What still amazes me are the massive power amplifiers of that era-people regularly bought amplifiers of up to 500 watts/channel-maybe they liked bass? What amazes me is how well the very small Bose speakers sound (I know, it is fashionable to mock Bose)-but they do deliver very good sound, from a very small enclosure.
That’s quite an important qualifier, but even when you factor in a subwoofer, the reality is that they are desgined for a certain type of sound, optimised for modern recording eq.
With modern electronic crossovers and enclosure design, you can do a lot with a little, but you still cannt get away from the fact that you do need significant sized speakers if you want a flat response.
Many listeners are not even aware of what such a set up would sound like, however when we talk of good sound reproduction, its worth remembering that almost everythng prior to 1985 was laid down on tape, and even today a signifcant amount of material is not digitally recorded.
I have reel to reel machines that can outperform CD very easily (all except on s/n ratio) on pretty much most music. The i-tunes method of consumption, although modern, is actually a step backward due to it being a lossy format, our home systems probably have more capacity to reproduce music than the sources that are most commonly in use.
Reel to reel is cumbersome and recording with it is not a matter of just pressing a few buttons, you have to ensure levels are set up correctly, along with making sure you have the right tape speed and tape bias, but when you get it right it is much more dynamic than almost every source that is available to the current consumer.
We are so concerned with some of these numbers, but when you are busting out music at any sort of volume, you really struggle to hear background noise below 40dB, or distortion at levels less than 5%.
The simple difference between speakers sizes eventually comes down to sound level. You can make a speaker that plays arbitrarily low bass in any size. What you can’t do is make it also play at an arbitrary loudness. To make sound you must shift air, and for the loudness at any given frequency is governed by how much air you shift. As the frequency drops you must sweep out more and more air for the same loudness. This is why a tweeter can be tiny, but a bass driver so much larger in diameter, and also why the bass driver must have a much longer throw. IN order to stop the swept air from simply flowing around the outside of the driver the front of the driver is isolated from the back, typically by a box. Since the air in the box will be compressed by the driver’s movement you need to take account of the physics of the air in the box, and a range of options comes about that creates a number of possible box to speaker design possibilities. There are a number of compromises that can be made in this, that can trade bass extension off against box size and the quality of bass. You can also apply active equalisation (the Linkwitz transform is commonly used to design such) and get a smaller box - albeit at the price of needing more power, and running into the thermal limitations of the bass driver. But eventually the dominant physics is that low loud bass needs a reasonable sized box. It needn’t be huge, but it can’t be small.
For higher frequencies clearly this doesn’t matter so much. But there is devil in the details. The ear can localise frequencies above 100 Hz, and a speaker capable of playing loud to 100Hz is still limited by the same rules and cant be stupidly tiny.
There are a whole slew of other issues in designing speakers. The division of labour between the drivers is never sharp (and there are good reasons why it shouldn’t be) and in the overlap region you must cope with energy coming from both drivers - yielding potential interference issues, plus often under-appreciated issues with differing dispersion patters - due to the different diameters of driver, yielding issues in the balance between direct and diffuse frequency response. The width of the speaker, and the shape of the speaker come into play at higher frequencies, and cause difficult to ameliorate ripples in the response. Cheap drivers typically have lots more issues with distortion and intrinsic frequency response problems. A clever designer can sometimes turn such an issue into an advantage, and tweak the design so one flaw partially compensates for another, but typically this only gets you so far.
A good speaker design has a lot of attention to detail paid to balance out a whole range of competing problems. But you can’t fight the physics forever. The difference in sound quality between a sizeable high quality speaker and a pair of tiny satellite speakers with a subwoofer is enormous. Not to say the tiny speakers are bad intrinsically, but most people have not had the chance to hear really good speakers in a good room. Sadly many people never go to listen to live music either, and have little idea what the music actually sounds like. (This allies to some audiophiles too - there are some ridiculous ides in some audiophile circles about what constitutes an accurate reproduction.)
casdave mentions tape and, probably inadvertently, brings up a critical point about perception of music and the medium. Tape can sound very dynamic. This is despite the fact that tape actually has a lower dynamic range than CD. Even a tweaked within an inch of its life professional tape machine with Dolby SR had just about the same range. However tape still sounds more dynamic. This is because, counter-intuitively, it compresses the peaks. As it compresses the peaks it generates distortion in a manner that the ear interprets as sounding “loud”. Sound engineers became adept at running the recordings hot, to get just the right amount of tape sound. Nowadays you can buy both analogue devices and digital plugins that attempt to emulate the same effect. Some even allow you to specify the tape formulation they try to emulate. The sound of rock and roll from the 80’s is in part tape compression.
Since the mid 90’s however the wheels have fallen off, and music is produced with stupidly low dynamic range, with the Loudness Wars wrecking the final result for everyone. I find much modern popular music almost un-listenable due to the wretchedly lousy production.
Just to emphasise - digital compression - as used in iPods, iTunes, MP3 players and the like has nothing whatsoever to do with dynamic range compression as discussed above. A VBR MP3 or 256kb/s AAC compressed digital audio stream will be indistinguishable from the original for most people on even quite high quality systems, and even more lossy formats still retain the dynamic range in the sense of musical dynamics.
And a final note - signal to noise ratio is the same measure as dynamic range when considering the recording technology. Musical dynamic range has no metric and is a different question. These two get confused in conversation occasionally leading to bizarre misunderstandings.
Both
If you can listen to Abbey Road without thinking “drugs”, you are entire too limited in your life experiences. Best trip album ever (yes, that one is likely to get this poor thread kicked back into cafe - please don’t, It took me 10 minutes to find it this time). With the ritual First Play of New Vinyl, the music WAS the reason for the gathering - crank the volume up to at least 60db.
As mentioned, I have a reel-to-reel/open reel deck. I had recorded a program of Beatles a local radio station played - when entertaining, those tapes would be playing at about 15db (vs 45db for conversation). It served as a social lubricant (1980’s) - everybody, from gen X to pre-boomer was comfy with it. At 15db, Abbey Road could pass without anyone needing to step out for a smoke.
I’m not sure what tape standards we are talking about here.
The ability of tape to reproduce sound is governed by the tape speed (write), the track width, and the tape head gap. These all influence frequency response, s/n ratio, dynamic range.
Tape itself has a very good linear response within its own hysteresis curve, and this is why bias is so important.
Studio tapes will be running at a minimum of 15i/ps but could well be double that and can readily exceed 26kHz, along with a conservative figure of 77dB s/n ratio.
The idea that tape sounds dynamic because of inherent limiting in tape headroom is misleading, tape can be plenty dynamic without that, and in any case is dependent on track width and tape speed. You are far more likely to inroduce that type of headroom distortion on lower speeds and narrower 4 track machines than on 2 track high speed machines, I just do not buy this as an unintentional effect of using tape.
As a measure of how great a differance write and replay speeds can make to dynamic range, try this out if you can.
Get yourself a 78 record, and the same title on 45 - this will limit you mainly to late 1950’s recordings, play each one in a back to back comparison.In terms of noise and sound quality the 45 wins it hands down, however in terms of dynamic range, the 78 will rule.
Something similar happens with tape, the higher the speed the better the dynamic range - I happen to have a few 78s along with the same recording on 45 and was surprised at the outcome - the 78s sound much more lively.
CD is inherently 98db, and with shaped dither can exceed 106db in the mid bands, where S/N is most critical. This is nearly 30db better than tape. As I wrote, the only tape system that ever came close to 16 bit digital’s dynamic range was Dolby SR, which was introduced just about in time to become obsolete as digital became the norm.
The hysteresis curve of tape is why tape has the ability to be run hot and deliver the slam so liked of some recording engineers. Rathar than instantly clip - as digital does, tape runs into the curved areas at the ends of the hysteresis S curve, and this naturally rounds off the peaks. So you don’t get a nasty clipping noise, but with care you get a nice level dependant effect. Curving off the peaks will translate to harmonic distortion, and in general will add a lot of even harmonics. So the sound fattens up nicely as it hits those peaks.
The choosing of tape’s reference level, and hence the claimed dynamic range, is a choice partly on how far into the curve you want to go. That “conservative” 77db is a choice not to go far at all. It yields a recording with good accuracy and little in the way of these effects. But rock and roll recording engineers became very adept at running the recordings hot and pushed much harder.
To approximate the effect you can buy such devices as analog tape emulators and DAW plugins
Hardware:
Software: