R/E/P Community

Please login or register.

Login with username, password and session length
Advanced search  

Pages: [1]   Go Down

Author Topic: Understanding Pre-Ringing  (Read 11175 times)

eightyeightkeys

  • Hero Member
  • *****
  • Offline Offline
  • Posts: 789
Understanding Pre-Ringing
« on: December 05, 2010, 10:55:26 AM »

How much do modern a/d d/a convertors still suffer from pre-ringing ?

Has it been eliminated or drastically reduced in modern convertors ?

If you drop more money on convertors do you necessarily get a better result, or, is pre-ringing simply a "fact" of digital audio that simply does not exist in analog ?

'Just trying to understand this particular issue.
Logged
Dave T.
D&D Music

John Roberts {JR}

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 266
Re: Understanding Pre-Ringing
« Reply #1 on: December 05, 2010, 01:23:02 PM »

Google "Gibbs phenomenon"..

JR
Logged

cgc

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 329
Re: Understanding Pre-Ringing
« Reply #2 on: December 05, 2010, 05:29:20 PM »

The classic Gibbs example is a square wave at full scale and close to one half the sampling frequency.  Are you recording a lot of stuff like that?  I think most of the filters used in converters handle all but the most extreme cases very well.

It comes more into play for image processing and FFT based lossy compression like MP3 where pre-echo can be a problem.

A previous discussion:

http://recforums.prosoundweb.com/index.php/mv/msg/20816/0/0/ 6936/

Bruno knows a lot about this and can explain how it affects the real world better.
Logged

bruno putzeys

  • Hero Member
  • *****
  • Offline Offline
  • Posts: 1078
Re: Understanding Pre-Ringing
« Reply #3 on: December 06, 2010, 03:21:16 AM »

Gibbs phenomenon indeed and fundamentally you're stuck with it. It's not just "fact", it's a mathematical truth. Pre-ringing is a necessary result of having a sharp low-pass filter with a linear phase response. Actually, very old converters had no pre-ringing because they did not have linear phase. They had tons of post-ringing instead.
Eliminating ringing, pre or post, requires shallow filters and/or non-linear-phase filters. If you still want a flat response to about 20kHz and no aliasing/imaging, this implies a higher sampling rate. Peter Craven pointed out that in a high sampling rate system, all filters could be made sharp at fs/2 and that a single slow-roll filter starting at 20kHz then reduces the ringing. You don't want all filters to be slow rolloff because a cascade of slow rolloff filters becomes a sharp filter so there would be no point in converter manufacturers including them as standard. If you want to control the impulse response of the system, you should use exactly one slow filter (and a release medium of at least 96kHz), preferably during mastering. He calls this type of filter "apodizing".

Of course the question is, with 96kHz, is pre-ringing at 48kHz an issue at all, what with most loudspeakers imposing an apodising filter of their own (not to mention ears)?
Logged
Warp Drive. Tractor Beam. Room Correction. Whatever.

Affiliations: Hypex, Grimm Audio.

KB_S1

  • Hero Member
  • *****
  • Offline Offline
  • Posts: 931
Re: Understanding Pre-Ringing
« Reply #4 on: December 06, 2010, 06:15:22 AM »

Bruno, is this something that is avoided by the DSD systems?
Logged
<a href="http://www.parklanerecordingstudios.com/" class="link3">Park Lane Studio</a> Where to find me most of the time<br /><br />

<a href="http://www.flickr.com/photos/kb_s1/" class="link3">Flickr</a>where to see what I have been up to  <br /><br />

John Roberts {JR}

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 266
Re: Understanding Pre-Ringing
« Reply #5 on: December 06, 2010, 09:39:54 AM »

KB_S1 wrote on Mon, 06 December 2010 05:15

Bruno, is this something that is avoided by the DSD systems?


I'm no Bruno...  Cool

But to understand Gibbs, you need to look at the spectral content of square waves. The pre-ringing (actually not a very accurate term IMO). Is in fact the clean removal of higher frequency components.

I wouldn't lose any sleep over it.

JR
Logged

bruno putzeys

  • Hero Member
  • *****
  • Offline Offline
  • Posts: 1078
Re: Understanding Pre-Ringing
« Reply #6 on: December 06, 2010, 10:16:39 AM »

Because DSD has such a high sampling rate you can live with much more modest filtering. This is basically the selling feature of DSD. Whether that should make it a buying feature is another matter. If you are worried about audibility of preringing, move up to 96kHz and rest assured that your speakers and ears will do the rest.

And if you are really really seriously worried about preringing, do listening tests using off-line sample rate converters (e.g. saracon) with 96kHz source material (going to 44.1kHz and back, and compare with the 96kHz original). You'll be amazed at how small the difference can be if the filters are good.

The difference between good and bad digital filters does not lie in the pre-ringing (that is only a mathematical necessity), but in just about every other aspect of it (choice of corner frequencies, inband ripple etc). It's not because pre-ringing is the most visible feature of digital audio that it is the most audible one.
Logged
Warp Drive. Tractor Beam. Room Correction. Whatever.

Affiliations: Hypex, Grimm Audio.

eightyeightkeys

  • Hero Member
  • *****
  • Offline Offline
  • Posts: 789
Re: Understanding Pre-Ringing
« Reply #7 on: December 06, 2010, 01:25:12 PM »

Is pre-ringing cumulative ?

That is, the more tracks that you record, the more transient smear that we get ?

Is this possibly why digital has always sounded so much better, more open, better depth, etc... with less tracks than with more ?
Logged
Dave T.
D&D Music

dcollins

  • Hero Member
  • *****
  • Offline Offline
  • Posts: 2815
Re: Understanding Pre-Ringing
« Reply #8 on: December 06, 2010, 02:31:10 PM »

bruno putzeys wrote on Mon, 06 December 2010 07:16


And if you are really really seriously worried about preringing, do listening tests using off-line sample rate converters (e.g. saracon) with 96kHz source material (going to 44.1kHz and back, and compare with the 96kHz original). You'll be amazed at how small the difference can be if the filters are good.



Fwiw, we have just gone through tests using apodising filters vs conventional FIR (in a DAC, mainly at 44k1) and I was surprised and how little difference there was.

I would give it a letter grade of "meh."


DC

Geoff Emerick de Fake

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 348
Re: Understanding Pre-Ringing
« Reply #9 on: December 06, 2010, 06:05:24 PM »

eightyeightkeys wrote on Mon, 06 December 2010 12:25

Is pre-ringing cumulative ?

That is, the more tracks that you record, the more transient smear that we get ?

Is this possibly why digital has always sounded so much better, more open, better depth, etc... with less tracks than with more ?
Cumulative effects is not about adding tracks. It is about increasing the number of process a signal is going through.  Going several times A/D/D/A is cumulative. Noise and rise-time, for example, will increase according to the square-root of the number of cumulative conversions.
Adding tracks is generally the contrary of cumulative, because the signals are more or less correlated, so they sum better than the artifacts, that are mostly non-correlated. The overall signal-to-noise ratio of a mix is generally better than the signal-to-noise ratio of an individual track.
Logged

bruno putzeys

  • Hero Member
  • *****
  • Offline Offline
  • Posts: 1078
Re: Understanding Pre-Ringing
« Reply #10 on: December 07, 2010, 03:14:07 AM »

Pre-ringing is not cumulative. A long string of linear-phase brick-wall filters is still a brick-wall filter. The impulse response remains a sin(x)/x function.

That said, inband ripple DOES cumulate, which is why skimping on inband ripple specs is a bad ideal.
Logged
Warp Drive. Tractor Beam. Room Correction. Whatever.

Affiliations: Hypex, Grimm Audio.
Pages: [1]   Go Up
 

Site Hosted By Ashdown Technologies, Inc.

Page created in 0.036 seconds with 19 queries.