IPB

Welcome Guest ( Log In | Register )

Resampling - Rounding Errors and Bit Depth.
zerowalker
post May 19 2013, 22:26
Post #1





Group: Members
Posts: 239
Joined: 6-August 11
Member No.: 92828



Okay so i recently got knowledge about the use of Bit Depth for Resampling.
I noticed that when i resampled old SNES music (which is pretty much, midi sound) i heard noise appearing on occasions.
At first i thought it was there originally, but then i rechecked, and well it wasnīt.

I then asked around about it, and got knowledge about Rounding Errors.
When resampling is done, there will be errors, and at 16bit (which was what i was working at) the errors are noticeable as there isnīt much do work with.
But when working at 32bit Float, the errors are so small that they pretty much donīt exist i guess, at least itīs not audible from my understanding.

But now to the thing that crossed my mind.

If 32bit float is great when working with resampling and such, how is 64bit float?
Is it much better, or isn't there any improvement?
Go to the top of the page
+Quote Post
 
Start new topic
Replies
bennetng
post May 21 2013, 15:53
Post #2





Group: Members
Posts: 215
Joined: 22-December 05
Member No.: 26587



http://www.hydrogenaudio.org/forums/index....howtopic=100951

Noise spectrum of avisynth's resampler (uploaded in post #25 by zerowalker) vs Adobe Auditon 1.5's resampler (convert to 32-bit float before resampling, then convert to 16-bit without dithering)
Red: Audition
Green: avisynth

This post has been edited by bennetng: May 21 2013, 15:56
Go to the top of the page
+Quote Post
zerowalker
post May 21 2013, 21:22
Post #3





Group: Members
Posts: 239
Joined: 6-August 11
Member No.: 92828



What samples should i resample at 24 bit?
And should i convert it to float before, or let it be?

And from looking at the Versus against Adobe Audition 1.5, isnīt it a bit better, or am i wrong?
And it doesnīt seem to be dithered by Avisynth resampler anyway, as it wouldnīt have those peaks if i understood the Instruction video linked earlier.


And just to confirm the SRC graphs, The less noise (Blue and colored lines) the better, is that correct?
Cause if so, the Adobe Audition CS6 seems to be one of the best, and Zinotope.


Will make when i can, but the site http://sox.sourceforge.net/ is currently blocked.

Go to the top of the page
+Quote Post
bennetng
post May 22 2013, 00:24
Post #4





Group: Members
Posts: 215
Joined: 22-December 05
Member No.: 26587



QUOTE (zerowalker @ May 22 2013, 04:22) *
And from looking at the Versus against Adobe Audition 1.5, isnīt it a bit better, or am i wrong?
And it doesnīt seem to be dithered by Avisynth resampler anyway, as it wouldnīt have those peaks if i understood the Instruction video linked earlier.

You are not wrong, when I listened to the 13579.wav you uploaded at high volume (which I feel pain), I can also hear noise at the end. From technical point of view (graphs and statistics) the resampler in SoX is even better then Audition 1.5.

Benefits of dithering depend on the singal to noise ratio of the original signal. If the original signal has enough noise and the signal (the rising tone of 13579.wav in this discussion) is strong enough, there is no need to add extra noise (dither) during processing, therefore I disabled dithering in Audition to show that there is no visible truncation artifacts even when dithering is disabled.

I don't know if avisynth can disable dithering, or use less noise, or use different kinds of noise shaping or not, but it is unfair to say a resampler is a bad one because of higher dithering level. A (relatively) bad resampler can show more than 1 bright line in the graph, like Ableton Live 8.2 shows up in the default page of
http://src.infinitewave.ca/
while a (relatively) good one is like Audacity 2.0.3. An extremely bad one is like Secret Rabbit Code 0.1.2 (ZOH) with bright lines all over the graph.
Go to the top of the page
+Quote Post

Posts in this topic
- zerowalker   Resampling - Rounding Errors and Bit Depth.   May 19 2013, 22:26
- - AndyH-ha   Better for what? Quantization errors are obviously...   May 19 2013, 22:51
|- - zerowalker   QUOTE (AndyH-ha @ May 19 2013, 23:51...   May 19 2013, 22:55
- - jensend   Edit: I wrote the below before seeing Andy-ha...   May 19 2013, 23:24
|- - zerowalker   Probably, i have tried reading, but i guess i shou...   May 20 2013, 00:51
- - saratoga   If its 8 bit then playing it at 16 bit is no diffe...   May 20 2013, 01:49
- - AndyH-ha   Quantization errors are differences between the fl...   May 20 2013, 01:53
|- - zerowalker   QUOTE (AndyH-ha @ May 20 2013, 02:53...   May 20 2013, 02:09
|- - bennetng   QUOTE (zerowalker @ May 20 2013, 09:09) h...   May 20 2013, 05:55
|- - zerowalker   QUOTE (bennetng @ May 20 2013, 06:55) QUO...   May 20 2013, 07:52
|- - bennetng   QUOTE (zerowalker @ May 20 2013, 14:52) Q...   May 20 2013, 08:07
|- - zerowalker   Windows 8, donīt know the resampler itself, but it...   May 20 2013, 08:17
|- - bennetng   QUOTE (zerowalker @ May 20 2013, 15:17) W...   May 20 2013, 10:19
|- - zerowalker   I donīt know how high volume i should have on 5432...   May 20 2013, 19:38
|- - bennetng   QUOTE (zerowalker @ May 21 2013, 02:38) 1...   May 21 2013, 01:57
|- - zerowalker   I only hear it after, so 2. I did upload it here:...   May 21 2013, 02:01
|- - bennetng   QUOTE (zerowalker @ May 21 2013, 09:01) I...   May 21 2013, 02:15
||- - zerowalker   QUOTE (bennetng @ May 21 2013, 03:15) QUO...   May 21 2013, 02:24
||- - bennetng   Some resampler comparisons: http://src.infinitewav...   May 21 2013, 02:38
||- - zerowalker   Can i make one such graph to try the resampler i u...   May 21 2013, 02:58
||- - phofman   QUOTE (zerowalker @ May 21 2013, 03:58) C...   May 21 2013, 08:50
|- - jensend   I don't hear any problem with your resampled a...   May 21 2013, 06:27
|- - zerowalker   Really? Donīt you hear any noise at all;S? Cause ...   May 21 2013, 07:37
|- - probedb   I'm not sure what you mean by noise in that cl...   May 21 2013, 08:10
|- - zerowalker   Increase your volume, listen at 4 sec towards the ...   May 21 2013, 08:22
|- - probedb   QUOTE (zerowalker @ May 21 2013, 08:22) I...   May 21 2013, 11:36
|- - zerowalker   QUOTE (probedb @ May 21 2013, 12:36) QUOT...   May 21 2013, 21:13
|- - lvqcl   QUOTE (zerowalker @ May 22 2013, 00:13) B...   May 21 2013, 21:25
|- - zerowalker   QUOTE (lvqcl @ May 21 2013, 22:25) QUOTE ...   May 22 2013, 01:14
|- - Brazil2   QUOTE (zerowalker @ May 22 2013, 02:14) Q...   May 22 2013, 09:14
|- - phofman   QUOTE (zerowalker @ May 22 2013, 02:14) W...   May 22 2013, 12:37
- - saratoga   I doubt you can even find a 16 bit resampler. Anyt...   May 20 2013, 03:18
- - AndyH-ha   QUOTE What do you mean that one 1 large error rema...   May 20 2013, 07:05
- - AndyH-ha   Quantization errors never "go away". The...   May 20 2013, 10:10
|- - zerowalker   QUOTE (AndyH-ha @ May 20 2013, 11:10...   May 20 2013, 19:41
|- - jensend   QUOTE (zerowalker @ May 20 2013, 12:41) I...   May 21 2013, 00:09
|- - zerowalker   By changing the noise and moving it so it doesnīt ...   May 21 2013, 01:18
- - AndyH-ha   Some companies have made a big advertising thing a...   May 20 2013, 21:23
- - bennetng   http://src.infinitewave.ca/faq.html QUOTE Absolute...   May 21 2013, 03:29
- - jensend   BTW I'm sorry if my earlier response sounded h...   May 21 2013, 07:04
- - bennetng   Source and result files in src.infinitewave.ca are...   May 21 2013, 10:23
- - lvqcl   Spectrogram for the file uploaded in post #25 (sox...   May 21 2013, 10:52
- - bennetng   http://www.hydrogenaudio.org/forums/index....howto...   May 21 2013, 15:53
|- - zerowalker   What samples should i resample at 24 bit? And shou...   May 21 2013, 21:22
|- - saratoga   QUOTE (zerowalker @ May 21 2013, 16:22) W...   May 21 2013, 21:38
|- - phofman   QUOTE (zerowalker @ May 21 2013, 22:22) W...   May 21 2013, 22:01
|- - bennetng   QUOTE (zerowalker @ May 22 2013, 04:22) A...   May 22 2013, 00:24
|- - zerowalker   Okay i compared Avisynths Resampler at both Float ...   May 22 2013, 10:11
- - Nick.C   Try this sample - if you hear anything except dial...   May 22 2013, 10:12
|- - Dynamic   QUOTE (Nick.C @ May 22 2013, 09:12) Try t...   May 22 2013, 10:35
|- - zerowalker   QUOTE (Nick.C @ May 22 2013, 11:12) Try t...   May 22 2013, 11:00
||- - bennetng   QUOTE (zerowalker @ May 22 2013, 18:00) I...   May 22 2013, 11:16
||- - zerowalker   QUOTE (bennetng @ May 22 2013, 12:16) QUO...   May 23 2013, 03:28
||- - bennetng   QUOTE (zerowalker @ May 23 2013, 10:28) W...   May 23 2013, 04:07
||- - zerowalker   QUOTE (bennetng @ May 23 2013, 05:07) QUO...   May 23 2013, 04:38
||- - probedb   QUOTE (zerowalker @ May 23 2013, 04:38) I...   May 23 2013, 08:04
||- - zerowalker   QUOTE (probedb @ May 23 2013, 09:04) QUOT...   May 23 2013, 08:12
|- - probedb   QUOTE (Nick.C @ May 22 2013, 10:12) Try t...   May 22 2013, 11:18
|- - Ferongr   QUOTE (Nick.C @ May 22 2013, 11:12) Try t...   May 22 2013, 12:49
- - bennetng   Just think that udial is more useful to detect cli...   May 22 2013, 10:44
- - bennetng   Then it seems that you can pass Ethan Winer's ...   May 23 2013, 06:00
|- - zerowalker   QUOTE (bennetng @ May 23 2013, 07:00) The...   May 23 2013, 06:16
- - bennetng   I think zerowalker's system should be fine. 13...   May 23 2013, 10:47


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: 20th April 2014 - 16:53