[Steve999]

Well this thread got me curious! I am surprised that AAC seems to have kept pace so well with more modern codecs, based on the results of the hydrogen listening tests linked above?

I also looked around a little and experimented a little this morning. So Airplay 2 mirroring can actually transmit lossless by audio and video via WiFi it seems? And Apple AAC seems to handily outperform Bluetooth at a given bitrate based on the hydrogen listening tests? And my Onkyo receiver seems to handle Airplay much better than it used to with a series of firmware updates over the last six years? Streaming from iPad to receiver is now very convenient and much more stable. And mirroring from my iPad to my Apple TV is apparently lossless?

These things were all new to me as of this morning when I looked in to stuff a little.

I’ve got all my analog stuff backed up to digital over various codecs over the years and in the end I settled down to Apple AAC, usually Vbr usually around 256 kbps. During my days when I actually ABX’ed codecs (15 to 20 years ago) I settled in on 192 kbps lame mp3, and even that was more than necessary for me personally.

But in the end, I just stream nowadays, but with all my stuff that’s out of print (not much anymore) backed up to Apple’s and Google’s music services so I can stream it anywhere.

 

[bigshot]

No difference at all; that wasn't my point. He said "there no reason to use MP3 these days" I simply provided him a viable one to consider. Blanket statements need to be challenged sometimes. Some people have older collections with older file format types.

Well, it would be a good idea to use AAC from here on out. You can probably encode a notch lower with the same results.

 

[Davesrose]

Well this thread got me curious! I am surprised that AAC seems to have kept pace so well with more modern codecs, based on the results of the hydrogen listening tests linked above?

AAC is also staying a modern format: it has various new standards (as it has been included with MPEG standards(. So with music delivery, Apple is a main backer of it. With video, it's pretty standard as an alternative to Dolby codecs (as Dolby requires a license and AAC is open source). The format itself offers high quality surround (48 96kHz channels plus 16 LFE channels). The only main thing it can't do is have the proprietary 3D audio metadata of 3D audio formats like Dolby Atmos or DTS:X. Any social media site you're watching videos off of are a new ACC codec: they're being used because they also utilize adaptive bitrates.

 

[Spork67]

I convinced myself I could hear a difference between 320MP3 and FLAC files if I concentrated rather than just listened to the music.
I then went through my digital music library, artist by artist, album by album, and split them into a HQ folder and an MP3 folder, and made 2 separate playlists based on codec.
Then I forked out ~ $18 for a "HQ" download of an album I thought was worth having in FLAC format. (PIL - This is what you want...)
Then I added it to my HQ playlist and enjoyed it. It really did sound better.
2nd last track, fiddling around on Foobar - and I see I'm listening to my original 320MP3 copy of album.

In my defense, I was listening for enjoyment rather than concentrating, but I'm probably not going to rush to buy CD's or HQ downloads of all the albums I like but "only" have in MP3 format.
Certainly on a cheap blutooth speaker I take camping there is no difference.
On my mid-fi headphone setup there is probably a very slight difference.
It will be interesting to see if that difference is any greater with the better gear I have coming.
I sort of hope it is to validate my ears and my purchase - but I also hope it isn't, as replacing all those MP3,s will get costly.

 

[sander99]

Then I forked out ~ $18 for a "HQ" download of an album I thought was worth having in FLAC format. (PIL - This is what you want...)
Then I added it to my HQ playlist and enjoyed it. It really did sound better.

(As explained in this forum many times
If it really sounded better (which is not a sure thing based on an uncontrolled listening test) then it probably is because a different master was used.
The only way to be sure that you are comparing the same master is by converting the FLAC to mp3 yourself (and ideally back to FLAC which you then compare to the original FLAC, this to exclude the possibility of a false positive due to something in the playback system treating one format "unfairly" different from another format).
Having said that, a better master can be a good reason to buy the more expensive format. Unfortunately it is not a given that a better master is always used for the more expensive format: it can also be the same or worse.

 

[Ryokan]

I bought an album I already had in mp3 in flac (format snob) and it sounded worse, couldn't believe it until I realised it was a different master. I sometimes load two versions by mistake on a Dap so the same track plays twice and try as I might I can't hear a difference.

 

[bigshot]

This is purely anecdotal, but it's based on my collection of tens of thousands of discs... When it comes to mastering of popular music and jazz, I find that half the time, there is basically no real difference between hires (in my case SACD) and plain vanilla CD. When there is a difference, half the time it's better, and half the time it's worse. A 25% success rate is crummy odds, so I've stopped buying SACDs until I read a review from someone who has done a comparison and says the SACD is better. For classical music, hires seems to make no difference at all, but some companies only release music on SACD, so I'm stuck. For convenience, I prefer CD as a format. I can do whatever I want with it and I have a nice hard copy to archive.

 

[71 dB]

This is purely anecdotal, but it's based on my collection of tens of thousands of discs... When it comes to mastering of popular music and jazz, I find that half the time, there is basically no real difference between hires (in my case SACD) and plain vanilla CD. When there is a difference, half the time it's better, and half the time it's worse. A 25% success rate is crummy odds, so I've stopped buying SACDs until I read a review from someone who has done a comparison and says the SACD is better. For classical music, hires seems to make no difference at all, but some companies only release music on SACD, so I'm stuck. For convenience, I prefer CD as a format. I can do whatever I want with it and I have a nice hard copy to archive.

It makes sense to only release a hybrid disc with cd layer, 2.0 SACD layer and multichannel SACD layer. I like those. If I put them in my CD-player, it plays as a normal CD. If I put it in my Blu-ray-player supporting SACD it plays a multichannel SACD.

 

[Blackwoof]

LAME MP3 at V0 can pretty much do 24-bit/44.1KHz but forced -Y would stop the 16KHz ~ 22KHz area steal bits from under 16KHz area.

 

[WoodyLuvr]

LAME MP3 at V0 can pretty much do 24-bit/44.1KHz but forced -Y would stop the 16KHz ~ 22KHz area steal bits from under 16KHz area.

Not trying to be argumentative, but can someone else pipe in and confirm this? Genuinely curious if this is indeed correct or not. A little bit more detail would be much appreciated. Respects.

 

[castleofargh]

MP3 doesn't have bit depth, so it means nothing. As a lossy format, we can find differences at around -60dB even under the best circumstances. And regardless of all that, if the MP3 file is extracted as a 16bit PCM, then it's a 16bit file that plays in the DAC no matter how optimistically you try to interpret it.

Trying to go in his direction a little(why am I doing this?), well dithered 16bit won't be inferior to 24bit for a listener, and as lossy is all about perception, we could at least consider that the all thing doesn't matter and that MP3 will do the job. Dynamic range isn't really a concern for MP3, even if quiet signals will tend to be scrapped by the psychoacoustic algorithm, when there is nothing else on the track, MP3 can keep quiet sounds.
If we consider that 16 and 24bit sound the same, and that V0 MP3 usually sounds like the 16bit lossless version, then we can go crazy and decide to extrapolate about it sounding like 24bit.

 

[gerelmx1986]

In AAC VBR 224 my entire music collection slimmed to 345GB (from 1.54TB in hi-res , 17-bit FLAC and DSD) tge sound is okay for those low bitrate situations. No song goes below 192kbps but some went higher than 124 reChing maximum 320

 

[Steve999]

Not trying to be argumentative, but can someone else pipe in and confirm this? Genuinely curious if this is indeed correct or not. A little bit more detail would be much appreciated. Respects.

I think your instincts may be good, as it may, at a minimum, be an oversimplification, but I think conceptually, as a generalization, it may be more correct than incorrect. But I would be happy to be corrected if I am wrong. The technical side of this stuff is truly over my head.

Here is the documentation:

https://svn.code.sf.net/p/lame/svn/trunk/lame/USAGE

Discussion of the -Y switch is right at the end.

 

[bigshot]

At higher data rates, using VBR it’s possible for AAC to actually go beyond 320 if it’s necessary. I can’t see how it would be. But it’s possible.

High data rate lossy, 16/44.1 and 24/96 can all achieve transparency, so for all intents and purposes, they would sound the same.

 

[Blackwoof]

----- deleted