[KeithEmo]

Let me try and explain it in simpler terms......

I USUALLY listen to my music in stereo - most often on speakers - but sometimes using headphones.
But, especially when I use the computer, sometimes I try out one or another processor - like the new Dolby Atmos plugin for Windows or the Dolby Headphone plugin.
And, sometimes, I decide to make some sort of adjustment - like removing noise from a poorly re-mastered old recording.
And, sometimes, I listen to my music on other equipment, either or loan or at other folks sites.
Therefore, I need my music collection in a form that I KNOW will work AS INTENDED for any of these purposes.
I may put some files on a USB stick to play in my car - and, for that, I may use lossy compression.
However, it would be much too complicated and time consuming to have to run a whole new series of validation tests on some lossy format every time I try a new player program...
And, since we all know the sort of measurable changes lossy compression causes, we also know that the result WILL be significant differences with at least some processing.
And it would be a very expensive mistake to purchase files today in a lossy format - and then find out that they don't decode properly on the new player program I purchase tomorrow.
All of these problems and risk will be present if I use lossy compression - and I avoid all of them by NOT using lossy compression.

If, every time a new decoder, or a new piece of gear, seems to sound a little unusual, I have to get out the lossless copy to confirm where the difference lies....
Then I've found it's much easier just to use the lossless version to begin with - and avoid the possibility of having to.
As far as I'm concerned, the "cost" of using lossy compression, in terms of issues like this, is too high.... and the benefit is trivial at best.

To be honest, I don't care what the engineers or musicians can hear; I only care about what I can hear.
I also have a strong suspicion that the engineers and musicians may not have actually personally auditioned the specific MP3 version they're offering to sell me on Amazon.
(Therefore, the fact that "many of them have difficulty hearing the difference sometimes" isn't really relevant as applied to a particular album they didn't hear.)
And I know for a fact that they didn't try that five-year-old album on the new Dolby Surround Upmixer (which is what you get on all new Atmos AVRs - but didn't exist five years ago).
Or, to phrase that differently, an actual lossless copy of the content is "future proof", while we can't know if the lossy version will work the same on something new until we try.
(We can be sure that 'the original will work like the original"; but, with any lossy copy, we can never know until we try it.)

Again, "difficult" doesn't count... and neither does "maybe".
When I listen to a song I want to KNOW FOR SURE that I'm hearing what I should be.
Something that's "probably the same as the original" isn't good enough...
And neither is something where "I think the differences will be difficult to notice"...
To me, saving a few cents, or a few bytes, is an awful high price to pay for all those doubts.

And, yes, I've heard one or two studios, and many of the top manufacturer's "showcase audiophile systems".
And, yes, many of them sounded very good.
And, yeah, a few of them actually didn't sound very good at all (but I'm sure somebody thought they sounded great).
For the record, you will NEVER hear me make any claims about "how things sound on expensive systems" - one way or the other.
(there is some "very expensive audiophile gear" that is nowhere near accurate - and nor would I listen to it.)

Are you honestly suggesting that the studios you work for plan to start releasing all of their products in lossy format?
After all, if NOBODY can EVER hear the difference, then that would seem to be the obvious thing to do.
However, if they're not planning to do so, then I guess they agree with me that the lossy version really is "pretty good but not quite the same".

I have two direct questions for @gregorio....

Have you discarded all of your lossless recordings and replaced them with perfectly adequate lossy copies?
If not; why not?

1. Huh? How self-contradictory is that?

2. None of this makes any sense. If you want to alter the music then of course it's not going to "play exactly like it should" and "as it was intended". Lossless is not the simplest OR "most reliable way of achieving that goal". If for example you want surround sound, a far better/more reliable way of "hearing the music exactly as it was intended" would be to buy a recording actually mixed in surround, in say Dolby Digital (5.1) which is a lossy format, than upmixing a lossless stereo mix!



1. What about the musicians and recording engineers who actually created the recording (or any other engineers or musicians), could none of them tell the difference either? The assertions about modern high bit-rate lossy codecs is not solely based on what an "audience with no experience" can or cannot discern, it's also based on what engineers (with a great deal of experience) can and cannot discern. Not for the first time, your analogy/anecdote is NOT applicable and if anything, to the moderately intelligent/logical person, it actually indicates the exact opposite of your own assertion!

2. You agree with whom? It's only your assertion that lossy files "are often sufficiently close to lossless to be difficult" to distinguish, therefore you're only agreeing with yourself!
2a. But this isn't the "What KeithEmo is Convinced of" forum! Especially when "what KeithEmo is convinced of", contradicts the actual evidence and even your own evidence! According to bigshot you've actually done the thread's title, you've actually tested for yourself the audiophile claim and failed to discern a difference. So you not only failed to support the claim but actually added to the wealth of evidence which provides very high confidence that the claim false, yet still you support the claim and are "not convinced". Why is that?



1. Indeed I would be amazed, as in 20 odd years I've never come across a single person who could easily tell the difference. Even in the late '90's, when the lossy codecs were far less effective, it was NOT easy to tell the difference!

2. Nope, even with "seriously expensive kit" it is never "immediately apparent". Have you ever actually heard "seriously expensive kit" or have you only ever heard extremely cheap or fairly cheap kit? I assume you've heard average consumer systems and probably one or more audiophile systems, which at less than about $100k are fairly cheap but have you ever heard a "seriously expensive" system, say a $10m world class recording studio? Fairly often we get audiophiles coming here, making claims about "expensive kit" but it's both nonsense and hypocritical because typically they haven't got the faintest idea what an expensive, exceptionally good system actually sounds like and assume that everyone here is even more ignorant than they are!

G

 

[KeithEmo]

It would also be nice if the sports shoe manufacturers included a disclaimer that: "our shoe probably won't enable you to play like Michael Jordan"...
And the guys who sell minivans would include one stating" "we can't promise that your kids will become well behaved little angels like the kids in our commercials"...
And the fine folks who sell all those exercise machines would start showing average folks with average bodies using them...

However, that seems most unlikely.

Perhaps audiophiles should develop a little common sense...
And realize that ads for audio gear are probably no more, and no less, accurate than ads for cars, exercise gear, and health supplements.

I should also point out that most of us don't NEED a wrist watch that's accurate to 30 seconds a month.
However, most of us own one, probably because we wouldn't save much by seeking out and purchasing a watch that was off by five minutes a day.

In the end, I'm perfectly happy to accept the conclusion that .05% of the people on this planet "might" be able to reliably hear the difference between a 256kb lossy file and a lossless one. I know I'm not in that .05% group. Nor are most of us...

It would be lovely if folks selling gear or services on the basis of lossless being better were to say something like ",05% of you might be able to hear improved sound quality thanks to our lossless audio"

 

[KeithEmo]

I agree.....

And, as you mention, in this case we know the information has been lost.

I see it sort of like making a choice between an original work of art and "a counterfeit that's so good I probably won't be able to notice the difference".
Or between a real Rolex for $50 and "a Chinese knock-off that's just as good" for $35.
If the difference in cost is trivial - I'd STILL prefer to have the real thing - even if I probably won't notice the difference.
(And that would be true even if I can't sell it - so resale value doesn't enter into my decision.)

I think this one is different from possible differences in DACs, amps, and cables because there's a known loss of information involved. And it's not just a little information that's lost, it's quite a large percentage. The audible effect (or lack thereof) of that loss of information needs to be established with listening tests, so that takes us back to all of the issues with listening tests. But I personally am satisfied that, to my ears, the audible difference between 320 and lossless is very subtle at most, and very unlikely to be of any practical importance.

 

[Sgt. Ear Ache]

It would also be nice if the sports shoe manufacturers included a disclaimer that: "our shoe probably won't enable you to play like Michael Jordan"...
And the guys who sell minivans would include one stating" "we can't promise that your kids will become well behaved little angels like the kids in our commercials"...
And the fine folks who sell all those exercise machines would start showing average folks with average bodies using them...

However, that seems most unlikely.

Perhaps audiophiles should develop a little common sense...
And realize that ads for audio gear are probably no more, and no less, accurate than ads for cars, exercise gear, and health supplements.

I should also point out that most of us don't NEED a wrist watch that's accurate to 30 seconds a month.
However, most of us own one, probably because we wouldn't save much by seeking out and purchasing a watch that was off by five minutes a day.

sure. The interesting thing about audiophiles is some of them will try to tell you the shoes DO allow them to run a 4 minute mile...then when you ask them to show you they seem to kinda find ways to not do so. They "choose not to run!" (a little Seinfeld reference )

As far as watches go, we aren't talking about a choice between one thing that works and one that doesn't. We're talking about a choice between two things that work functionally identically (both these watches keep perfect time) - and it's been shown pretty conclusively that thats true in spite of your strange obfuscations to the contrary.

 

[castleofargh]

I think this one is different from possible differences in DACs, amps, and cables because there's a known loss of information involved. And it's not just a little information that's lost, it's quite a large percentage. The audible effect (or lack thereof) of that loss of information needs to be established with listening tests, so that takes us back to all of the issues with listening tests. But I personally am satisfied that, to my ears, the audible difference between 320 and lossless is very subtle at most, and very unlikely to be of any practical importance.

to return your caution in statements to you, do you actually have conclusive evidence that "to my ears, the audible difference between 320 and lossless is very subtle at most"? or are you just presenting it like that because at the back of your head you can't accept that so much done to a signal isn't audible to you? because if you do have a test showing that you noticed subtle differences at 320, bigshot, I and a few others would love to see it.

 

[SonyFan121]

1. Indeed I would be amazed, as in 20 odd years I've never come across a single person who could easily tell the difference. Even in the late '90's, when the lossy codecs were far less effective, it was NOT easy to tell the difference!

2. Nope, even with "seriously expensive kit" it is never "immediately apparent". Have you ever actually heard "seriously expensive kit" or have you only ever heard extremely cheap or fairly cheap kit? I assume you've heard average consumer systems and probably one or more audiophile systems, which at less than about $100k are fairly cheap but have you ever heard a "seriously expensive" system, say a $10m world class recording studio? Fairly often we get audiophiles coming here, making claims about "expensive kit" but it's both nonsense and hypocritical because typically they haven't got the faintest idea what an expensive, exceptionally good system actually sounds like and assume that everyone here is even more ignorant than they are!

G

In the late 90's/early 2000's most people had yet to be accustomed to lossy file formats. CD's and tape's where the norm back then. I actually bought one of the first ever MP3 players (it was a Philips) back in 2003 and the maximum bitrate it could play was 128Kbps. I was just a kid back then and couldn't tell the difference in sound quality between lossy mp3 and CD format, but now whenever I hear a 128kbps audio file, it's instantly recognisable. There are certain signs to listen out for to identify a lossy file format, it is called "artificats".

You do not need a $100,000 hifi system to be able to hear these artifacts. That is a silly and somewhat bizzare/crazy assumption.
This is my setup; (total cost of cd player + Linn power amp; $800. I bought the Linn amp for $330 in 2013 but it was actually worth $4,000 in 1993 - it's 25 years old.)
Whenever I hear a low bitrate file through this system, it's very obvious. You can say what you like, but I am very experienced in this hobby, I suspect I take it more seriously than you do, I know what i'm talking about. I do not need someone with less experience to tell me what I can and can't do.

Cheers!

 

[Sgt. Ear Ache]

In the late 90's/early 2000's most people had yet to be accustomed to lossy file formats. CD's and tape's where the norm back then. I actually bought one of the first ever MP3 players (it was a Philips) back in 2003 and the maximum bitrate it could play was 128Kbps. I was just a kid back then and couldn't tell the difference in sound quality between lossy mp3 and CD format, but now whenever I hear a 128kbps audio file, it's instantly recognisable. There are certain signs to listen out for to identify a lossy file format, it is called "artificats".

You do not need a $100,000 hifi system to be able to hear these artifacts. That is a silly and somewhat bizzare/crazy assumption.
This is my setup; (total cost of cd player + Linn power amp; $800. I bought the Linn amp for $330 in 2013 but it was actually worth $4,000 in 1993 - it's 25 years old.)
Whenever I hear a low bitrate file through this system, it's very obvious. You can say what you like, but I am very experienced in this hobby, I suspect I take it more seriously than you do, I know what i'm talking about. I do not need someone with less experience to tell me what I can and can't do

lol...BOOM!

btw, none of us are arguing that nobody can tell the difference between a 128kb lossy and a lossless file. Most of us can do that. This discussion primarily revolves around 256kb and up lossy files...

 

[castleofargh]

I think the results from this site speak for themselves...

http://soundexpert.org/encoders

I downloaded some tracks, and I'm both positively surprised and a little disappointed by how they went to do it.
first the bad:
there is too much delay for at least a guy like me to achieve the best discrimination I can. I've more than confirmed that I was way better with extremely short samples+back and forth "instantaneous" switching. here when I fail to notice something, I'm not actually sure if I could or couldn't pass in an abx.
the good:
I sort of get why they put it in a single file, it's easier to handle and marginally harder for the listener to cheat. I don't happen to think it was worth lowering the quality of the test for that, but I understand the choice.
the samples are pretty good choices IMO, speech, acoustic stuff, instruments known to be tricky for some codecs.... I've seen so many online tests where the track would have been fine at 96kbps that it's a good surprise for me.
they avoided the typical gain mistake!!!! all the files I tried were several dB below full scale and none of what I tried resulted in instersample clipping thanks to that. while if I remove my foobar setting(EQ, replaygain, avoid clipping,...) it doesn't take long to find clipping tracks, a few of them in a noticeable way for me. my point is they give a chance for the encoder to show what it can do, instead of trying to find any mean necessary to create audible difference.

 

[SonyFan121]

lol...BOOM!

btw, none of us are arguing that nobody can tell the difference between a 128kb lossy and a lossless file. Most of us can do that. This discussion primarily revolves around 256kb and up lossy files...

It is indeed harder to tell the difference between 256 kbps and 320 kbps, but you would not need a $100,000 hifi system to do so..

 

[castleofargh]

It is indeed harder to tell the difference between 256 kbps and 320 kbps, but you would not need a $100,000 hifi system to do so..

money is a silly reference in audio no matter what. but as always in mystic audiophile arguments, it is rooted in some legitimate idea and only later becomes perverted to make rubbish points about spending money or having superhuman hearing. depending on the audio cue that changes audibly, there will be situations where one playback system will make noticing harder. it could be as simple as a matter of frequency response. but it could also be about some solid noise, or distortions caused by the headphone that will end as loud or maybe louder than the artifact we're trying to perceive. possibly masking it entirely for us. so in some cases, having a very high fidelity system could probably let us go further when looking for our own hearing threshold.
but yeah it's a matter of circumstances, a matter of what we're testing, and most of all, money is not a measure of signal fidelity. based on Harman tests with headphones, it's not even a good measure of subjective fidelity.

 

[gregorio]

[1] In the late 90's/early 2000's most people had yet to be accustomed to lossy file formats.
[2] I was just a kid back then and couldn't tell the difference in sound quality between lossy mp3 and CD format ...
[3] You do not need a $100,000 hifi system to be able to hear these artifacts. That is a silly and somewhat bizzare/crazy assumption.
[4] You can say what you like, but I am very experienced in this hobby, I suspect I take it more seriously than you do, I know what i'm talking about.
[5] I do not need someone with less experience to tell me what I can and can't do.

1. We are not still in the late 90's/early 2000's though!
2. I wasn't, I was an adult, already an experienced professional and I could easily tell the difference between a 128kbps file and CD. Now, 20 years later and with much more efficient codecs and 320kbps or 256 VBR, I can't tell the difference and neither can any of my colleagues, either on cheap systems like yours or in the multi-million dollar studios where we work.
3. Hey, you're the one who said "But if using some seriously expensive kit, the difference can be immediately apparent.". Now that you've been called out on it, apparently have no idea what "seriously expensive kit" actually sounds like and only have cheap kit yourself, you're saying you DON'T need expensive kit. So which is it?
4. Just the fact that you call it a "hobby" demonstrates that you take it less seriously than me, because for me it's not just a hobby, it's how I make my living and it has been since before you were "just a kid"! So, you are OBVIOUSLY way less experienced than me, obviously take it way less seriously than me and clearly do NOT know what you're talking about! Again, it's the same old audiophile nonsense, you try the "I've got better kit than you" and when that's shown to be false, you change your tune and try the "I take it more seriously" and "I have more experience", both of which are also false. All you're doing is just making yourself look more foolish, surely that's not why you posted here is it?
5. If you "don't need someone with less experience to tell you what you can do", what makes you think that we do?

Your old audiophile tactics might work in other forums but not in this one, this is the sound science forum, NOT the "I've got cheap kit, less knowledge and experience but I'm going to state the opposite" forum! Why don't you do what the thread is titled and actually try a controlled test, instead of using false appeals to your (lack of) authority?

G

 

[Phronesis]

to return your caution in statements to you, do you actually have conclusive evidence that "to my ears, the audible difference between 320 and lossless is very subtle at most"? or are you just presenting it like that because at the back of your head you can't accept that so much done to a signal isn't audible to you? because if you do have a test showing that you noticed subtle differences at 320, bigshot, I and a few others would love to see it.

I was saying that I didn't notice obvious differences at all, so any differences I could have perceived, consciously or subconsciously, were presumably subtle at most. It's possible that I couldn't perceive any differences at all. I don't know. I didn't spend more time delving into, since I have both Tidal lossless and Spotify 320, and usually use Tidal "just in case" it sounds a little better, though sometimes I'm lazy and just keep listening using Spotify. My sense is that any audible difference between the two is tiny compared to differences in recording quality, my headphones, and how my perception is varying within the day and from day to day.

Another point to consider with all of this: maybe it's better NOT to have hearing which goes up to near 20 kHz. If differences in audio reproduction related to differences in DACs, amps, lossy, etc. are mainly at higher frequencies closer to 20 kHz, not being able to hear those high frequencies would mean that differences in gear matter less, or maybe they don't matter at all. It could be argued that not being able to hear those higher frequencies means missing out on part of the music, but such a person wouldn't hear those frequencies in normal life either, so presumably their brain would adapt and recalibrate what sounds natural and realistic to them based on the frequencies they can hear. Doing a FR hearing test is probably the first thing any budding audiophile should do.

And not being able to hear higher frequencies does NOT mean that someone has worse auditory perception. Auditory perception ability is very much influenced by experience, hence my being to distinguish between sounds of cymbals and judge whether cymbals sound realistic because I've heard a lot of cymbals in real life. I don't have that ability with the sound of the clarinet because I've never played the clarinet and not often heard it live except in the context of large orchestras. There are probably people out there who can hear up to about 20 kHz, but have mainly listened to crappy music without acoustic instruments, not heard much live music, and have never played instruments themselves, and therefore have relatively poor auditory perception skill.

 

[Phronesis]

It would also be nice if the sports shoe manufacturers included a disclaimer that: "our shoe probably won't enable you to play like Michael Jordan"...
And the guys who sell minivans would include one stating" "we can't promise that your kids will become well behaved little angels like the kids in our commercials"...
And the fine folks who sell all those exercise machines would start showing average folks with average bodies using them...

However, that seems most unlikely.

Perhaps audiophiles should develop a little common sense...
And realize that ads for audio gear are probably no more, and no less, accurate than ads for cars, exercise gear, and health supplements.

I should also point out that most of us don't NEED a wrist watch that's accurate to 30 seconds a month.
However, most of us own one, probably because we wouldn't save much by seeking out and purchasing a watch that was off by five minutes a day.

I don't place much blame on manufacturers. The root of the problem is the fallibility of our perception, and our lack of conscious awareness of that fallibility. We're sitting ducks. And most of the hype about products is generated by fellow audiophiles, not by manufacturers. Some manufacturers just get lucky (and their products may look good, have cool names, and use technology which sounds impressive to laypeople) and their products get "hot," which tends to be a self-reinforcing process which makes them ever more popular.

 

[SonyFan121]

1. We are not still in the late 90's/early 2000's though!
2. I wasn't, I was an adult, already an experienced professional and I could easily tell the difference between a 128kbps file and CD. Now, 20 years later and with much more efficient codecs and 320kbps or 256 VBR, I can't tell the difference and neither can any of my colleagues, either on cheap systems like yours or in the multi-million dollar studios where we work.
3. Hey, you're the one who said "But if using some seriously expensive kit, the difference can be immediately apparent.". Now that you've been called out on it, apparently have no idea what "seriously expensive kit" actually sounds like and only have cheap kit yourself, you're saying you DON'T need expensive kit. So which is it?
4. Just the fact that you call it a "hobby" demonstrates that you take it less seriously than me, because for me it's not just a hobby, it's how I make my living and it has been since before you were "just a kid"! So, you are OBVIOUSLY way less experienced than me, obviously take it way less seriously than me and clearly do NOT know what you're talking about! Again, it's the same old audiophile nonsense, you try the "I've got better kit than you" and when that's shown to be false, you change your tune and try the "I take it more seriously" and "I have more experience", both of which are also false. All you're doing is just making yourself look more foolish, surely that's not why you posted here is it?
5. If you "don't need someone with less experience to tell you what you can do", what makes you think that we do?

Your old audiophile tactics might work in other forums but not in this one, this is the sound science forum, NOT the "I've got cheap kit, less knowledge and experience but I'm going to state the opposite" forum! Why don't you do what the thread is titled and actually try a controlled test, instead of using false appeals to your (lack of) authority?

G

Listen pal, - I do not appreciate your extremely aggressive tone!! You seem very touchy! You sure you're okay?!
If you want me to debate with you then calm down!!
I do not care what you say, you can try to irritate me with your aggresive tone, but you will fail.
Do you know anything about Linn? they certainly do not make cheap systems, if they read your statement they would laugh. Honest they would.
I highly, highly doubt you are what you say you are, otherwise you would not be saying the things you do.
I am not going to reply to your extremely aggresive and child-like manner! you have a good day now you hear !?!

 

[gregorio]

Listen pal, - I do not appreciate your extremely aggressive tone!!

You are the one who came here with the aggressive and condescending tone. You are the one who brought up your supposedly expensive kit and your supposedly superior experience and now you're getting YOUR OWN tactics thrown back at you, you get all upset and call it a "child-like manner". The solution is simple, if you don't want to be treated in such a manner, do not treat me or others in that manner yourself. Surely that's not too difficult a concept for you to grasp? AGAIN, why don't you stay on the topic of this thread, actually try a controlled test for yourself and STOP with your aspersions and doing exactly what you're complaining I'm doing?!

G