• WANTED: Happy members who like to discuss audio and other topics related to our interest. Desire to learn and share knowledge of science required. There are many reviews of audio hardware and expert members to help answer your questions. Click here to have your audio equipment measured for free!

High Resolution Audio?

daftcombo

Major Contributor
Forum Donor
Joined
Feb 5, 2019
Messages
3,687
Likes
4,068
Well, you definitely have golden ears. I'm unable to distinguish between a well encoded 128kbps mp3 VS WAV/FLAC.
 

maty

Major Contributor
Joined
Dec 12, 2017
Messages
4,596
Likes
3,160
Location
Tarragona (Spain)
That is what have accused me in the past.

Many find it hard to believe that they have a problem with their equipment, especially those who have spent a lot of money on them. The limiting factor can be many things. When I removed one, another one appeared that the first one hid until I finally had the current sound. It has taken me a lot of time and effort to get it!

In other systems, much more expensive, I was unable. I returned to my house and did the test quickly and had no problems.
 

watchnerd

Grand Contributor
Joined
Dec 8, 2016
Messages
12,449
Likes
10,408
Location
Seattle Area, USA
Why is it when I viewed this on youtube the next suggested video is about the Battle of the Coral sea? And I've neither been looking at military topics lately or anything related to navies. I expected to see where the USS HiRes was part of the action.


That's probably my fault for watching videos on high rez audio and the Battle of Midway (from the same video maker) in the same day.
 

JJB70

Major Contributor
Forum Donor
Joined
Aug 17, 2018
Messages
2,905
Likes
6,148
Location
Singapore
I used to work on warship designs, admittedly the bits I worked on were the simple and less complex bits (engines and shafting and associated bits like gear boxes and propellers) but despite the things being many orders of magnitude more complex than a DAC or amplifier I can't say I recall anything like the hidden mysteries that we are assured are the key to designing decent audio gear.
 

maty

Major Contributor
Joined
Dec 12, 2017
Messages
4,596
Likes
3,160
Location
Tarragona (Spain)
[ Si bien, para oídos no entrenados, la diferencia entre un MP3 codificado a 192kbps versus un formato Lossless escuchados en un sistema casero no es notoria, si los escuchamos y comparamos en un sistema de alta fidelidad, en un MP3 se notará la ausencia de agudos y los graves y medios tendrán un ruido, por lo tanto la coloración del sonido será diferente…

La representación fidedigna empleada comúnmente como referencia es la de un CD, codificado a una frecuencia de muestreo de 44,1 kHz y 16bits en formato tipo WAV PCM, donde si comparamos sus espectros de frecuencia con los de un formato lossless son iguales.

Por lo mismo, si un archivo tipo FLAC luego creamos una copia codificada en MP3, le invertimos la fase y los mezclamos, podemos escuchar los rastros que deja la codificación en MP3, donde todas las frecuencias altas superiores a 15,5khz fueron cortadas y, aun más notorio, las frecuencias criticas que están dentro del rango de frecuencias medias… ]

To English with Google translator (I am lazy):

Although, for untrained ears, the difference between an MP3 encoded at 192kbps versus a Lossless format listened to in a home system is not known, if we listen to them and compare them in a high fidelity system, an MP3 will show the absence of treble and the bass and middle will have a noise, therefore the coloration of the sound will be different ...

The reliable representation commonly used as a reference is that of a CD, coded at a sampling frequency of 44.1 kHz and 16 bits in WAV PCM format, where if we compare its frequency spectra with those of a lossless format they are the same.

For the same reason, if a FLAC type file then we create a copy encoded in MP3, we invert the phase and we mix them, we can hear the traces left by the encoding in MP3, where all the high frequencies higher than 15,5khz were cut and, even more notorious, the critical frequencies that are within the range of average frequencies ...
 

daftcombo

Major Contributor
Forum Donor
Joined
Feb 5, 2019
Messages
3,687
Likes
4,068
Many find it hard to believe that they have a problem with their equipment, especially those who have spent a lot of money on them. The limiting factor can be many things. When I removed one, another one appeared that the first one hid until I finally had the current sound. It has taken me a lot of time and effort to get it!

What is your system again? Sound incredible.
 

mansr

Major Contributor
Joined
Oct 5, 2018
Messages
4,685
Likes
10,700
Location
Hampshire
Year: 2012

MP3 vs AAC vs FLAC vs CD Page 2

https://www.stereophile.com/content/mp3-vs-aac-vs-flac-vs-cd-page-2

308MP3fig03.jpg


Fig.3 Spectrum of 500Hz-spaced multitone signal at –10dBFS, 16-bit linear PCM encoding (linear frequency scale, 10dB/vertical div.).


308MP3fig07.jpg


Fig.7 Spectrum of 500Hz-spaced multitone signal at –10dBFS, MP3 encoding at 320kbps (linear frequency scale, 10dB/vertical div.).




308MP3fig09.jpg


Fig.9 Spectrum of multitone signal with frequency gaps at –10dBFS, 16-bit linear PCM encoding (linear frequency scale, 10dB/vertical div.).

308MP3fig10.jpg


Fig.10 Spectrum of multitone signal with frequency gaps at –10dBFS, MP3 encoding at 320kbps (linear frequency scale, 10dB/vertical div.).


With decent recordings it is very easy to differentiate between MP3 320kbps (LAME) and FLAC 16/44. If it is not the case, it is that you have a bottleneck in your system that prevents it.
That is a pointless test. MP3 was designed to encode music or speech with little perceptible distortion. A steady-state multi-tone signal is actually fairly easy to encode, as these graphs show. The somewhat higher noise floor is perceptually masked by the much stronger tones. In the second sample, the noise level drops in the gaps, exactly as it should. The assertion in the last paragraph is in no way supported by the figures, nor by anything else.
 

maty

Major Contributor
Joined
Dec 12, 2017
Messages
4,596
Likes
3,160
Location
Tarragona (Spain)
This is an interesting observation. MP3 compression destroys the phase relationships between fundamental tones and harmonics. When this processing is applied to music, percussive sounds tend to become bursts of noise instead of identifiable actions. I find that this is the most disturbing artifact produced by MP3 compression. You seem to be describing a somewhat similar effect. Accurate phase response is very important when reproducing transients.


A second hand AV Marantz SR4500 tweaked and modded. Same with KEF Q100 5.25" coaxial speakers. And the other components too. Few DIY devices to clean the electrical grid (DC and RF/EMI). And...

https://www.soundandvision.com/content/marantz-sr4500-av-receiver-ht-labs-measures

205marantz.3.jpg
 

daftcombo

Major Contributor
Forum Donor
Joined
Feb 5, 2019
Messages
3,687
Likes
4,068
A second hand AV Marantz SR4500 tweaked and modded. Same with KEF Q100 5.25" coaxial speakers. And the other components too. Few DIY devices to clean the electrical grid, RF/EMI. And...

https://www.soundandvision.com/content/marantz-sr4500-av-receiver-ht-labs-measures

205marantz.3.jpg
Do you mean that with the Marantz, Kef and a few 'tweeks', you can hear subtilties in WAV that make you distinguish them from mp3, whereas you can't in other homes with expensive systems?
I don't understand how it is possible, but why not.
 

maty

Major Contributor
Joined
Dec 12, 2017
Messages
4,596
Likes
3,160
Location
Tarragona (Spain)
In 2011 I have not the KEF and the Marantz and...

My main system sounded great but ended up being kidnapped by television, family and loud music After a few years I went back to listen to music but it turns out that it sounded much worse than I remembered! I speak of excellent recordings. So I decided to buy a second-hand amplifier to be able to experience improvements that I had read out there.

By the way, modern AV sounds much worse than before, I mean for music.
 

Calexico

Senior Member
Joined
May 21, 2019
Messages
358
Likes
72
In 2011 I have not the KEF and the Marantz and...

My main system sounded great but ended up being kidnapped by television, family and loud music After a few years I went back to listen to music but it turns out that it sounded much worse than I remembered! I speak of excellent recordings. So I decided to buy a second-hand amplifier to be able to experience improvements that I had read out there.

By the way, modern AV sounds much worse than before, I mean for music.
Hello
Do you hear differences between dac their filter and their output stages ? And which kind do you like?
 

JJB70

Major Contributor
Forum Donor
Joined
Aug 17, 2018
Messages
2,905
Likes
6,148
Location
Singapore
I think there are two questions:

-is there a discernible difference in sound quality between hih res files and MP3, RBCD etc; and

-in cases where a difference can be discerned, does it affect enjoyment of music?

In the case of the first, I think it varies according to listener and file. I can discern differences sometimes, but other times not. I suspect that whether I can or can't depends on my familiarity with a recording in that if it is a track I listen to a lot and which I know very well then I can zoom in on those sections where the difference may be most readily identifiable. However, I wouldn't call the difference obvious, and if I was to do a proper statistical analysis I'm pretty sure my results would be consistent with what you'd expect for guessing.

For the second question, I have never found any difference in enjoyment between listening to MP3 or FLAC or CD or those high-res files I have. The differences are subtle and not obvious, I always think that if you have to really push yourself and risk overheating the brain from the concentration required to identify differences in an ABX test then that tells it's own story, that the differences aren't worth worrying about. If it was that obvious we'd get 100% without having to think much about it when doing double blind tests. This is the same for hardware, I think it's true that people can discern the difference between some amplifiers, DACs, CD players etc but the effort required to discern such differences would indicate that for normal listening purposes then any discernible differences are irrelevant.

Some things are indeed obvious. I think poor recording quality can be extremely obvious. Differences between speakers and how they are set up in a room can be very obvious, as can headphones (try going from a very analytical model such as the Etymotic ER4SR to a consumer oriented model with deep V tuning and back again) but I really don't think MP3 is obvious. Out of curiosity, I've just been playing some MP3 files on my DAP in place of the FLAC files I normally use. OK, it's not a particularly expensive DAP (Shanling M3s) but it is a nice one in my opinion, using particularly revealing IEMs (Etymotic ER4SR), if anything will highlight subtle differences it is this particular IEM model in my view. I have to say I thoroughly enjoyed the music and felt absolutely no disappointment or sense of missing anything because of the file format. However, that clearly wasn't a double blind test and could have been the result of expectation bias just as much as listening to high res and finding that it sounds much better is influenced by expectation bias.
 

Willem

Major Contributor
Joined
Jan 8, 2019
Messages
3,654
Likes
5,276
I think we must distinguish between lossless compression such as FLAC and lossy compression such as MP3. Lossless compression is indeed lossless and indistinguishable because on decompression the file is the same as the original. Lossy compression does degenerate the sound quality, although by how much depends on the compression rate. There is quite a bit of research that shows that for most people 320 kbs is indistuinguishable from lossless. Below that, sound quality gets progressively worse, of course. Lower bitrates than 320 kbps are most common in internet radio, and even at these somewhat bitrates internet radio can still be sonically superior to much FM radio that has its own limitations.
The sonic benefits from high resolution are a different matter again. This has nothing to do with compression, but with more dynamic range and higher frequency reproduction than is possible at 16/44.1. The debate is out on whether this will make a sonic improvement. The downside is obvious: larger files requiring more bandwith required for streaming services, and hence also higher costs and more energy consumption. For Hi Res files lossless compression is obviously advantageous, but lossy compression of Hi Res material does not seem to make much sense to me.
 

shanecoughlan

Member
Forum Donor
Joined
May 13, 2019
Messages
72
Likes
85
Location
Japan
Riffing on what @Willem posted while I was typing away...

I think we might need to hit reset on where this is going. There is a hard limit on human perception. Despite many subjective opinions, there is little to gain in going beyond 16 bits 44.1kHz. Dogs and cats, of course, have a different perspective, though it would not apply to recorded music. Check this out to see how stuff above 30-40kHz is white noise.

The numbers show high bitrate MP3 and AAC (320kbs and 256kbs respectively) will be fine for your music. Let’s pull a baseline from an industry that needs to get this right: broadcasting. Here is a guide showing the bitrates you need with various formats to get transparent sound (perceptually CD quality from a CD original).

Anyway, back to humans. Of course there may be a few outliers who perceive above 20kHz, but it won’t be by much, and it will drop off dramatically with age. Which is fine. There is no particular harm in collecting music at bit depths and sample rates similar to studio master recordings, but let’s not assert this impacts playback positively without detailed and comprehensive ABX testing being cited.

Note: I have a bunch of 24 bit 92kHz stuff lying around. I could get it so I got it. For fun. After all, this is our hobby :)
 
Last edited:

Daverz

Major Contributor
Joined
Mar 17, 2019
Messages
1,294
Likes
1,451
Well, you definitely have golden ears. I'm unable to distinguish between a well encoded 128kbps mp3 VS WAV/FLAC.

A lot of streaming radio uses 128 kbps, and it sounds terrible to my ears. I guess they could be poorly encoded.
 

BillG

Major Contributor
Joined
Sep 12, 2018
Messages
1,699
Likes
2,266
Location
Auckland, New Zealand
A lot of streaming radio uses 128 kbps, and it sounds terrible to my ears. I guess they could be poorly encoded.

Yeah, I can tell the difference with 128 kbps MP3 streams as well. However, 128 kbps AAC/M4A sounds pretty good to me... :cool:
 

maty

Major Contributor
Joined
Dec 12, 2017
Messages
4,596
Likes
3,160
Location
Tarragona (Spain)
I barely hear MP3. The few lossy files are OGG -q9 that I have recoded from FLAC for the phone. I have saved this EQ (JRiver MC) for a long time:

JRMC-DSP-Equalizer-MP3.png


And SoX 96 in Output Format, I think.
 

daftcombo

Major Contributor
Forum Donor
Joined
Feb 5, 2019
Messages
3,687
Likes
4,068
I barely hear MP3. The few lossy files are OGG -q9 that I have recoded from FLAC for the phone. I have saved this EQ (JRiver MC) for a long time:

JRMC-DSP-Equalizer-MP3.png


And SoX 96 in Output Format, I think.
Interesting. So do you consider that all mp3 have a bass boost and lack treble? Will hear again to check.
 

maty

Major Contributor
Joined
Dec 12, 2017
Messages
4,596
Likes
3,160
Location
Tarragona (Spain)
Usually, MP3 are from modern commercial music, badly recorded and with DR < 9 dB and less!!! And the bass are very enhanced, so to the current taste.

MP3s have a low pass filter at about 15 kHz.

The phase is modified when equalizing. Some EQ do it better than others.

If we retouch up to 180 - 200 Hz that phase change will not appreciate it, as happens when equalizing the subwoofers.

Instead of enhancing more the high frequencies I prefer to lower the low frequencies.

By the way, with classical music, the difference is where the difference is most noticeable.

I also notice the difference between FLAC 16/44 and OGG Vorbis -q9. With -q10 it is very difficult but it takes up too much space for the phone. -q9 is an excellent compromise.

[Spanish] https://nauscopio.wordpress.com/tag/ogg/
 
Top Bottom