How much have speakers improved over the past decades?

[Koeitje]

For DACs and amplifiers ( Purifi, AHB2, etc.) it is clear a lot of improvement has been made over the past decades. I can see how a lot of active Hifi speakers are pushing limits, but that is also being carried by developments in DSP and class D amplifiers. But what about passive loudspeakers and the drivers themselves?

Currently I have DIY speakers (didn't build them myself) using Scan Speak Revelator drivers. This design in 20 years old, and even though I know it is pretty good (the design was sold to a commercial company so they could make their own version.), I always wonder what 20 years of development has brought us. These speakers cost about 600 euro a piece to build and could be bought pre-build for 1000 a piece or something. If I would spend the same money today, how much better would it be? Have people gotten better at crossovers? Has distortion been lowered? Higher SPL? Better design due to better measurements? I find it hard to find any concrete data on these things. All I know is that off axis response should be smooth...but didn't we already know that decades ago?

Measurements from my speakers can be found here:
http://www.markinton.nl/Scan-Speak Reference Monitor.htm

Should something like the Revel Performa3 M106 or Kef R3 wipe the floor with this? My gut says it is still going to be pretty close...

 

[Blumlein 88]

Send them to Amir and find out.

 

[goldark]

If it measured well then, it should still be competitive with speakers that measure similarly today. A lot of advancements have come in the budget price brackets, where you can get more competently designed/measuring speakers at less expensive prices than before.

 

[Koeitje]

Send them to Amir and find out.

All the speakers I mention have already been measured (by Harman, Stereophile and in the link I posted...but that has no off axis data). But I don't know how they compare (I can't read waterfall graphs), and there is also no THD data.

Besides shipping over the pond is going to be hard.

 

[Willem]

My modern Quad 2805 electrostatic speakers are better than my old Quad ELS57s (I still have both), but the difference is not night and day. But of course the ELS57 was well ahead of its time. I also have late model 11 Ohm Rogers LS3/5as that I can compare directly with their modern incarnation, the Harbeth P3ESR. Again, the modern one is better, and in this case more obviously so. However, in their own way these are all speakers at the more expensive and ambitious end of the market.
I am indeed convinced that the biggest advances have been at the budget end of the market, where modern materials and computer modelling have simplified the design and manufacturing process. I owned a pair of Wharfedale Diamond 9.0s in a bedroom system, and until one of them recently died they were quite pleasant. Their replacements, the Q Acoustics 3010, are rather better again, and still very cheap.
Proper and fair comparison is not necessarily easy, however, since many older speakers will have deteriorated, though not all of them. No early 15 Ohm LS3/5as will still be on spec, but the newer and improved 11 Ohm version with different materials can still be quite like the original (Harbeth´s Alan Shaw recently measured one and it was as fresh as it ever was).
As for electronics, there too there have been improvements, but we must not forget that already by the 1980s designing a transparent power amplifier was a solved problem. The big improvement in electronics was the introduction of the CD. In one stroke this gave users perfect source quality.

 

[Blumlein 88]

Longer answer is I'm sure there have been steady improvements in diaphragm travel, power handling etc. from newer materials technology.

I helped a friend build some ambitious speakers some 25 years ago. Since we weren't paying retail, he splurged on the best Dynaudio drivers. He went with a three way so the crossovers were very easy to tame, and we built a herculean cabinet. The drivers were something you'd see in a $25k speaker back then. Though our labor and materials cost was a small fraction. They were excellent speakers better than many expensive commercial ones we heard. Software modeling and measurement would let you fine tune the design even better were you using the same drivers. I don't know if less expensive drivers come closer to those, but I don't know that your basic drivers are better than those were. It is however software and measurement that let makers do more with less expense when they want to be in that market.

You wanted concrete data, and the best way would be to measure what you have. I realized that wasn't likely or easy (and I didn't know you were in Europe). But that is the way to find out. You could do some measuring yourself with a PC, REW (free) and one of the inexpensive measurement microphones ($50 to $100). Though tedious you can even get some useful data off axis above 1 khz.

My extra comment, I do believe Harman is onto something very important about how good speakers work in our rooms. It lets them make a very nice speaker without getting too outlandish on materials. Beyond power handling or loudness, distortion and all that isn't so important as flat smooth response on and off axis. So I expect some of the Harman designs might best what you have even though the parts quality of what you have is excellent. I don't if that is enough to call it wiping the floor better.

 

[gfx_1]

Don't know about modern stuff but I'm not a fan of powered speakers. Listened to a couple of bookshelf speakers at the nearby DIY speakerstore and went home with an older Vifa premium 11 design. I really like them. Vifa is usually more affordable than Scanspeak.
They still sell the Scanspeak reference monitor, it's a good no compromise design.

 

[Koeitje]

.

You wanted concrete data, and the best way would be to measure what you have. I realized that wasn't likely or easy (and I didn't know you were in Europe). But that is the way to find out. You could do some measuring yourself with a PC, REW (free) and one of the inexpensive measurement microphones ($50 to $100). Though tedious you can even get some useful data off axis above 1 khz.

I have a UMIK but don't know where to begin to make measurements that are worth anything. Sounds like a lot of work .

 

[Blumlein 88]

Here is a good place to start. RoomEQ Wizard a free, but very capable bit of software.
https://www.roomeqwizard.com/

Putz about with it a little and you can easily get it to work with the Umik. You need a Windows, Mac, or Linux computer.

Here is my own measurement with distortion of some LSR305 mk I speakers.

Here is a waterfall plot from the same speaker and measurements. There are of course other things you can do with REW.

 

[digitalfrost]

Since when have speaker measurement devices been around?

I can imaginge with the affordability of computers and especially PCs, measurement tools became widespread, even in smaller companies. But what did people use before there were PCs? How was frequency response measured, and how would you calibrate the devices? Who were the first to invest in such technology? And what did companies that didn't yet have access to it? Was there ever a time where you couldn't measure, but you still developed speakers (chassis)?

With todays technology, everything seems so easy and still I wouldn't say measuring speakers is easy. Just look at all the REW guides. And making sense of the measurements is another issue

 

[JohnBooty]

I've not seen anything to suggest that driver construction has improved too much in the last 20 years, though of course I'll be happy to be shown that I'm wrong! Woofer materials get fancier, but at the high end, people still love their paper cones.

Much of the progress has been kind of a trickle-down from high end stuff to lower end stuff as far as I can tell.

1. Active speakers with DSP-based crossovers that can do things passive crossovers just can't. This is also one reason why car audio can be shockingly decent these days.
2. Unless I am mistaken, waveguide construction has gotten a lot more sophisticated.
3. As cheap, clean power has become more abundant thanks to Class D, manufacturers have been less afraid to go for low-efficiency designs that manage to wring some halfway decent bass from compact speaker designs.

 

[Willem]

Speakers have been measured for a long time, although not necessarily as precisely as is possible now. Microphones are nothing new, and instead of digital gear pen ploters were used. See here for a BBC publication of some results.: http://www.keith-snook.info/wireles...ld-1968/New B.B.C. Monitoring Loudspeaker.pdf See also here: https://www.harbeth.co.uk/our-story/bbc-heritage.php As far as measurement gear goes, B&K have been in the business for a long time, and well before the age of digital instrumentation: https://www.bksv.com/en/about/History In a sense, the real importance was not so much with the measurement gear, but with the creation of the right conditions for those measurements with the use of anechoic chambers.

 

[Wes]

a more interesting question is what ideas (Thiel-Small alignment math; diffraction near the tweeter) or test methods (laser interferometry to assess cone breakup) occurred and when

also cabinet rigidity, etc.

 

[PeterZui]

I've not seen anything to suggest that driver construction has improved too much in the last 20 years, though of course I'll be happy to be shown that I'm wrong! Woofer materials get fancier, but at the high end, people still love their paper cones.

Lot of progress has been kind of a trickle-down from high end stuff to lower end stuff as far as I can tell.

1. Active speakers with DSP-based crossovers that can do things passive crossovers just can't. This is also one reason why car audio can be shockingly decent these days.
2. Unless I am mistaken, waveguide construction has gotten a lot more sophisticated.
3. As cheap, clean power has become more abundant thanks to Class D, manufacturers have been less afraid to go for low-efficiency designs that manage to wring some halfway decent bass from compact speaker designs.

Hi John Booty,
If so, I would like to get the proof that at high end audio people still love paper cones; I guess this goes from woofer units to tweeter units ? And please give a list of all high equipment supplier who design this way.

I do not think that the audio engineers in speaker audio would agree with your statement as it would mean they all have a meaningless job isn't it ? Can you name one who agrees with you ?

As far as I'm concerned active speakers can sound impressive with all kind of music but don't play anything acoustic; you can hear the instruments are transformed in an electric violin, an electric piano or an electric ........This perception is of 30 years ago and didn't change when f.e. listening to a Kii Three set nowadays.
Nothing wrong if you do not reject this effect but for me this is not high end.

At last an example of one of the bigger speaker designers that did and still does a lot of research;
Did you see and compare the measurement report of the B&W 804 D2 and the newest 804 D3 ? Did you listen to them in a comparison ? Can you tell what design change caused the biggest change in sound perception ?

 

[Hugo9000]

Since when have speaker measurement devices been around?

I can imaginge with the affordability of computers and especially PCs, measurement tools became widespread, even in smaller companies. But what did people use before there were PCs? How was frequency response measured, and how would you calibrate the devices? Who were the first to invest in such technology? And what did companies that didn't yet have access to it? Was there ever a time where you couldn't measure, but you still developed speakers (chassis)?

With todays technology, everything seems so easy and still I wouldn't say measuring speakers is easy. Just look at all the REW guides. And making sense of the measurements is another issue

Regarding using computers in loudspeaker testing and design, as far as I know (I can't find any earlier references for any other company. There are articles about Bell Labs using computers in the early to mid-1960s for other things, including creating the first "computer art"), KEF was the first:

Since its founding in 1961, KEF has been unapologetically driven by technology rather than audiophile fashion. In many ways, the result of that focus on technology has driven the art and fashion of the audio industry. KEF’s position as a technological leader rather than a follower is largely due to its early embrace of computer analysis and simulation. Everybody does it now – we started doing it in 1969.

Back in the late 1960s using a Hewlett-Packard Fourier Analyzer, KEF began a joint venture with Bradford University in the UK, centered around two questions: How does sound reflect around a rectangular room and what EQ is required to design a speaker with the perfect shape for audio response? By 1975, KEF had integrated Fourier Analysis into its design efforts, with the Model 105 being the first mass-produced speaker in the world designed through the use of waveform analysis. In the early 1970s, there were few H-P Fourier analyzers in the commercial market, mainly due to their cost – around $60,000 each. We saw that the investment was a wise one.

Laurie Fincham, the acclaimed audio engineer who started working at KEF in 1969, explains it this way: “We started doing the Model 105 and [using computer analysis] was all very approximate, but what it did do was give [the design] a very broad distribution pattern, which gave a sort of nice ‘airy’ sound, so that the stereo image seemed to float in space rather than appear to be coming from a couple of speakers.

“We started to make the Model 105 and we thought we’d use the computer to make it very accurately because it was the beginning of computer matching [of left and right speakers in a pair]. But at the time our attempt at computer matching was simply to print the frequency responses of the first fifty or so and we would look at one another and say, ‘do you have one like this?’ We decided that wasn’t particularly effective so [one of our engineers] wrote us a matching program, and that was the beginning of computerized measuring in production.”

Music is an art we are all passionate about, that’s why for six decades we’ve been at the forefront of using science to make the best possible loudspeakers you can buy.

https://us.kef.com/blog/1969-kef-pioneers-the-use-of-computers-in-the-design-of-loudspeakers


Prior to the rise of the "audiophile" leisure class and subjective reviewers/influencers, there was incredible work by people at Bell Labs and a number of other companies, which was always driven by science. Dr. Harry Olson at RCA Laboratories did some incredible work with transducers in particular (microphones as well as loudspeakers). Brief educational highlights: Bachelor's in Electrical Engineering, Master's with a thesis on acoustic wave filters in solids, followed by a doctorate in Physics.

https://en.wikipedia.org/wiki/Harry_F._Olson

Shortly after World War II, Dr. Olson conducted an experiment, now considered a classic, to determine the preferred bandwidth for the reproduction of music. Previous experimenters had found that listeners seemed to prefer a high-frequency cutoff of 5000 Hz for reproduced music. Dr. Olson suspected that this was likely due to imperfections in the sound, especially in the higher frequencies, as reproduced by equipment in common use at the time. These imperfections included clicks and pops (from 78 rpm recordings), added noise (from AM radio broadcast static), hiss and harmonic distortion (from amplifier circuits), and non linear frequency response from primitive loudspeaker designs. If the sound was free of these problems, he reasoned, listeners would prefer full frequency reproduction.

In his experiment, he set up a room which was divided diagonally by a visually opaque but acoustically transparent screen. The screen incorporated a concealed low-pass acoustical filter having an upper frequency cutoff of 5000 Hz. This filter could be opened or closed, allowing either the full range of frequencies to pass or the range only below 5000 Hz. At first, a small orchestra sat and performed on one side of the screen, while a group of test subjects sat on the other and listened. The listeners were asked to select their preference between two conditions: full bandwidth or restricted bandwidth. There was overwhelming preference in favor of the full bandwidth. Next, the orchestra was replaced with a sound-reproduction system with loudspeakers positioned behind the screen instead. When the sound system was free of distortion, the listeners preferred the full bandwidth. But when he introduced small amounts of nonlinear distortion, the subjects preferred a restricted bandwidth, thus demonstrating clearly the importance of high quality in audio systems.

As a result of this experiment and the work of others, such as Avery Fisher and later Edgar Villchur, high fidelity sound recording, transmission, and reproduction equipment saw increased investment, development, and public acceptance in the following decades. The design and manufacture of everything from microphones, to tape recorders, vinyl records, amplifiers, and loudspeakers were impacted.

Please check out this fantastic article from the February, 1954 issue of Radio News, and be amazed at all the data! hahaha!:

Radio-News-1954-02-R Harry Olson and John Preston article.pdf

RCA published frequency response graphs, directivity plots, harmonic distortion, impedance, sensitivity, etc. etc. on Dr. Olson's designs back in the late 1940s and throughout the 1950s.

Oh, look, it's Elvis, listening to his latest record played back over the RCA LC-1 (not sure which variant of the loudspeaker it is, it was nicknamed "the furnace" because of the cabinet appearance haha!):

 

[Thomas savage]

Well there's new designs then there's what's been in your hard disc for fifteen years that's your finally getting round to putting together.

 

[Purité Audio]

Hi John Booty,
If so, I would like to get the proof that at high end audio people still love paper cones; I guess this goes from woofer units to tweeter units ? And please give a list of all high equipment supplier who design this way.

I do not think that the audio engineers in speaker audio would agree with your statement as it would mean they all have a meaningless job isn't it ? Can you name one who agrees with you ?

As far as I'm concerned active speakers can sound impressive with all kind of music but don't play anything acoustic; you can hear the instruments are transformed in an electric violin, an electric piano or an electric ........This perception is of 30 years ago and didn't change when f.e. listening to a Kii Three set nowadays.
Nothing wrong if you do not reject this effect but for me this is not high end.

At last an example of one of the bigger speaker designers that did and still does a lot of research;
Did you see and compare the measurement report of the B&W 804 D2 and the newest 804 D3 ? Did you listen to them in a comparison ? Can you tell what design change caused the biggest change in sound perception ?

Yes it’s well known that active speakers can only play active music and passives , passive music.
Keith

 

[thewas]

Radiation pattern, structural analysis of the casings, numeric simulation and optimisation of crossovers were all stuff known and used already by some companies in the late 70s, so even from that time there exist some speakers which sound even with today standards good, something you can't say for 40 year old TVs, computers, telefphones etc.

Here for a example is a 1979 AES paper about a loudspeaker I am glad to own today, with modal analysis and a flat 13" woofer which has its first breakup mode at 3 kHz!
http://jahonen.kapsi.fi/Audio/Papers/Loudspeaker with honeycomb disk diaphragm/

The biggest improvement in the last years was the optimisation of loudspeaker chassis drives with Klippel etc. especially in regards of low distortion and maximum SPL, so it could be said that a good loudspeakers of a specific dimension may not got so much better, but quite louder.

 

[JohnBooty]

Hi John Booty,
If so, I would like to get the proof that at high end audio people still love paper cones; I guess this goes from woofer units to tweeter units ? And please give a list of all high equipment supplier who design this way.

Whenever I get a reply such as this I find myself wishing that "the internet" had some central location upon which a meteor would fall.

Do you really think I meant that all high-end drivers are made out of paper and that's the only thing people like?

Come on.

The list of high-end paper cone woofer manufacturers is extensive: Morel, Scan-Speak, SEAS. And so on. And those woofers are used by many other companies in turn: for example you'll find Scan-Speak drivers in Wilson Audio speakers and those from many other high-end manufacturers. If you'd like to know more, check out the high-end drivers available at Madisound. Many, not all, are paper cones.

ASR deserves better than the kind of toxic internet discourse where you assume everybody else is an idiot and jump all over one slightly vague sentence with four nasty paragraphs of your own. Good discussions don't happen that way.

Please treat your fellow ASR members with respect unless you have some kind of reason not to.

I do not think that the audio engineers in speaker audio would agree with your statement as it would mean they all have a meaningless job isn't it ?

I said, "I've not seen anything to suggest that driver construction has improved too much in the last 20 years."

I didn't claim there's been no progress. Of course there has been some. But hi-fi systems were already reproducing audio with excellent fidelity 20+ years ago. There aren't a lot of breakthroughs left to be made in traditional loudspeaker design. Unless somebody comes up with a radically new type of speaker, the primary progress we'll see over the next 20+ years will be a continuation of the trend of hi-fi audio features trickling down to lower price points. The JBL 3-series speakers are a good example of performance in a $149 speaker that would not have been imaginable 20 years ago.

As for what all the audio engineers of the world are doing, the vast majority are not working in high-end audio designing big fancy woofers for home hi-fi setups, I can tell you that much. Ultra high-end drivers are an extremely niche product and that is one reason why progress has not been revolutionary in that area. How many drivers did Scan-Speak ship last year? Probably less than the number of iPhones Apple sells in an hour.

For decades now the majority of audio engineering resources have been poured into embedded audio. Car audio, bluetooth speakers, the speakers in your phone and laptop, etc. These markets utterly dwarf big home hi-fi speakers by multiple orders of magnitude. And, while those miniature speakers will never match a nice set of proper home hi-fi speakers due to their size constraints, progress in those areas has been truly astonishing in the last 20 years.

 

[mhardy6647]

Virtually all of the important stuff was worked out in service to the development of "talkies" in the late 1920s -- and, as mentioned above, Harry Olson probably did most of the rest.

Anyone in this thread who has not heard an RCA LC-1A or an Altec Duplex... well... you should.
A Quad "ESL-57", too.



I don't have a pair of LC-1As -- but I know a fellow with two pairs of 'em. One in period enclosures, one in modern enclosures. Remarkable loudspeakers are the LC-1As.

source: http://www.itishifi.com/2011/02/rca-lc-1a.html