perfectly it’s beaten by the SACD’s dynamics.
    Greetings from Belgium,
    Yours sincerely,
    Marc

    1. Paul’s explanation makes sense to me but I don’t understand your preference. Whenever I visit my friend with his dedicated room and excellent set up for vinyl, SACD and streaming, SACD DSD always wins over hiz-res streaming with vinyl last overall. We’re both critical classical music listeners and agree on what we hear, which is why I have resisted streaming so far. SACD and CD recorded from about the last 10 years does it all for me.

    2. We should ask Paul if the DR range of the Bailey recording even exceeds the one of a chrome cassette 😉 Remember it’s no symphony orchestra but a cello…I guess dynamic range isn’t relevant here.

  5. The ear is more sensitive to changes at lower volumes. That’s the reason for logarithmic taper pots vs linear taper are used in audio ckts. Its more suitable to the way our ears receive increasing or decreasing sounds.

  6. This example along with an understanding sound pressure levels ( SPL ) explains why many people prefer the compressed dynamic range. How loud is soft or low conversation? It is 50 dB SPL. So you adjust the setting of your system so that the lowest notes can be heard at a 50 dB SPL ( you can hear the notes but they sound like quiet conversation ). So with this adjustment how loud are the loudest notes in your 90 dB dynamic range music. They are 50 dB plus 90 dB ( this is how dB’s work ) to produce a 140 dB SPL. How loud is 140 dB SPL? It sounds like a gun or a firecracker going off right next to your ear. 140 dB SPL may damage your ears and may destroy your speakers! Nobody needs a 90 dB dynamic range in their music.

    Crazy people like my dear friend FR listen to their music so the peaks are 90 to 95 dB SPL. On rare occasion I crank my stereo up to peak levels of 85 dB SPL ( I have an SPL meter ). If I go any louder my wife leaves the room.

    The magic of vinyl is not in its 60 dB dynamic range. I am not sure what the magic is due to, but it is there and for those of us who were raised on the sound of vinyl we will always like that sound.

    1. When you define 40–50 dB as adjusted (assuming you mean by volume control) for the lowest level sound that’s audible that this number is for speaker output as SPL, correct? Does this number change when you’re wearing a pair of well coupled headphones?

      Hopefully, FR is not wearing headphones at these levels.

        1. I was looking for a number Tony. I know it’s obvious. Since you laid out everything so nicely with speakers I figured you could give me a number with headphones as well.

          1. A SWAG ( scientific wild ass guess ) is that if you can hear very low level sounds ( a whisper ) on your headphones you are at no more than 30 dB SPL on your headphones

            1. No you cookin Tony.., ever heard of that acronym when I was an engineer. I’ll have to remember that.

              Kluge was the best term that I heard around my department. It was an absolute no-no.

      1. stimpy2,
        I very rarely listen to music through cans anymore, as I much prefer the openness of loudspeakers in front of me.
        My Ultrasone – ‘Signature DJ’ headphones are rated, by the manufacturer, to 115dB.

    2. I use a cartridge which is delivered with a measuring chart of tracking 100dB…but I doubt any record going near there is available, this reading isn’t important for musical quality anyway 😉

      The thing that makes all these absolute dynamic range discussions a theory anyway is, that if you look into the forum how loud most listen at maximum (around 70-80 dB) and how high normal quiet environment noise level is (a quiet bedroom is 25-30dB), most would be perfectly fine with the dynamic range of a cassette deck (if it would sound good and dynamic enough within that range) 😉

      1. CtA, I use C Weighting because I think that the bass frequencies should be used in an SPL measurement. How do you check calibration of an SPL meter? I have test records that claim the have a 0 dB or 30 dB test tones but how loud that sound is controlled by where I set the volume. What do you use as source to calibrate?

        1. This is a difficult question to answer. The complication is the speed of the measurement. Most Db meters can measure with some delay so you are never sure if the pick up the very short burst.

          I tend to use always the same “app” for this because this way I know I have the same underlying bias. I can pick which weighting I want, but I know it is not laboratory precise. It just gives you a decent idea. At home, the UMIK mic will be better at it, but for what appears as “constant” sound levels, the app and the UMIK via REW are reasonably overlapping.

          Again, I am not attempting laboratory conditions.

    3. Tony,
      As I sit in my silent room a certain logic tells me that should be 0dB yet the dB app on my iPad, Decibel X, is reading 38dB. I have no idea where that 38dB is coming from. Is that the sound level of silence? When I listen to music it typically registers say 85dB. Is it therefore correct to assume I am enjoying 47dB of dynamic range in room? If so it doesn’t really matter that the cd has a theoretical dynamic range of 96dB even if the actual recording manages that because I am only experiencing 47dB. To experience the full dynamic range I need to turn up the volume till the app registers 134dB (96+38) which is not something I would do. Re reading your post that’s pretty much what you said.

      More generally, one of the goals of hi-fi is to recreate the live or real experience in your home but I agree no one needs 90dB of dynamic range at home so it’s a goal that will never be achieved. Okay, in theory it could be achieved but does anyone actually listen at such levels. There’s bound to be at least one, there always is, but for the majority I think not.

      So to me it seems we are all talking about two connected but different issues here, one being the theoretical and practical dynamic range available on whatever the recorded medium is and the other being the dynamic range we actually enjoy when listening at home in room. One thing I have always thought, if you want to improve the sound of your system and enjoy a little more dynamic range, turn up the volume. Not too much though, we need to protect our ageing ears.

      1. The dB app on my wife’s iphone is worthless. Anything louder than talking registers 90 dB. My SPL meter is old, I got it from Radio Shack at least 20 years ago. It stilo works, but I have no idea if it is accurate.

    4. Tony,
      The needle in my ‘Radio Shack’ dB meter (also set to ‘C’ weighting) usually indicates (flickers) between 92 – 106 dB.
      A while back ‘Aeroaudio’ explained, here on Paul’s Posts, that the bass notes don’t sound realistic until you’re listening at around 100dB, which is why I listen at the volume level that I do; otherwise,
      I might as well just listen to Rock ‘n Roll music through a BOSE – ‘Wave’

        1. YMMV…absolutely agree with you there Tony.
          I have to feel the bass as well as hear it, for me such is the nature of deep bass.
          On most recordings from the ’60s, ’70’s & the ’80s, 85dBs just doesn’t cut it in the ‘Wow, feel that bass’ department.
          I had two subwoofers for about a year, but I was having to constantly adjust them up or down depending on how much bass was in the recording & I was told that that is not how one is supposed to use subwoofers.
          Sometimes there’s a lot to be said for having tone controls…sigh!

  7. [Paul=Now when the two tracks are played back the loudest portions on the higher dynamic range version is much more dynamic than the compressed one and succeeds in scaring the crap out of us.]

    Even though I set “softest passages” typically between 50-60db, dynamic headroom with my great recorded acoustical CD redbook sources (typically 30db+ peaks) often catches me off guard, manifesting in quick breath-catching and emotional goosebump responses!

  8. In a live seting with an amplified bluegrass band the sound board person (?) had cranked it so loud that we couldn’t stay. Very dissapointed as it was The Outlaw Festival and two hours later we would have enjoyed Willie Nelson. But our pleasure had been ruined.

    1. Funny, as I also struggled with listening to amplified bluegrass that was being ruined by the soundboard.
      It was a backyard/garage venue a block away from a big annual artfair two years ago.
      The ‘regular’ soundguy couldn’t attend, presumably because of Covid.

      One of the musicians was standing in for the soundguy, but had no idea how to set gain structure; he wanted to hear his monitor and kept increasing his instrument mic gain until he was totally overloading the mixer input and wouldn’t listen to constructive criticism.

      I went home early, even though there was free beer supplied by a local brewery.

  9. Vinyl masters do not benefit from brick wall limiting while digital masters require it in order to sound loud and “competitive.” I always turn it off for files being sent for vinyl and yes, it sounds more dynamic and better.

    1. Yes, as Cookie also pointed out and as the one or other listening experience and dynamic range database evaluation tells…even when vinyl in the past was limited for cutting, digital was compressed even more 😉

  10. I agree that the San Francisco Mahler recordings, with volume set so the softest passage (say, the tympani and solo string bass at the beginning of movement III in the first symphony) has clear pitches, are a good example of dynamics done right. In my experience, some engineers try too hard to capture an orchestra’s full dynamic range, resulting in either soft passages where pitch and expression are lost, or peaks that blow you out of the room-for example, the Wigglesworth Shostakovich symphonies on BIS SACDs.
    It’s interesting that the dynamic range you sense while playing in an orchestra does not match measured dynamics. From my seat in left center of the orchestra I measure peaks at 110 db yet never feel that the music is “too loud”. I suspect that it’s distortion in the recordings and equipment, and in room interactions, that cause listeners to sense that the music is too loud, not the db levels in the recording. In my experience, eliminating phase problems in my room has enabled my to play recordings at much higher volume levels, achieving a better sense of the power of the music without fatigue.

    1. My 110 db measure was in a rehearsal hall from the middle of the orchestra. Concert hall levels measured from the audience would be significantly lower. I measure about 34 db in my music room with the door closed and nothing playing (but equipment on).

      1. This is similar at home, but more common at night.

        But if this is true, a 90 dB dynamic range would be almost unbearable. I also don’t know of any music with acoustic instruments that would reach 90 dB dynamic range. I also don’t think that analog recording can reach this dynamic range and also few amplifiers have a SINAD that would allow this 90 dB. Forget about tubes, and any phono preamp.

        What the “producers” here say is that THEY reduce the dynamic range, not the technology! But if they don’t reduce it in their analog or vinyl versions it definitely means the master had less than 90 dBs to start with.

  11. Great explanation!

    “A great example of this can be found in the San Francisco Symphony Mahler recordings with Tillson Thomas. Most of the music is at a softer level and so we crank up the BHK preamp from its normal listening number of 30 to 45. We hear everything from the pin drop to the crash of the tympanis and the furious bowing of the basses.
    The dynamic range on that recording is stunning.
    If you set the level correctly.”

    Exactly, I’ve been trying to get my wife to understand and appreciate these very recordings!

  12. I try to get through the todays’ jungle of different understanding, perception, facts and implications …

    What’s a fact and where I agree with todays’ post is that vinyl (or analog in general) has a reduced dynamic range compared to digital and that it often sounds more dynamic (although I’d say this strongly depends on the vinyl and digital rig used).

    Most of the rest imo needs clarification once more and I don’t really get the basic explanation of this post.
    But let’s start with what I’d call the initial misconception of todays’ post, which makes me think Paul stopped the analysis of his perception too early:
    “vinyl can sound more dynamic” (that’s correct). But it does never sound more dynamic in a way as it would have a bigger absolute dynamic range…it often just sounds more dynamic as some DAC’s or amps sound more dynamic than others or some recordings sound more dynamic than others, without incorporating a different dynamic range. The real topic is the different dynamic perception of both in a given dynamic range. That is also, why the attempt to explain it with the dynamic range effects as in today’s post, doesn’t make sense. It especially doesn’t make sense because the effect that vinyl often sounds more dynamic also and especially happens with recordings which were not compressed at all (as most audiophile releases or reissues are not).

    ———
    “To our ear/brain, this higher dynamic range version doesn’t sound more dynamic, in fact, it likely sounds less dynamic because much of the information we use to mentally measure dynamics are simply lost.”
    ——-

    I tried but this also doesn’t make sense for me. Even if we’d speak of a really compressed version of an LP. If the uncompressed higher dynamic range digital version would have less loud sound at the low level area, this would mean a blacker background compared to the compressed version and would mean the dynamic impulses would stand out even more…it should then sound more dynamic (in terms of a bigger range), not less.

    The actual effect of compression is different as far as I understand it. It means, we get a perception of better detail, as the too soft sounding parts in a higher dynamic range version are lifted more above noise level by compression. That’s the opposite of a more dynamic sound, it just simulates more low level detail in the compressed version.

    For me today’s post is an attempt to explain a positive but hard to explain experience … “vinyl often sounds more dynamic” (not as if it had a bigger dynamic range)…with a disadvantage (compression), that neither applies to every LP affected by the above mentioned positive experience, nor has a comprehensible connection to a more dynamic perception.

    I don’t get started on today’s wrong implications, but the SFS Symphony recording is a great (again wrong) example for today’s story:

    When I bought the vinyl box (I also have the SACD’s), I asked Kevin Gray about his mastering of this box, as it sounded so far superior to the SACD’s (and more dynamic, but that’s the rather minor part).

    He said he didn’t have to apply any compression or limiting (as he doesn’t use to for nearly all his masterings), except for a slight gain riding on 1 or 2 points of the set. Gain riding certainly also means reduction of dynamic range, but he did so on 1-2 points of a whole! symphony set only and this not for 5 or more dB but clearly below that. He tried to do the whole thing better than the SACD and the SFS producer staff and the few audiophile orientated musicians confirmed it to him.

    As Paul described, this set is an extremely dynamic sounding one and a symphony recording is generally among the most demanding in terms of dynamics. So you can imagine, if a whole LP set (except of 2 single peaks) of such a recording is pressed on vinyl without any limitation, that hardly any other recording actually needs any kind of compression, Paul speaks of (but certainly normal recordings of the past often have it applied anyway, as have even more CD’s).

    So the SFS set at the end is the perfect example why today’s theory of an explanation doesn’t fit.
    Because compression doesn’t apply to most audiophile grade LP’s and so the explanation doesn’t.
    And because “vinyl often sounds more dynamic” isn’t perceived as “a greater range of loudest and softest” anyway. It’s perceived as bigger impact, impulses within a part of the dynamic range only. Just as if one plays a song, ranging from just 10-60dB level on different DAC’s, of which the one sounds more dynamic than the other. It has nothing to do with a different absolute dynamic range, we perceive different dynamic performance independent of it and even within a limited dynamic bandwidth.

    Why vinyl often sounds more dynamic?
    I don’t have an explanation yet, but it sure depends on the gear used, too.

    1. Well, a few points of this essay further question the relevance of excessive dynamic range (although it’s definitely never bad), so somehow fit into today’s topic 😉

      The below mentioned 12 bit, with 72 dB by the way are roughly the typical dynamic range of analog, tape or vinyl 😉

      —————————-

      On top of all that: as loudness increases, higher frequency masking takes effect in your ear. At low volumes of 20 to 40dB, masking doesn’t occur except for sounds close in pitch. However, at 80dB sounds below 40dB will be masked, while at 100dB sound below 70dB are impossible to hear. The dynamic nature of the ear and listening material makes it hard to give a precise number, but the real dynamic range of your hearing is likely in the region of 70dB in an average environment, down to just 40dB in very loud environments. A bit depth of just 12 bits would probably have most people covered, so 16-bit CDs give us plenty of headroom.

      Surprisingly, 12 bits is probably enough for a decent sounding music master and to cater to the dynamic range of most listening environments. However, digital audio transports more than just music, and examples like speech or environmental recordings for TV can make use of a wider dynamic range than most music does.

  13. Dynamics depend on the system, not so much the source or recording. The system needs to be able to reveal it or it doesn’t matter how much dynamics are on the source recording. Some systems compress it to various degrees.

    1. Actually Joe, it depends on both.
      If the recording is ‘flat’ or compressed to the max, then it
      doesn’t matter how dynamic your home-audio rig is.
      You can’t create what’s not there to begin with.

      1. I agree you cannot create what’s not there but what is there can sound it’s best on the right equipment on the right speaker’s that don’t compress the dynamics that are there.

Leave a Reply

© 2022 PS Audio, Inc.

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