Is it just me that finds Flac or Wav downloads ruined by the sound of bluebottles in the background!!
orbis72 said:
:rofl: Only hear it on 24/192........so not that much stuff available to be ruined.
Craig M. said:
I think this is important. If a business is reliant on the higher quality files sounding better, then it's not too much of a conspiracy theory to perhaps guess that they maybe made artificially better (although I am definitely not claiming this to be the case).
I have a few Linn downloads. They all sound stunning. But if I resample to 320 kbps MP3, they still sound stunning (as do their streaming radio stations (320 kbps mp3). I tried the same thing with the HDtracks sampler last night...same result. Great quality, not in any audible way (to me) compromised by the compession.
Time for a syringe maybe?
But then I was reading up on the science of digital audio, and as has been mentioned, the 24 bits over 16 is irrelevant, as only a small portion of the available dynamic range of 16 bits is actually used, and the only difference with 44.1 and 96 Khz, is that the upper frequencies are extended to double ie over 40Khz. Now call me a deaf old fart, but the last time I checked, I couldn't hear anything above around 16 Khz.
Time for me to stop worrying about hi-res downloads I think.
hi i am not a techie but i had a sony minidisc player and i found theMajorFubar said:How did Minidisc work? Didn't that use some kind of lossy compression?
Reason I ask is I've still got one of those and actually I prefer its sound to MP3.
Not sure what that says about my ears as I'm sure someone's going to tell me that 320K MP3 is lightyears ahead of the Victorian technology using to compand the audio to/from MDs.
recordings were great,i then fed it through mf dac and really could not tell the difference
as long as it was digital out to the dac,, playing the minidisc players audio out it was thin and bright sounding to my ears.
snivilisationism said:
In case it's of interest, Linn's description of why their Studio Master sounds better, can be read here: http://www.linnrecords.com/linn-what-is-a-studio-master.aspx
CnoEvil said:
It's not a very good explanation. The fact is, CD is 16 bit. That gives a dynamic range of 96 db... Now there hasn't been a piece of music made that can utilize that, and if it did, in order to hear the quietist sounds, you would have to have the volume so high that your head would explode when the loudest sounds kicked in. So dynamic range increases are not needed.
The 96Khz (or more) simply adds frequencies that are inaudible. There is no "filling in" of data, the same algorithm is used to create the tracks...
http://en.wikipedia.org/wiki/Nyquist%E2%80%93Shannon_sampling_theorem
So the 2 advantages of 24/96 "for playback" are 1. Higher dynamic range - Totally unnessary and 2. Frequencies above 22 Khz - Dubious advantage...unless you believe that these ultra high frequencies can be "felt"...maybe they can, but not by me so far.
snivilisationism said:
I don't feel qualified to argue about the analysis of the technical side. What I can comment on is what I hear (imagined or otherwise)....which is that the higher the sample rate, and the greater the bit depth (for a given recording), the less information is lost (ambient and otherwise), leading to a more natural and realistic presentation.
I come from the "vinyl generation", and have been waiting a long time for the sound to get back to the quality that I enjoyed from my TT in the 80s and early 90s..... this is of course a personal viewpoint. IMO.With the advent of high-res music, this is now happening.
On AV, you only have to compare DD with True HD to hear what these hi-res formats can do. Anyway, I thorougherly respect and understand your stance on this, and am certainly not claiming to have the answers.
Cno
To add to that, the human ear can not detect volume differences of less than 0.5dB or even 1dB, according to some studies. Let's be generous and say that it's closer to 0.1dB. Now, even assuming a much too wide acoustic range from -20 to 140dB SPL (complete silence to well above the pain threshold), a resolution of 1600 steps (11 bits) is all that is required to represent the entire range of human hearing! Sadly, the pcm signal on CDs is linear, not logarithmic. So in order to achieve this with pcm, you must allow enough resolution to represent a 0.1dB increase (1% increase in amplitude) across the entire range. And for that, you need to discard values above 65000 (95dB) and below 100 (40dB). So it seems that the usable dynamic range of 16-bit pcm is actually closer to 55dB. For 24-bit signals, the usable range is closer to 104dB.snivilisationism said:
Regarding the sampling rate, I've read several studies that suggest that the human brain does respond to supersonic frequencies as high as 30kHz, although we are not consciously aware of it. Another argument is that the higher sampling rate causes most of the aliased harmonics to fold out of the audible spectrum, leading to a cleaner sound.
I'm not a sound engineer, so take the above with a grain of salt please. But I can see the rationale for going above and beyond 16/44.1 without needing to go into the "personal interpretation" minefield. But like others have said, the quality of the production matters much more than the format in which it comes. Steely Dan's Gaucho in 128kbps mp3 still sounds better than RHCP's Califonication in flac.
tremon said:
I totally agree, which is why I qualified my above comment with - "for a given recording".
Linn Radio sounds better than my poorly produced CDs.
steve_1979
Well-known member
tremon said:
Sorry for going off topic a bit. But is it possible to get good quality versions of the Red Hot Chili Peppers music? I have most of their albums but the CD's are to clipped and distorted to enjoy properly.
CnoEvil said:
The thing is, that is all to do with recording and mastering quality. A CD is already "potentialy" higher resolution and far better quality than any vinyl. The only reason a vinyl version of an album would sound better is the way it's been made. All things being equal and CD trounces vinyl. But as Steve says above, there are some shockingly unlistenable CDs out there.
It's strange nobody's mentioned that WAV Lossless uses between 940kbps - 1200kbps compared to MP3 320kbps
snivilisationism said:
The resolution of Vinyl is (in theory) infinate. Soundwaves, by their nature are Analogue, so this can only be turned into digital by taking a series of "snapshots", which approximate the soundwave with a series of steps. Apparantly, some sounds that have very quick transitions, such as a drum beat or a trumpet's tone, can be distorted because they change too quickly for the sample rate.
Vinyl, provided its kept in the Analogue domain through the mastering, needs no conversion and reproduces the waveform intact.....though this method also has inherent problems.
This is probably a rather simplistic overview, which I'm sure others who are more knowledgeable can elaborate on.
steve_1979
Well-known member
TALON1973 said:yes....theres this round thing called vinyl
Hmmm vinyl eh. Is that a new lossless codec? What bit rate is it?
CnoEvil said:
"in theory" is very valid. But then 24/96 files do not give you any more resolution. They use the same algorithm. There is no filling in, just extra Hz above what we can hear, and extra potential dynamic range (and of course vinyl has a far lower DR than CD anyway). It is quite possible we can feel it, I couldn't argue against that...I seem to remember a range of Pioneer CD players that tried to add such frequencies, but whether it is relevant...who knows.
All I know is that I "personally" cannot differentiate between a 24/96 and a downsampled version "from that file"...which leads me to believe it's the mastering, not the resolution, bit depth or whatever.
ie... When a record company go out of their way to supply hi res files (at a grossly inflated price), they make sure the recording is a good one...
As has been previously mentioned, Linn online radio sounds amazing. Much better than most CDs, and it's 320 kbps MP3...
It's not too difficult to justify a sample-rate higher than 44.1kHz. While only the young and privileged can hear upto CD's 'Nyquist limit' of 22kHz, most of us can probably hear up to 10kHz+, frequencies which are at most represented by just four distinct samples per second (there’s no such thing as .41 of a sample). That's not very many at all and is the reason why, as far as I can tell, when I have directly compared 44.1kHz to 96kHz, it's the high frequencies (8k+) which 96kHz seems to reproduce more cleanly.
MajorFubar said:
They don't use the extra to sample more per second at a given frequency though. The sounds you hear at 8khz(for example) in a 24/96 recording are exactly the same as those in a 16/44.1.
and where you got 4 samples per second is beyond me. It is, by its very name, 44.1 thousand samples per second.
while 44khz is above the nyquist frequency for human hearing a higher sample rate is certainly beneficial. The nyquist frequency is the minimum you need to produce those frequencies, that doesn't mean that everything below 22khz it is unaffected. You will get all sorts of spectral smearing which you'll need filters to get rid of, potentially removing frequencies/altering the sound quality. When producing a DACit should be easier to produce smoother curves.
If you want to prove this to yourself grab a piece of paper and a pencil and draw a sine wave and put a few dots down, rub out the sine wave and leave the dots then connect up the dots. Now do the same on a another piece of paper with double the number of dots, you will instantly have a more accurate curve!
Probably harder to provide a case for 24 bit over 16 bit, logic would dictate that it doesn't have quite the extra benefit. I don't know of any 16 bit @96khz music!
If you want to prove this to yourself grab a piece of paper and a pencil and draw a sine wave and put a few dots down, rub out the sine wave and leave the dots then connect up the dots. Now do the same on a another piece of paper with double the number of dots, you will instantly have a more accurate curve!
Probably harder to provide a case for 24 bit over 16 bit, logic would dictate that it doesn't have quite the extra benefit. I don't know of any 16 bit @96khz music!
A 96kHz digital file has over twice as many samples-per-second at any given audio-frequency than 44.1kHz no matter how you look at it. So no, the sounds are not exactly the same. They in fact become more proportionately disparate as the audio frequency increases.snivilisationism said:
You've totally misread what I typed.snivilisationism said:
MajorFubar said:
Possibly, what do you mean with 4 distinct samples per second? Unless for some reason you're dividing 44.1 Khz "sample rate", by 10 KHz "frequency"... That's a bit like dividing 8 by fish. If I sample a track made on say a sequencer, that is a flat tone of 12 KHz...it still has 44100 samples taken per second, not 3.
And, the Nyquist-Shannon theorum states simply that you need twice the sampling rate than the highest frequency to get "perfect reconstruction of the original wave". I understood that above that, it makes no difference to the end result as the algorithm will yield the same (perfect result). ie the sound at a particular frequency will be exactly the same. If it isn't, then the theorum is wrong, and someone needs to go back to the drawing board.
Either way, I can't tell the difference.
Not really, though I do confess I made a typing error due to being distracted (by my kids), and for that I apologise: I meant samples per cycle not samples per second, but the basis of what I said still stands: 4 samples per cycle of a 10kHz sound is not very much.snivilisationism said:
The theorem is correct that you need to at least double the sampling frequency, because audio waves have a 'plus' and 'minus' value which one sample per cycle wouldn't capture. But just two samples per cycle most certainly don't give a 'perfect reconstruction'.snivilisationism said:
The below screenshot shows two 1/1000th of a second snippets of a 10kHz sine wave. The top one is sampled at 44.1kHz and the bottom one at 96kHz. As you can see, the top wave is not very good at all while the bottom one more accurately represents a truer shape of the wave.
...and once dithered, both waves will be identical. There is plenty of information in the top one to perfectly reconstruct the original analogue signal.
And I get your drift now.
And I get your drift now.
Sure. But have you performed a frequency analysis of both signals? Connecting the dots with straight lines as you did will result in many high-frequency components that will be both inaudible and expensive to generate.MajorFubar said:
Can you pass both waveforms through a 20kHz low-pass filter, and show those results as well?
Similar threads
TRENDING THREADS
-
-
-
I test OLED TVs for a living – and trust me, different sizes perform differently- Started by What Hi-Fi?
- Replies: 3
-
-
-
-
Facebook
Twitter
Instagram