• WANTED: Happy members who like to discuss audio and other topics related to our interest. Desire to learn and share knowledge of science required as is 20 years of participation in forums (not all true). Come here to have fun, be ready to be teased and not take online life too seriously. We now measure and review equipment for free! Click here for details.

Is lossy outdated in 2019 & onwards?

BDWoody

Addicted to Fun and Learning
Patreon Donor
Joined
Jan 9, 2019
Messages
585
Likes
740
Same here I'm getting fed up with him nagging DBT stuff when even quite few science areas admit its flawed.
What science areas?

And, it's usually just blind testing being requested, not double blind.

This is a science site after all...don't get angry when people are asked to back up what are typically very strident and 'night and day differences' types of subjective posts, with no attempt to control anything.
 
Joined
Sep 1, 2019
Messages
12
Likes
1
What science areas?

And, it's usually just blind testing being requested, not double blind.

This is a science site after all...don't get angry when people are asked to back up what are typically very strident and 'night and day differences' types of subjective posts, with no attempt to control anything.
I'm not that angry just bit peeved when i only said i stuck with lossless. Since I'm too lazy to see if lossy can work with my FLAC collection. Two 256GB SD cards isn't that awkward since i kept it and my AW45 in my large ER4SR case.
 

BDWoody

Addicted to Fun and Learning
Patreon Donor
Joined
Jan 9, 2019
Messages
585
Likes
740
I'm not that angry just bit peeved when i only said i stuck with lossless. Since I'm too lazy to see if lossy can work with my FLAC collection. Two 256GB SD cards isn't that awkward since i kept it and my AW45 in my large ER4SR case.
Well, that isn't all you said. You basically tried to negate the value of blind testing.
 
Joined
Sep 2, 2019
Messages
108
Likes
89
Location
1 mile east of Sleater Kinney Rd
What science areas?

And, it's usually just blind testing being requested, not double blind.

This is a science site after all...don't get angry when people are asked to back up what are typically very strident and 'night and day differences' types of subjective posts, with no attempt to control anything.
Well, I wasn't talking about "night & day" differences, just enough to notice a loss of low-level resolution. I don't hear a difference between Flac and Apple Lossless, FWIW. I do appreciate that gear gets measured properly here.

I have a hard time listening to LPs because of the continuous loss of fidelity as the stylus approaches the deadwax. Michael Fremer seems to think that is "no big". But compared to lossy vs lossless, that is a gross distortion. I haven't spent time ABXing anything, don't have the means or the desire. My audio comparisons have been all "sighted", but having spent a decade as a recordist [recording engineer is too elevated a term for what I was doing] I can tell you that taking the headphones off and walking into the room where the music was playing was like walking into a different world. What I was recording was "Classical" music [weird term, properly speaking the term only applies to Haydn, Mozart, Beethoven and contemporaries], so usually just two microphones in ORTF configuration, and yes, the whole chain would be altering the sound. But the most sonically obvious difference was via swapping microphones. Microphones, like phono cartridges, like speakers, are transducers, and they are all obviously coloring and distorting the sound.

I would expect that ABXing lossy vs. lossless would usually result in "I dunno" as they would be very close, particularly as most recorded music is compressed and sonically compromised in the first place. ABXing "Florence & The Machine" would probably result in "no difference", an uncompressed recording of Mahler's Third Symphony would be more likely to reveal a difference.

In any case, the version I store on the computer is the version I store on the DAP as the cost of memory is so cheap right now.
 
Last edited:
Joined
Sep 1, 2019
Messages
12
Likes
1
Well, that isn't all you said. You basically tried to negate the value of blind testing.
Got angry there since that quote reminded why i stopped using HA forum's, Since no one there can handle casual/subjective views. Like above i wasn't making a massive claim just that i use flac since 85% is industrial, goth, ambient which i listen to, I've not got the attention span(Adhd) to test them to see if i can walk away with whatever lossy codec.

To be fair i get lossy on ios stuff and old stuff, But on android/DAP's? why bother when 512GB memory cards are £85 which he ignores.
 

Soniclife

Major Contributor
Patreon Donor
Joined
Apr 13, 2017
Messages
1,615
Likes
1,193
Location
UK
I would expect that ABXing lossy vs. lossless would usually result in "I dunno" as they would be very close, particularly as most recorded music is compressed and sonically compromised in the first place. ABXing "Florence & The Machine" would probably result in "no difference", an uncompressed recording of Mahler's Third Symphony would be more likely to reveal a difference.
Do the test, you only need foobar and a laptop, you will learn things. The bit about different types of music does not match my experience, complex messy modern stuff stresses encoders.
 
Joined
Sep 2, 2019
Messages
108
Likes
89
Location
1 mile east of Sleater Kinney Rd
Do the test, you only need foobar and a laptop, you will learn things. The bit about different types of music does not match my experience, complex messy modern stuff stresses encoders.
My point about Mahler's Third has to do with the extraordinary wide dynamic range of that music, with a lot of time spent at the bottom of the dynamic range. My laptop is too old and too untrustworthy for this sort of thing. In any case, I've already wasted too much time tweaking, don't wanna anymore.
 

Julf

Addicted to Fun and Learning
Joined
Mar 1, 2016
Messages
550
Likes
386
Location
Amsterdam, The Netherlands
Got angry there since that quote reminded why i stopped using HA forum's, Since no one there can handle casual/subjective views.
It is not that we can't handle them, it is just that they are useless and pointless distractions. OK, you have a subjective perception of X. So what? How is that of any value or interest to the rest of us?
 
Joined
Aug 24, 2019
Messages
35
Likes
10
For me I came here a bit ago to learn a different point of view. There is more science involved in digital reproduction of sound than I ever imagined.

Coming from the Analog world it seemed that science was a tool to achieve an end result which was pleasing to the ear. There was and is differences of opinion as to what sounds better-Neve, API, SSL, Harrison and many more. In digital science is king as the goal is that there is absolutely no difference in the end result.

From a pure logical thought process, any compromise with a goal of minimizing storage space of a file by altering the composition of the file is suspect. How to quantify if this compromise is audible and discernable may be futile.

As for blind listening tests I think that the implementation of them is flawed. In mixing we use comparison tracks to keep our ears on course. Trying to determine any difference in say a snare hit by playing the same hit in different formats back to back without a comparison in between to use as a baseline is pretty difficult for even the best trained ears.

Pretty simple. Store and playback the files in the format that has the least amount of compromises unless storage is an issue but realize a compromise has been made.
 

Soniclife

Major Contributor
Patreon Donor
Joined
Apr 13, 2017
Messages
1,615
Likes
1,193
Location
UK
My point about Mahler's Third has to do with the extraordinary wide dynamic range of that music, with a lot of time spent at the bottom of the dynamic range.
I'm not aware of anything in the popular codecs that should be challenged by dynamic range, it's about complexity in the short music frames.
 

Julf

Addicted to Fun and Learning
Joined
Mar 1, 2016
Messages
550
Likes
386
Location
Amsterdam, The Netherlands
Coming from the Analog world it seemed that science was a tool to achieve an end result which was pleasing to the ear. There was and is differences of opinion as to what sounds better-Neve, API, SSL, Harrison and many more. In digital science is king as the goal is that there is absolutely no difference in the end result.
Let me refine that a bit - the goal is no audible difference. Coming from an analog world, I am sure you are comfortable with that concept, as analog is never perfect, but often "good enough".

From a pure logical thought process, any compromise with a goal of minimizing storage space of a file by altering the composition of the file is suspect. How to quantify if this compromise is audible and discernable may be futile.
Not really. The audibility can most definitely be tested. Here is one good method: BS.1116: Methods for the subjective assessment of small impairments in audio systems.

As for blind listening tests I think that the implementation of them is flawed. In mixing we use comparison tracks to keep our ears on course. Trying to determine any difference in say a snare hit by playing the same hit in different formats back to back without a comparison in between to use as a baseline is pretty difficult for even the best trained ears.
That is exactly how the popular ABX and MUSHRA tests work.
 
Joined
Aug 24, 2019
Messages
35
Likes
10
We are back to audible or discernable or able to be heard. Be sure to strap your head into a device which restricts all movement when you triple or quadruple blind test to determine if it is audible.

And be sure to have at least 10 others strapped in with you.

Analog is "good enough"? What a snobbish thing to say. Please provide your triple blind tests to prove that statement.
 

Sal1950

Major Contributor
The Chicago Crusher
Joined
Mar 1, 2016
Messages
5,733
Likes
3,220
Location
Central Fl
Coming from the Analog world it seemed that science was a tool to achieve an end result which was pleasing to the ear. There was and is differences of opinion as to what sounds better-Neve, API, SSL, Harrison and many more. In digital science is king as the goal is that there is absolutely no difference in the end result.
Science has always been king for either. The goal has always been the same for analog vs digital, High Fidelity. Recording and amplifier technology that doesn't distort the source, something analog recording was never quite able to accoplish. Of course the sound of purposeful signal minipulation at the mixer is a whole different subject.
 

Blumlein 88

Major Contributor
Joined
Feb 23, 2016
Messages
6,495
Likes
6,224
My point about Mahler's Third has to do with the extraordinary wide dynamic range of that music, with a lot of time spent at the bottom of the dynamic range. My laptop is too old and too untrustworthy for this sort of thing. In any case, I've already wasted too much time tweaking, don't wanna anymore.
I share your preference for lossless. I prefer when possible to get the native formats. Some of the modern lossy codecs are impressively good and usually inaudible. But like you say, memory is cheap why bother. I might do FLAC or ALAC to get more music per gig of memory, but these days I'd rarely bother to convert anything to lossy. And yes I'd care not one whit even if I can't hear the difference.

I also don't see the big deal with ER4S's main thing. Memory is cheap, I'm not bothering to see if there is a difference. Just keep it lossless and don't worry about it.

Now for myself, I've tested a few formats blind on some of my own music. Just from curiosity. In one case I used to have a car radio that only played MP3 and WAV from CD_R. So my solution was highest VBR MP3 it would handle so I could get 5 hrs a disc. With something playing from memory sticks, I just buy a big one. Then just dump the lossless stuff on there as WAV or FLAC.
 
Last edited:
Joined
Aug 24, 2019
Messages
35
Likes
10
Science has always been king for either. The goal has always been the same for analog vs digital, High Fidelity. Recording and amplifier technology that doesn't distort the source, something analog recording was never quite able to accoplish. Of course the sound of purposeful signal minipulation at the mixer is a whole different subject.
The signal chain is quite similar up to the actual recording to tape which does add distortion or saturation to the signal. Mixing ITB or in the box in the digital realm has many advantages with the biggest being able to recall a mix at a specific point in time by simply bringing up a session rather than having to consult notes and photos of the individual settings of a ton of knobs and sliders.

The next biggest advantage is having the ability to closely replicate the sonic signature of thousands of dollars of classic analog gear with plugins that are 1/10th the cost or less. Just placing an eq or a compressor in the path affects the signal before any manipulation. The science of capturing the unique sonic signature of this classic analog gear has reached the point where the difference is minute- some say they can hear an audible difference, some say they cannot.

I can assure you that most, if not all, of the music available was either produced with analog gear or with plugins that replicate that analog gear.

Yes it is not perfect and yes analog adds saturation and distortion and yes it raises the noise floor but you will miss it if it is not there.

My last post here. I am grateful for the knowledge I have absorbed.
 

Julf

Addicted to Fun and Learning
Joined
Mar 1, 2016
Messages
550
Likes
386
Location
Amsterdam, The Netherlands
Analog is "good enough"? What a snobbish thing to say. Please provide your triple blind tests to prove that statement.
So are you claiming analog is perfect?
 

Julf

Addicted to Fun and Learning
Joined
Mar 1, 2016
Messages
550
Likes
386
Location
Amsterdam, The Netherlands
I think good enough and perfect are different things inherently. He claimed good enough.
No, I claimed good enough. He responded:

Analog is "good enough"? What a snobbish thing to say. Please provide your triple blind tests to prove that statement.
OK, he also came with all sorts of silliness against double blind tests...
 

solderdude

Major Contributor
Joined
Jul 21, 2018
Messages
2,448
Likes
3,903
Location
The Neverlands
I can assure you that most, if not all, of the music available was either produced with analog gear or with plugins that replicate that analog gear.
Indeed it is for the vast majority of pop/rock etc. studio recordings.
Don't know if many of the best Jazz and classical recordings are processed in a similar way.
AFAIK far less processing done there.

In studios most sound engineers have quite different (sometimes even opposing) views from electronics engineers.
Hi-Fi reproduction and what's done in a studio usually are totally different things.
In Hi-Fi the goal is to either reproduce as accurately as possible or deliberately add a 'personal flavor' where in a studio the goal is to record as good as possible with often purpose designed microphones that often aren't 'flat', modify that recorded signal so it sounds the way the engineers/producers/artist want it to be and to make a mix for the masses.
Different goal posts.

The gap between those worlds is quite large.
Fact remains, we can only try to reproduce a finished 'product'.
 
Last edited:
Top Bottom