Skip to main content

Topic: How do I test whether a burned CD sounds the same as a purchased CD? (Read 3684 times) previous topic - next topic

0 Members and 1 Guest are viewing this topic.
How do I test whether a burned CD sounds the same as a purchased CD?
How do I test whether the burned CD sounds the same as the bought CD? Is there a program that shows graphs of each CD so I can compare peaks?
I used EAC, and set my offset values.


Thanks!

  • pdq
  • [*][*][*][*][*]
How do I test whether a burned CD sounds the same as a purchased CD?
Reply #1
As I recall, EAC has a tool that lets you compare the digital data of two wav files (haven't used EAC myself in many yaers). If the digital data is the same then the discs sound the same.

  • db1989
  • [*][*][*][*][*]
  • Global Moderator
How do I test whether a burned CD sounds the same as a purchased CD?
Reply #2
And even if there are some differences, you need to perform a double-blind test to prove that they are audible; subjective impressions based upon sighted comparisons are completely unreliable.

Is there a program that shows graphs of each CD so I can compare peaks?
This is also no use as a way of comparing different instances of (what should be) the same audio.

This is based specifically on your use of the phrase “sounds the same”, so I apologise if you were just using that to mean ‘has the same data’. If not, #8 of our Terms of Service applies.
  • Last Edit: 17 March, 2013, 12:47:58 PM by db1989

  • kennedyb4
  • [*][*][*][*][*]
How do I test whether a burned CD sounds the same as a purchased CD?
Reply #3
I believe it was Sebastian that determined the output was bit perfect by some spdif capture and analysis procedure. It was settled a long time ago in any event.

  • db1989
  • [*][*][*][*][*]
  • Global Moderator
How do I test whether a burned CD sounds the same as a purchased CD?
Reply #4
What output? Are you sure you replied to the right thread? I find it hard to believe that someone determined in advance that all CD rips ever are bit-identical to each other.

  • knutinh
  • [*][*][*][*][*]
How do I test whether a burned CD sounds the same as a purchased CD?
Reply #5
I believe it was Sebastian that determined the output was bit perfect by some spdif capture and analysis procedure. It was settled a long time ago in any event.

I once tested the spdif output of an old Sony dvd player playing a CD-R copy and compared it with rips of the original CD. They were exact (after aligning them).

Not sure if this answers the OP though. Accuraterip should be a pretty good judge if it is a popular recording.

-k
  • Last Edit: 18 March, 2013, 04:30:18 AM by knutinh

  • Porcus
  • [*][*][*][*][*]
How do I test whether a burned CD sounds the same as a purchased CD?
Reply #6
I once tested the spdif output of an old Sony dvd player playing a CD-R copy and compared it with rips of the original CD. They were exact (after aligning them).


Is there any software that auto-aligns? Even, auto-aligns to least errors? Or did you have to do the dirty job manually?
  • Last Edit: 18 March, 2013, 12:07:03 PM by Porcus

  • greynol
  • [*][*][*][*][*]
  • Global Moderator
How do I test whether a burned CD sounds the same as a purchased CD?
Reply #7
If the output is saved to a wave file, I don't see how time-aligning is particularly difficult.  Back when I used to measure the offset between different pressings of various CD titles using CEP/Audition this was actually quite trivial.

Comparing data between these pressings (or burned copies against the original for that matter) was quite trivial as well.

I hope it doesn't come as a surprise to anyone, but every single burned copy I've ripped and verified against the original source material revealed no differences.  With this in mind, I've never seen the point in verifying S/PDIF output from a standalone CD player.  If people want to erroneously believe that there are differences in audio playback between discs that rip the same (in terms of "air," "sound stage," frequency response, dynamic range, "PRAT" or any other typical bs audiotomfoolery) that's their business.
  • Last Edit: 18 March, 2013, 12:45:56 PM by greynol
Your eyes cannot hear.

  • dhromed
  • [*][*][*][*][*]
How do I test whether a burned CD sounds the same as a purchased CD?
Reply #8
What if I have a bunch of pristine FLACs from a previous rip, and just lackadaisically tell my burninator to make them into an audio CD?

I thought things like pregap, indexes, frame padding in the files and more such format details would make bit-perfect copies difficult. But given the replies in this thread, I'm probably wrong about that.

  • greynol
  • [*][*][*][*][*]
  • Global Moderator
How do I test whether a burned CD sounds the same as a purchased CD?
Reply #9
We're talking about time-aligning streamed audio.  All that other stuff really doesn't apply.  The only thing you should worry about is whether your burning program is altering the PCM data prior to burning.
Your eyes cannot hear.

  • db1989
  • [*][*][*][*][*]
  • Global Moderator
How do I test whether a burned CD sounds the same as a purchased CD?
Reply #10
What if I have a bunch of pristine FLACs from a previous rip, and just lackadaisically tell my burninator to make them into an audio CD?

I thought things like pregap, indexes, frame padding in the files and more such format details would make bit-perfect copies difficult. But given the replies in this thread, I'm probably wrong about that.
Burning software could only get things like this wrong if it was hopelessly badly coded, in which case, you’d have plenty other problems besides. Indexes and pregaps (and offsets) don’t affect the burned audio, only the seek-points written to the ToC/PQ subchannels. As for lossless formats, any worthwhile ones will by definition store completely accurate lengths to enable restoration of the original waveform, making concerns about frames and padding completely unfounded.

  • greynol
  • [*][*][*][*][*]
  • Global Moderator
How do I test whether a burned CD sounds the same as a purchased CD?
Reply #11
There have been rippers that have not properly reproduced correct track lengths or would trim silence by default (not necessarily to frame boundaries) but again, this really isn't going to affect sound quality in the ways I previously mentioned.
Your eyes cannot hear.

  • krabapple
  • [*][*][*][*][*]
How do I test whether a burned CD sounds the same as a purchased CD?
Reply #12
I once tested the spdif output of an old Sony dvd player playing a CD-R copy and compared it with rips of the original CD. They were exact (after aligning them).


Is there any software that auto-aligns? Even, auto-aligns to least errors? Or did you have to do the dirty job manually?


Audio Diffmaker

http://www.libinst.com/Audio%20DiffMaker.htm

  • mitchco
  • [*]
How do I test whether a burned CD sounds the same as a purchased CD?
Reply #13
I once tested the spdif output of an old Sony dvd player playing a CD-R copy and compared it with rips of the original CD. They were exact (after aligning them).


Is there any software that auto-aligns? Even, auto-aligns to least errors? Or did you have to do the dirty job manually?


Audio Diffmaker

http://www.libinst.com/Audio%20DiffMaker.htm


+1 on Audio DiffMaker.  While not exactly the same scenario, here is one example: http://www.computeraudiophile.com/content/...ity-comparison/

  • db1989
  • [*][*][*][*][*]
  • Global Moderator
How do I test whether a burned CD sounds the same as a purchased CD?
Reply #14
Oh boy. While he arrived inevitably at the right conclusions, and I suppose it’s good to see fanciful causes of differences debunked in a publication with “Audiophile” in its title, I think that guy wasted a lot of his own time by giving the proposed differences any credibility at all.

  • knutinh
  • [*][*][*][*][*]
How do I test whether a burned CD sounds the same as a purchased CD?
Reply #15
Is there any software that auto-aligns? Even, auto-aligns to least errors? Or did you have to do the dirty job manually?

There might be software for this. It was so simply to do manually that I frankly did not consider it. The graph below shows the first few samples after aligning, one shifted for visual clarity. It really is easy to find the right shift.



This graph shows the difference-vector after alignement:


(using MATLAB you can do something ala err = sum(abs(vectA - vectB)) to get a numerical answer).

Finally, I did a check using the analog output and quickly figured out that everything becomes more complicated. The plot below shows the low-frequency time offset using xcorrelation within bins of a wav-file vs the same data played out of the analog terminals of my DVD player into the analog terminals of my sound card:



I guess what one really want for this last test is a sort of PLL that separates (according to some criteria) differences between the two waveforms as either "jitter/clockdrift" or "noise, distortion ++".
  • Last Edit: 19 March, 2013, 05:03:06 AM by knutinh

  • Porcus
  • [*][*][*][*][*]
How do I test whether a burned CD sounds the same as a purchased CD?
Reply #16
http://translate.google.no/#en/no/Thank%20...e%20information @ krabapple and knutinh.

(It is easy for one pair with leading silence, of course, but I was after something for batch processing.)
  • Last Edit: 19 March, 2013, 06:10:12 AM by Porcus

  • knutinh
  • [*][*][*][*][*]
How do I test whether a burned CD sounds the same as a purchased CD?
Reply #17
(It is easy for one pair with leading silence, of course, but I was after something for batch processing.)

To a man with a hammer, every problem is a nail: I would make a Nx2 structure of filenames in MATLAB (or generate is using the appropriate file commands such as 'dir'), traverse it in a loop, load pairs of files, use xcorr(a,b) to find the index of the peak, pre/postpad by silence so they are equal length, then do whatever aligned processing I wanted to do.

To someone with more Linux scripting skills than myself, I guess that the solution might look differently.

-k

  • krabapple
  • [*][*][*][*][*]
How do I test whether a burned CD sounds the same as a purchased CD?
Reply #18
Oh boy. While he arrived inevitably at the right conclusions, and I suppose it’s good to see fanciful causes of differences debunked in a publication with “Audiophile” in its title, I think that guy wasted a lot of his own time by giving the proposed differences any credibility at all.



For the CA crowd, that's necessary,  because they *do* find those differences credible.