• 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.

What kind of sound differences do you hear in your system between FLAC and streaming?

Blake Klondike

Active Member
Patreon Donor
Joined
Jan 20, 2019
Messages
134
Likes
79
#1
Was listening to the highest streaming-rate Amazon Music this evening to test out new speakers. A/B'd the same tune (Cinema Show by Genesis) in FLAC through Foobar and the resolution/imaging was greatly improved, but the highs were noticeably harsher. My ears started to burn out after only a couple minutes. It seems to me that the compression in the Amazon algorithm was rolling off the highs, which were audible via the FLAC? The difference was shocking!

Has anyone else had similar experience? May be a separate topic, but have you folks struggled with painful high-end when putting systems together?
 

Blake Klondike

Active Member
Patreon Donor
Joined
Jan 20, 2019
Messages
134
Likes
79
#3
Thanks for the response-- is that high-end filtering done for a reason or is it just a side-effect of the decoding process?
 

BillG

Addicted to Fun and Learning
Joined
Sep 12, 2018
Messages
834
Likes
754
Location
Auckland, New Zealand
#4
Thanks for the response-- is that high-end filtering done for a reason or is it just a side-effect of the decoding process?
It's done for a reason: LAME, a widely used freeware encoder, filters at 18 - 20kHz to reduce compression artifacts.


have you folks struggled with painful high-end when putting systems together?
No, but I understand that can be an issue with certain types of tweeters Iike horns and ribbons. Mine are a composite of ceramic and metal, and while they're bright, they're certainly not harsh... :cool:
 

digitalfrost

Active Member
Patreon Donor
Joined
Jul 22, 2018
Messages
114
Likes
128
Location
Palatinate, Germany
#8
Correctly encoded lossy files are generally transparent at 128kbps since 2004/2005:

2004: http://listening-tests.freetzi.com/html/Multiformat_128kbps_public_listening_test_results.htm
2005: http://listening-tests.freetzi.com/mf-128-1/results.htm
2008: http://listening-tests.freetzi.com/mp3-128-1/results.htm

If you hear a difference, you either have golden ears, hearing damage that doesn't agree with the encoder, or somebody made a mistake. I spend money on audio equipment, but I probably couldn't ABX a 128kbps file from FLAC to save my life....

That said. When encoding is ******, it's generally metallic, mushy sounding hihats that give it away for me. While most encoders cut off content above 16khz, this is usually not noticeable with real music. I tried to ABX this when I could still hear up to 18.5khz and was unable to.

Thanks-- is there a program I can use to test FLACs and streaming audio?
For files only: http://losslessaudiochecker.com/#downloads

Thanks for the response-- is that high-end filtering done for a reason or is it just a side-effect of the decoding process?
In MP3, it is done because the last scalefactor band's quality can only be adjusted by changing the global scale factor. This will waste a lot of space. See: https://wiki.hydrogenaud.io/index.php?title=LAME_Y_switch
 
Last edited:

andreasmaaan

Major Contributor
Joined
Jun 19, 2018
Messages
3,517
Likes
2,759
#10
No, but I understand that can be an issue with certain types of tweeters Iike horns and ribbons.
It’s not a problem that’s specific to any tweeter type per se. Any tweeter of any material (within reason ofc) can be made to sound overly bright if the implementation is poor, and any decently-designed tweeter can be made to sound neutral.

Think about the question from this angle: What causes “brightness”?

Considering all plausible answers to that question, it becomes clear that tweeters of a wide variety of types and materials needn’t be inherently bright.

Basically, it is a question of implementation rather than materials.
 
Joined
May 8, 2019
Messages
54
Likes
55
#11
I directly compare Pandora which streams at 192kps on the highest setting with Tidal which of course is in 1411kps FLAC on the Hifi setting. I can pick out better high's that are sharper and defined, a tightened and defined low end and the midrange becomes a bit more broad and boosts the imaging, the speakers sort of disappear where as with Pandora the music sort of becomes more boxed-in. I use Pandora like the old FM stations only better without commercials and Tidal when I want to listen to something specific. Like both services and have tried to drop Pandora more than once but just keep letting it ride.
 
Joined
Jun 18, 2019
Messages
56
Likes
64
#12
I directly compare Pandora which streams at 192kps on the highest setting with Tidal which of course is in 1411kps FLAC on the Hifi setting. I can pick out better high's that are sharper and defined, a tightened and defined low end and the midrange becomes a bit more broad and boosts the imaging, the speakers sort of disappear where as with Pandora the music sort of becomes more boxed-in. I use Pandora like the old FM stations only better without commercials and Tidal when I want to listen to something specific. Like both services and have tried to drop Pandora more than once but just keep letting it ride.
Just curious - by "directly compare" do you mean blind A/B testing?
 
Joined
May 8, 2019
Messages
54
Likes
55
#13
Just curious - by "directly compare" do you mean blind A/B testing?[/QUOTE

Not really possible to do this blind as I am doing the switching/selecting between the services and there is no quarantee that I am actually using tracks from identical sources. Tidal gives me a definite edge in sound qualty while Pandora's music genome program is really quite uncanny in finding new music.
 
Last edited:
Joined
Jun 18, 2019
Messages
56
Likes
64
#14
I know for myself, I can always hear how fantastic lossless files are - until I try and pick them blind. lol...I don't trust what my ears tell me anymore unless I I hold it up to harsh scrutiny.
 

Soniclife

Major Contributor
Patreon Donor
Joined
Apr 13, 2017
Messages
1,643
Likes
1,213
Location
UK
#15
Was listening to the highest streaming-rate Amazon Music this evening to test out new speakers. A/B'd the same tune (Cinema Show by Genesis) in FLAC through Foobar and the resolution/imaging was greatly improved, but the highs were noticeably harsher. My ears started to burn out after only a couple minutes. It seems to me that the compression in the Amazon algorithm was rolling off the highs, which were audible via the FLAC? The difference was shocking!
I've no experience of Amazon files, or reason to believe they have messed things up, but comparing streaming to FLAC is complicated. I would suggest you start by turning your problem FLAC file into high bit rate MP3 and listen to that, and see what you think. Start by listening to them knowing which is which, as you are now, if you think the MP3 sounds like the Amazon MP3 then compare your 2 files using an ABX tool (e.g. foobar).
 
Joined
May 8, 2019
Messages
54
Likes
55
#16
Luckily my ears are still pretty good but definitely seeing a reduction in my high frequency and my right ear is a bit better than left. If I did the test blind it would be a wash I am sure. I go back to the perceived soundstage that I get from Tidal vs Pandora especially through my speakers, the image is no doubt broader. Unfortunately, I don't think we can find a way to measure imaging.
 

GrimSurfer

Senior Member
Joined
May 25, 2019
Messages
459
Likes
357
#18
I can hear an audible difference between 128Kbps mp3s and uncompressed CDs at 14400kBps. The difference would be informally described as a lack of dynamic range, which I don't believe is true. So. I would describe 128kBps mp3s as sounding as if they had a restricted frequency range, a hint of smearing at high frequencies, and a trace of treble harshness.

Things at 320 kBps would require younger ears, though I would not doubt if a properly trained young woman could hear a difference (young people having greater frequency range hearing, women having slightly better high frequency acuity).

For my money, it's lossless CD rips all the way. CDs are cheap, they last a very long time (I have ones that are over 30 years old that play without errors), they don't require a lot of space to store, and they're something that is owned (not rented).
 
Joined
Jun 18, 2019
Messages
56
Likes
64
#19
i doubt if any but the most highly trained ears could reliably identify lossless vs 320k mp3. Actually, I doubt even highly trained ears can.
 

Sal1950

Major Contributor
The Chicago Crusher
Joined
Mar 1, 2016
Messages
5,755
Likes
3,232
Location
Central Fl
#20
Things at 320 kBps would require younger ears, though I would not doubt if a properly trained young woman could hear a difference (young people having greater frequency range hearing, women having slightly better high frequency acuity).

For my money, it's lossless CD rips all the way. CDs are cheap, they last a very long time (I have ones that are over 30 years old that play without errors), they don't require a lot of space to store, and they're something that is owned (not rented).
Agreed. I recently got a free month of Qobuz and paid for a second at the highest tier. After that trial period I couldn't hear any difference of note between the high rez Qobuz stream and my Spotify 320kbps.
Spotify wins on all counts for me, it's user interface is great and the catalog is the largest around.
For the music I love the most I do CD rips to the hd so I know they will always be there. I have a lifetime of aquired music in CD rips and needle drop recordings going all the way back to the 60s.
I'm happy.
 
Top Bottom