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 7

post #91 of 2109
Quote:
Originally Posted by JaZZ View Post
...and don't even exclude a benefit from 192 kHz. Your reasoning isn't entirely convincing to me.

What? His reasonings for greater than 96khz recording isn't convincing?

Let me guess, only when you've heard it?

Say this never stops, sampling frequency only increases with time. Next year, Sennheiser discovers some new microphone construction that can recording upto 200khz! Wow, it's incredible, perfect straight frequency response right upto 200khz. Would you think a theoretical 384khz recording has some benefits? Even though those high frequencies are used today for finding fish? No, that would be ridiculous. Then why doesn't that apply for 192khz recording?




EK
post #92 of 2109
Quote:
Originally Posted by evilking View Post
What? His reasonings for greater than 96khz recording isn't convincing?

Let me guess, only when you've heard it?
Indeed. I also need more convincing arguments than yours.
.
post #93 of 2109
I am afraid I have to disagree.
"There is no difference between 24bit and 16bit other than higher dynamic range".
The potential for a larger dynamic range is great and nothing to be scoffed at. As for there being "no additional resolution", nothing really could be further from the truth. Have you listened to any 8bit recordings recently? There is little high pitches to speak of (and definitely no sparkle), and in most encodes there is a pleasant rolloff.
Why is this so? Well, high frequencies are harder to put in the digital realm and need more data to be represented accurately (or at all in some extreme cases). The spectrum is expanded from 16 bit when it is in 24bit. The resolution is "higher".
16bit can extend beyond the range of human hearing, but there is little study on whether frequencies above conventional hearing have an impact on the overall sound on the rest of the song through constructive interferance where as removing a lot of highs will leave just the fundamental and things will sound nearly identical to eachother - you lose the character that is violin or flute (produce a similar wave form and have similar attacks decays so on). Logically, and in my opinion audibly, you have extra clarity and detail in 24bit PCM resolutions because of this.
Here i become a little less certain of what I am talking about but i try to express normally. I think 16bit PCM is limited to 96khz sampling, and from a technical perspective you can see aliasing in lower frequencies than in 24bit PCM which allows 192 khz sampling. Even if you can not hear the harmonic being played by the headphones you will be able to hear the alias
correct me if i am wrong on any of this, but I am quite sure this is how it works in the digital
post #94 of 2109
Quote:
Originally Posted by CDBacklash View Post
Have you listened to any 8bit recordings recently? There is little high pitches to speak of (and definitely no sparkle), and in most encodes there is a pleasant rolloff.
He's not arguing that there isn't a benefit between going from 8 bit to 16 bit, he's saying there isn't any benefit from going between 16 bit to 24 bit. 8 bits represents 2^8= 256 possible levels, 2^16= 65536, and 2^24 = 16777216. Those are totally different things. The benefit you get from increasing resolution (bit depth) is asymptotic. There is a maximum resolution that you are capable of discerning- whatever that may be- and storing the data (music in this case) at any higher resolution doesn't give you any additional benefit because you cannot discern it. The second issue is that as your resolution increases, background noise caused by the various electronic components sets a certain practical maximum level of resolution. This is the difference between accuracy and precision. gregorio is arguing that going from 16 bit to 24 bit provides no benefit because of both of these reasons. The nyquist-whatever sampling theory (and theory = law, it doesn't mean some random unproved ideas: e.g. the theory (law) of gravity). says that you need to sample a frequency at approximately 2X the frequency you wish to reconstruct, so if you want to be able to reconstruct a 20 kHz tone, you need to be able to sample that tone at 40 kHz (and CDs sample at 44.1 kHz). Unless you're like 12 years old, you can't hear much beyond 20 kHz.

Now come the arguments of people professing that they can hear greater than 20 kHz. Sorry, you probably can't - even if you think you can hear it. Take a tone generator, set your system to a given volume, and listen to tones generated at 100, 200, 500, 1000, 2000, 5000, 10000, 12000, 15000, 20000. Set the volume to a level where you would normally listen when playing either the 500, 1000, or 2000 hz tone, and then proceed through the rest without adjusting the volume. Now being honest, can you really hear 20,000 hz? Science suggests that if you're an adult male you probably can't. Simply saying that you can won't convince me either. I'm 31, and seven years ago when I had my ears tested I could hear maybe up to 20 kHz, but the sound was so faint that it didn't really matter much.

This is why medical researchers study the effect of drugs using double blind studies- so that neither the individual being measured, nor the individual measuring knows whether they are being given the real treatment or being given a placebo. Just wanting something to be true is enough in order to convince yourself that it is. Some small fraction of the adult population maybe able to hear 20 kHz without drastically increasing the volume, but it will be a very small percentage.

Quote:
Why is this so? Well, high frequencies are harder to put in the digital realm and need more data to be represented accurately (or at all in some extreme cases). The spectrum is expanded from 16 bit when it is in 24bit. The resolution is "higher".
Dude, he understands this - don't be patronizing.

Quote:
16bit can extend beyond the range of human hearing, but there is little study on whether frequencies above conventional hearing have an impact on the overall sound on the rest of the song through constructive interferance where as removing a lot of highs will leave just the fundamental and things will sound nearly identical to eachother - you lose the character that is violin or flute (produce a similar wave form and have similar attacks decays so on).
If you're going to engage in a scientific discussion and reference a paper supporting your position you are remiss if you do not cite it specifically.

Quote:
Logically
You haven't made any arguments from which any conclusions can be deduced 'logically'.
post #95 of 2109
.
a. double blind testing isnt a good testing method
b. Interference - Wikipedia, the free encyclopedia this applies to all independant frequencies and also harmonics within the same system (even those outside the hearing range no matter how subtle the impact is there).
Got a headphone that can produce 5hz? Try boosting 5-15 hz really loud for me and see how it sounds It wont be the same.
I dont know about you, but since I am hearing the effect of harmonics and the like in real life I want to hear them at home, no matter how audible the effect is.
c. please do not be rude to me because of my poor english next time
d. I never mentioned anything abotu being able to hear above 20khz, although 2 years ago at last test i was reported to have had a range of 17hz-22.2 khz (the latter of which extends beyond the rule you provided for 44.1khz). Not that this really affects the "music". I am not listening for > 13 khz sounds when i listen to a song, but i hear them and the effect they have on other frequencies.
post #96 of 2109
Quote:
Originally Posted by CDBacklash View Post
.
a. double blind testing isnt a good testing method
According to whom? It is the gold standard in biomedical research. What do you propose as a better alternative?


Quote:
Got a headphone that can produce 5hz? Try boosting 5-15 hz really loud for me and see how it sounds It wont be the same.
I dont know about you, but since I am hearing the effect of harmonics and the like in real life I want to hear them at home, no matter how audible the effect is.
But we're not talking about 5-15 hz, so this is irrelevant. 44.1kHz sampling already can deal with 5-15 hz. We're talking about high frequencies. I don't understand your point.

Quote:
c. please do not be rude to me because of my poor english next time
I wasn't rude to you because of your poor English. I wasn't rude to you at all. I was rather pointedly arguing that you hadn't supported your conclusions.

Quote:
d. I never mentioned anything abotu being able to hear above 20khz, although 2 years ago at last test i was reported to have had a range of 17hz-22.2 khz (the latter of which extends beyond the rule you provided for 44.1khz). Not that this really affects the "music". I am not listening for > 13 khz sounds when i listen to a song, but i hear them and the effect they have on other frequencies.
Except the whole point of higher frequency sampling rates is in order to get a better representation of higher frequency sounds.
post #97 of 2109
Quote:
Originally Posted by Clutz View Post
(and theory = law, it doesn't mean some random unproved ideas: e.g. the theory (law) of gravity).
No sir, theories are theories. Based on facts and provable experiments, yes. But they are not complete laws as we do not have a complete understanding of these things. Today's theories are tomorrow's jokes (some times).

Quote:
Originally Posted by CDBacklash View Post
.
a. double blind testing isnt a good testing method
b. Interference - Wikipedia, the free encyclopedia this applies to all independant frequencies and also harmonics within the same system (even those outside the hearing range no matter how subtle the impact is there).
You need to read the first lines:
Interference usually refers to the interaction of waves which are correlated or coherent with each other, either because they come from the same source or because they have the same or nearly the same frequency.

Two non-monochromatic waves are only fully coherent with each other if they both have exactly the same range of wavelengths and the same phase differences at each of the constituent wavelengths.

Quote:
Got a headphone that can produce 5hz? Try boosting 5-15 hz really loud for me and see how it sounds It wont be the same.
I dont know about you, but since I am hearing the effect of harmonics and the like in real life I want to hear them at home, no matter how audible the effect is.
c. please do not be rude to me because of my poor english next time
d. I never mentioned anything abotu being able to hear above 20khz, although 2 years ago at last test i was reported to have had a range of 17hz-22.2 khz (the latter of which extends beyond the rule you provided for 44.1khz). Not that this really affects the "music". I am not listening for > 13 khz sounds when i listen to a song, but i hear them and the effect they have on other frequencies.
I haven't heard of any headphones that can do 5Hz-15Hz. Of course, I haven't ever looked at >$1000 dollar phones or electrostats. Also "really loud" may end up breaking your phones because they weren't designed to... play those frequencies "really loudly".

I don't think anyone is taking a jab at your english. I think it's just fine.
post #98 of 2109
@aleatoris there are plenty that will. Ultrasone proline, and dt770. I think HD650 goes down to 7 or 12 within the scale they used for their measurement (but will produce lower than taht i think).
@clutz double blind is a bad test when it comes to audio. No one hears exactly the same as someone else (both in terms of actual hearing and what they perceive which is where subjectivity comes into it). If you use the same person they will hear differences that arent there (and also not hear ones that are) because they will know in most cases that it is a double blind test or not be paying attention to that area of the music enough. Results from it are never to be taken seriously, really whether they be "i heard a difference" or "I could not hear a difference".
@aleatoris the lines about the waves being similar frequencies involves the aspect of "beats". Beats do not exist nearly as much as they do in that case (because they are out of sync for a longer period of time due to doppler-like effects when their frequencies are comparable). Beats always exist from a complex source but become "inaudible" (i.e. the smooth sound we hear in our ears). Superposition applies to every sound frequency in a given system. even wikipedia knows this "interferance >usually<" etc not to mention the sounds are coming from the same source when you listen to music
@clutz the low frequency is an extreme example of what high frequencies will do to the sound because there is more energy behidn the production of lower frequencies and the result will be more extreme. A similar thing happens with high frequencies only to a lesser degree.
post #99 of 2109
Quote:
Originally Posted by CDBacklash View Post
@aleatoris there are plenty that will. Ultrasone proline, and dt770. I think HD650 goes down to 7 or 12 within the scale they used for their measurement (but will produce lower than taht i think).
<..snip..>
@aleatoris the lines about the waves being similar frequencies involves the aspect of "beats". Beats do not exist nearly as much as they do in that case (because they are out of sync for a longer period of time due to doppler-like effects when their frequencies are comparable). Beats always exist from a complex source but become inaudible. Superposition applies to every sound frequency in a given system.
EDIT: Yes, those phones can produce those signals, but the point of my post was that if you EQ the crap out of the low frequencies, the signals will most likely get distorted.

huh? Superposition doesn't mean that high frequency signals affect lower frequency signals... the waveforms are just added together to produce something more complex. The audible bits are still audible (without much noticable change) and the inaudible bits are... inaudible. If you take out the highs altogether... well, nothing really happens to the audible signal.

Edit, I'm a retard.
post #100 of 2109
Dont worry about it, it's confusing stuff. The effect when listening to recorded music is a bit more extreme than the real world on recordings where it was one room and omnidirectional microphones because it exists in each mic with sound bleeding and then it is multiplied really.
post #101 of 2109
Quote:
Originally Posted by Aleatoris View Post
No sir, theories are theories. Based on facts and provable experiments, yes. But they are not complete laws as we do not have a complete understanding of these things. Today's theories are tomorrow's jokes (some times).
I disagree because there are no such things as 'scientific laws'. The basic idea of science is that all ideas (hypothesis or theory) must be disprovable. Calling something a 'law' means that it impossible to disprove, and is therefore inherently a-scientific. What the public colloquially call 'scientific laws', are in fact 'scientific theories'. The 'Law of Gravity' results from Classic Newtonian Mechanics or General Relativity which are theories. I stand by my original statement.
post #102 of 2109
Quote:
Originally Posted by Clutz View Post
I disagree because there are no such things as 'scientific laws'. The basic idea of science is that all ideas (hypothesis or theory) must be disprovable. Calling something a 'law' means that it impossible to disprove, and is therefore inherently a-scientific. What the public colloquially call 'scientific laws', are in fact 'scientific theories'. The 'Law of Gravity' results from Classic Newtonian Mechanics or General Relativity which are theories. I stand by my original statement.
Not that I disagree in general, but I thought law was applied only to those theories that were proved mathematically. Newton came up with calculus just so he could prove this theories. I could be completely wrong
post #103 of 2109
Thread Starter 
Quote:
Originally Posted by Aleatoris View Post
No sir, theories are theories. Based on facts and provable experiments, yes. But they are not complete laws as we do not have a complete understanding of these things. Today's theories are tomorrow's jokes (some times).
I think you may have misunderstood the nature of digital audio and the theeory behind it. In many cases, theories are developed to explain how the universe around us works. This means that as our understanding and observational abilities of the universe improves, so sometimes the theories need to be changed or modified. However, this is not the case with digital audio. Digital audio was not discovered and then a theory invented to explain it, it's the other way around. The Nyquist theory was created in the 1920s and then when technology had improved in the early 1970s Digital Audio was invented based on the theory. In other words, if the Nyquist theory is wrong, digital audio would not exist.

Quote:
Originally Posted by CDBacklash View Post
I am afraid I have to disagree.
"There is no difference between 24bit and 16bit other than higher dynamic range".
The potential for a larger dynamic range is great and nothing to be scoffed at. As for there being "no additional resolution", nothing really could be further from the truth. Have you listened to any 8bit recordings recently? There is little high pitches to speak of (and definitely no sparkle), and in most encodes there is a pleasant rolloff.
Why is this so? Well, high frequencies are harder to put in the digital realm and need more data to be represented accurately (or at all in some extreme cases). The spectrum is expanded from 16 bit when it is in 24bit. The resolution is "higher".
16bit can extend beyond the range of human hearing...
Sorry, but I'm afraid you have misunderstood how digital audio works. The bit depth in digital audio encodes the dynamic range, it is the sample rate that is responsible for the frequency range. 8bit 44.1kFs/s has exactly the same frequency range as 24bit 44.1kFs/s, both max out at 22,050Hz. This cannot be disputed as it is a basic tenet of the Nyquist Sampling Theorem. With lower bit depths (say 8bit) the dynamic range (48dB) might be such that the dither noise is noticeable enough to mask the high frequency content. But with 16bit the noise floor is so low that it cannot be heard and therefore cannot mask the high frequency content. So in fact when you say 8bit recordings have no high frequencies you are mistaken (unless there is also a lower sampling rate). 8bit 44.1kFs/s actually has more high frequency content than 16bit 44.1kFs/s because there is the program material plus 48dB more noise distributed throughout the 22,050Hz frequency range.

I also think you need to read my original post to understand why the greater dynamic range of 24bit compared to 16bit is neither desirable nor even possible.

Hope this clears things up a bit,

G
post #104 of 2109
Quote:
Originally Posted by scompton View Post
Not that I disagree in general, but I thought law was applied only to those theories that were proved mathematically. Newton came up with calculus just so he could prove this theories. I could be completely wrong
Kind of.

You can 'prove' laws within theories. For example, conservation laws follow from symmetries (Noether's theorem). For example, the law of conservation of energy exists because the laws of physics are (assumed to be) time invariant, i.e. the same today as they were yesterday and will be tomorrow. Momentum is conserved because of translational invariance. Angular momentum because of rotational invariance. Etc. However these 'laws' are just corollaries of the theory we started with, and aren't truer than the original theory in any sense.

Proof in physics ultimately comes from experiments. Newton had lots of astronomical data and the famous Keplererian laws, which were based on empirical data and seemed very accurate, but were not really understood. Newton didn't need calculus to prove his theory of mechanics and gravitational law, but it was necessary for the mathematical description of mechanics (velocities, accelerations etc.), and from the simple starting points of his three laws, he could finally derive Kepler's laws. In a way he 'proved' them, but the point was that it was a test of the validity of his theory.
post #105 of 2109
If a mosquito sneezes in a noisy factory, has it contributed anything to the noise problem?

Mastering engineers claim to hear differences between various dithering algorithms and choose the best one for that particular job. These noise-shaped dithering algorithms are referenced in the original post. How is the mastering engineer supposed to choose something he cannot hear? If the mastering engineer can hear coloration to the piece caused by the dither, surely an audiophile with top line gear and a treated listening room could as well. If coloration from the dither is audible, and distortion from truncation if one chooses not to dither even worse, then one must conclude that an audible difference can be heard between 24 bit and 16 bit audio without raising peak levels so high as to cause deafness or deathness. That dither is even mentioned in this discussion suggests an acknowledgment that 16 bits worth of signal to noise ratio is not foolproof. If 96 db were a great overkill, truncation distortion would be a purely academic topic and dither would be some irrelevant relic of 8 bit audio.

Don't get me wrong. If we pick a random pop song and make a proper blind test for this, my money says we all fail. But then, almost all of us would fail the -V0 mp3 test too...
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!