or Connect
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 118

post #1756 of 3426
Quote:
Originally Posted by RazorJack View Post
 

 

Feeling, or perceiving, how exactly, with which of the five senses?

 

This has nothing to do with science. 

 

That said, I think people feeling things is what keeps 99,9% of Head-Fi and the audiophile community thriving, and there's nothing wrong with that.

 

"Feeling" a difference (at least in this context), would be picking up on minute differences on a subconscious level.  These minute differences would accumulate over a listening period and would make the listening experience overall less satisfactory when compared to the alternative.

 

I *believe* there are documented studies that show small volume differences, at a1 db or maybe less, will be enough to persuade a listener they "feel" A is better than B, even though they are the same recording.  That might be one example.

post #1757 of 3426
Quote:
Originally Posted by Rooster81 View Post
 

 

"Feeling" a difference (at least in this context), would be picking up on minute differences on a subconscious level.  These minute differences would accumulate over a listening period and would make the listening experience overall less satisfactory when compared to the alternative.

 

When it comes to human perception, differences in sound don't "reveal themselves over time". Our auditory memory for comparing two similar sounds is around 3-4 seconds at most. Our minds do *adapt* to sounds though. What sounds muffled initially might sound more normal over an extended period of listening. So the exact opposite to your theory is the truth of the matter. You don't become more perceptive over time, you become less perceptive and more forgiving.

 

All of the tests of frequencies above the range of human hearing that I've ever seen come to the same conclusion... they add absolutely nothing to sound quality. They can only detract if they create distortion at lower frequencies within the audible range, or cause irritation if the volume is too high. They use high volume super-audible frequency blasts on rioters to induce headaches and nausea.

 

You're better off with plain old redbook.

post #1758 of 3426


I've another question regarding equipment, again playing devil's advocate.  And again, I'm trying to increase my knowledge.

 

 

If a DAP has a frequency response of, say 10Hz ~ 20kHz, wouldn't that mean a recording in 192 kHz is meaningless?

 

Thanks.

post #1759 of 3426
Quote:
Originally Posted by Rooster81 View Post
 


I've another question regarding equipment, again playing devil's advocate.  And again, I'm trying to increase my knowledge.

 

 

If a DAP has a frequency response of, say 10Hz ~ 20kHz, wouldn't that mean a recording in 192 kHz is meaningless?

 

Thanks.

high sample rate on the track has no real purpose IMO. but high sample capabilities from the dac can be good( again for filters more than anything).

 if we're talking only frequency response, then I would answer your question by another one, if a human can hear between 10hz ~ 20khz, wouldn't it be meaningless to record higher frequencies? ^_^

post #1760 of 3426

It would be very difficult to design a study that relied on people reporting their 'feelings' without specifically asking them leading questions that might lead to confirmation bias. I do know of studies like testing of Tsutomu Oohashi's idea of the 'hypersonic effect' where they did things like monitor brain waves etc. The results have not been replicated in similar experiments though.

 

http://en.wikipedia.org/wiki/Hypersonic_effect

 

As far as testing of 24 bits goes Tomscy2000 suggested that maybe the ear can selectively tune out the noise floor in an environment (could be possible, the auditory system is pretty complex) and so perhaps would be able to make use of the extra dynamic range 24 bit audio provides.

 

This still doesn't solve the problem that most content does not even use the 96dB dynamic range of 16 bit audio, and end users have a tendency to turn down the volume rather than have 96dB peaks.

post #1761 of 3426
Quote:
Originally Posted by Rooster81 View Post
 

If a DAP has a frequency response of, say 10Hz ~ 20kHz, wouldn't that mean a recording in 192 kHz is meaningless?

 

Yes. And even if the DAP is capable of reproducing super audible frequencies the answer is the same.

post #1762 of 3426
Quote:
Originally Posted by bigshot View Post
 

Now that you've had the scales pulled off your eyes when it comes to 24 bit, would you like us in Sound Science to turn you on to just as surprising truths about high bit rate lossy files, headphone amps and external DACs?

 

I would. I'm slowly making my way through this ENTIRE thread. Been very enlightening, to say the least. Seriously though, from here I'd love recommendations for reading up on the 'surprising truths' you mentioned.

post #1763 of 3426
Quote:
Originally Posted by thievesarmy View Post
 

 

I would. I'm slowly making my way through this ENTIRE thread. Been very enlightening, to say the least. Seriously though, from here I'd love recommendations for reading up on the 'surprising truths' you mentioned.

the truths really aren't that surprising...unless you're a salesman for cables or something.

 

a good place to start reading is here, some nice links to listening tests regarding flac/mp3 and stuff like that   http://www.hydrogenaud.io/forums/index.php?showtopic=82777

 

there are numerous threads here on headfi, just scroll around sound science and pick a topic of interest.  there's a little bit of good info in most threads.

 

ethan winer (just google) is an excellent source of info, he has some good videos which are easy to watch and understand and a lot of info on his site.  archimago is one of my favorites, he's kind of an everyman tester, doing testing on normal (as in not the highest end) equipment, using normal equipment.  and it's a fun read   http://archimago.blogspot.ca/

post #1764 of 3426
Quote:
Originally Posted by thievesarmy View Post
 

I would. I'm slowly making my way through this ENTIRE thread. Been very enlightening, to say the least. Seriously though, from here I'd love recommendations for reading up on the 'surprising truths' you mentioned.

 

Ah! That is easy. At this point in digital audio, the solid state electronics part (CD players, DACs, amps) are all audibly transparent if they are operating to spec and not broken. That means that a $40 Coby CD player sounds just as good as a high end one, and as long as an amp has enough power to do the job and matches the impedance of the transducers, you are set to go. Many headphones work fine right out of an iPod and don't require amping at all. All modern DACs should be audibly transparent, whether they are external or built into a DAP or disk player. None of these things are what you should focus your attention on. The most important consideration in electronic components is features and usability. You shouldn't spend a lot of money on this part of your system.

 

As long as the bitrate is high enough and the codec is recent enough, even lossy files will sound as good as any higher bitrate formats. Don't depend on numbers on a sheet, do your own controlled listening tests and figure out for yourself what works. For me, it's LAME 320 or AAC 256 VBR... and I am VERY picky. Anything beyond that is fine, but not an audible improvement.

 

Transducers, however, make a BIG difference. They are the wild card of the system. Audition them before you buy them and buy the ones that sound the best to you. In general, good headphones and speakers are quite expensive. Here is where you should put your money.

 

Once you have functional electronics and high quality headphones or speakers, you can fine tune them with equalization- or in the case of speakers, a combination of room treatment and EQ. Balancing the response can make any system sound better. Get a good digital equalizer and experiment. Learn what different frequency bands sound like.

 

Once you have set up your system the way you think it should work well, spend time carefully listening to really good recordings. Analyze what you hear and try to identify problems. If you clearly define a problem and understand what is causing it, you are most of the way to solving it. Always refine your settings until you hit perfection. If you have speakers, consider going to 5:1 and implementing DSPs. If you already have a solid two channel system, 5:1 will hit it out of the park. Then sit down for a few decades and enjoy great music.

post #1765 of 3426
Quote:
Originally Posted by bigshot View Post
 

 

Ah! That is easy. At this point in digital audio, the solid state electronics part (CD players, DACs, amps) are all audibly transparent if they are operating to spec and not broken. That means that a $40 Coby CD player sounds just as good as a high end one, and as long as an amp has enough power to do the job and matches the impedance of the transducers, you are set to go. Many headphones work fine right out of an iPod and don't require amping at all. All modern DACs should be audibly transparent, whether they are external or built into a DAP or disk player. None of these things are what you should focus your attention on. The most important consideration in electronic components is features and usability. You shouldn't spend a lot of money on this part of your system.

 

As long as the bitrate is high enough and the codec is recent enough, even lossy files will sound as good as any higher bitrate formats. Don't depend on numbers on a sheet, do your own controlled listening tests and figure out for yourself what works. For me, it's LAME 320 or AAC 256 VBR... and I am VERY picky. Anything beyond that is fine, but not an audible improvement.

 

Transducers, however, make a BIG difference. They are the wild card of the system. Audition them before you buy them and buy the ones that sound the best to you. In general, good headphones and speakers are quite expensive. Here is where you should put your money.

 

Once you have functional electronics and high quality headphones or speakers, you can fine tune them with equalization- or in the case of speakers, a combination of room treatment and EQ. Balancing the response can make any system sound better. Get a good digital equalizer and experiment. Learn what different frequency bands sound like.

 

Once you have set up your system the way you think it should work well, spend time carefully listening to really good recordings. Analyze what you hear and try to identify problems. If you clearly define a problem and understand what is causing it, you are most of the way to solving it. Always refine your settings until you hit perfection. If you have speakers, consider going to 5:1 and implementing DSPs. If you already have a solid two channel system, 5:1 will hit it out of the park. Then sit down for a few decades and enjoy great music.

 

Heh. Ok, interesting starting point. I have noticed some of the things you're talking about already. For example…

 

Last week I finally was able to obtain some good quality HD 650's (used, at a great price). I had long been looking forward to hearing some high-end Senn's, as they are very praised on this site. Well, I wasn't disappointed, and I really am loving them, but I was a little surprised by something. First, I tried using them with my computer as the source (retina MBP) to a Dragonfly DAC @ 16, 44.1. The amp was a new one I just got, a S.M.S.L. sAp II that set me back $60. That rig sounded awesome and I was thinking, WOW, I can't wait to hear how it sounds at home on my Bifrost + Vali! This is gonna be AWESOME.

 

Well, it did sound awesome, but that isn't the important part. After these more "capable" rigs, I decided to try an experiment. I plugged them straight into my iPhone 5, and to my surprise, these 300 Ohm high-end phones sounded pretty great from there, too. I don't know if they sounded QUITE as good as the other rigs, and I didn't listen quite as long, or to the same pieces of music (for a more accurate test), but I was pleasantly surprised at how well they sounded straight out of the iPhone. I almost came here asking about it, but I figured there was a lot I still didn't fully understand about impedance & sensitivity, etc.

 

I still need to finish this thread, I'm only on page 105…

post #1766 of 3426
Quote:
Originally Posted by bigshot View Post
 

<snip, snip>

 

What sounds muffled initially might sound more normal over an extended period of listening.

 

Yeah... just ask anyone with Senn HD-650s... oh, did I say that out loud? :L3000: 

post #1767 of 3426
Quote:
Originally Posted by Krutsch View Post
 

 

Yeah... just ask anyone with Senn HD-650s... oh, did I say that out loud? :L3000: 

Yes, if you come from Grados.  :D

post #1768 of 3426
Quote:

Originally Posted by a_recording View Post

 

As far as testing of 24 bits goes Tomscy2000 suggested that maybe the ear can selectively tune out the noise floor in an environment (could be possible, the auditory system is pretty complex) and so perhaps would be able to make use of the extra dynamic range 24 bit audio provides.

 

Well, it could do the same to the noise floor of dithered 16-bit PCM audio. After all, a -110 dBFS tone can still be heard in that format "under" the noise, if the volume is turned up.

post #1769 of 3426

I will probably regret posting anything on Sound Science, not being a "Sound Scientist", but I feel compelled to mention why I have been buying 24-bit tracks on HDTracks (I don't work for anyone in the industry).

 

They cost the same, roughly speaking, as the physical CD for newer releases or remasters.

 

For example, Lana Del Rey's newest release (deluxe edition) lists at $US 17.99 and the same CD + bonus tracks lists at $US 15.99 on Amazon and, BTW, I can download it now instead of waiting for shipment.  Yes, I know you can buy used CDs for less - I am ignoring that, for new stuff, although I do buy used CDs for older releases.

 

Anecdotally speaking, as an amateur enthusiast, I have been impressed with the mastering on many of the tracks I've purchased from HDTracks; especially compared with the same music purchased from iTunes (AAC), as mastered for iTunes and/or AAC-encoded.

 

Comparing lossy AAC with lossless versions may be viewed as an unfair comparison of apples and oranges, but many on this forum don't believe that one can reliably hear the difference - so maybe the comparison isn't unfair, I don't know.

 

I recently did this comparison with Lorde's Pure Heroine - I purchased the iTunes version and downloaded the HDTracks version (24-bit, 48 kHz).  There's a section at the start of the second track (400 Lux) that sounds like an electronic sweep, if you will, and the differences between the AAC version and the HDTracks version was really apparent.  The lossless version sounds like it's sweeping across from left to right and even with headphones has sort of a 3-D imaging effect.  The AAC version, in my amateur-sighted-evaluation-opionion, was very flat and didn't convey that sense of sound staging or movement of the sound source.  Is that from mastering differences, is that from compression?  I have no idea.

 

So, for me, my ears and my wallet, whether or not I can hear the difference between 16 and 24-bit isn't totally the point (I probably can't, based on what I've read and my ad-hoc experience of taking 24-bit tracks and then re-sampling and dithering using iZotope 64 in my Mac).

 

But it's also probably not hurting anything, either, and for the mastering and the convenience, why not?

post #1770 of 3426
Quote:
Originally Posted by Krutsch View Post
 

I will probably regret posting anything on Sound Science, not being a "Sound Scientist", but I feel compelled to mention why I have been buying 24-bit tracks on HDTracks (I don't work for anyone in the industry).

...

There's nothing to regret! If you are curious about the topic, interested in learning, and honest and open minded, you will fit in just fine here! Welcome :)

 

Comparing lossy AAC with lossless versions may be viewed as an unfair comparison of apples and oranges, but many on this forum don't believe that one can reliably hear the difference - so maybe the comparison isn't unfair, I don't know.

...

The fair comparison is to take the hi-res lossless version and compress that to the AAC bitrate of your choosing. Then you compare the high-res to the lossy version which came from the same master.

 

In the case of iTunes vs HDtracks, you are likely comparing different masters, so you learn nothing about the differences in  bitrates and compression.

 

I recently did this comparison with Lorde's Pure Heroine - I purchased the iTunes version and downloaded the HDTracks version (24-bit, 48 kHz).  There's a section at the start of the second track (400 Lux) that sounds like an electronic sweep, if you will, and the differences between the AAC version and the HDTracks version was really apparent.  The lossless version sounds like it's sweeping across from left to right and even with headphones has sort of a 3-D imaging effect.  The AAC version, in my amateur-sighted-evaluation-opionion, was very flat and didn't convey that sense of sound staging or movement of the sound source.  Is that from mastering differences, is that from compression?  I have no idea.

 

See my previous paragraph about different masters.

 

So, for me, my ears and my wallet, whether or not I can hear the difference between 16 and 24-bit isn't totally the point (I probably can't, based on what I've read and my ad-hoc experience of taking 24-bit tracks and then re-sampling and dithering using iZotope 64 in my Mac).

 

I think this is the relevant evidence. You can make your decision based on tests like this, ideally using an ABX comparator like the free one that works with Foobar2000.

 

But it's also probably not hurting anything, either, and for the mastering and the convenience, why not?

 

The point of the discussion here is not whether or not folks should be allowed to purchase hires audio for their own personal listening pleasure. Rather, this discussion is focused on whether or not there are actual tangible audible benefits to using 24 bit audio for listening.

 

There exist a large group of supports of hires formats that insist that unless one has a 24bit playback chain (or other esoteric "hires" format) with hires sources that people are missing out on music. There is a lot of noise  out there that uses false logic and nonscience to convince oblivious, budding audiophiles that they must repurchase their audio catalog in high res formats and shell out thousands of dollars for new equipment to play those high res formats in bit-perfectness, or otherwise they will miss out on enjoying musical nuances. This confuses a lot of innocent folks into paying a lot of money which nets zero real-life improvement.

 

Here, we are trying to get the facts straight so that people can make informed decisions regarding where and how to spend their money (or not spend it!). It is to inform the consumer, so they can make their own choices. If you want to buy hires audio, you should feel free to do so; however, you should understand that perceived quality it not a valid reason to do so.

 

 I have and I may continue to purchase highres tracks from sources like HDtracks if I feel that in doing so I am a) supporting artists that I like and b) using my money spent as a vote to influence more artists, producers, and record companies to focus on quality recording, engineering, and production practices rather than loudness.

 

If buying the latest Zepplin remaster supports Page, Plant, and Jonesy, then I am inclined to do so (however, if the remaster is a butchered, processed turd, I'll opt to stick with my boxed set until a better version comes along).

 

 

Cheers


Edited by ab initio - 6/19/14 at 10:50am
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!