Why dynamic range compression can actually be good for our health

Fiddle Faddle

Member
Aug 7, 2015
548
2
16
Australia
I am sure that I am not alone here in that during the 1970s (and I guess for the 60s for those of you who grew up then), I unknowingly did not look after my hearing. As a youngster, I was charged with mowing the family lawn - all 1400 square metres of it - each and every weekend over summer and perhaps every other weekend in Spring and Autumn. That is two hours of an engine going full throttle less than two metres from my ears. Back in the mid 70s, there was simply no awareness of the dangers of noise exposure. I don't even think you could go anywhere to buy ear protection. No one ever wore it even if you could get it. And certainly no one ever warned me of the dangers - parents, doctor, school - no one. And I remember that after mowing the lawn, I couldn't hear properly for well into the same afternoon - it wasn't till late afternoon that my hearing felt "right" again.

And this went on for year after year until suddenly one day someone woke up to the dangers, it became "public health" knowledge and - at least the smart ones which now included me - kitted up with hearing protection. Whilst I also studied and listened to recorded music, it was never ever remotely loud enough to actually cause a threshold shift in my hearing (which is actually the correct term though I never knew that in the 70s). It was only ever when using power tools such as the lawn mower. Recorded music was actually a "quiet" hobby believe it or not.

In much later life I had a hearing test and given the damage I'd unknowingly done cutting the grass decades before, I was surprised to see the losses had been thankfully relatively minor (about 5 dB in the left ear and 10dB in the right at the "noise notch" frequencies). It could have been far worse and I was told that pretty much everyone has damage to some extent. Suffice to say that ever since I had that hearing test a decade ago, I've been extra careful, though nothing is ever going to undo the damage I caused myself - probably along with millions of other young people of that era and earlier ones.

There has only ever been one instance in my entire life where listening to actual music itself caused a threshold shift. And that was at a live concert in Sydney of a Shostakovich Symphony. I decided there and then that this was my last ever live concert - at least for the foreseeable future - and I cancelled my yearly subscription after that. I felt that having to wear musician's ear plugs at a live classical concert was half-defeating the purpose of being there in the first place. It was there and then that I became serious once again about great sound reproduction in my home, even though it had been an on and off hobby of mine since the age of around 8.

So what does the state of my hearing have to do with dynamic range in music? It is simply that in order for the softest bits to be heard with absolute clarity, the loudest bits can and do reach uncomfortable levels. In fact, I've measured some very good domestic audio systems with high dynamic range classical material and if you go by the notion that 30 dBA is about as soft as you are prepared to go in order to hear, say, a solo part in an orchestral work, then you are pushing past 110 dBA when everyone is going flat out. This is not only uncomfortable listening. It is also dangerous - literally. These levels are common in an actual concert hall but clarity at such extremes is much more difficult to achieve with a domestic audio system in a domestic setting as most of us would know.

With all that said then, I have always struggled with the immense dynamic range inherent in classical music. So far as recorded classical music is concerned, the dynamic range is as much about the actual work and composer as it is the orchestra, conductor, hall and recording techniques used. I suspect that some labels such as EMI and DGG used subtle compression to tame dynamic range routinely, though not in such a way that you are overtly aware of it. EMI, for example, had some wonderful audio recording consoles with some killer limiter and compression modules onboard - more about that later. And for the most part, when listening to DG recordings from the late 50s to early 70s, I am never overwhelmed by the dynamic range because the softest parts are clear and simply significantly louder than they are on other labels such as Decca (where I believe signal processing was absolutely taboo). I think RCA did a far bit of gain riding on their mixers even in the Living Stereo days - there is just no way I hear what I am hearing on those recordings unless there were additional spot mikes that were pushed in and out as needed.

Anyway, up until last week this had been a problem I had begrudgingly lived with. My only "solution" was to either listen to our national Classical FM station (which obviously needs to use compression), or limit my listening sessions to only a couple of one-hour sessions per week. With popular music, it simply isn't a problem at all. With such a small dynamic range it is very easy for me to find a low volume level that is satisfying - even at a level measuring about 56 dBA, that is more than enough for popular music so far as I am concerned. But with classical, it has just been a losing battle.

Here is graph below to show what I am talking about. This is a wave form representation of a Reference Recordings vinyl LP (Rachmaninoff Symphonic Dances - RM1504) I bought a couple of years ago. Look at the peaks versus the softest bits - it's nuts. And this is the half-speed mastered vinyl pressed at Quality Record Pressings - as good as it is, the soft bits "include" all the analogue noise - turntable rumble, vinyl surface noise, etc!!:

rr.jpg



And here is a far older recording. Ravel's Bolero - an early 1960s Decca reissue by Speakers Corner - Ansermet with the Swiss Romande:

bolero.jpg


So why have I now "solved" this problem? Well, remember earlier in this post I mentioned the compressor / limiter modules used by EMI? Two years ago, a company called Waves collaborated with Abbey Road to release a plug-in emulation of the famous EMI TG12345 channel strip. Plug-ins are software that can be used on a digital audio workstation and they are able to replicate - within technological constraints of course - the workings of the original piece of actual hardware. Typically the original piece of analogue hardware is exhaustively put through all of it's paces and the results painstakingly measured. This particular TG12345 strip was used in the recording chain for the Beatles and (if I recall correctly) Pink Floyd amongst other famous artists. I am not sure if it was used with classical but since many classical recordings were made at Abbey Road right when this console was in use, I think it may well have been.

I decided last week to download the demo version of it (7 day limit) and started to experiment. Bearing in mind that the vast majority of users are processing popular music, it did take me quite some time to find the best setting for classical. But once I found it, what can I say other than I am hugely impressed. I have mucked with plug-ins of various sorts for years now and for the most part they are very underwhelming. Most of them - in doing what they are designed for - sling proverbial mud all over the pristine audio being processed. Inother words, the colourations are so great, they simply cannot be considered of any use to a purist.

But this Abbey Road TG12345 plug-in is a completely different ball game. Think of the very best (but subtle) compression that the best classical music FM stations might have used back in the old days where sound quality really mattered above everything else (I am thinking here of ABC FM in Australia back in the late 1970s to early 80s) - but then up the ante again a couple of notches in terms of sheer audio performance and transparency. The bottom line is that I have found a way to process my music in such a way that dynamics are no longer excessive and yet the essential quality of the music is unaffected.

Below is the aforementioned Reference Recordings graph after applying the best compression setting I could come up with - bearing in mind that I wanted the effect to never actually draw attention to itself and that the whole point is to be able to reduce the volume control setting. With this new file, I can turn the volume down by about 7 dB versus the original. That is a huge improvement:

rr_emi.jpg


Same with the Bolero. Again, a "saving" of around 7 dB on the volume control yet you'd only be aware of the compression if you directly compared it to the original:

bolero_emi.jpg


Of course, it is quite possible to add a limiter / compressor to the actual audio system itself but then you run into practical issues that high resolution software processing avoids. Apart from that, you would have to spend huge amounts of money to employ a limiter of the same calibre.

Anyway, if nothing more, I hope this post will give people cause to consider how dynamic range in music and the associated volume control setting needed to accommodate it can be rather unfriendly to our ears. As I say, this is not really a problem with popular music but it is a very big one with classical music. And the "better" the recording, usually the bigger the "problem" is.
 

Folsom

VIP/Donor
Oct 25, 2015
6,024
1,490
520
Eastern WA
I don't blame you for wanting a bit. It's not really like loud parts of a song suddenly sound quiet, they typically still sound dynamic and loud but won't rack the db's so darn high. You get, the majority of the impression of dynamic fun, but without hurting your ears as bad. When I use my inefficient speakers I can't break 102db, no way, no how, there literally isn't enough power (may not even break 100db) but it sounds plenty loud and dynamic for my smaller room. But he parts of my setup are highly optimized to get the most out of what power there is, so it sounds like a high powered setup most of the time.
 

NorthStar

Member
Feb 8, 2011
24,305
1,323
435
Vancouver Island, B.C. Canada
I have that first recording you just mentioned above:



...On CD (RR-96CD) and with [HDCD] encoding.
I don't know how the dynamic range would compare with the LP?
It's also avail in that format (hi-res HRX DVD-R - 176/24): http://www.elusivedisc.com/Eiji-Oue-Rachmaninoff-Symphonic-Dances-HRX-DVD-R/productinfo/RRDR96/

But this is cool; that Abbey Road TG12345 plug-in I guess works with digital as well.
________

Some classical music recordings we can barely hear the soft passages, so we tend to turn the volume up...and then, suddenly with the louder passages we quickly run to the preamp and turn down that master volume level control knob. :eek:
Lol, when that happens...
 

Fiddle Faddle

Member
Aug 7, 2015
548
2
16
Australia
But this is cool; that Abbey Road TG12345 plug-in I guess works with digital as well.

It will only work with digital PCM as it is a workstation plug-in. It can process up to 24/192 and works with 64 bit internal precision. That is fine by me as my entire vinyl library has been transferred to 24 bit digital. For pure analogue systems there are of course hardware compressors out there but the good ones are quite expensive and anything remotely worth considering requires balanced connections - entry level is several thousand US for anything remotely decent. Apparently the actual hardware used to model this plug-in is worth far more than that again.
 

Fiddle Faddle

Member
Aug 7, 2015
548
2
16
Australia
You get, the majority of the impression of dynamic fun, but without hurting your ears as bad.

That is precisely it in a nutshell. You can enjoy the massive dynamic swings actually even more than before because you get the "fix" but it is never uncomfortable to the ears. This compressor / limiter is actually quite unique because depending upon the virtual input voltage, it will raise the RMS level and above that level it reduces it. The "crossover" point is a virtual 1.5 volts I believe. That makes it perfect for classical because of the extra flexibility. Unlike a normal limiter where everything is squashed down and then you have to apply make-up gain because of the level drop, with this the really soft bits are brought up rather than compressed but the loud bits are subtly compressed.

I actually tied about 6 different limiters over the last week and none of them were even remotely in the same class when it came to classical.
 

Don Hills

Well-Known Member
Jun 20, 2013
366
1
323
Wellington, New Zealand
Good dynamic range compression is pretty much essential when you bring a performance from a "large" acoustic into a small one. I have a study saved (somewhere, still looking) which looked at what dynamic range sounds "right". After attending a live performance in a hall, the subjects were given a recording of the performance and adjusted the dynamic range in their listening rooms to give a subjectively similar range. They all chose a smaller dynamic range than that measured at the live performance.

Regarding the use of compressors in radio broadcasting, years ago the BBC used compression for live broadcasts. For events such as The Proms, an engineer with a copy of the score would gain-ride the signal sent to the broadcast chain. The ultimate intelligent look-ahead compressor... :)
 
Last edited:

Fiddle Faddle

Member
Aug 7, 2015
548
2
16
Australia
Good dynamic range compression is pretty much essential when you bring a performance from a "large" acoustic into a small one. I have a study saved (somewhere, still looking) which looked at what dynamic range sounds "right". After attending a live performance in a hall, the subjects were given a recording of the performance and adjusted the dynamic range in their listening rooms to give a subjectively similar range. They all chose a smaller dynamic range than that measured at the live performance.

Regarding the use of compressors in radio broadcasting, years ago the BBC used compression for live broadcasts. For events such as The Proms, an engineer with a copy of the score would gain-ride the signal sent to the broadcast chain. The ultimate intelligent look-ahead compressor... :)

Hi Don,

Great post. Thanks! In my opinion nothing beats manual gain riding but of course as you say it requires great familiarity with the work and preferably a score in front of you. I think this is possibly what RCA did a lot of back in the 1960s. Even the very best limiters - this one included - cannot be 100% "transparent" so to speak. For example, if I take an old, pre-Dolby A analogue concerto recording then you will hear the hiss floor slowly come up noticeably (though still with reasonable subtlety) during a cadenza for example. But you wouldn't really notice it if that same recording was digital or made with Dolby A (or was simply a "quiet" tape without any noise reduction) because it is more the hiss level modulation that is noticeable and not so much the compression of the music content itself. With manual gain-riding you'd have the levels set at the start of the cadenza and leave them for the duration and you would not hear the background hiss level slowly rising.

Our national radio station ABC FM (as it was known in the 70s and 80s) were really committed to quality sound back then but from the early 90s onwards sound took a back seat. But in the 70s and 80s they used the same sort of compression (or perhaps a tad more) that I am using with the TG12345 plug-in. And they sounded absolutely fantastic whether they were playing vinyl or - in the later years - CD. I remember actually preferring their sound to the actual record (and CD in particular) itself because the compression just made it more listenable. I also remember having a preference for good quality pre-recorded cassettes back then again because they employed subtle compression.

I have to say that listening this past few days has been a real treat - really relaxing now that I can hear the softest parts with the same sort of clarity I would get in the concert hall but not having to worry about how loud it gets when it is maxed out.

I know record companies go on and on about signal purity and boast about not using EQ and compression, but I honestly wish that consumer audiophile equipment companies would bring out quality limiters and compressors that work just as well as this plug-in does. I would much rather have put a hardware compressor into my home audio reproduction chain but as a mainstream equipment user, I cannot go to my local Rega, Creek or Musical Fidelity dealer and buy one! And their sound quality (and compression quality of course) really varies in any case. I also tried the Waves emulation of the classic dBX 160 compressor. Although I was able to achieve similarly satisfying levels of compression and the RMS level of the output file was roughly comparable to the TG12345 results, the dBX 160 sucked the life and subtlety out of the music and just made it too dull and emasculated. The revered Fairchild 600 compressor plug-in did not do any better either! Too many rose-coloured nostalgic glasses being worn in the digital plug-in industry methinks!
 

About us

  • What’s Best Forum is THE forum for high end audio, product reviews, advice and sharing experiences on the best of everything else. This is THE place where audiophiles and audio companies discuss vintage, contemporary and new audio products, music servers, music streamers, computer audio, digital-to-analog converters, turntables, phono stages, cartridges, reel-to-reel tape machines, speakers, headphones and tube and solid-state amplification. Founded in 2010 What’s Best Forum invites intelligent and courteous people of all interests and backgrounds to describe and discuss the best of everything. From beginners to life-long hobbyists to industry professionals, we enjoy learning about new things and meeting new people, and participating in spirited debates.

Quick Navigation

User Menu

Steve Williams
Site Founder | Site Owner | Administrator
Ron Resnick
Site Co-Owner | Administrator
Julian (The Fixer)
Website Build | Marketing Managersing