IPB

Welcome Guest ( Log In | Register )

 
Reply to this topicStart new topic
Bit Depth and aac *quicktime aac*
Mix3dmessagez
post Jan 26 2013, 17:10
Post #1





Group: Members
Posts: 87
Joined: 22-March 09
Member No.: 68252



I have some needledrops, 24/96 *forget the hdtracks now cause i'm learning some new stuff*

andn i use izotope src and mbit to resample and dither to 48/16

i've apparently learned that there is no depth in aac *bit depth*

and i'm curious, is it better to simply just leave all my drops at 24 bits so when encoded for portable use it will be used?

In comparisons i've done when dithered and when just left normal the dithered files *when both are encoded to aac* result in slightly larger file sizes and higher bitrates *of maybe 1,2 o even 3 larger, than when converting from 24 bit sources.

I figured with that being the case it must ignore bit depths and always use 16 bit anyway so I left it at that.

If any experts could provide me insight it would be greatly appreciated as i'd like to know which is better and how much difference it makes *not in terms of listening tests but scienticially*
Go to the top of the page
+Quote Post
Mix3dmessagez
post Jan 26 2013, 22:04
Post #2





Group: Members
Posts: 87
Joined: 22-March 09
Member No.: 68252



when i said expert it was jokingly, really anyone who knows anytype of anything is welcome to say something wacko.gif
Go to the top of the page
+Quote Post
saratoga
post Jan 26 2013, 23:11
Post #3





Group: Members
Posts: 4868
Joined: 2-September 02
Member No.: 3264



Bitrate goes up slightly because the dither is harder to compress. I probably wouldn't bother dithering down to 16 bit, but it won't make much difference.
Go to the top of the page
+Quote Post
greynol
post Jan 26 2013, 23:36
Post #4





Group: Super Moderator
Posts: 10000
Joined: 1-April 04
From: San Francisco
Member No.: 13167



Will this silly notion that a listening test does not somehow constitute "science" when dealing with perceptual coding ever die?

This post has been edited by greynol: Jan 26 2013, 23:36


--------------------
Placebophiles: put up or shut up!
Go to the top of the page
+Quote Post
Mix3dmessagez
post Jan 27 2013, 05:00
Post #5





Group: Members
Posts: 87
Joined: 22-March 09
Member No.: 68252



QUOTE (greynol @ Jan 26 2013, 17:36) *
Will this silly notion that a listening test does not somehow constitute "science" when dealing with perceptual coding ever die?



I apologize for my inspecification, I meant, I was looking for a reason as to if one choice was better than the other why, so I could learn, as I'm very interested in audio fidelity and all its perks too, not, that's the only and listening tests don't prove anything.

i tried googling this and found no topics on it so i came here


Bitrate goes up slightly because the dither is harder to compress. I probably wouldn't bother dithering down to 16 bit, but it won't make much difference.

When you say it won't make much difference which choice, are you referring to the fact of with lossy compression the differences between dithering or just converting from the 24 bit source will be insigificant?
With that being said, i remember reading when you resample you should always dither
The files in hand are resamples with izotope, then i dither with mbit dither+ for portable use.

How does qt aac handle 24 bit compared to dithering it in izotope?
Which is better *As in, better, not what i prefer sonically* and why?
Is the difference significant between the 2 options? and why?

I hope i don't come off as annoying, i'm genuinely interested in knowing these things.
Go to the top of the page
+Quote Post
[JAZ]
post Jan 27 2013, 10:10
Post #6





Group: Members
Posts: 1767
Joined: 24-June 02
From: Catalunya(Spain)
Member No.: 2383



Correction: Dither is not used when resampling. It is used when reducing the bit depth, to prevent the addition of unwanted distortions by the quantization error and add broadband noise instead. This also means that applying dither without doing bit depth reduction is not a good idea.


I haven't used Quicktime encoder for quite some years, but I assume it supports 24bit input. In that case, you only need resampling, and this does not need any dithering. The encoders usually change the bit depth from fixed point to floating point anyway, and as such, it does not matter what it originally is. (Note: this does not mean that it reduces the 24bit to 16bits. both end being represented in 32bit float).


So, which is better? Evidently, given what I've said above, the correct way is with the 24bit file, since the 16bit dithered one has added noise. (Which the encoder might ignore or not, depending on its innerworkings)
Go to the top of the page
+Quote Post

Reply to this topicStart new topic
1 User(s) are reading this topic (1 Guests and 0 Anonymous Users)
0 Members:

 



RSS Lo-Fi Version Time is now: 1st August 2014 - 07:55