HiFi Technology Flatlined Last Century

[Theta]

HiFi is pretty much a solved problem. speaker

HiFi is pretty much a solved problem.

With the exception of speakers

 

[egellings]

With the exception of speakers

True.

 

[ferrellms]

With the exception of speakers

and the rooms they are in

 

[DEF]

Class D design is completely and radically different than class AB. The switching power supplies that power it are also the same with respect to linear power supplies. The two combined bring high efficiency, small size and ability to produce a ton of power. And with advent of SMD and better power transistors, combined with proper feedback design, is able to produce an incredibly unique offering than that of the past. It is not at all a refinement.

Klasik 2020's are using class A/B and Klasik claims because it sounded better for that particular case.
Meaning, they are less efficient at the cost of sound quality. Can you comment on that?

 

[Willem]

Infinity IRS, Altec 604's, Quad 57 come to mind. As for monitors The large ATc's.

I have had Quad 57s from the mid seventies, and am now playing their 2805 successors. The 57s are still a remarkable speaker, and one that I could easily live with. At the same time the 2805s are better in all respects: more power handling, more bass, less directional with a far more realistic spatial representation. I also have a pair of the venerable LS3/5as and their modern successors, the Harbeth P3ESR. Again, the modern incarnation is much better. In both cases audiophile romantics claim that the older models are somehow better, but this is just plain nonsense.
I also own a succession of Quad power amplifiers: a 303, a 405-2 and a 606-2. At low volume I cannot tell them apart, but when pushed, the much more powerful 606-2 is audibly cleaner. The 2x45 watt of the 303 does not cut it with the less efficient 2805s in my much larger current listening room.

 

[mocenigo]

Klasik 2020's are using class A/B and Klasik claims because it sounded better for that particular case.
Meaning, they are less efficient at the cost of sound quality. Can you comment on that?

These are about 1200USD per pair, so the cost of the components is critical. In the 100W output level (which is what the internal amp must deliver) a small AB amp will cost less than a class D solution. So I guess the reason is price. They can twist it to mean sound quality, because they would have had to use a lower quality class D amp, but we are talking of a ~50USD budget for two amps, tops.

 

[Snarfie]

View attachment 354383
On 26 April 1939: Test pilot Fritz Wendel flew a prototype Messerschmitt Me 209 V1, registered D-INJR, over a three-kilometre, closed course at Augsburg, Germany, setting a new Fédération Aéronautique Internationale (FAI) world record with an average speed of 755.14 kilometres per hour (469.22 miles per hour). The aircraft was fitted with a Daimler-Benz DB 601 engine which was a supercharged, liquid-cooled inverted V12 with direct fuel injection.
View attachment 354384
As such, this engine represented the peak of internal combustion engine development. Apart from the use of computerised fuel management, the internal combustion engine has not developed in any significant way since. The engine in a 2024 Mercedes still uses a crankshaft, camshafts, poppet valves, fuel injection, pistons etc, just as the DB601 did in 1939.
I feel there is a corollary between this and HiFi.
HiFi as it stands has developed very little in the last few decades. The stand-out development has been in digital technology but as far as reproduction equipment is concerned, name me one big development. Let's explore this assertion.

Sources:

Turntables. When did TT technology peak? It’s difficult to say for sure but my assertion is that a turntable has such a simple job to do, the peak must be when super accurate speed and negligible rumble was achieved. I’m sure there will be much controversy here but my pick is the Technics range of direct drive units. Having worked with the SL1000 and carried out repairs on them, the quality of manufacturing and the execution of the design has never been topped IMHO.
View attachment 354385
Ridiculously overengineered TT’s have been manufactured by niche outfits but if you take the OMA unit that has been widely discussed on ASR, you will find that the Technics SL 1500 turntable handily beats its rumble figure and is a fraction of the price.

Tonearms. The pivoting tonearm arguably reached it’s present state of technical development with a unit like the SME 3012, introduced in 1958.
View attachment 354386
This tonearm had a counterweight, an anti-skating mechanism and a soft lift. No real significant advancements since this model only differences in pivots, bearings and materials. Linear tonearms have made brief appearances but they are niche compared to pivoting arms. Some novel mechanisms have been developed for linear arms with air bearings and so forth but based on popularity, they seem to be a dead end.

Tape. Both analog and digital tape systems are effectively dead. Domestic analog tape reached a peak in the 1980’s with the advent of Dolby noise reduction. This development wasn’t really followed in professional analog tape equipment as it had already reached an optimum probably in the late 70’s when manufacturers like Studer were producing high quality multitrack and stereo mastering recorders.
View attachment 354387
However as a domestic technology, the compact cassette was the only tape format that was ever really commercially successful. While DAT and ADAT enjoyed a brief moment in the sun, both are now obsolete thanks to the dominance of HDD and high capacity memory digital recording. I don’t believe that there have been any new tape devices designed or manufactured for decades.

Digital Disk. Since the development of the SACD format in the 1990’s, there has been no significant technical advance in digital disk technology. Even though SACD was a flop commercially, it was a significant improvement on the original Digital Compact Disk from a technical perspective. As the format determines the specification of the playback device, there has been no significant improvement in this technology since then.

Digital file/streaming. From a technical and quality perspective, this is the only source medium that has had recent development. The advent of 32 bit float recording has pushed recording technology into the stratosphere and its capabilities far exceed anything that has preceded it. 32 bit float can record audio data +770 dB above 0 dBFS and -758 dB below. This gives 32-bit float recordings an incomprehensible dynamic range of 1528 dB. This figure is hard to fully grasp because the dynamic range between the quietest sound on Earth (an anechoic chamber) and the loudest sound possible (194 dB) is only 185 dB. With over 1000 dB of headroom above the quietest and loudest sounds on Earth, clipping is impossible. Distorted audio above 0 dBFS can easily be recovered in post by attenuating the signal. So, in theory, digital recording peaked in the last couple of years, certainly with reference to the capabilities of human hearing.

Components.

Amplifiers. Have amplifiers really advanced functionally since the 70’s? Amplifiers really hit their stride in the 1970’s. Full-bandwidth 20-20khz power at extremely low distortion became commonplace. Whether it was the modest amplifier section in a mid-priced receiver like the Kenwood KR-5400 (35 watts/ch RMS from 20-20kHz at <0.5% THD) from 1974 or the Pioneer Spec 2 power amplifier from 1976 rated at 250 watts/ch 20-20kHz at <0.1% THD, amps in the 1970’s delivered the goods.
View attachment 354388
Since then, most development has been incremental based on tweaking circuits and incorporating modern components. Some may say “what about class D, that’s new”. Yes, it’s the latest development of an audio amplifying device that operates in the range of human hearing. Functionally, a modern class D amplifier is so similar to a Kenwood KR5400 that it really only rates as a refinement, not a fundamental redesign.

Speakers. One area of HiFi that has had so much time and effort poured into it for so little effect is speaker design. There’s an ocean of speaker manufacturers and designs stretching to the horizon and apart from some obvious differences, open versus closed for example, speakers from the cheapest to the most ludicrously expensive share the same fundamental mechanisms of operation, the moving coil, cone loudspeaker. The moving coil loudspeaker was developed by C.W. Rice and E.W. Kellogg in the early 1920’s.
View attachment 354389
Since then, the basic design has been refined and improved using modern materials as they became available. These improvements have brought the design to a plateau where the physical limitations of the device have been reached. Put simply, moving coil loudspeakers are as good as they will ever be. Someone with more knowledge than me might take a stab at when this plateau was reached but my guess would be the 80’s or 90’s when materials like carbon fibre and Kevlar were incorporated into the construction, thereby allowing the mechanism to get as close to its ideal as is practical. “What about electrostatics?”
View attachment 354391
Kudos to QUAD and the others who explored this technology as it was one of the few times that a fundamental change was achieved in speaker design. The shame is that, while electrostatic speakers showed much promise, the reality was that they lacked the properties that were already common in moving coil speakers. Electrostatic speakers did not reproduce low frequencies as effectively as conventional speakers and the SPL that was achievable was well behind also. It’s a bit like the piston engine vs the rotary engine. Rotary engines have some notable qualities when compared to piston engines, simplicity, reduced reciprocating mass and compactness. However, the rotary design has some built-in problems that can’t be refined out like the combustion chamber shape which lowers the efficiency of the engine.

DAC’s. As I pointed out in another of my posts, there are only a few manufacturers of high quality DAC chips for audio reproduction. Subsequently, only the supporting circuitry is different between manufacturers.
View attachment 354392
A quality DAC chip costs around US$50 so a $10,000 DAC will have the same practical performance as a $200 dollar unit with the same chip. DAC chips probably also represent another true advancement in audio reproduction technology inasmuch as some of them include digital signal processing (DSP) that can be employed to compensate for room acoustics amongst other things. This technology probably still has some room left for development but for the HiFi stereo crowd, the prospect of digitally processed multi-speaker systems brings out their inner Luddite. The “High End” acolytes will never accept this sort of meddling and just want to slink off and try to tweak their pure two speaker equipment closer to perfection.

At this point in time, just about anyone can have an audio reproduction system that does everything so well that there’s no point in trying to improve it. In fact, for most components, the listener could put together a system composed entirely of devices made last century and be assured that they perform just as well as anything they could purchase in 2024. Without any question, speakers are the last link in the chain and the performance of those components will have the greatest perceivable effect on the quality of the sound being reproduced. Speaker choice has many variables, budget, available space and subjective performance. However, speaker technology is at the same place it was in 1970 in all but detail.

I think that this situation has produced all of the laughable tweaks and snake-oil products that are currently swamping the HiFi scene. As there’s nowhere to go as far as the basic equipment is concerned, once you’ve reached the limit of what you can spend on components, if you desire more from your gear, you’re a prime target for hokey products that claim to be able to improve the unimprovable. The old quote “a lie repeated loud and long enough becomes the truth” has never been more accurate when applied to HiFi components and accessories. A lot of time, money and effort has been put into creating products that do nothing but thanks to the malleability of human perception, masses have been hoodwinked into believing that these things are having a positive effect on the performance of their audio equipment. Has this held back true development and improvement in audio equipment? I don’t think so. The DAC chip is a prime example of how there are still genuine, engineering-driven developments going on in the field of audio. Notice though that something like a DAC chip is not being developed by and for “High-End” HiFi applications but for broad application in entertainment devices like televisions and media centres. HiFi equipment manufacturers just ride on the coattails of these developments and repackage them to try to convince the punters that they’re getting something “special” for the inflated price.

Mabey worthwhile (for many probably insignificant) mentioning development power consumption as a example by replacing my 1TB HDD for a 2TB SSD has increased speed considerably an brought back laptop wattage considerably from around 30 watts to 7 watts consumption as a stationary NAS (Foobar2000 media server) with the screen closed. The laptop fan hardly needs to cool anymore + SSD, no mechanical parts, dead silent. The Raspberry Pi3 (used as server/NAS housewide) in combination with WD 3TB Backup HDD consumes (combined) around 15 watts, so is now close to 100% less in consumption and therefore declared obsolescent. So a decent laptop combined with SSD as audio media storages an server/NAS consumption comes close to a LED lamp.

https://imgur.com/AI0YZGw

 

[ahofer]

These are about 1200USD per pair, so the cost of the components is critical. In the 100W output level (which is what the internal amp must deliver) a small AB amp will cost less than a class D solution. So I guess the reason is price. They can twist it to mean sound quality, because they would have had to use a lower quality class D amp, but we are talking of a ~50USD budget for two amps, tops.

Although there are cheaper chip-based amps that would fit the bill? Not up to the specs of discrete Class D yet, tho.

 

[antcollinet]

Although there are cheaper chip-based amps that would fit the bill? Not up to the specs of discrete Class D yet, tho.

Probably up to the specs of a typical AB though. If you can get one with a decent cooling solution.

 

[DEF]

These are about 1200USD per pair, so the cost of the components is critical. In the 100W output level (which is what the internal amp must deliver) a small AB amp will cost less than a class D solution. So I guess the reason is price. They can twist it to mean sound quality, because they would have had to use a lower quality class D amp, but we are talking of a ~50USD budget for two amps, tops.

They draw 30w each 24/7 doing nothing

 

[dlaloum]

They draw 30w each 24/7 doing nothing

So do many class D amps... (my Crown XLS2500 seems to draw circa 35W at idle)

 

[DEF]

So do many class D amps... (my Crown XLS2500 seems to draw circa 35W at idle)

Per speaker or together? The 30w is meant per speaker.

 

[dlaloum]

Per speaker or together? The 30w is meant per speaker.

circa 35W for a stereo power amp - so two channels / two speakers....

But there are plenty of AB designs that have similar idle power draws - or lower... my Quad 606 draws around 14W at idle for a 135W@8ohm/ch class AB power amp.

But many class AB designs try to make a virtue of having a high Bias current to maximise the time spent in Class A mode - needless to say, that will boost the idle power consumption! - The Quad current dumping design is a very nice sounding amp, while also being very power efficient....

 

[Willem]

Energy efficiency is increasingly important in this age of higher energy prices and global warming. In particular, anything that is switched on 24/7 demands close scrutiny. I know some keep their audio systems on 24/7, but I see no reason for such waste. 24/7 35 watt at idle is 300 kWh in a year, or more than 10% of a typical Dutch household's electricity use.

 

[DEF]

Energy efficiency is increasingly important in this age of higher energy prices and global warming. In particular, anything that is switched on 24/7 demands close scrutiny. I know some keep their audio systems on 24/7, but I see no reason for such waste. 24/7 35 watt at idle is 300 kWh in a year, or more than 10% of a typical Dutch household's electricity use.

I can't reasonably put my 200TB server to sleep as cost of wear and tear would be higher than just keep it on.
Imagine on every time I sleep I put possibility for 1 out of 20 motors to break.

If I had a 30TB SSD, I could offload a reasonable amount so that I only need the server on commits.
But those are prohibitively expensive. Or even a 60TB one e.g Western Digital Ultrastar DC SN655

 

[dsprito]

Interesting there has not been a single mention of the digital class D technology which potentially could be the most advanced audio path concept.
Class D now equals analog class D.
Bruno P. has managed to influence and convince the whole industry the more feedback the better for the audio quality perception which is controversial to put it mildly.
But what is not controversial it is much more cost efficient to deliver inflated and at some point negligibly relevant frequency domain parameters advantages.

 

[antcollinet]

which is controversial to put it mildly.

Not to people who understand the nature of feedback, it isn't

 

[Digital_Thor]

I think now that it's common to see detailed off-axis measurements (I had never seen many / any until a couple years ago personally) it's starting to dawn on more listeners exactly how much a waveguide can do for the dispersion.

I also think the CAD tools for developing good WGs are getting better and more widespread. There's even a guy on DIYAudio.com that's published free 3D printing files for waveguides for maybe a dozen tweeters or so.

So basically the means to create a good WG and the knowledge of their actual, quantitative importance is only just now becoming mainstream, I think.

Definitely more love to Augerpro for his great work
https://www.somasonus.net/waveguides

An SB ADC or CDC in a printed waveguide... just makes DIY so much easier.

 

[dsprito]

Not to people who understand the nature of feedback, it isn't

And what follows from this understanding ?

 

[antcollinet]

And what follows from this understanding ?

(correctly applied) Feedback linearises an amplifier (removes distortion), getting it closer to the ideal "straight wire with gain". It doesn't "damage the music" or any of the other audiophile myths that exist about it.