Skip to main content

Notice

Please note that most of the software linked on this forum is likely to be safe to use. If you are unsure, feel free to ask in the relevant topics, or send a private message to an administrator or moderator. To help curb the problems of false positives, or in the event that you do find actual malware, you can contribute through the article linked here.
Topic: CD quality v. 24bit (HD) observation (Read 5520 times) previous topic - next topic
0 Members and 1 Guest are viewing this topic.

CD quality v. 24bit (HD) observation

My personal preference for fine listening is high definition digital, a headphone amp and studio headcans. That said, I always downsample for having my music either with me with earbuds or for listening in the car. For downsampling, and because my portable player is an iPhone, I choose the best setting for QAAC which generally creates M4As for well engineered audio with a bitrate of 300-400kbps.

What should I believe if a CD or a HD flac downsamples to a bitrate of 200-250kbps? I don’t come across these often, but on occasion both the CD and HD flac for the same album will give me nearly the same low bitrate. Am I to believe the audio engineering wasn’t even good enough for 320kbps MP3s?

Cheers from the Avalon :)


Re: CD quality v. 24bit (HD) observation

Reply #1
What should I believe if a CD or a HD flac downsamples to a bitrate of 200-250kbps?

A 200 kbps flac file? That sounds too low. If you mean AAC that is normal. Not sure what you are asking.

Re: CD quality v. 24bit (HD) observation

Reply #2
A 200 kbps flac file? That sounds too low. If you mean AAC that is normal. Not sure what you are asking.
The result of converting the flac was 200kbps, and as per my observations, is NOT normal for the QAAC setting used (see post). I’m asking if the original digital recording was worth paying more for the CD (or HD) than for the inexpensive 320 MP3(?) To me, the implication is the original audio was poorly engineered(?)

Re: CD quality v. 24bit (HD) observation

Reply #3
Quote
What should I believe if a CD or a HD flac downsamples to a bitrate of 200-250kbps?
I'd say it's "easy to compress".  ;)  An easy-to-compress file doesn't necessarily have poor sound quality.   I notice you didn't say, "some bad-sounding files compress to low bitrates."

Quote
I don’t come across these often, but on occasion both the CD and HD flac for the same album will give me nearly the same low bitrate.
If the CD and "HD" file compress to the same bitrate I'd guess the "HD" file was upsampled from the CD (in a way where no data was added).  Increasing the sample rate, dithering, or making a very-slight volume change will make more data, resulting in a bigger (higher bitrate) FLAC.

Quote
Am I to believe the audio engineering wasn’t even good enough for 320kbps MP3s?
Again, low bitrates don't necessarily mean poor sound quality and you can't compare lossy & lossless compression.   The higher-bitrate MP3 might sound worse than the FLAC, or they might sound the same, etc.

Re: CD quality v. 24bit (HD) observation

Reply #4
A 200 kbps flac file? That sounds too low. If you mean AAC that is normal. Not sure what you are asking.
The result of converting the flac was 200kbps, and as per my observations, is NOT normal for the QAAC setting used (see post). I’m asking if the original digital recording was worth paying more for the CD (or HD) than for the inexpensive 320 MP3(?) To me, the implication is the original audio was poorly engineered(?)

Could be from a mono audio source, analog tap with limited high frequency bandwidth, have  a lot of silence etc. 

Re: CD quality v. 24bit (HD) observation

Reply #5
I'd say it's "easy to compress".  ;)  An easy-to-compress file doesn't necessarily have poor sound quality.   I notice you didn't say, "some bad-sounding files compress to low bitrates."

Thanks for your thoughts.

Today’s thoughts come only from comparing the bitrates for conversions without serious A-B listening tests... and I  don’t hear you saying “a low bit conversion result of a poorly engineered recording won’t necessarily sound worse than the original—it may sound the same” ...

I think what I may be referring to is something Neil Young was pushing a few years back, about Pono bringing us all back to the age of analog... and all along, I came to believe many of his recordings, by listening to his “HD” tracks, comparing his to others (e.g., Alison Krause), and coming to the conclusion that very few of Neil’s recordings were worthy of HD, and paying the higher price for them (although some were definitely worthy)...

I’m not arguing—just throwing out thought. And, I’m not saying HD is a marketing ploy—I’m simply wondering if some HDs are worthy of the extra cost, because some recordings definitely are...

Re: CD quality v. 24bit (HD) observation

Reply #6
Quote
without serious A-B listening tests...
HydrogenAudio promotes level-matched, blind, lABX tests.    Sighted tests can be misleading even with "serious" and "careful" listening.

Quote
And, I’m not saying HD is a marketing ploy—I’m simply wondering if some HDs are worthy of the extra cost, because some recordings definitely are...
The odds are, you can't hear a difference between an HD original and a copy downsampled to CD quality in a proper ABX test.    But sometimes the HD track is from a different or mix or master so it might be worth it.    Sometimes an MP3 can sound identical to an HD original depending on the program material and your ability to hear compression artifacts.    ABX tests can be humbling!  

Quote
Neil Young was pushing a few years back, about Pono bringing us all back to the age of analog...
Neil Young might have  great  hearing but the odds are against him...   He's been exposed to lots of loud music and he's at the age where people frequently experience some hearing loss.   

And, analog was not as good as digital.    The main reason studios converted to digital and old guys like me replaced all of their vinyl was for improved sound quality.    (DVDs & Blu-Rays are also better than VHS...  Audio as well as video... ;) )

Re: CD quality v. 24bit (HD) observation

Reply #7
[…]
And, analog was not as good as digital.    The main reason studios converted to digital and old guys like me replaced all of their vinyl was for improved sound quality.  ...

I’m not going anywhere near that argument because, to a large part it is subjective, and quite dependent on hardware and the money we spend on the hardware. I didn’t invest in my speakers, but I did invest in my headcans... and I’m happy with the result...

Re: CD quality v. 24bit (HD) observation

Reply #8
[…]
And, analog was not as good as digital.    The main reason studios converted to digital and old guys like me replaced all of their vinyl was for improved sound quality.  ...

I’m not going anywhere near that argument because, to a large part it is subjective, and quite dependent on hardware and the money we spend on the hardware. I didn’t invest in my speakers, but I did invest in my headcans... and I’m happy with the result...
Well the reason not to go near the digital vs. analog argument is not because it is subjective, but because it's been beaten to death and people who are tired of it already chose their side. But whether it's subjective or not, people who like to think analog is better may try to convince you the difference is subjective, cause if it's objective they have no argument. It is objective because it can be tested scientifically if people can tell if there's degradation in either digital or analog.

I suppose you can say it is subjective in the sense that one prefers it cause of the nostalgia factor or the "ritual" of playing vinyl or reel-to-reel tape which I have nothing against and you'll find most here don't either, but you mentioned it's dependent on hardware and money so I'm not sure that's what you meant.

Quote
I’m not arguing—just throwing out thought. And, I’m not saying HD is a marketing ploy—
I don't think you'll offend anyone here by saying this  ;)

Re: CD quality v. 24bit (HD) observation

Reply #9
[…]
I suppose you can say it is subjective in the sense that one prefers it cause of the nostalgia factor or the "ritual" of playing vinyl or reel-to-reel tape which I have nothing against and you'll find most here don't either, but you mentioned it's dependent on hardware and money so I'm not sure that's what you meant.
I think I agree with you, but I’m allowing it to be subjective while not knowing what listeners expect from their music. I spend quite a bit of my money on my music. Personally, I expect to get what I’m paying for because most of that money is in my library. I’m certainly not going anywhere near the cost of a turntable and what it costs for worthy vinyl, ... subjective or not...

Re: CD quality v. 24bit (HD) observation

Reply #10
I wouldn't doubt that most 24 bit is from analog looping.  :o

 

Re: CD quality v. 24bit (HD) observation

Reply #11
I wouldn't doubt that most 24 bit is from analog looping.  :o
I don’t know why. It might improve signal over noise, but I cannot imagine it improving voice/instrument inter-modulation(?)

Re: CD quality v. 24bit (HD) observation

Reply #12
> Increasing the sample rate, dithering, or making a very-slight volume change will make more data, resulting in a bigger (higher bitrate) FLAC

I'd rather say, it will lose some data (because it's not a reversible transform) by adding some noise. Then, the encoder has no realistic way to 'see' what this signal is 'really' based on, so it has to exactly encode this more complicated signal, even if it's worse than the original in any practical meaning and contains less useful information.

Also, there are some special cases: if the volume increase is an exact multiple by a power of 2 , then the bitrate increase will be insignificant, because FLAC detects 'wasted bits'. (it doesn't 'make more data' too, because in this case the volume change is trivially reversible with no error, as long as it didn't clip after volume increase)
a fan of AutoEq + Meier Crossfeed

Re: CD quality v. 24bit (HD) observation

Reply #13
Is it correct to assume that a downsampled conversion to 48kHz will be ~10% larger than the same music sampled at 44.1kHz?

Re: CD quality v. 24bit (HD) observation

Reply #14
If everything remains the same (2 channel, 16 bit) you will have 48000 samples instead of 44100 per second.
Indeed ~10% larger
TheWellTemperedComputer.com

Re: CD quality v. 24bit (HD) observation

Reply #15
Is it correct to assume that a downsampled conversion to 48kHz will be ~10% larger than the same music sampled at 44.1kHz?
Uncompressed you would have a ratio of 48/44.1, that is around ten percent up.
Longer story: When compressing with say FLAC, you could have two effects which pull in opposite directions. The "extra" content is treble from 22.05 to 24 kHz, and
* if there is nothing there, then the 48 file will be about same size; some recordings have hardly anything there;
* if what is there compresses worse than the file's average, the file will grow by more, and that treble signal is often just noise (which is not much compressible except it isn't noise at full volume).

Re: CD quality v. 24bit (HD) observation

Reply #16
I’m not saying HD is a marketing ploy—I’m simply wondering if some HDs are worthy of the extra cost, because some recordings definitely are...

I tend to be of the mindset like Markuza97's signature says, "Everything above 16/44 is scam." ; because it basically plays inline what what DVDdoug said that even downsampled music from "HD" to standard 16/44.1, it's very unlikely you will hear a difference simply because the sound quality at standard CD rate is already more than enough for our hearing limit basically and as a bonus it's easier to convert standard 16/44.1 to lossy formats as it will just work since it's at a standard we have had since the 1980's (although probably closer to around 30 years or so on a mainstream level now). side note... hell, I still got the first CD player I had when I was basically a kid (likely a bit prior to my teenage years) which is a Panasonic RX-DS620 which I was using in the early 1990's (probably 1991 or 1992 when we got it) which is still the best all-around CD/cassette tape player I got. a quick look online it appears one sold on Ebay for $120+36 ship in Feb 2021 as that person listed it as 1991, which sounds right from my memory of it.

p.s. but that whole standard audio CD vs vinyl... ill back standard audio CD's hands down as they are all-around better with convenience and more durable and you don't have any noise from the vinyl playing either and a lot of people can play audio CD's in their homes, unlike vinyl. so unless someone has some nostalgia thing for vinyl, standard audio CD's are the obviously better all-around choice. even "if" vinyl was proven better on some level, I doubt it would enough of a difference to matter given CD's are just easier-to-use and more than high enough sound quality. there is a reason people have been using audio CD's for a long time now (call it 30+ years) as they are the all-around best choice for high quality music playback and convenience and there is no real reason to upgrade from that technology unlike stuff in relation to video which has changed (and improved) quite a bit since the 1980's when audio CD's came out.
For music I suggest (using Foobar2000)... MP3 (LAME) @ V5 (130kbps). NOTE: using on AGPTEK-U3 as of Mar 18th 2021. I use 'fatsort' (on Linux) so MP3's are listed in proper order on AGPTEK-U3.

Re: CD quality v. 24bit (HD) observation

Reply #17
I tend to be of the mindset like Markuza97's signature says, "Everything above 16/44 is scam." ...
Then I hope we can agree to disagree. I tend to describe the difference between listening to a well engineered 24bit track with studio headphones via DAC/AMP as listening to the actual instrument versus listening to a well engineered recording of the instrument. The differences are difficult to pick out and describe, but the feeling of being there versus in my living room is real to me.

Your points regarding 16bit/44kHz are understood, but to my hearing I don’t believe the sampling rate (2x human hearing) is good enough to accurately sample more than 3 instruments that have a tendency to modulate each other.

#jussayin

Re: CD quality v. 24bit (HD) observation

Reply #18
Your points regarding 16bit/44kHz are understood, but to my hearing I don’t believe the sampling rate (2x human hearing) is good enough to accurately sample more than 3 instruments that have a tendency to modulate each other.
Choosing between the sampling theorem or a belief system i'd prefer the theorem.
It means that all modulations below at least 20kHz are perfectly captured at a 44.1kHz samplerate even from 21 instruments and more.
You may even add unnecessary Intermodulation with playing back inaudible captured HF content at higher sample rates.
Next stop: TOS 8
Is troll-adiposity coming from feederism?
With 24bit music you can listen to silence much louder!


Re: CD quality v. 24bit (HD) observation

Reply #20
@m.shaffer ; while I don't know all of the technical details (maybe Wombat does though and there is the whole TOS 8 thing to etc), but for the sake of argument even "if" what you said is somewhat true... I figure when listening to music in general the listener is probably primarily focused on the singers voice, and maybe the core of the basic song to on some level(like the instruments etc that are more obviously in your face), more than every little detail of instruments which seem to mostly take a backseat those things would be the gist of it when just sitting back and enjoying the music (unless there are people do don't do that(?), but I kind of assume what I said there is typically the norm for most people(?)).

but that stuff aside... if you can't tell by ABX test between standard lossless 16/44.1 and your "HD" track, then there is really no difference. it's more of a difference because you think it's better more than there actually being a difference, unless of course you can ABX it. because I would not even attempt to claim I can hear a difference between standard 16/44.1 and "HD" audio. because even with decent bit rate lossy audio most people will struggle to hear a difference. or another way I could put it... when the expert listening testers around here struggle to hear a difference between standard 16/44.1 and 192kbps (or so) lossy audio files, it's all that more unlikely one could hear a difference between lossless 16/44.1 and "HD" audio since the difference there would be even smaller, especially given human hearing limits (even if we assume someone with ears as good as humans can get). so even if there was by some very small chance someone could ABX standard lossless 16/44.1 vs "HD" audio, the difference would have to be very small since, at least from my observations, I think people who do listening tests around here with decent bit rate lossy files and can notice very subtle differences are pretty close to splitting hairs already (no offense to any of them as there contributions are appreciated by those around here I am sure). so when just sitting back and enjoying ones music, which even those types who do those tests around here admit, they would never(or at least rarely) notice the difference that they very slightly can when doing ABX testing (and ABX testing requires quite a bit of focus it appears after a certain bit rate for many to where it's unlikely they would notice it when just listening to random songs straight up to enjoy them), which makes it even less likely your going to notice any different between 16/44.1 and "HD" audio, especially when just sitting back and enjoying the music.

with that said... you can still see random posts online here and there over the years who buy into that kind of thinking that lossy audio is crap, or at least some who might claim 128kbps is crap(but these types might be still stuck in the old days when MP3 was say 128kbps CBR and not more like the modern standards it's had for quite sometime now and AAC/Opus etc fairs even better than MP3 which I assume you probably already know that), and that lossless is much better. but if they actually do a blind listening test, it's almost guaranteed they won't do as well as they think they will ;) (i.e. the differences between lossless and your typical decent bit rate lossy file are much less than they think they are).

p.s. another possible factor... most of the equipment (i.e. headphones/speakers etc) the common-ish person is likely to use (of which I would consider myself among the common person here) is probably not going to cost more than $200-300 tops in my estimations (I would imagine even if I am wrong here it surely can't be much beyond that price point). so even assuming there are cases where $500-1000+ sound equipment is better, I am more of the mindset that after a certain price point, the gains are no where near enough to justify spending hundreds of dollars more etc. or... even for those who don't mind dropping say $1k on sound equipment, even for these types there has to be a cut off point to where after a certain point it's largely wasted $, especially if your not some professional who does it for a living and needs every little advantage they can get. one last thing, to speak for myself... I got what's probably in the ball park of 'average' headphones (Sony MDR-NC7) which are nothing special and I won't be surprised if there are headphones noticeably better than what I have, but, without having really tested any fancy headphones, my hunch is there won't be a significant enough of a difference to justify spending say $100-300 as, off the top of my head, I would imagine a fair amount of people around here have headphones in the $150-300 range(?) and while it would be nice to hear those myself just to see if I can notice a clear improvement or not over what I currently use, I am definitely not going to drop $150-300 to find out and I have my doubts about there being enough of a difference on those to justify the premium price. hell, even my computer speakers (i.e. Klipsch Pro-Media), which are above average, can still be bought for $140 or so online right now and for that price range, I can't imagine finding something noticeably better to the point I would want to spend a lot more $ to get it as those speakers have a good all around sound to them and, while not what I would call cheap, they are still within a reasonable price for the common-ish person so that they might consider getting something like those unlike if you start spending say $300+ it's starting to become a decent investment, pretty much.
For music I suggest (using Foobar2000)... MP3 (LAME) @ V5 (130kbps). NOTE: using on AGPTEK-U3 as of Mar 18th 2021. I use 'fatsort' (on Linux) so MP3's are listed in proper order on AGPTEK-U3.

Re: CD quality v. 24bit (HD) observation

Reply #21
I believe you’re all pretty much correct regarding ABX’ing 95% of the music out there, and especially true for 95% of the home audio reproduction electronics and speakers available at reasonable costs... but that  has nothing to do with my original question, or why I asked it.... #jussayin

Re: CD quality v. 24bit (HD) observation

Reply #22
I believe you’re all pretty much correct regarding ABX’ing 95% of the music out there, and especially true for 95% of the home audio reproduction electronics and speakers available at reasonable costs...

So what hifi equipment do you have?

Re: CD quality v. 24bit (HD) observation

Reply #23
m.shaffer, I assume you have some 96/192 kHz files, which program are you using to play these files?
gold plated toslink fan

Re: CD quality v. 24bit (HD) observation

Reply #24
Regarding my audio hardware and software, you’re missing the point of my op. I simply posed an observation regarding some HD and CD quality tracks being encoded by codecs at settings that should’ve resulted in kbps rates as high as nearly 400 if well engineered, but which the result rate was lower than 250. I posed the question... were these audio tracks worthy of CD quality in the 1st place? The question was pretty much answered above by DVDdoug and saratoga.

I’ve done my share of ABX testing, and have determined under which circumstances I can truly appreciate 24bit over 16bit (see above), and which of my favorite artists are likely to produced music worthy of 24bit and being able to hear differences. I am not going to ABX every album I intend to purchase just for deciding if the extra expense might be worthwhile. Most of my listening via the hardware I use the most often (car audio, workout earbuds, family room theatre) isn’t going to result in the differences I’m hearing with headcans & DAC/AMP anyway. My op was about questioning the original quality based on results of codecs on CDs yielding bitrates lower than expected. The answer received... no simple rule of thumb because its complicated.