[K2Grey]

Of course it is interesting that someone feels this way and it may normally be worth investigating to see if there is some unknown problem with the iPod. However, threads of this kind have come up numerous times before with absolutely no supporting evidence for the claim and absolutely no valid & successful ABX comparions (not even any invalid successful ones either*). The most pyrotechnic example of this is one thread in which the poster bashed LAME mp3 as being completely horrible in comparison to WAV and ALAC as being slightly inferior. After an entertaining argument, he managed to arrange some kind of pseudo-blind test (by holding the screen so it was facing away from his eyes) and completely failed to distinguish between LAME alt-present-standard and WAV. However, with good grace, he made a thread giving these results and admitted the role of psychology in his previous listening.

* Valid being a trial over a good number of attempts in which the user doesn't know how what his current percentage correct is. It also helps if it can be done consistently over multiple trials, to minimize the role of luck e.g. 1 case in which a person at HydrogenAudio had a successful trial by randomly clicking, as his sound was on mute, or his headphones were off, or some other situation like that.

 

[tyrion]

I tried to clean this thread up a bit. Lets stay on topic and not attack each other.

 

[Febs]

(edit: in light of Tyrion's cleanup of the thread, my original first paragraph is not necessary.)

Quote:

Originally Posted by Riboge So why wouldn't others of you try it out? And if you heard a difference, why wouldn't you want to report it and abx test yourself as much as wanting the initial reporter here to do it?

What makes you think that we haven't? I've done many ABX tests, and they have confirmed to me that there is no perceptible, audible difference between a well-encoded MP3 and the original, let alone a lossless file and the original.

 

[Riboge]

Quote:

Originally Posted by Febs (edit: in light of Tyrion's cleanup of the thread, my original first paragraph is not necessary.) What makes you think that we haven't? I've done many ABX tests, and they have confirmed to me that there is no perceptible, audible difference between a well-encoded MP3 and the original, let alone a lossless file and the original.

What makes me think it is that you didn't say so until now. Was I supposed to have pearls you strew before us in memory from other threads? Thanks for finally coming out with something experience-based instead of all the 'reasoning'. I have appreciated many other posts of yours I have come across so was disappointed in how you reacted to this one. But then you were highly provoked ad hominem after that so it all got even worse.

 

[Febs]

Quote:

Originally Posted by Riboge What makes me think it is that you didn't say so until now. Was I supposed to have pearls you strew before us in memory from other threads? Thanks for finally coming out with something experience-based instead of all the 'reasoning'. I have appreciated many other posts of yours I have come across so was disappointed in how you reacted to this one.

Fair enough. I assumed that it was implicit in some of my other posts in this thread, but I could have been more clear about it. So, for the record, I have done quite a few ABX tests. I do not perceive a difference between even a well encoded MP3 and the original file. I have in the past thought that I could hear a difference between certain lossy files and the original, and through ABX testing I have discovered that whatever differences I heard were either (a) attributable to something else, like a volume difference, (b) the result of files encoded using inferior encoders, rather than a systemic problem with the format or quality level. (In other words, I realized that some observations that I had made and attributed to the mp3 format or to a bitrate had more to do with a particular poorly-encoded file than limitations in the format.)

I do believe that there are people here at Head-Fi who can hear a difference between MP3s and lossless, even at relatively high bitrates. However, I also believe that those people are exceedingly rare, and that for every one person who can do so, there are probably 100 people who think that they can do so, but cannot.

Having learned for myself, through listening, that lossy compression schemes are better than I once thought they were, I am skeptical when I hear someone claim that they can hear a difference between a well-encoded lossy file and an uncompressed file. That is not to say that I don't believe them; however, I understand from my own experience that what people perceive to be an audible difference can result from many things, including differences in volume and the dreaded "placebo effect," rather than actual differences in sound quality. Thus, I believe that it is not helpful to attempt to qualify differences in sound until it is established that what the listener is hearing is an actual audible difference, and not the result of some other factor.

I am more skeptical when I hear someone claim that there is an audible difference between a lossless file and the uncompressed original. Lossy encoders, by definition, discard data and use perceptual encoding schemes to discard data that is inaudible. As it is beyond dispute that lossy encoders discard information, it is reasonable to believe that there are instances when there are an audible differences between lossy files and the original, and that such differences will be perceptible to a portion of the population (albeit a small one).

With lossless compression schemes, on the other hand, it is easy to demonstrate that the compression algorithim discards nothing, and that the compressed file can be uncompressed to a bit-perfect version of the original. Therefore, I believe that it is reasonable to be skeptical of claims that there is an audible difference between a lossless file and the uncompressed original. Nevertheless, I am willing to acknowledge the possibility, however unlikely, that someone can hear a difference between the two, as the result of a problem with the decoder or something similar. In light of the fact that lossless encoding schemes are demonstrably lossless, I do think that it is reasonable for me to remain extremely skeptical of those claiming to hear a difference until they can show that they can do so in a double-blind test.

Quote:

But then you were highly provoked ad hominem after that so it all got even worse.

Indeed. But we can only hope that is done and we can carry on a more productive conversation.

 

[ObiHuang]

http://en.wikipedia.org/wiki/Placebo_effect

 

[jjcha]

Quote:

Originally Posted by gdg Do you know what EAC is? It's a ripper that can read and reread a CD up to 50 times in order to get the data correct. Your CD player can only read once and when there is an problem (due to a scratch, a poor pressing, laser tacking error, warps on the CD etc etc etc) error correction algorthms extrapolate a best guess for the missing data. As a result, a CD ripped using EAC may actually be more accurate (and often does sound better) than the data stream from the single pass your CDP is capable of . You need to keep an open mind. Ps Thes read errors that occur during real time CD playback are the reason mega buck transports worth $5000-$10000 were developed.

I hope people reading this thread don't take this post seriously. The redbook CD standard has a robust error correction standard within it. I own a mega-bucks transport, and it's not because of read-errors. It has a lot more to do with jitter, upsampling, form factor than anything else. Quite honestly, any transport worth 5 bucks is quite capable of reading CDs without any read-errors. Spending good money on transports is about something other than that.

As long as we're on the topic, I can discern no difference between apple lossless and wav. I find the proposition that there's a difference to be silly.

Best,

-Jason

 

[Scrith]

Uninstall iTunes. Problem solved.

 

[Nospam]

Quote:

Originally Posted by Riboge It seems to me this is a sad sort of thread--like many others, including one or two where I too offered a puzzling observation hoping others would share their experience in exchange so we could see if there was enough corroboration to warrant careful testing. Instead we get many versions of scornful and doctrinaire explanations about why it couldn't be so or other forms of peremptory explaining it away in lieu of observations by others....

Well, let's see now. The original poster starts a thread with a contentious title, dismissing established principles that Lossless means you don't lose a single bit of data, and rejects multiple, very polite requests to provide something to back up his claim.

Gee, I wonder why people seem a bit dismissive...

If he had presented this as something along the lines of "Hey guys, I'm hearing what seems to be a distinct difference between ALAC and WAV. Anyone else notice this?", I think the tone of the responses probably would've been quite different.

 

[gdg]

Quote:

Originally Posted by jjcha I hope people reading this thread don't take this post seriously. The redbook CD standard has a robust error correction standard within it. I own a mega-bucks transport, and it's not because of read-errors. It has a lot more to do with jitter, upsampling, form factor than anything else. Quite honestly, any transport worth 5 bucks is quite capable of reading CDs without any read-errors. Spending good money on transports is about something other than that. As long as we're on the topic, I can discern no difference between apple lossless and wav. I find the proposition that there's a difference to be silly. Best, -Jason

I hope no reading the above comment takes it as gospel from a person with ZERO technical backround. Where do you even come up with a comment like that? Do you even know what "robust" error correction means? The error correction in a CDP is done after the fact with mathematical extrapolation software because it can't be done in real time. Explain to me how the "robust" error correction of a CDP goes back and "corrects" for a tiny scratch or blemish on the surface of the disc or a mistrack due to vibration? You can't go back and reread a damaged section on the fly. Computers can accomplish this when ripping and playing back because they is not accessing the data in real time. As for jitter, that is certainly part of what high end transports improve on but it is not the whole story by any means. In fact when the first high end transports came out little was even known about jitter. Furthermore many of the top transports invest heavily on in improved vibration control and better quality mechanical drive and read mechanisms. If this wasn't the case they'd all be using Philip CDM12 mechanisms.

 

[gdg]

Quote:

Originally Posted by Riboge But then you were highly provoked ad hominem after that so it all got even worse.

You must be joking. Who provoked who. First this guy Febs demands corroboration even though he has yet to explain how a AB test on a computer will prove anything about what I observed on an Ipod. Then he posts this:
Quote:

Originally Posted by Febs Hey, guess what? I just developed x-ray vision. No, really. I can read what's written on the other side of that wall. What? No, of course I'm not going to read it to you. Why should I? I know that I can see it. If you want to see it, use your own eyes.

I ask again who provoked who?

Quote:

Originally Posted by Nospam Well, let's see now. The original poster starts a thread with a contentious title, dismissing established principles that Lossless means you don't lose a single bit of data, and rejects multiple, very polite requests to provide something to back up his claim.

First of all I don't give a flying screw about "established" principles in your little tea party. Secondly exactly who are you trying to kid with the implication that I started off being rude to people making pleasant and well meaning comments? I have been rude only in response to ridicule and attack. Find a single instance were I responded inappropriately to anyone making a polite comment.

Quote:

Originally Posted by Nospam Reminds me of a thread last year where someone claimed WAV files didn't sound as good as playing off of the original CD

What is clear is that you don't mind dishing out the snotty attitude but don't like it when it's fired back.

Finally, since you can't seem to read and like to twist my comments to suit your purposes I'll repeat myself one last time...

I have accepted that in theory, mathematically there should be no difference between WAV and ALAC. I then went on to point out that an AB test on a computer (whether it's a listening test or an actual comparison of digital data) does not prove or disprove what I heard on an Ipod. In theory all amps should behave like straight wires with gain and all interconnects should sound the same and blah blah blah...
By the way you remind me of the guy who insisted that it was commonly "established" that the earth was flat.

More to the point...what is "established" and "accepted" here can be quite different from that at the more hard core Audio Asylum or at the even more hardcore Tact Hackers Group. The fact that something is widely accepted generally means #*&%.

 

[russdog]

Reading this thread makes me feel like I'm in bar, watching 2 or 3 drunk guys have a really loud, really dumb argument that just keeps going on for a really long time.

 

[SH4RKY]

Haha, I'm just glad I dont understand most of the stuff people have mentioned

Surely listening to discover errors with the files or equipment isnt what this is all about... high end home equipment is there so you can maximise your enoyment with the music, its not made so you can pick the minute differences between one type of cable to another, or one type of compression to another.

Be thankful you can hear at all and are in a postion to be at the 'hi end' of the audio market.

 

[ObiHuang]

Quote:

Originally Posted by gdg I hope no reading the above comment takes it as gospel from a person with ZERO technical backround. Where do you even come up with a comment like that? Do you even know what "robust" error correction means? The error correction in a CDP is done after the fact with mathematical extrapolation software because it can't be done in real time. Explain to me how the "robust" error correction of a CDP goes back and "corrects" for a tiny scratch or blemish on the surface of the disc or a mistrack due to vibration? You can't go back and reread a damaged section on the fly. Computers can accomplish this when ripping and playing back because they is not accessing the data in real time. As for jitter, that is certainly part of what high end transports improve on but it is not the whole story by any means. In fact when the first high end transports came out little was even known about jitter. Furthermore many of the top transports invest heavily on in improved vibration control and better quality mechanical drive and read mechanisms. If this wasn't the case they'd all be using Philip CDM12 mechanisms.

Oh my. You my friend, GRG, are not up to par with the exciting new technology of today! First of all, CDs have a beautiful amount of redundancy built in explained by this wonderful article. http://en.wikipedia.org/wiki/Cross-I...Solomon_Coding

Furthermore, you are not actually listening to a CD “on the fly” and the CD drive does not need to “[go] back and ‘correct’” because there’s a frighteningly clever thing called a buffer, kinda like how computer harddrives have a small 4-8MB buffer it reads to. The joy of reading to a buffer is that it buys a lot of time for the CD player to recheck the CD, trying its hardest to give you a bit-perfect reproduction of your favorite CD! You know that little spinny up period when you first select a track and click play? That’s the CPU giving the CD some time to fill up the buffer. That’s how a CDP goes back and corrects for tiny scratches! No CD player is “on-the-fly”… at least not one I would pay for.

The only time something’s going to mathematically fill in that gap is with heavy scratches, and in that case, you’re not going to make it any better using EAC or whatever. Even cheap CD players have a good 40-seconds of anti-skip technology. That means the player has 40 seconds to figure out what that missing .2 seconds contains... and that CD player is definetely not operating at 1x speed. If it's not readable by then, it's not going to be readable by EAC. That big block of missing data’s going to be stored onto the WAV file on your computer no matter how many times you try to read it, so it won’t make a difference. You’re either going to get a skip/pop or your fancy software will try to fill in that gap, but only in the situation where there is a truely unreadable sector.

I got a huge gripe with all this fuss about "jitter" but I'll save that for another day. Remember folks. Don't buy a fancy CD player for "jitter" worries... buy it for looks, fancy features, and quality parts... not jitter protection

 

[Brent Hutto]

Quote:

Originally Posted by wanderman replaygain is there for a reason

Is replaygain supposed to work on WAV files? I download Foobar v0.9.3.1 plus a couple of command-line encoders for OGG and Musepack. So I ripped a CD to WAV and those two lossy formats and wanted to do an ABX. I could use replaygain on OGG and MPC but not on the WAV file.