or Connect
Head-Fi.org › Forums › Equipment Forums › Sound Science › Hi res....no audible difference?
New Posts  All Forums:Forum Nav:

Hi res....no audible difference? - Page 2

post #16 of 42

Here is a great article showing that there is no point to higher bitrate or lossless files. It is brand new:

 

http://xiph.org/~xiphmont/demo/neil-young.html

post #17 of 42
Thread Starter 

Wow...thanks all for the insight! For me, it's very educational.

Obviously there is much to be said for better engineering. If I understand it correctly, there seems to be a measurable difference with higher bit-rate recordings. whether or not those differences are audible seems to be hotly debated. But just as an amplifier (a good one, anyway) doesn't just "make it louder", it seems that to have a better file (i.e. higher bit-rate) will make a difference at some level. Whether or not it's worth it is a personal decision.

post #18 of 42

Not exactly...

 

Hires files have the ability to reproduce frequencies beyond the range of human hearing and provide detailed resolution down to super low volume levels that are too quiet to be heard at normal music listening volumes. There is absolutely no reason for it for playing back music in the home. But a recording engineer who is mixing a recording might appreciate having a little more latitude for raising the volume of an instrument in the mix without dragging up the noise floor with it.

 

High bit rates are for recording, not playback. In fact, super high frequencies if they exist in recordings with high sampling rates can cause distortion in lower frequencies on a lot of home audio equipment. Frequencies you can't hear can cause distortion in frequencies you can hear. Not good.


Edited by bigshot - 4/9/14 at 7:47pm
post #19 of 42

Higher sampling rate allows higher frequencies to be stored. CDs use 44.1kHz which can store up to 22.05kHz, higher than any human can hear.

High bit depth allows higher signal to noise ratio, which means the possibility of more dynamic range without noise in the quiet sections. However, 16bit (used on CDs) allow up to 96dB (even more with modern dither techniques) which is far more than any recording needs.

post #20 of 42

I would like to chime in....  I think the intent of the thread has perhaps been missed....

 

If one were comparing a CD to a hi-res download, then no there would not likely be an audible difference.

 

I think what the question and what I would like to see discussed is when is a digital file ""Close"" enough.

 

I can clearly hear the difference between a 128MP3 and a FLAC or wave file.  I think I can hear the difference between a 320MP3 and a FLAC or wave file.

 

My question is where is the point of diminishing returns?  Its certainly been reached by 24/96....

post #21 of 42

Go and listen to the reduced bit-depth files on Ethan Winer's site:- http://ethanwiner.com/BitsTest.html

 

w

post #22 of 42
Quote:
Originally Posted by gatorengineer64 View Post
 

If one were comparing a CD to a hi-res download, then no there would not likely be an audible difference.

 

Theoretically yes but in practice this is unfortunately not true. As already demonstrated by my example on the first page. But again, this really has very little to with the claims of "hi-res is a new enabler which wasn't possible before" that is being thrown around these days.

Quote:
I think what the question and what I would like to see discussed is when is a digital file ""Close"" enough.

 

Again theoretically speaking, 16bit 44kHz 320kbps mp3 or 256kbps AAC is pretty damn close.
Quote:
I can clearly hear the difference between a 128MP3 and a FLAC or wave file.  I think I can hear the difference between a 320MP3 and a FLAC or wave file.

 

That's what a lot of people say too, why not try the ABX test for yourself and find out for certain? I did it before and found out that it's pretty hard, and came to the conclusion that it's not the technology that is the limiting factor (which is double confirmed by my signal engineering training), but that bad recording/mixing is where the blame should be put.
Quote:
My question is where is the point of diminishing returns?  Its certainly been reached by 24/96....

 

If it's done properly I think 16bit 44kHz is plenty good enough for playback. Anything over is not even diminishing returns but waste. But then the problem is that things aren't done properly most of the time, and I'm afraid that I'll have to buy "hi-res" downloads just to get songs which should have been done properly in the first place.


Edited by nanaholic - 4/13/14 at 9:03am
post #23 of 42
Let's call "the law of diminishing returns" by a better name... "audible transparency". The point where a sound file or format sounds exactly the same as the master is the point of transparency. Beyond that, nothing matters any more because you just can't hear it.

I'd suggest that the point of transparency is somewhere between 256 VBR to 320 / MP3 LAME or AAC.

As a hifi nut, I only worry about sound I can hear. Audiophools care more about sound they can't hear.
post #24 of 42

Oh yeah, if you recall the rockwell review of iphone 5.  He states he can hear page turn and all kind of details of a orchestra piece because of the 009.  The funny thing is, that's how it was recorded.  It's most likely not the headphones that's bringing out the details, but any decent headphones can hear it because of the way it was recorded.

 

Recently I had a chance to hear a binaural recording, blew me away.  Now I need to collect them all.  :biggrin:  So, yeah the way it was recorded has lots to do with it sounding really great whether it's losses or compressed at a decent bitrate.  

 

This Binaural amber Rubarth album at 24/96 I have is nearly a gig in size, it's probably the way the it was recorded that makes it sound so incredible.  

 

post #25 of 42

Another thing about the binaural albums is, the soundstage that people want so much is there, and it's controlled in the way it's recorded.  The way the stage is perceived is dependent on the recording.

 

Not sure how much you can compress these recording and retain decency of the sound


Edited by SilverEars - 4/13/14 at 1:38pm
post #26 of 42

I can't hear any difference between 128kbps MP3 and FLAC, so I'm 100% sure that anything above 16/48 won't make any audible improvement. Perhaps with blind testing some people could hear intermodulation distortion, and I guess some may even like it, like many people like distorting their sound with tubes, but that's not the goal of high fidelity I think.

post #27 of 42
Sound stage in headphones is a shadow of sound stage in speakers, even with binaural recordings.

Perception of small details in headphones is something I've noticed too. Does anyone know why? My theory is that it is easier to get louder in headphones. Having the tranducers right up next to your ear swallows up some of the volume. With speakers, the sound "blooms" around you and reflects off the walls creating a corporal feeling, particularly in bass. Because of this, you tend not to turn up the sound as loud, so small sounds remain small. They aren't blown up like with headphones.
post #28 of 42
Quote:
Originally Posted by bigshot View Post

Sound stage in headphones is a shadow of sound stage in speakers, even with binaural recordings.

Perception of small details in headphones is something I've noticed too. Does anyone know why? My theory is that it is easier to get louder in headphones. Having the tranducers right up next to your ear swallows up some of the volume. With speakers, the sound "blooms" around you and reflects off the walls creating a corporal feeling, particularly in bass. Because of this, you tend not to turn up the sound as loud, so small sounds remain small. They aren't blown up like with headphones.

This is why I love iems.  It's very intimate and the noise isolation, this I think makes it so that  you can hear lots of details.

post #29 of 42
For me, it's all about natural balance. I don't want something to sound "warm" or "detailed". I want it to sound naural and real. My speaker system gives me a sense of natural presence that can't be achieved any other way. When I put on a recording with a really well organized sound stage, I can sit on the couch and not only point at each musician in the band, I can feel their height and the depth of the room they are performing in. Sometimes it raises the hair on the back of my neck it's so uncanny... particularly with chamber music where the different voices are so clearly placed and defined.
post #30 of 42

Thanks Guys for the thoughts...

 

I think two things come out of this for me....  No need to head to DSD land... 24/96 is the limit.

 

You may say that is too high, but I also take 24 things away from the thread

 

1)  If its put out at 24/96, then it is at least in theory well mastered.

 

2)  I think alot of stuff that has been ripped, may not have been ripped well.

 

So perhaps what I was hearing was a better master, or the difference of a poor rip.....

 

Thanks for the input.  Will check out he Emily Rubarth.... I own alot of Chesky....

New Posts  All Forums:Forum Nav:
  Return Home
  Back to Forum: Sound Science
Head-Fi.org › Forums › Equipment Forums › Sound Science › Hi res....no audible difference?