Head-Fi.org › Forums › Equipment Forums › Sound Science › 24bit vs 16bit, the myth exploded!
New Posts  All Forums:Forum Nav:

24bit vs 16bit, the myth exploded! - Page 30

post #436 of 836
8/21/09 at 12:58pm
Quote:
Originally Posted by D. Lundberg View Post
Dither does not reproduce waveforms at all, it is used to decorrelate quantization errors from a signal.
Yes, I should have been more clear there.

Quote:
Originally Posted by D. Lundberg View Post
"[...] if the quantisation is performed using the right dither, then the only consequence of the digitisation is effectively the addition of a white,
uncorrelated, benign, random noise floor. The level of the noise depends on the number of the bits in the channel – and that is that!"
-- J. Robert Stuart, Coding High Quality Digital Audio, Meridian Audio Ltd.
Robert makes a pretty good case for higher than 16/44.1 bit-depth and sample rate in that paper.

Jim
Reply
post #437 of 836
8/21/09 at 2:34pm
I understand that more bits doesn't mean higher quality. It's obvious because DSD is only a 1bit signal yet it's much higher fidelity than 16bit CD.

My question to this is how does a 1bit DSD recorded stream compare then to these 16bit or 24bit streams.

Direct Stream Digital - Wikipedia, the free encyclopedia
Reply
post #438 of 836
8/28/09 at 10:52am
Hello Everyone,

For anyone who is interested, the second in our series of 24-bit comparisions has gone out through Twitter. This time we even have a 192kHz Studio Master!

This one is from the Beethoven Piano Concertos by the Scottish Chamber Orchestra, with Sir Charles Mackerras and Artur Pizarro.

Download the 16-bit/44.1kHz, 24-bit/96kHz and the 24-bit/192kHz for free and compare.

Please bear in mind that these are the files exactly as we sell them for download on the store - and for the 16-bit, exactly as it is sold on hybrid SACD. Hence the reason we are calling it a real-world test.

When reporting back in, let us know your listening setup. But most importantly, ENJOY!

Jim - Linn Records
Reply
post #439 of 836
8/28/09 at 3:41pm
Thanks, Jim for a superb piano concerto.

It was so peaceful that I almost fell asleep while trying to AB between the two files.

My system do not support 192khz, so my AB testing was between 16/44.1 vs. 24/96khz.

My setup is listed on my signature, so no need repeating here.

I have done repeated tests hearing from one passage from start to finish, and several places where I thought I hear the differences in multiple times.

Without going out in full detail, here is what I got to stay.

When I AB between two files after several minutes in between, I could tell apart very much. That's being my honest impression. However, if I listen right after the other, then I notice more lively, less fatigue sound from 24/96. I have no clue if that's just a placebo or ego here.

I was leaning toward OP, but after hearing this X number of times, I can't explain, but I favor the 24/96 by a small margin. The base seems to be slightly fuller, the piano is less harsh, and overall less fatigue in 24/96. Other than that, I can't notice a thing.

If done blind test, I don't know how many can actually distinguish one from the other. They are both exceptional quality. If I must though, I must say that I favor the 24/96 a small margin.
Reply
post #440 of 836
8/29/09 at 2:41am
alot of that might have to do with an affinity for your dac to be geared toward one than the other. if it upsamples and such.
Reply
post #441 of 836
8/29/09 at 2:46am
That's a comparison of different masters, rather than a comparison of just sample and bit rates, of course, and a great marketing tool. Give me the original masters, and I could do the sample rate conversion in reverse and convince people that 44/16 was better than 96/24, I'd wager...
Reply
post #442 of 836
8/29/09 at 10:10am
I haven't read the whole thread but it seems to me that the only reason to go beyond 44,1 kHz sampling rate is if frequencies higher than 22 kHz have some effect on sound quality.

I've read that some instruments like the violin can have harmonics as high as 40 kHz or more and they affect the timbre of the sound. So basically a violin in a 192 kHz recording would sound slightly more realistic due to these subtle timbre and spatial cues.

While humans can not directly hear frequencies over 20 kHz, they can still be perceived through skin and bones, so there might be some truth to that.

There's a good video about sound quality and why it matters with big names from the industry in a round table. Find "Deep Listening: Why Audio Quality Matters" on this list and press "watch video." It lasts for two hours but is very interesting.
Reply
post #443 of 836
8/31/09 at 11:17am
Gregorio,
Thank you for taking the time to post throughout this thread. It was quite an interesting read and I didn't know as much about digital audio as I thought I did.
Reply
post #444 of 836
8/31/09 at 11:24am
excellent read. i will not spend so much on audio from now on.
Reply
post #445 of 836
8/31/09 at 9:10pm
Quote:
Originally Posted by xnor View Post
Seriously, on the 28th page people still don't understand what the OP said.. no wonder that he left ...
This is the actual reason.
Reply
post #446 of 836
9/3/09 at 8:11pm
"So, can you actually hear any benefits of the larger (48dB) dynamic range offered by 24bit? Unfortunately, no you can't."

Why then, with a lot of listening, can I distinguish my own 24 bit master tracks from the dithered 16 bit versions in a blind test? Perhaps I’ve just learned to pick out the noise shaping. Or perhaps it’s something more. Subjectively, the dithered tracks never sound quite as good as the originals.

I’m not disputing the theory you’ve articulated, as I wouldn’t dispute the Nyquist frequency. Unfortunately, in practice, these things turn out to be more complicated than any of us would like. In the case of Nyquist, roll off below the cut off frequency leads many to sample at extremely high rates. I have no explanation for why 16 bit files don’t sound as good as the 24 bit originals, but in practice, they don’t. Mind you, the differences are extremely subtle.
Reply
post #447 of 836
9/3/09 at 8:51pm
Quote:
Originally Posted by nnotis View Post
"Why then, with a lot of listening, can I distinguish my own 24 bit master tracks from the dithered 16 bit versions in a blind test? .
Can you post a couple of short samples of 16 and 24 bit verions ?
Reply
post #448 of 836
9/6/09 at 9:34pm
I've never heard 24-bit postulated over 16-bit other than in audiophile circles. But, that is beside the fact. We will never, ever have a proper debate here at head-fi about anything which is remotely fact/fiction, the platform simply doesn't exist. In fact, internet isn't a good place to debate because people are too emotional as they suck down coffee (or in my case, alcohol) and argue.

One of the problems here is that many people came in to prove something which I will leave out of fact, or fiction. Anything can be proved, whether it be 16 bit or 24 bit is superior; whether analogue or digital; whether bose or skullcandy - everything can be proved by debate, and depending on who has the weaker argument, or the quieter voice.

If Gregorio got flack, I would imagine it has nothing to do with reality, just the same as his or anyone's arguments are based on heat and temper; above all a desire to prove someone else wrong and oneself right; to appear more knowledgeable is paramount.

Nice read though
Reply
post #449 of 836
9/7/09 at 4:11pm
Quote:
Originally Posted by shigzeo View Post
I've never heard 24-bit postulated over 16-bit other than in audiophile circles. But, that is beside the fact. We will never, ever have a proper debate here at head-fi about anything which is remotely fact/fiction, the platform simply doesn't exist. In fact, internet isn't a good place to debate because people are too emotional as they suck down coffee (or in my case, alcohol) and argue.

One of the problems here is that many people came in to prove something which I will leave out of fact, or fiction. Anything can be proved, whether it be 16 bit or 24 bit is superior; whether analogue or digital; whether bose or skullcandy - everything can be proved by debate, and depending on who has the weaker argument, or the quieter voice.

If Gregorio got flack, I would imagine it has nothing to do with reality, just the same as his or anyone's arguments are based on heat and temper; above all a desire to prove someone else wrong and oneself right; to appear more knowledgeable is paramount.

Nice read though
Hi,
Just postulating here... There are some here that freely admit we do not know everything there is to know in ANY given field of endeavor. I will go so far as to say that while I have compiled a body of knowledge, and even some wisdom in selected fields of study, I am fundamentally a student at heart and forever will be. I can and do learn from virtually everyone I interact with. I am sure that there are many others who frequent these forums who feel the same way.

An intellectually honest discussion places some responsibilities on its participants. Some participants accept those responsibilities, some do not.
We can even learn from those who are too immature or intellectually dishonest to accept those responsibilities.

BTW, the term debate assumes opposing sides of an argument, whereas discussion assumes more of a position of neutrality in the quest of greater knowledge.
Reply
post #450 of 836
9/7/09 at 7:50pm
Quote:
Originally Posted by b0dhi View Post
This is the actual reason.
And you even supported that, you should be ashamed of yourself.

Btw, does this forum even have an ignore feature?


On topic: There's no reason to debate/discuss this anymore.

The only difference you hear is the difference between different masters. This has been shown, I've tested it (and provide the files free for download if you want to try it yourself) and we got confirmation that the files are mastered differently.
Nothing more to add.


So who has lost the debate, in several aspects?
Not gregorio's side, that's for sure.
Reply
New Posts  All Forums:Forum Nav:
  Return Home
  Back to Forum: Sound Science
Head-Fi.org › Forums › Equipment Forums › Sound Science › 24bit vs 16bit, the myth exploded!