[bigshot]

Regardless of the player, the tuning varies from one to another.

Well, I only know Apple, but going back to the original TV set iMacs, AAC 256 and MP3 LAME 320 both sound exactly like lossless. It doesn't matter if it's a computer, iPod or phone... transparent is transparent. And that's the way it should be. If the funky format is one you never use, I suppose it's fine, but if I bought a player that played one format worse than another, I'd probably return it. But that's just me. I test every piece of equipment I buy carefully to make sure it works properly so I can return it in case I get a bum one.

 

[StanD]

Well I totally understand the need for 24 or 32 bits, but only within a production environment.

Everything, and I mean everything is EQ'd and mixed on a computer these days, and that means that the sound (or signal) remains entirely within the digital domain. Having those extra bits for dynamic range allows for far more accurate EQ, summing, and effects. If you were only working with 16/44 then quantisation noise would gradually creep in with more and more operations on the source. With 32/384 this is simply not a problem - even 24/192 would suffice for the vast majority of productions.

When the project is finalised and bounced down to 16/44 it will sound identical to that which the producers have worked with. Nothing more is needed for the delivery to the end user.

People seem to jump onto the fact that studios use this technology and therefore believe that they must have it to experience the sound as it was originally. However they fail to grasp that that the reason that studios use this technology is to allow editing and adjustments with minimal impact to the sound quality. For simply playback it's totallly not necessary to be sitting on a 32/384 master, or even a 24/192 master - 16/44 is more than is required.

So with that in mind I agree with you, but 32 bit DACs are still needed.

Most of the technically oriented posters to this thread have said everything you just stated, except for the bit about 32 bit DACs. After doing all of the signal processing, etc. at 32 bits, one does not need a 32 bit DAC to produce listenable output, especially since a deliverable format of 16/44 does not require 32 bit DACs. Next most important item, what DAC actually delivers a DR or SNR that begins to approach 1 LSB of 32 bits? To get a feel for the scale, take 1 Volt and divide that by 2^32 and you get 0.00000000023 which is less than a nano-volt. If you want to fiddle with the MSB as a sign bit, it's still less than a nano-volt. Good luck finding a DAC that can deliver that kind of a signal that is not swamped by noise. What is the DR that you expect a 32 bit DAC to be able to deliver to be called 32 bit?

 

[sonitus mirus]

What is responsible then? Please tell me

The music player. The application that ultimately sends the PCM (in most cases) signal to your DAC to convert it to analog. The DAC does not take raw AAC, MP3, OGG, Opus, ALAC, or FLAC without first being decoded to a format the chip can read.

 

[castleofargh]

What is responsible then? Please tell me

https://www.head-fi.org/f/threads/tidal-lossless-streaming.733233/page-236#post-13539596

 

[Cold Train]

Well, I only know Apple, but going back to the original TV set iMacs, AAC 256 and MP3 LAME 320 both sound exactly like lossless. It doesn't matter if it's a computer, iPod or phone... transparent is transparent. And that's the way it should be. If the funky format is one you never use, I suppose it's fine, but if I bought a player that played one format worse than another, I'd probably return it. But that's just me. I test every piece of equipment I buy carefully to make sure it works properly so I can return it in case I get a bum one.

Agree that both AAC 256 and MP3 320 sounds like lossless on some devices, agree that originally they may be transparent but they are not transparent through any setup, and surely not from the output of an Apple device

 

[sonitus mirus]

Agree that both AAC 256 and MP3 320 sounds like lossless on some devices, agree that originally they may be transparent but they are not transparent through any setup, and surely not from the output of an Apple device

You stated earlier that a "proper ABX" was performed and that you converted a 24/192 WAV into multiple formats. Maybe start there. How did you convert the file. What tools and settings were used to make the conversion? How did you perform the ABX to verify your claims? How did you volume match the different versions? Seems like a few steps along the way could be responsible for any noticeable differences that might be heard. Without providing a bit more information, there is no way for any of us to know.

 

[bigshot]

Agree that both AAC 256 and MP3 320 sounds like lossless on some devices, agree that originally they may be transparent but they are not transparent through any setup, and surely not from the output of an Apple device

I'm not sure what you're saying here. "Sounding like lossless" is the same as "audibly transparent" and the audio output of Apple devices is generally as good or better than most dedicated audio components.

 

[castleofargh]

I know nothing of Apple gears so some of what I say might not apply:
some devices mess up a codec, in my experience it has too often been typically a problem with .OGG, but it's a non null possibility. with mp3 it happened to me only twice on DAPs with the firmware still in progress. I don't know if Apple is the type to sell gears while still beta testing the firmware.
also one encoder could do stuff like lower the loudness a little, and that in turn could be audible because it's clearly quieter, or result in removing the audible intersample clipping you get on the other file. that I've experienced several times while trying random encoders or crappy files from a buddy. the song needs to have first been mastered without a brain, but loudness war is still going strong.
and that's about it as to why I imagine one would clearly identify aac 256 or 320mp3.

oh maybe if they are at different sample rates and the device sucks at 44.1khz with default rate at 48khz? but I wouldn't consider that a proper test of formats anyway.


can I complain about how it's not on topic now that I've participated to the off topic? you know, "do what I say not what I do" kind of modo.

 

[gregorio]

Agree that both AAC 256 and MP3 320 sounds like lossless on some devices, agree that originally they may be transparent but they are not transparent through any setup, and surely not from the output of an Apple device

Actually, Apple devices are generally very transparent, with a virtually flat FR throughout the audible range. This raises two observations I've noticed over the years:

1. Audiophiles often like to cite the rule/law of diminishing returns. What's interesting is that not only does this rule not apply in many areas of audiophilia but often it's even actually inverted! IE. Increasing cost often gives either effectively the same performance or in some cases worse performance. For example, audiophile cables typically provide no performance gains, at any price, USB "de-crapifiers" typically do the opposite and audiophile DACs perform equally or less transparently than bog-standard Apple devices.

2. Audiophiles typically reference everything to their personal preferences and simply misappropriate terms and invent new ones, to get around this fact. If an audiophile prefers some bit of kit they tend to describe it as: "accurate", "neutral" or "transparent", regardless of the fact that it may actually be less accurate, neutral or transparent than a far cheaper/mass produced version. In my experience, present most audiophiles with a very accurate/neutral/transparent listening experience and they'll be shocked/surprised and often critical. A typical amalgam of comment would be "Very nice but too clinical, I prefer my system because it's much more musical." - The misappropriated term "musical" apparently means coloured/inaccurate and "clinical" apparently means accurate. I don't see how a system can be too accurate though, so I suspect the term "clinical" has been invented by audiophiles to counter/defend a personal preference for a certain level of inaccuracy. I'm not saying audiophiles cannot have a personal preference, just that they should call it what it is and not lie about it! I've got no problem for example, with someone preferring vinyl to digital, I do object to them asserting that it's more accurate or transparent though.

G

 

[bigshot]

If Shakespeare was alive today, he wouldn't be writing sonnets. He'd be writing audiophile equipment reviews!

"The quality of tweeters are not strained." "To PRAT or not to PRAT, that is the question."

 

[StanD]

If Shakespeare was alive today, he wouldn't be writing sonnets. He'd be writing audiophile equipment reviews!

"The quality of tweeters are not strained." "To PRAT or not to PRAT, that is the question."

Whether 'tis nobler in the mind to make up Schitt.
The slings and arrows of outrageous forum chatter.
Or to take arms against a sea of myths
And by opposing end them.

 

[Cold Train]

Whether 'tis nobler in the mind to make up Schitt.
The slings and arrows of outrageous forum chatter.
Or to take arms against a sea of myths
And by opposing end them.

A bove ante, ab asino retro, a stulto undique caveto

 

[Cold Train]

I'm not sure what you're saying here. "Sounding like lossless" is the same as "audibly transparent" and the audio output of Apple devices is generally as good or better than most dedicated audio components.

Lossless and transparent, the same? I can read that you like Apple so much....

 

[bigshot]

If lossless is the unaltered source and a compressed version is audibly transparent (adding or subtracting nothing you can hear) then they sound exactly the same. Are you talking about theoretical sound or sound you can actually hear?

 

[Cold Train]

If lossless is the unaltered source and a compressed version is audibly transparent (adding or subtracting nothing you can hear) then they sound exactly the same. Are you talking about theoretical sound or sound you can actually hear?

Only sound that a human can hear, theoretical graphs and numbers are for engineers and are somewhat useless in real world.

We shall only focus on what we can hear !