• 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.

Does lossless really matter?

Atanasi

Senior Member
Forum Donor
Joined
Jan 8, 2019
Messages
400
Likes
380
Gapless has nothing to do with codecs or formats.
Codecs and formats do matter for gapless. Some formats have native support for gapless, so they can cut at sample boundaries, and these boundaries have minimal glitches. Other formats don't support this natively, so they need extensions to determine sample boundaries. Without these extensions, they can only cut at frame boundaries.
For example, MP3 doesn't support gapless natively, and this applies to some other MPEG codecs, too. MPEG is a movie consortium, so they are not very interested in gapless playback. Vorbis, on the other hand, has a good support.
 
OP
A

abdo123

Major Contributor
Forum Donor
Joined
Nov 15, 2020
Messages
3,753
Likes
3,282
Location
Brussels, Belgium
Codecs and formats do matter for gapless. Some formats have native support for gapless, so they can cut at sample boundaries, and these boundaries have minimal glitches. Other formats don't support this natively, so they need extensions to determine sample boundaries. Without these extensions, they can only cut at frame boundaries.
For example, MP3 doesn't support gapless natively, and this applies to some other MPEG codecs, too. MPEG is a movie consortium, so they are not very interested in gapless playback. Vorbis, on the other hand, has a good support.

Oh cool! thanks for sharing!
 
OP
A

abdo123

Major Contributor
Forum Donor
Joined
Nov 15, 2020
Messages
3,753
Likes
3,282
Location
Brussels, Belgium
FYI, class D amps are highly efficient and draw power only as needed.

So unless @watchnerd's casual listening habits involve square waves at 120dB, the subs are not going to be using more than a few (or maybe a few dozen) watts.

Am I that poor in expressing myself? :facepalm: I meant that i hope he had Class D amp but I didn't know they can reach such capacities (I know hypex has a 1200W unit but not 1300W or more).

Then he mentioned that he does use Class D so i'm interested to know what it is.
 

waynel

Addicted to Fun and Learning
Forum Donor
Joined
Nov 14, 2019
Messages
751
Likes
883
Could you explain how this would be? I'm not sure I follow..
My questions are:
1) Is there any guarantee that lossless compression artifacts will remain inaudible once DSP room correction (or EQ) is applied?
2) When using only a digital volume controlled DAC directly into an AMP will the lossless compressed noise floor always be inaudible?
 

andreasmaaan

Master Contributor
Forum Donor
Joined
Jun 19, 2018
Messages
6,652
Likes
9,119
Am I that poor in expressing myself? :facepalm: I meant that i hope he had Class D amp but I didn't know they can reach such capacities (I know hypex has a 1200W unit but not 1300W or more).

Then he mentioned that he does use Class D so i'm interested to know what it is.

Aaaaah, sorry, misunderstood.

There are lots of >1000W class D amps on the market though FYI. All the big class D amp manufacturers (Pascal, ICEPower, Hypex, etc) have models.
 

andreasmaaan

Master Contributor
Forum Donor
Joined
Jun 19, 2018
Messages
6,652
Likes
9,119
My questions are:
1) Is there any guarantee that lossless compression artifacts will remain inaudible once DSP room correction (or EQ) is applied?
2) When using only a digital volume controlled DAC directly into an AMP will the lossless compressed noise floor always be inaudible?

The noise floor and other artefacts of the lossily compressed audio will always track with the audio itself. So if you lower the volume of the system, you will lower the level of the noise and distortion along with the signal (i.e. no problem).

OTOH, I could imagine in theory that if EQ or time-domain processing is applied, the integrity of the psychoacoustical calculations on which the compression algorithm is based may become less valid. For example, if you boost a particular frequency range and reduce the level of another range, distortion in the boosted range may become more audible. It's not something I'd expect to be a real-world problem, though.
 

StevenEleven

Addicted to Fun and Learning
Forum Donor
Joined
Dec 1, 2018
Messages
566
Likes
1,123
I have Spotify (serves the whole family), Amazon, Qobuz, YouTube Music (comes with my YouTube Premier subscription), and Apple Music, because. . . Coronavirus. They keep me amused and support the Music industry. :p

I would say I prefer lossless for upmixing, just in case the compression affects the effects (see what I did there?) of the upmixing algorithm, but the truth is I don’t even worry about that anymore. Good lossy seems to provide a very pleasant upmix, whatever the differences might be.

I don’t worry about the sound anymore. I did ABX before ABX was cool. I know the CODECs and how they performed at various bitrates years ago, and they’ve only improved since then. Assuming the same master, at 160 kbps any difference in sound to me is probably very occasional and less than moving my head three inches in one direction or another, IMHO, and at 192 kbps and up things get even tighter. You’ve really got to try the ABX to get an idea how insignificant the changes may be for you. For me, it’s insignificant if it’s discernible at all.

What does distinguish the services is the road they lead you down for music discovery, the UI, the metadata they provide, the versatility for streaming with various gear, the responsiveness of the search, etc. For all of the above, for me, all of the above services are pretty heavy hitters and provide a very enjoyable and diverse set of experiences.

I don’t use Tidal because of what seems like a bit of a price premium and MQA, and what I perceive as some somewhat deceptive marketing at times, but I do think they put some pressure on the market for lossless streams and high quality masters, so I appreciate that. Credit where credit is due. Still not gonna plunk my money down for it.

Does lossless matter? For me, maybe for upmixing and quality assurance, but as a practical matter, nowadays, really only as compared to services like Sirius/XM, where the sound is obviously degraded (if things have not improved over there recently).
 

Blujackaal

Active Member
Joined
Feb 7, 2020
Messages
221
Likes
99

MP3 has pre echo issues where 4 albums i tested sound bad even at 320kbps. Yet when i tried 160k Vorbis it was transparent & some albums sounded iffy on V2 and V0 was only slight gains. I have zero idea how the FHG group green light MP3 when it got outperformed by MP2 let alone AAC/Vorbis.

I don't like the fact when MP3 fails or gives up it just spits out a 320k file and the LAME devs have put zero effort trying fit stuff like PNS, New perceptual model, much more advanced noise shaping etc without breaking standards.

--allshort mode actually fixes it pre echo issues by 95% but i can still tell eig.wav at 320kbps.
 
Last edited:

waynel

Addicted to Fun and Learning
Forum Donor
Joined
Nov 14, 2019
Messages
751
Likes
883
For the digital volume control I was more concerned about noise floor when the volume is all the way up during a quiet passage.
The noise floor and other artefacts of the lossily compressed audio will always track with the audio itself. So if you lower the volume of the system, you will lower the level of the noise and distortion along with the signal (i.e. no problem).

OTOH, I could imagine in theory that if EQ or time-domain processing is applied, the integrity of the psychoacoustical calculations on which the compression algorithm is based may become less valid. For example, if you boost a particular frequency range and reduce the level of another range, distortion in the boosted range may become more audible. It's not something I'd expect to be a real-world problem, though.
 
Last edited:

M00ndancer

Addicted to Fun and Learning
Forum Donor
Joined
Feb 4, 2019
Messages
700
Likes
687
Location
Sweden
would say I prefer lossless for upmixing
Same here, all music I make (no, it's only for personal listening, but I'm on soundcloud) is lossless, same goes for all my CDs.
Just to have a pristine copy for messing around with.

Export / encode to Ogg 320kbps 16/44.1 with dithering.

MP3 128-160 kbps and Opus 160 kbps that I can detect sometimes.
MP3 192? perhaps if it's classical piece or lots of cymbals. 256 and up? No.
AAC? perhaps 128Kbps 192-256kbps no not a chance.
Ogg? That's hard @ 160Kbps. Perhaps. 192 and up. No
 

andreasmaaan

Master Contributor
Forum Donor
Joined
Jun 19, 2018
Messages
6,652
Likes
9,119
For the digital volume control I was more concerned about noise floor when the volume is all the way up during a quest passage.

Ok, I see now.

But this is not a potential problem specific to use of lossily compressed files with digital volume control and/or DSP. The relationship between the noise floor of a compressed file and the signal is fixed at the point of compression. Whether you use digital or analogue volume control upon playback, the noise floor of the file relative to the signal is unaffected.
 

KeithPhantom

Addicted to Fun and Learning
Forum Donor
Joined
May 8, 2020
Messages
558
Likes
529
Me, too. I have been loving the no ads on any YouTube videos. And it looks like YouTube Music and regular YouTube will be one single app before too long.
Not only for that, but also background play without having to sideload something like YouTube Vanced.
 

waynel

Addicted to Fun and Learning
Forum Donor
Joined
Nov 14, 2019
Messages
751
Likes
883
Ok, I see now.

But this is not a potential problem specific to use of lossily compressed files with digital volume control and/or DSP. The relationship between the noise floor of a compressed file and the signal is fixed at the point of compression. Whether you use digital or analogue volume control upon playback, the noise floor of the file relative to the signal is unaffected.
that's a good point, so it's just a question of whether the noise floor is audible when the gain is up and music signal is low whether analog or digital volume control is used.
 

andreasmaaan

Master Contributor
Forum Donor
Joined
Jun 19, 2018
Messages
6,652
Likes
9,119
that's a good point, so it's just a question of whether the noise floor is audible when the gain is up and music signal is low whether analog or digital volume control is used.

Exactly :)
 

Katji

Major Contributor
Joined
Sep 26, 2017
Messages
1,530
Likes
973
Location
Mzansi
Yes because hi-fi. (Short for high fidelity.)

Case closed.

The rest is irrelevant, or beside the point.
 

617

Major Contributor
Forum Donor
Joined
Mar 18, 2019
Messages
1,489
Likes
3,138
Location
Somerville, MA
The thing which concerns me about different services is the master used, not the codec. I'm a big david bowie fan and I have flac files of all the japanese rykodisk gold releases (44.1/16) and they are waayyyy better than the more recent 'abbey road' remasters. Well maybe not waayyyy better but noticeably better.
 
Top Bottom