The audio quality with streaming can be very good (i have seen full on studio masters in 24-192 with clean spectrals, twice out of like 100 albums tested).
-
Welcome to ASR. 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!
Shouldn't we upgrade the 20-20 audible range ?!
- Thread starter lashto
- Start date
-
- Tags
- hearing range
I have not ever disputed that. I just pointed out that there is zero or minimal correlation between market success and audio quality in those charts. I don't really bother with lossless on Spotify, and probably most of the tracks in my Spotify playlists I own as FLAC files in my library. I can listen to either and be perfectly happy with sound quality - what truly matters is the quality of the recording, and really not the format. Anyone that claims Motown songs fom the 60s are far more detailed if played as 24/192 is deluded.
Last edited:
It's been following these exercises in mental masturbation for something like half a century now.
Once we got to RedBook and beyond, I've never seen a single properly performed listening test able to repeatably show identifiable differences.
If any are real but so tiny they can only be shown using 50 different special source properties, who really gives a rodents behind?
With the real easily audible differences between speakers, rooms, 2ch vs multich, more; I gave up playing these useless listening tests long ago.
Back in the 1980s Sony/Philps told us the CD was bringing us "Perfect Sound Forever". Well, maybe not quite "perfect or forever" but over and over RedBook
has shown itself to be more than "close enough" with and exceptions being such an outlier as to be almost completely irrelevant.
Wouldn't it be nice if all the time and effort that's been put into looking for audible differences in Hi-Rez had been put into speaker development, DRC-Treatments,
microphone design, more.
AMEN my friend.
My biggest issue with streaming is you never really know the provenance of the recording you're listening to. Or if the one you listen to today will be the same that you hear tomorrow? Hopefully someday they will start at least including metadata that will give you some idea of where that file came from. Guess I'm just too old-school but for the music I love I want to know I have the best sounding version available. My current drives contain something like 4500 albums of the music I love most and to which I add constantly, I use my Apple Music sub so little sometimes I wonder why I continue to pay the $15 a month but it does come in handy to sample new releases before purchase, specially all the multich stuff they offer that's not available most anyplace else.
Last edited:
I don't know about that. Most of the time they're comparing different masters without being aware of it, or they're simply mistaken.
Bear in mind there's deliberate attempts made to deceive them like the DSOTM SACD with the hobbled CD layer.
Plus you have the audio journalists claiming it's obvious to them. Some of them no doubt are just lying, but they have credibility to many.
Same original track in 320k, 16/44, 24/192 - same track implies identical master, otherwise the comparison is between different tracks and hence a waste of time to establish if the formats are distinguishable.
Deleted
I'm fairly certain that 320k mp3 may be audible under perfect playback conditions but from lossless Redbook on, I'd fully agree.
"The results showed no significant difference among the sound stimuli, but that the correct response rate for three sound stimuli was close to the significance probability (5% level). Furthermore, it showed that one subject attained to a 75% correct response rate, which indicated a significant difference. In order to confirm the reliability of this result, a strict statistical supplementary test with this subject also was conducted. This subject evaluated 20 times over six kinds of sound stimulus. As a result, no significant difference was found among the six sound stimuli. Therefore, it is concluded that this subject could not discriminate between these sound stimuli with and without very high frequency components."
Yes, we know that, but never over-estimate the knowledge of the typical enthusiast.
Same original track, and preferably encoded/converted by yourself so you can be sure, and then level-matched (some lossy encoders/decoders change gain).
I think we've already established that we don't give a hoot. LOL
Fair enough.
When i test sampling rates i use audacity to get the difference between the versions and ensure there is no difference below 9/10 of the nyquist frequency.
For 24-96 vs 24-54
Great!
RexrothPigeon
Major Contributor
- Joined
- Mar 21, 2025
- Messages
- 1,543
- Likes
- 3,207
I said the format is obsolescent, so I'm not sure how you could draw any conclusion about sound quality.
That's a bit ambiguous, at least it was for me. "Format" could refer to the CD physical medium (your def) or 16-bit linear PCM + 44.1 kHz sampling rate + 2-channel stereo (mine). The former has certainly fallen out of favor, the latter has not, that's what I meant.
I still buy CDs from my fav artists when they come out, and rip them and make a high quality scan of the "album notes" - clearly I am a minority. I guess the music industry and the artists don't really care much about album notes these days, since neither Amazon downloads nor Spotify include them. So if they don't want me to learn more about them, why would I bother...
ZolaIII
Master Contributor
- Joined
- Jul 28, 2019
- Messages
- 5,710
- Likes
- 3,417
First of all basic of any lossy format is bandwidth limiting in top of the spectrum from 12~16 KHz, some cut entirely what's above 20 KHz as Opus. Some change encoding as Mp3 and so on. In materials where there is very little such containt you won't be able to hear the difference and vice versa. So some things and instruments (like saxophone, pipe organs and so on) are more prone to what's might be lost. It's not that much about relative bit rate but how they are designed. You can hear 17 to 20000 Hz 19 bit (if you are luckily) and there won't be many instruments in the first place that will have much to say (beyond second harmonic) after let's say 12.5 KHz. Redwood 16 bit + dithering comes close to 19 bit and you don't nead more. With some lossy formats it's posible to control shaving and bit rate reduction and in relatively lower bit rate (650~700 KB/s) is posible to pack 24 bit master reduced to 19 bits and top intact. It is same for redwood but meaningful reduction in size for such is small compared to standard compression loseles format as flac. In case of some such lossy formats it's posible to save part that's bean shaved off in hybrid mode and glue it back again to get the intact master back 1/1. I did comparation cuple years back hire so feel free to find it if you wish aiming that then in full swing MQA streaming would go to some sense full direction and reduction in bandwidth for keeping costs down even if it's only about 25% to standard flac and keep the costumers satisfied, so far I didn't get results.
Sorry but that's not accurate. MP3 at 320k CBR goes all the way to 21kHZ, as does ~250k VBR. It's only at 128k that you get the 16kHz upper limit. Red book takes you to 22.05. If you think you heard the differences because of that cut off, you were listening to the wrong thing. It's sharp transients that may give away the difference.
If you refer to hi-low-pass filters used in the encoding it is another topic altogether, but it's completely untrue it just chops off higher frequencies.
Many of my FLACs for acoustic jazz or classical solo are less than 500kbps, so 320 doesn't have to do very heavy lifting to compress.
And OggV at 320 is darn near impossible to tell apart from RedBook, I admit I had much higher failure rate than with MP3 when comparing with lossless. Very easy to do simply switching the stream from lossless to 320... and I don't care about the microscopic differences...
Last edited:
ZolaIII
Master Contributor
- Joined
- Jul 28, 2019
- Messages
- 5,710
- Likes
- 3,417
Mp3 resasemples, it's graph analyser sean and heard when shown (otherwise you probably won't). I whose less interested in decades old codec to hybrid losy Opus and my starting assumption it will fair much better where wrong. Used Power DB to minimise resasampling artifacts and did transcoding locked to single thread. For polifonic up to 32 bit aprouch and when both codecs are used along with smart bandwidth limit it's better on average 128~160 kbps (including all variants of Farens Mpeg and other encoders and same about AAC). You nead to bare in mind shaping and cutting entirely off are different things. You where lazy too search (so am I). So I went evoluting WavePack in it's hibrid form for the task. Representatively I used 96 KHz 24 bit real master and with specific noise flour and range (including wiled saxophone) & on purpose. Again more than 19 bit we don't need and 16 bit with dithering is fine. Have a nice time.
I assume English isn't your first language, but did you use some translation software?
Anyway, it is not really accurate to state that MP3 (or other perceptual codecs) "resample". They re-encode, into a rather different representation from which the sound waveform is recreated.
Similar threads
