Any audiophiles here?

http://en.wikipedia.org/wiki/MP3#Encoding_audio

Why should I explain what MP3 compression does? The point of the test is to decide how - and if - these mp3 files sound differently from the original.

^ ok, let's go back to your first post on this page:

Dean wrote: That test has always bothered me and I'm not sure it is as representative or definitive as they would like to think  it is. The problem for me is that the three formats were converted back to Red Book standard (16-bit @ 44.1Hz sampling) so they could be played back on the Marantz CD14 CD-Player - for that I don't find it that surprising that people found it difficult to tell one from the other.

I can't see the logic in that statement. Suppose for the sake of the argument that what you think is happening there during the mp3 encoding ... that would degrade the signal and make it more distinguishable from the original, not less as you suppose.

BTW: The third CD was created by converting the original CD to WAV and then burning it back to CD-R unchanged ... and only the worst audiophiles would claim that this process could affect the audio quality in any way.

As far as your other question is concerned:

Dean wrote: Then what do the 128kbps and 256kbps bitrates do?

These files contain a compressed audio stream. If the source was 16bit/44.1khz before the mp3 compression, the result after decompression will again be 16bit/44.1khz. There is no downsampling happening during encoding, and no upsampling happening during decoding.

^ sorry, that doesn't make any sense at all.

When you rip a CD, the first step is that the 16bit/44.1khz stream is extracted from the CD. Then, a 16bit/44.1khz MP3 is created from that stream. I don't know why you think that there is any upsampling happening ... the MP3 bitrate doesn't have anything to do with the resolution of the signal (16bit) or the sampling frequency (44.1khz).

They really just ripped the CD, created the mp3s from them and then decoded them back to WAV and burned that to CD.

^ what do you mean by "converted back"? The mp3s were simply decoded and the resulting WAV burned to CD ... there is no signal degradation involved in that process. And of course they had to do that ... how else could you conduct an unbiased test? The sources had to look totally identical to the contestants.

http://www.geocities.com/altbinariessoundsmusicclassical/mp3test.html

I'm never getting tired of posting links to this test ...

I think that it's very difficult to conduct a truly objective test on this. So far I've heard many people say that they can hear those differences, but frankly I don't believe any of them. Please don't be offended by that ... I have my own ears, I think they're in quite good shape and I'm a skilled listener, and I can't hear a difference between CD and properly ripped mp3s. Neither could the guys that did the c't test. My own experience, combined with the test, far outweigh your statements - from my perspective of course.

BTW: I think it's entirely possible that an expensive CD player can sound better than FLACs played on a computer. It doesn't necessarily mean that the expensive player is "more audiophile" though. Often those expensive players included filters and "upsampling" circuitry that enhances the signal (makes it more "smooth"). Be that as it may, such enhancements are not part of the original audiophile philosophy, which is to preserve the original signal as accurately as possible.

I got a great sound card for my pc about a year ago; that along with my fav winamp sound enhancement plugins.... i can hardly stand to listen to music in my truck anymore.


My precious 24 bit copies of PFM's Storia Un di Minuto and Per Un Amico Gold.... you can hear them breathing for goodness sake.

Mr ProgFreak wrote: There have been many studies that show that even on very good equipment it's often not possible to tell the original (CD/FLAC) apart from high bitrate MP3.

I suppose i sound like a typical audiophile diva here, but i dl MP3's (320 kbps) and purchase CD's quite frequently, and i can always tell the difference

Of course, i've had many people tell me that they can't hear the difference between mp3 (even high bit rate) and lossless quality. Perhaps i'm taking crazy pills or something.




Edited by Dominic - July 03 2009 at 02:29

Mr ProgFreak wrote: Dean wrote: Of course as the signals get even smaller the digital distortion will get bigger (and the effective resolution smaller), while the analogue will remain undistorted. Well, I'm not so sure about that. Even if that is the case, I've yet to hear this distortion. And I listen to many classical recordings on CD, which have some *very* silent parts. If the distortion happens like you describe, I'd say that those parts are still well above those 30dB distance from the noise floor. While I certainly agree that this distortion is not acceptable to a true audiophile (any degradation of the signal is not acceptable) ... but I have to ask: What difference does it make if it only happens for parts of the signal that we cannot hear?

That's kind of the point: all these numbers that people throw around to show how good a system is are ideal numbers created under unrealistic conditions and do not bear any resemblence to reality. The values are based upon single tones - music is a complex waveform that is naturally rich in harmonics so adding to those is not going to distract from your listenning pleasure unless it is extreme and harsh (eg clipping or excessive crossover distortion) - in my hyperthetical example the quiet passages are "only" 50dB below peak - but that is actually quite a lot (it is a power ration of 100,000:1) when most music has a range of no more than 20dB (100:1) and typically 10dB (10:1). If you distort a single piano tone by 3% it will still sound like a very accpetable piano - it won't sound like a Steinway or a Bosendorfer anymore but it will still sound like piano and not an electric organ.

Mr ProgFreak wrote: One comment regarding the distortion of digital signals approaching the noise floor: Those signals are also very, very silent. I think that a vinyl pressing would have similar problems?

Wow Dean, thanks for the schooling. I never understood the science behind what my ears were telling me. It's interesting to note the proliferation of DAC and digital audio on the audiophile sites. The vast majority agree that LPs sound better, but the convenience of a music server with FLAC files for many is worth the tradeoff, and high end DACs are getting very, very good.

Personally, I'll keep what I have on vinyl, but FLACs are "good enough" for me. At a certain level, it's all about tradeoffs. The guy with $500,000 into his system is still making tradeoffs. Loudspeaker and amplifier design is all about tradeoffs as well. There is no "perfect," we just choose which shortcomings are least offensive.

Mr ProgFreak wrote: ^ the fact remains though that vinyl has inferior dynamic bandwidth ... I'd say that a digital medium with 14 bit resolution would be a good analogy.

Not without qualification it wouldn't.  Expressing it like that is misleading.