Head-Fi.org › Forums › Equipment Forums › Sound Science › 24-bit audio a con, according to Gizmodo
New Posts  All Forums:Forum Nav:

24-bit audio a con, according to Gizmodo - Page 12

post #166 of 210
Quote:

a simplified, very brief explanation presenting more correct info can be found here.

 

 

I looked back again at Giorgio's post from a couple of years ago, and at the textbook link above, and I didn't really see a lot of fundamental disagreements.

 

I've referred others to the same link you posted above to understand the effect of quantitization error, as it does a very good job of breaking down the story behind the 1/2 LSB signal that can represent the noise that comes from it.  The text loses me a bit when it jumps to the conclusion that a listener could "hear" this error in quiet passages or at the end of echo trails.   Even without dither, someone would need to have the volume set to "death by audio" to quote a previous poster, in order to hear the raw error, AND be listening to a file with essentially zero noise floor or "hiss" to begin with.  If you're working with any kind of analog tape hiss or less-than-perfect microphone preamp, the noise will be there, and it will do a fair amount of dithering by itself just being there.  This is not to say dither is bad - it's free and can't hurt - just that it's not going to make the difference between artistry and suffering, as the text suggests.

 

People often demonstration quant-noise by synthesizing low-level test tones, then downconverting them from 24 bits to a lower number, but this is really going to have limited application to the real world, as these signals will have no noise, and will allow/force the listener to crank the volume to the point where what we're really doing is examining error around a 3-6 bit magnified range.

post #167 of 210
Quote:
Originally Posted by danroche View Post



 

I looked back again at Giorgio's post from a couple of years ago, and at the textbook link above, and I didn't really see a lot of fundamental disagreements.

 

I've referred others to the same link you posted above to understand the effect of quantitization error, as it does a very good job of breaking down the story behind the 1/2 LSB signal that can represent the noise that comes from it.  The text loses me a bit when it jumps to the conclusion that a listener could "hear" this error in quiet passages or at the end of echo trails.   Even without dither, someone would need to have the volume set to "death by audio" to quote a previous poster, in order to hear the raw error, AND be listening to a file with essentially zero noise floor or "hiss" to begin with.  If you're working with any kind of analog tape hiss or less-than-perfect microphone preamp, the noise will be there, and it will do a fair amount of dithering by itself just being there.  This is not to say dither is bad - it's free and can't hurt - just that it's not going to make the difference between artistry and suffering, as the text suggests.

 

People often demonstration quant-noise by synthesizing low-level test tones, then downconverting them from 24 bits to a lower number, but this is really going to have limited application to the real world, as these signals will have no noise, and will allow/force the listener to crank the volume to the point where what we're really doing is examining error around a 3-6 bit magnified range.


Thanks for being clearer about this than I was when I posted the quotes from gregorio's initial post.

 

The point is not whether or not there's a difference, but whether that difference is the difference audible. BTW I can't get the "Principles of Digital Audio" to display individual pages.

post #168 of 210

Hey danroche. I was reading your posts over at Stevehoffman.tv forums a week or so ago as well. I found your "arguing" with Barry Diament pretty funny, and that's not the first time I've found him saying absolutely ridiculous things. I recall a thread where he explains why he refuses to allow FLAC downloads because of possible audible differences. Are you kidding me?! Anyway, keep spreading the good word... ;-)

post #169 of 210
Quote:
Originally Posted by Vkamicht View Post

Hey danroche. I was reading your posts over at Stevehoffman.tv forums a week or so ago as well. I found your "arguing" with Barry Diament pretty funny, and that's not the first time I've found him saying absolutely ridiculous things. I recall a thread where he explains why he refuses to allow FLAC downloads because of possible audible differences. Are you kidding me?! Anyway, keep spreading the good word... ;-)



 

Thanks man.  For what it's worth I totally didn't intend that conversation thread to go in that direction, nor to end up in the position of stone thrower on that particular thread.  I actually am a fan of Barry's work, and think he's given us the best-sounding copies of many albums out of the Led Zeppelin catalog, not to mention his fantastic work on Bob Marley's stuff.   I think there's a real difference between the "art" of mixing and mastering and the science behind audio, and the assumption that the command of one implies expertise in the other is the source of a lot of misconceptions and blown cash.   The best pilot in the world isn't automatically a great aeronautical engineer.

post #170 of 210

I think if Apple or anyone large (like Amazon) began selling 24-bit downloads it would really Revitalize the music industry like the Movie industry does with every new format...Infact i would love to see it happen!

 

You had Star Wars on VHS, then you had to buy Star Wars on Laserdisc because it was better, then you had to buy it on DVD because it was more portable and had more features, And then you had to buy it all again on Blu-ray Because of 1080p! Yes it sucks buying the same movie over and over but everytime it was on a new format it was better then the last so it was worth buying. Problem with the mp3 market and the music industry today is mp3's are not really any better then what you already had on cd. They are more portable and easier to obtain but you can easily turn your cds into great mp3/lossless for free. You can upscale a dvd but it won't be no Blu-ray, So like the Blu-ray you would "want" to buy the better 24bit version of your favorite music.

 

It seriously would be the first time  the music industry has something to sell that is a upgrade from the cd in like forever!

 

 

And for the record Gizmodo is crap, always was and always will be!

 

post #171 of 210
Meant to post in this thread. So folks who rip their old/new vinyl records and then "resample and dither to 24/96" are mostly just wasting space? Meaning they could have ripped to redbook 16bit/44.1kHz and it would pretty much sound the same?
post #172 of 210
Quote:
Originally Posted by lessblue View Post

Meant to post in this thread.So folks who rip their old/new vinyl records and then "resample and dither to 24/96" are mostly just wasting space?Meaning they could have ripped to redbook 16bit/44.1kHz and it would pretty much sound the same?

depends what they do with them. For applying cleaning up the rips, 24/96 is great (but slower to process). After that you could well downsample/downconvert to 16/44.1



>I think if Apple or anyone large (like Amazon) began selling 24-bit downloads it would really Revitalize the music industry like the Movie industry does with every new format...Infact i would love to see it happen!


Personally, I think it's a complete con for releasing old masters of classical music CDs (same goes for cheap 24b DACs that use 24b/96k or 192 as a selling point). For classical music, 90% of the HD masters I have, are only notable for having next to appalling recording quality and next to 0 artistic value (3rd rate orchestra, conductor, soloist).
I maybe own 2-3 good ones, but then for well mastered mixes it gets substantially harder to tell the difference between the master and red-book.

What you do want is something like: http://www.arkivmusic.com/classical/album.jsp?album_id=443055 - recently recorded, full hd, (not sure whether the sound is 24b or not but it's properly mastered and miked. That said Achucarro is errr.... Well, out of form by now). Then I'd be more than happy to reach for my wallet.


The article seems to be on the money to me. When people pump the bad recording/poorly mastered 96k/24b master (or the crazier ones will no doubt buy the 24b/192k if available) into their cheap 24b HD bs in-built card and put on their dre beats it doesn't make it any better.

Props to the marketing dept.
Edited by svyr - 4/19/11 at 4:22am
post #173 of 210

Seeing that LP noise floor is about -65dB at best, I'd say that yes they aren't benefitting from the dynamic range a huge lot.

post #174 of 210
Quote:
Originally Posted by pocketrocket View Post

Seeing that LP noise floor is about -65dB at best, I'd say that yes they aren't benefitting from the dynamic range a huge lot.


seeing how it's not uncommon for modern recordings to have less than 8db of DR http://dynamicrangeday.co.uk/challenge/ not relevant. http://www.dr.loudness-war.info/index.php?sort=year&order=desc
For old classical LPs, sometimes people manage to acquire mint LPs of recordings that aren't available on CD, or better mastered LPs than the currently available CDs wink.gif . For older recording artefact's like clicking or higher noise floor would still be there for other mediums (tape, etc). and again, for processing the rip and cleaning it up, 96k24b is not inappropriate.
Edited by svyr - 4/19/11 at 5:01am
post #175 of 210

To the people who have spent some time listening to 24bit Vs 16bit:

 

I need some clarification.  It seems that unless you exceed 96dB in dynamic range, there is no audible difference between 24 bit playback and 16 bit playback?  Is this the bottom line?  (Recording benefits in 24 bit understood) 

 

USG

post #176 of 210

Bump for an answer to upstateguy, I am interested as well, thanks.

post #177 of 210

The only well controlled listening tests that have been done so far seem to indicate that 24 bit doesn't matter for playback unless you listen at very high volume levels to material that uses a lot of dynamic range.  I don't think it needs to use all of the 16 bit 96dB range either because you can hear the noise floor in soft passages without using all of the available range.  I don't have access to the AES papers myself so I can't give any specifics or offer any potential criticisms.  I think nick_charles explained the gist of it somewhere earlier in the thread. 

post #178 of 210

Oooooo! I am an 'interesting link' on this site!

 

http://www.audiocheck.net/blindtests_index.php

post #179 of 210
Quote:
Originally Posted by dalethorn View Post


They say that you can hear a big difference between the best CD quality and a 24/192 track. But that's about hearing, not measuring.

They aren't hearing it, they are seeing what they are playing. I'd like to see anyone blind test an ordinary 16/44 vs 24/192 and tell the difference. 

post #180 of 210
Quote:
Originally Posted by Marleybob217 View Post

They aren't hearing it, they are seeing what they are playing. I'd like to see anyone blind test an ordinary 16/44 vs 24/192 and tell the difference. 

 

I've been doing just that and noticing discernible differences between a few of my 16/44 and 24/96 songs with my SR-507.  Any of my other headphones I would say it's much harder to hear a noticeable change but with the Stax it can be easy to figure our which track is which.

 

A great example is "Firth of Fifth" by Genesis.  I have an Apple Lossless version and one that's 88.2K.  The difference between the two is immediate and easy to hear.  I know it's probably not the case for every hi-rez vs. redbook file but there are plenty of tracks where the different is substantial.  


Edited by tdogzthmn - 8/5/13 at 10:08pm
New Posts  All Forums:Forum Nav:
  Return Home
  Back to Forum: Sound Science
Head-Fi.org › Forums › Equipment Forums › Sound Science › 24-bit audio a con, according to Gizmodo