Are we at the pinnacle of audio?

[user 37518]

Seems to me that everything when it comes to sound reproduction is already well established and invented, even cheap power amplifiers are in the 0.00x% THD range, electronic components are so good and accesible. Audio is a mature technollogy, seems like everything analog has been invented and digital is, in my opinion, as good as  will ever be, or should I say, as good as it needs to be for our ears, a 32bit 384KHz converter is something completely useless if you ask me.  DSP is growing but the theory was already well established in the 70s! My book on DSP by Oppenheim is from 1975...

In my opinion, the Achiles heel of audio are speakers, when it comes to THD and frequency response they are several orders of magnitude worse than the amplifiers driving them, there is something so primitive about a coil wound on a cardboard cone, I know that there are many modern speakers with many advancements, and perhaps thats the next frontier in audio,  designing a completely revolutionary and different speaker, as for the rest of the stuff, I think we are just re-inventing the wheel.

 

[living sounds]

I think music sounded a lot better before Delta Sigma converters with steep linear filters. Our ears cannot recognize the difference between 0.001 and 0.0001 % THD, but they sure can detect how life is sucked out of the music by converters. Since most music these days is produced and auditioned via these converters few people have a clue what they are now missing.

And don't get me started about plugins, vintage vs modern synths, the drawbacks of the now-ubiquitous ported speakers, loudness wars, lossy compression formats, music produced for smartphones and headphones etc...

We're long past the pinnacle IMO and it will be quite some time until we get this good again...

 

[JohnRoberts]

Seems to me that everything when it comes to sound reproduction is already well established and invented, even cheap power amplifiers are in the 0.00x% THD range, electronic components are so good and accesible. Audio is a mature technollogy, seems like everything analog has been invented and digital is, in my opinion, as good as  will ever be, or should I say, as good as it needs to be for our ears, a 32bit 384KHz converter is something completely useless if you ask me.  DSP is growing but the theory was already well established in the 70s! My book on DSP by Oppenheim is from 1975...

The cliche is so old that there was a joke in an 1899 edition of Punch magazine.

In that edition, the comedy magazine offered a look at the "coming century." In colloquy, a genius asked "isn't there a clerk who can examine patents?" A boy replied "Quite unnecessary, Sir. Everything that can be invented has been invented."

 
While a little overstated, I was impressed about how much was known from my father's engineering notes and old texts from the 1930s.

In my opinion, the Achiles heel of audio are speakers, when it comes to THD and frequency response they are several orders of magnitude worse than the amplifiers driving them, there is something so primitive about a coil wound on a cardboard cone, I know that there are many modern speakers with many advancements, and perhaps thats the next frontier in audio,  designing a completely revolutionary and different speaker, as for the rest of the stuff, I think we are just re-inventing the wheel.

Loudspeakers are difficult but have also enjoyed huge advancements over the decades. Loudspeaker distortion is not very offensive like some active electronic path distortion and we have learned to ignore it. I have heard about cases where improved very low distortion bass speakers, were perceived as having weak bass output, because the missing bass distortion occurring at higher, more audible frequencies was considered normal and correct.

The vast majority are materials improvements like higher temperature adhesives that support higher temperature voice coils, but indeed mature technology. IMO the current SOTA for loudspeakers involves work in the area of advanced horn design, and advanced DSP applications for reducing distortion and protection.

Modern marketers invent differences between SKUs that don't exist to sell more poop... Consumers want to believe that audio is not that well understood (magic), so they can discover some new breakthrough by trial and error.

The reality is that there is always room for improvement but it already works pretty damn well...

JR

 

[Gold]

Recording equipment and the average recording  medium have become very low distortion. Simultaneously  more distortion Is being added to productions.  Equipment that would have been considered too noisy and high distortion when recording to tape have gained new life when recording to digital.

 

[boji]

Next might be in-ear adaptors paired to Musk's neural-link implants. Once that happens it's wide open again.

Examples:

Music that has ultra high frequency content might (might!) be appreciated later.  ;D

Using stereo field that depending on position of head changes song into another song that is in complimentary key, or perhaps phasing can hide/reveal song content depending on if you're standing still or running (software layer interacts with phone and/or implants).

Then there's 5.1 for the brain via neural-link training.  Adaptive training.  The brain expects what the ears neglect. What happens when the brain can receive/interpret more info that what the auditory nerve can currently supply?

The frontier is there if you imagine it.  =D

 

[pucho812]

This argument is decades old..

Read all 6 parts of “what have they done to my art” by Walter e. Sear

It’s an eye opener for sure

 

[user 37518]

I think music sounded a lot better before Delta Sigma converters with steep linear filters. Our ears cannot recognize the difference between 0.001 and 0.0001 % THD, but they sure can detect how life is sucked out of the music by converters. Since most music these days is produced and auditioned via these converters few people have a clue what they are now missing.

And don't get me started about plugins, vintage vs modern synths, the drawbacks of the now-ubiquitous ported speakers, loudness wars, lossy compression formats, music produced for smartphones and headphones etc...

We're long past the pinnacle IMO and it will be quite some time until we get this good again...

I disagree, I believe that music has never sounded any better, and by that I mean the medium is the best there is, not all music sounds good that is for sure but I'll take 44.1KHz/16bit over vinyl or 15 IPS tape every time. To me the whole analog/vintage gear revival and emulation is just a fad, let me tell you, I've seen it in the very young generations, they dont care about old equipment, for them its all about using technollogy and digital is the way to go.  I believe that the whole analog VS digital debate will end and it seems like digital is going to be the winner.

 

[john12ax7]

I would say we're not even at a local maximum.

In the future new things will get invented and things will improve,  just not necessarily in a straight line.

Recorded sound is still distinguishable from live sound,  regardless of medium.

 

[user 37518]

I would say we're not even at a local maximum.

In the future new things will get invented and things will improve,  just not necessarily in a straight line.

Recorded sound is still distinguishable from live sound,  regardless of medium.

I believe that is the case due to speakers like I mentioned in my first post, speakers do not sound like real live music, but the reproduction system itself is in my opinion as good as it should be.

 

[JohnRoberts]

Recorded sound is still distinguishable from live sound,  regardless of medium.

I try not to argue with people about what they say they can hear but I vaguely recall an experiment done last century in a recording studio where they worked to eliminate obvious tells.... IIRC they had live musicians in a studio, some premium recoding medium, and double blind listeners in the control room with A/B/X switch. As I recall they couldn't tell.

Another classic "is it live or memorex demonstration" was performed back even further...(1960s?). They had a chamber music ensemble (4 players?) on stage in Symphony Hall (Boston).  Listeners were in the audience a reasonable number of rows away from the stage. In this case the sound was dominated by the hall acoustics and they couldn't tell if it was live or a cassette tape recording (as long as it was memorex tape).  ;D

JR

 

[john12ax7]

In the early days of recorded sound people couldn't distinguish live from recorded either,  at least initially.  Now those old recordings are obvious to us as things have improved.

The cycle continues to repeat itself,  there were the memorex ads,  early CDs were regarded as "perfect ". Over time the ear learns the medium deficiencies.

And I also agree speakers have some ways to go.  My prediction is that a completely new technology will emerge to replace the current mechanical system.

 

[Tubetec]

I have a set of Quad Esl-57's designed in the 50's . The transparency of the sound from them is hard to equal even by todays standards . The other thing about them is they have a figure of 8 dispersion front to back and so dont excite room resonances in the same way as regular box speakers .They do have a few downsides ,  low output and limited power handling ,maybe a little bass light compared to honky ported box speakers , still the quality they manage in some respects beats many modern 'coloured' sounding  box speakers.

 

[bluebird]

but they sure can detect how life is sucked out of the music by converters. Since most music these days is produced and auditioned via these converters few people have a clue what they are now missing.

Whenever you say something like this regarding converters it always throws me off. Your so scientifically inclined, but this is pure audiophoolery talk. Surprises me coming from you.

There are no scientific tests proving this so you should be careful with statements like that.

I've said it on here before,  I've made exhaustive ABX tests in a real recording studio with real engineers that worked the rooms for many many years (probably 8 or so years ago now). We did tests on converters and sample rates. And in the end No one could tell the difference between any of the converters or any of the sample rates 16bit  44.1k up to 24bit 192K.

I can't expect you to believe me because I don't have an official record of the tests. The company wanted to figure out which A/D converter and tape machine was the best for archiving.  We were given a couple months to provide an answer and we were as non biased and scientific about it as possible.  There was no reason to choose one converter or sample rate over the other. We ended up going with the grey and blue 192K Digidesign converters at 24bit 192K (and Studer machines). The Digidesign converters because we had a bunch of them already and no one could tell the difference between them and any other expensive converter we tried. 24bit 192K just because it was "high"  and the bosses decided it was better to use 192K "just in case". Not because anyone could tell the difference between a master tape encoded at 44.K or 192K on a million dollar playback system among professional sound engineers young and old.

And yes people could definitely tell the difference between the ATR's and the Studer's...

Yes! digital audio technology has long reached its pinnacle with regards to what we can hear.

Mixing, plugin processing, encoding is another story I don't have enough info on to say anything but the listening of stereo digital files I feel confident on what I've found in these tests. And add to that I'm a mastering engineer that deals with converters and sample rates for a living, I can tell you converters and 16bit 44.1K don't suck any life out of anything.

It sounds like you are really giving into psycho-acoustics and voodoo stuff which is weird for what I know about you on this forum.

Just sweep a sign wave generator up till you can't hear it anymore. At even 16K it get so quiet... aliasing at 20khz just can't make much a difference to humans. Let alone suck the life out of anything! Kids listen to mp3's with no problems!

Anyhow sure I'm not going to change your mind but maybe I've planted a seed for further investigation or more deliberate and careful tests on your equipment. If something is sounding like the life is sucked out of it... its not the converter, something in your system is broken.

I mean... I have good friends who say they can tell the difference, and maybe they can, But it has nothing to do with Delta Sigma converters with steep linear filters. That technology is very advanced and I would say "perfected". Its now all about getting the same old performance into a smaller package with less power consumption. But that's not even that important anymore because (most) people just don't care!

No disrespect my friend, just had to say something because what you say may be misleading to others.

-Ian

 

[Scodiddly]

I would say that we're at a point (perhaps to be surpassed and looked back on differently) where recording high quality audio is no longer difficult or expensive.  So there's no longer any excuse for getting one's music recorded and distributed to others.

No limitations, aside from maybe personal time.  You can even get a pretty good orchestral sample library cheap enough these days, if your music requires one.

But then a lot of art seems to be stimulated by limitations, so maybe we're not in that great a situation after all.

 

[Tubetec]

I have a Yamaha aw16g which I found going cheap around ten years ago so snapped it up , I did a few sessions on it ,limited to 44.1 16 bit , I definately notice a difference between that and the Akai DPS24 running at 48k 24 bit .
I do accept Ian(Bluebirds) point that when your archiving something off tape , where you know precisely what your max level is going to be the difference wont be huge between 44.1khz/16bit and higher sample/bit rates , likewise relating to Ruairi's experience from a mastering perspective , you know precisely what levels your dealing with , a recording session is a different ball game as you have to anticipate levels and allow a margin in terms of headroom just in case you get a transient , that definately is a situation where the extra dynamic range of 24 bit is a bonus . I liked the point LivingSounds made about music being produced today in the knowledge that playback via smart speaker /bluetooth /mobile phone /MP3 etc , so no one cares a damn  eitherway.

I mean even the speakers in modern tv sets nowadays have crap bass end compared to the old hat crt tv's where you maybe had space for a four inch driver in the cabinet , in general terms consumer sound gear has gotten worse in the last two decades ,not better . What did you have in the 80's , maybe a sony or sharp brushed 'silver' finish phono tape and radio combi with speakers made from chip board , they still sound better now than what the average person listens to music on  . I have a Pye radio from the 50's and the six inch alnico driver and tube amp gives more realistic and pleasing  tone for spoken word to my ears.
Forget about the metering and measuring ,close your eyes and allows the ears be the final judge, distortion becomes objectionable at many times whats measurable anyway , modern amplification topologies distort more nastily than old ,thats not an issue if your judicious with the volume control ,but when the drunk guy at the party cranks it ,it does matter ,or in the case of tracking a recording where transients are likely .

I'm getting back into national radio in the evenings here , RTE Radio 1 , Creedon , Kenny and Ó Braonáin , I dont need my visual senses hijacked by imagery , I have a log fire to look at 

Regards ,
David.

 

[living sounds]

I disagree, I believe that music has never sounded any better, and by that I mean the medium is the best there is, not all music sounds good that is for sure but I'll take 44.1KHz/16bit over vinyl or 15 IPS tape every time. To me the whole analog/vintage gear revival and emulation is just a fad, let me tell you, I've seen it in the very young generations, they dont care about old equipment, for them its all about using technollogy and digital is the way to go.  I believe that the whole analog VS digital debate will end and it seems like digital is going to be the winner.

You misunderstand me, I think digital is great, I was criticizing certain aspects of most of the currently availible converters.

 

[living sounds]

Whenever you say something like this regarding converters it always throws me off. Your so scientifically inclined, but this is pure audiophoolery talk. Surprises me coming from you.

There are no scientific tests proving this so you should be careful with statements like that.

I've said it on here before,  I've made exhaustive ABX tests in a real recording studio with real engineers that worked the rooms for many many years (probably 8 or so years ago now). We did tests on converters and sample rates. And in the end No one could tell the difference between any of the converters or any of the sample rates 16bit  44.1k up to 24bit 192K.

I can't expect you to believe me because I don't have an official record of the tests. The company wanted to figure out which A/D converter and tape machine was the best for archiving.  We were given a couple months to provide an answer and we were as non biased and scientific about it as possible.  There was no reason to choose one converter or sample rate over the other. We ended up going with the grey and blue 192K Digidesign converters at 24bit 192K (and Studer machines). The Digidesign converters because we had a bunch of them already and no one could tell the difference between them and any other expensive converter we tried. 24bit 192K just because it was "high"  and the bosses decided it was better to use 192K "just in case". Not because anyone could tell the difference between a master tape encoded at 44.K or 192K on a million dollar playback system among professional sound engineers young and old.

And yes people could definitely tell the difference between the ATR's and the Studer's...

Yes! digital audio technology has long reached its pinnacle with regards to what we can hear.

Mixing, plugin processing, encoding is another story I don't have enough info on to say anything but the listening of stereo digital files I feel confident on what I've found in these tests. And add to that I'm a mastering engineer that deals with converters and sample rates for a living, I can tell you converters and 16bit 44.1K don't suck any life out of anything.

It sounds like you are really giving into psycho-acoustics and voodoo stuff which is weird for what I know about you on this forum.

Just sweep a sign wave generator up till you can't hear it anymore. At even 16K it get so quiet... aliasing at 20khz just can't make much a difference to humans. Let alone suck the life out of anything! Kids listen to mp3's with no problems!

Anyhow sure I'm not going to change your mind but maybe I've planted a seed for further investigation or more deliberate and careful tests on your equipment. If something is sounding like the life is sucked out of it... its not the converter, something in your system is broken.

I mean... I have good friends who say they can tell the difference, and maybe they can, But it has nothing to do with Delta Sigma converters with steep linear filters. That technology is very advanced and I would say "perfected". Its now all about getting the same old performance into a smaller package with less power consumption. But that's not even that important anymore because (most) people just don't care!

No disrespect my friend, just had to say something because what you say may be misleading to others.

-Ian

I own a  Soekris DAM1024, that's a semi-DIY DAC based on high-precision surface mount resistors. Pure PCM, sign-magnitude (which more or less means you get the same THD regardless of amplitude) with an FPGA for filters. You can design you own filters if you like and load whatever filter the DSP can handle into it.

You wouldn't believe what tiny differences in the filter configurations make to our ears. I ended up with a non-oversampling configuration, with a passive transitor-coupled stage after the resistor ladder. Everyone who had a listen couldn't belive how good the DAC with no digital filters sounds. They were hearing stuff in the music they never recognized before because the separation is so good.

There are other people who swear by the textbook linear-phase filters, but to my ears these always destroy the definition of the signal.

One of the problems with delta-sigma is that the high clock frequencies makes the converter very sensitive to jitter. The other is that you cannot do without the digital filters.

BTW, I think 16 bit is enough, a little more than 44,1 khz would be good to make filtering easier, but the medium can sound great indeed.

As for scientific tests, no converters to date are accurate enough to actually null to a sufficient degree with the non-converted signal. It obviously isn't doable at lower sample rates (because the filter always affects the audible range) and there appear to be other problems at higher sample rates. Most converters, even DC coupled ones, produce pretty obvious differential signals in a null test (see the diffmaker thread for a list at Gearslutz). I've tested the phase response of some popular converters and they showed strange responses, and different ones between sample rates (like switching from 44,1 to 48) no less.

Again, the problem is not THD, aliasing, frequency response. It's getting a clean step response. Same with ported speakers. I really dislike auditionion through those.

Optimizing for ultra-low THD or linear-frequency response up to 22 khz within 0.01 dB while sacrificing the step response doesn't make much sense, if the human hearing is incapable of differentiating the former but is actually sensitive to the latter.

We haven't spent anywhere near the money on audio research, a lot of pychoacoustic stuff isn't properly understood. And what we understand is often not applied. So don't jump to conclusions.

 

[bluebird]

Ruairi's experience from a mastering perspective ,

Me and Ruairi are close friends and sometimes talk about this stuff. He's more on the subtle almost spiritual side of sound, than I am. But he has one of the most expensive well designed listening environments that I have ever seen or heard so when he tells me this so and so converter is better than another, I have to believe something may be going on,

But I feel its important to put things in context when talking of such subtle things.


If you have all the best stuff and are at the top of your game in your field, perhaps waxing poetic about this or that AD codec is not a waste of breath, but for 99% of the people it can start them down an expensive path that's not going to make them better engineers or make they're music sound better.

I fear a lot of time is wasted for most people fiddling with converters, and worse... clocks. Younger engineers will ask what converters they should upgrade to and I tell them to invest in better microphones. Converters should be the last thing to be upgraded in a system and only when you have disposable income. I always like to point out how many amazing albums were recorded on ADAT. Even more on the crappy pro tools 888/24 systems.


Like in the world of fine wine. Wine tasters can talk all day about the intricate flavors and notes of whatever super expensive top shelf wine but for most people you can get a very good tasting bottle of wine for under $20 and the effects on the brain are the same...

 

[bluebird]

Everyone who had a listen couldn't belive how good the DAC with no digital filters sounds. They were hearing stuff in the music they never recognized before because the separation is so good.

But that's just silly to say. Who is everyone? did you do blind ABX tests? Or just friends over for a listen and a beer? On what system?

Just saying:
"I own a  Soekris DAM1024, that's a semi-DIY DAC based on high-precision surface mount resistors. Pure PCM, sign-magnitude (which more or less means you get the same THD regardless of amplitude)"

Already has me thinking your system is going to sound amazing! Having people over to hear your amazing system is going to have them going home thinking they heard an amazing system, Lol.

"One of the problems with delta-sigma is that the high clock frequencies makes the converter very sensitive to jitter."

That's just not a problem, especially one you can hear. But the way you say it makes it  sound like a big problem and why people believe this kind of stuff.

" It's getting a clean step response."

The wave is perfectly reconstructed from any "steps". They are invisible. People always try to compare digital audio to pixels in a picture but the similarities are only metaphoric.

Only in super low resolution or really cheap converters is all this stuff a thing to hear. And possibly if you are recording 100 tracks of audio, things might build up but just listening to stereo wav files should not audibly reveal any of the problems you mentioned.

Yes those things are things, I'm not saying your crazy or anything, but they just don't have anything to do with what people hear these days with modern converter technology. So again, in my opinion, in that context, we have reached a pinnacle.

 

[living sounds]

Only in super low resolution or really cheap converters is all this stuff a thing to hear. And possibly if you are recording 100 tracks of audio, things might build up but just listening to stereo wav files should not audibly reveal any of the problems you mentioned.

Yes those things are things, I'm not saying your crazy or anything, but they just don't have anything to do with what people hear these days with modern converter technology. So again, in my opinion, in that context, we have reached a pinnacle.

Look, I've got all kinds of converter, from super cheap to super expensive. Every one of them sounds different. I can listen to synths and drum machines with and without conversion at the flick of a switch in my studio. Differences are easily audible to my ears and many other's. But it's not much use talking about these things on the internet, even uploading files doesn't do much since people audition via converters and different systems.