Head-Fi.org › Forums › Equipment Forums › Dedicated Source Components › iTunes does not rip accurate audio data
New Posts  All Forums:Forum Nav:

iTunes does not rip accurate audio data - Page 10

post #136 of 199
Quote:
Originally Posted by hciman77 View Post
I just compared those two files in Audacity - they are not perfectly aligned - sadly I cannot post a comparison gif but in Audacity they are slightly misaligned
I've noticed that sometimes when I'm working on a track in Peak and save a few times, it will add a tiny fraction of a second of blank at the beginning of the track. I wonder if one or the other of the ripping programs is doing that.

See ya
Steve
post #137 of 199
Quote:
Originally Posted by bigshot View Post
I've noticed that sometimes when I'm working on a track in Peak and save a few times, it will add a tiny fraction of a second of blank at the beginning of the track. I wonder if one or the other of the ripping programs is doing that.

See ya
Steve
That seems to be the case. EAC has some options for determining and dealing with drive offsets during the ripping process including an option entitled "Fill up missing offset samples with silence" which sounds like it would case exactly what we are seeing.

I think iTunes and EAC will rip the same data given an error free CD. With scratched CDs EAC will be a bit more careful to get it right.
post #138 of 199
Quote:
Originally Posted by IPodPJ View Post
As nearly identical as they may look, I assure you they sound very different. But thank you two for confirming to everyone that they do have different waveforms. I already knew I wasn't crazy, but I guess some people (wink wink Pete) think you are just because they may not be able to hear the differences.
Actually if there were audible differences, I'm probably more capable of hearing them than you are. I aced aural music theory in college. I'm a musician.
post #139 of 199
Quote:
Originally Posted by IPodPJ View Post
As nearly identical as they may look, I assure you they sound very different. But thank you two for confirming to everyone that they do have different waveforms. I already knew I wasn't crazy, but I guess some people (wink wink Pete) think you are just because they may not be able to hear the differences.
I just confirmed using GoldWave that these two files DO NOT HAVE DIFFERENT WAVEFORMS. They have exactly the same waveform offset by six samples.

To test this, I used GoldWave to remove the six extra samples from file 1. As a result, file 1 and file 2 have different lengths, and I therefore could not use Foobar's binary comparator to compare them. Instead, I loaded both the adjusted file 1 and the original file 2 into Goldwave. I inverted the phase of file 2, copied it to the clipboard, and then used GoldWave's "mix" function to mix the contents of the two files. Since the phase of 2 was inverted, the result of the mix function will be to eliminate anything that the two files had in common, leaving only those differences between the two files.

Here is a screenshot of the resulting file in GoldWave, zoomed to the highest Y-axis resolution:



As you can see, the result was absolute silence. Once the six extra samples were removed from the beginning of file one, there was not a single sample difference between file 1 and file 2. In other words, with those six samples removed, the two files are EXACTLY THE SAME.

I don't know what you're hearing, but it is not the result of a difference in waveform between those two files.
post #140 of 199
Quote:
Originally Posted by Febs View Post
I just confirmed using GoldWave that these two files DO NOT HAVE DIFFERENT WAVEFORMS. They have exactly the same waveform offset by six samples.

To test this, I used GoldWave to remove the six extra samples from file 1. As a result, file 1 and file 2 have different lengths, and I therefore could not use Foobar's binary comparator to compare them. Instead, I loaded both the adjusted file 1 and the original file 2 into Goldwave. I inverted the phase of file 2, copied it to the clipboard, and then used GoldWave's "mix" function to mix the contents of the two files. Since the phase of 2 was inverted, the result of the mix function will be to eliminate anything that the two files had in common, leaving only those differences between the two files.

Here is a screenshot of the resulting file in GoldWave, zoomed to the highest Y-axis resolution:



As you can see, the result was absolute silence. Once the six extra samples were removed from the beginning of file one, there was not a single sample difference between file 1 and file 2. In other words, with those six samples removed, the two files are EXACTLY THE SAME.

I don't know what you're hearing, but it is not the result of a difference in waveform between those two files.
Dang, you beat me to it! As I was lying in bed last night and thinking about things (I tend to do this a lot) I realized this would be a good analysis to make and was going to do it this morning. Anyway, thanks for doing it, and the results are what I expected. After spending a good amount of time ABX'ing the tracks, I couldn't tell a difference.
post #141 of 199
You can also check if two wav files are different only by offset by using EAC's wave compare tool. Go to Tools -> Compare Waves.

Also, IIRC, if your CD has pre-emphasis then iTunes will automatically de-emphasize it. EAC will not, hence keeping a bit-perfect copy. This could make the results different.
post #142 of 199
Quote:
Originally Posted by Shiki View Post
You can also check if two wav files are different only by offset by using EAC's wave compare tool. Go to Tools -> Compare Waves.
Interesting--I never knew that tool existed in EAC!

EAC's wave compare tool confirms that the only difference between the files is 6 repeated samples at position 0:00:00.005.
post #143 of 199
Thread Starter 
Quote:
Originally Posted by Pete7 View Post
Actually if there were audible differences, I'm probably more capable of hearing them than you are. I aced aural music theory in college. I'm a musician.
More capable? Oh come on now. What is this, a kindergarten class with tit-for-tat dialogue? I'm a musician too. I've been playing guitar for 15 years. Neener Neener. Many of us are capable of hearing perfect pitch and tone. But that has absolutely nothing to do with hearing a difference in the audio coming out of my computer. Any one of you would be able to hear it.

Quote:
I don't know what you're hearing, but it is not the result of a difference in waveform between those two files.
Febs and dkjohnso:
If the waveforms are identical, perhaps the DAC is interpreting them differently due to timing error or jitter OR the transport is outputting them differently due to timing error or jitter.
But there are obvious sonic differences in the two files on my system which ANY of you would be able to hear. That is how I discovered this whole mess in the first place and why I am spending two weeks re-ripping all of my CDs.

But yes, it is interesting that the six sample offset (which I'm sure is from EAC's AccurateRip) would have any effect on the sound at all. I don't think it does because like I mentioned, the sound quality of an EAC ripped WAV vs. playing the CD in the drive are identical. It's only the iTunes rip that sounds different (and rolled off).

And thanks to everyone for your great detective work!
post #144 of 199
Quote:
Originally Posted by IPodPJ View Post
If the waveforms are identical, perhaps the DAC is interpreting them differently due to timing error or jitter OR the transport is outputting them differently due to timing error or jitter.

Occam's razor is getting awfully dull. If you're playing both files back through the same playback software, there is no reason that any of these things should be happening.

I don't think that there is any difference at all. Try ABXing the files in Foobar.
post #145 of 199
Thread Starter 
Quote:
Originally Posted by Febs View Post
Occam's razor is getting awfully dull. If you're playing both files back through the same playback software, there is no reason that any of these things should be happening.

I don't think that there is any difference at all. Try ABXing the files in Foobar.
You can think what you want. In a thread some time ago I ABXed lossless vs. WAV and got 7 in a row correct and people still doubted (which I eventually concluded had to do with memory allocation in Vista). So you don't have to believe me. I'm perfectly okay with that. I'm not a mission to convince anyone of anything, I only share what I discover. Whether my computer is outputting faulty signals (could be due to Vista) or not, it is still outputting an audible difference. You asked for proof and I showed you the bit-compare. You asked me to upload the files and I did. I can only do so much to satisfy you, and if you're not happy with that there's nothing I can do about it. I'll ABX them later when I get a chance.

And out of curiosity, what headphone equipment are you using to listen to your music on?
post #146 of 199
Quote:
Originally Posted by IPodPJ View Post
You asked for proof and I showed you the bit-compare.
Yes, and then we demonstrated through irrefutable evidence exactly why the binary comparison was different and provided the results of two separate tests, each of which independently and conclusively demonstrates that every single audio sample in these two files is exactly the same, subject only to an offset that is approximately 1/10,000 of a second long.

Quote:
You asked me to upload the files and I did. I can only do so much to satisfy you, and if you're not happy with that there's nothing I can do about it.
It's not a question of me being happy or unhappy. So far, you have come up with at least three different faulty conclusions about why you are supposedly hearing a difference between these two files that contain identical audio data, and we have systematically refuted everyone of your theories and all of your faulty logic. You still claim to hear a difference. Unless you can provide a reasonable explanation or some credible evidence that there is actually a difference, we are at the stage where we are justified in concluding that this is a perfect example of placebo at work.
post #147 of 199
I can picture it now...the next generation of heated debates which follow in the footsteps of "cables make a difference" are going to be about ripping and playback on PC
post #148 of 199
...
post #149 of 199
Thread Starter 
Quote:
Originally Posted by Febs View Post
It's not a question of me being happy or unhappy. So far, you have come up with at least three different faulty conclusions about why you are supposedly hearing a difference between these two files that contain identical audio data, and we have systematically refuted everyone of your theories and all of your faulty logic. You still claim to hear a difference. Unless you can provide a reasonable explanation or some credible evidence that there is actually a difference, we are at the stage where we are justified in concluding that this is a perfect example of placebo at work.
If timing errors and jitter don't contribute to audio differences, why do people choose I2S over SPDIF if they are both bit-perfect? Why do people buy products from companies like Empirical Audio that re-clock? Why do people buy digital or analog RCA cables that yield different sonic flavors? They do so because it makes digital music (which is already compressed from analog) sound more like an original, live, "analog" performance. Whether you feel it sounds better or not is subjective, but there are audible differences that people can hear.

I can't say conlusively what the reason is for the sonic differences that I can hear, but I know it isn't placebo. I'm not an engineer so I can only deduce what might be causing it. Through trial and error, we learn what it might be and what it might not be. You showed that the waveforms were identical, so obviously we can rule that factor out. The only thing left that I can think of is timing error and jitter. But it gets really annoying when someone posts about something they hear and then always has it labeled as placebo just because the other person either can't hear it or repeat the results on their system. We would all be content with iBuds if audio quality wasn't subjective. There's a reason people spend tens of thousands of dollars to buy high-end audio equipment, and it isn't solely for the purpose of emptying their bank account.

One thing I've learned in this hobby is that none of us know everything there is to know about audio at any one given point at time, but that doesn't mean that we are delusional because we can hear subtle differences. Quantum physicists don't even fully understand the nature of electrons (as they have observed single electrons in multiple places at once), so surely you aren't going to tell me that what flows from one end of your system to the other is going to be identical every single time just because a file is called "bit-perfect".
post #150 of 199
Question since we are on the topic. Could having to decompress the audio hurt the digital playback for the computer or transport? Or at worst would it just have breaks in the sound? Could that factor into jitter?
New Posts  All Forums:Forum Nav:
  Return Home
  Back to Forum: Dedicated Source Components
Head-Fi.org › Forums › Equipment Forums › Dedicated Source Components › iTunes does not rip accurate audio data