MP3 DIGITAL DOWNLOAD vs CD

I assume Tom your comparison was using a 320kbps mp3?

This isn't a fair test. You have to compare the same track WITHOUT knowing which is an MP3 and a which is a lossless file. Plus, you can't compare a downloaded track to a CD that you own. You have no idea what equipment the download provider used, you have no idea what software they used and what encoding was used.

I'd be more than happy to provide a sample track for you and then come back and tell me if you can tell the difference.

And no cheating by looking at the file as a waveform.

-Erik-

We don't care about "fair tests," we don't care whether there are humanly audible differences between 320k and CDs, this is simply about whether pay downloads are a suitable replacement for CDs. Pay downloads rarely use 320k, and as you say, we have no control over the quality of the encode done by the pay download sites.

The majority of internet connections can now easily support downloading in lossless CD format, in fact they could easily support MORE than CD quality, i.e. the Lord of the Rings box sets, if they wanted to, so its disappointing that they don't even bother to use 320k. And there are clear differences with the VBR because VBR algorithms are not smart enough to be perfect. Transient details tend to be lost. Tom was just pointing out which are some of those transient details in a specific recording.

Also, I could argue that a blind test is less fair than what Tom is doing, because in a blind test your brain must focus on deciding which recording is which, whereas when you already know which recording you're listening to, you can focus on hearing the subtle differences between the two recordings.

Since I don't feel like digging up the old threads on this topic, does anybody have a link to studies that "prove" that there is no audible difference between 320k and lossless 16bit 44hz?

It is going to get to the point that if you don't buy the digital download you don't have the music. In fact, it is already happening with less well-known soundtracks.

If we don't support the transfer of the availability of film music from CD to digital the labels just won't release the music. I am sure it is easier for them to make the decision when the music is in a digital format rather than going to the trouble of pressing CDs, etc.

With me being happy with what I usually hear on a digital music file, it is complete liner notes that I miss most from digital downloads.

Like I said I had little time to discuss the topic so I posted the first Goddamn thing I saw. But it looked like a comprehensive article on the subject whether it supported my argument or not.

-Erik-

As I said through my computer there was no difference. When I purchased the expensives cans from Grado the differences between a download vs a CD became as I described. In all fairness I frankly can't hear any difference on a later release House on Telegraph Hill. I've got both and they sound identical. I believe this was due to the analog source that was found.
Marantz CD5004 CD Player
by Marantz
4.6 out of 5 stars See all reviews (16 customer reviews) | Like (11)
List Price: $349.99
Price: $349.00 & this item ships for FREE with Super Saver Shipping. Details

$295.00
Quick Overview
The Rega Ear headphone amplifier benefits from an all new extruded aluminum case and sleek styling. But the real story behind the Rega is its superior circuitry, which delivers a fully satisfying headphone listening experience.

Grado Prestige Series SR325is Headphones by Grado
Buy new: $295.00


I've no doubt the CD is already pretty much gone and will go the route of vinyl. DVD is going to go the same way also. I've included the material I have. I decided since I listen several hours a day I wanted a nice system.

And I really appreciate the fact that I was merely pointing out the differences that I heard. Remember an analog source or use of the computer there is no difference. As Erik pointed out and I agree the 24/96 Chandos material is really nice CD or download.
Tom

Erik, why bother?
There are some people who consider themselves capable of superhuman feats.
Why would it bother anyone else?
They pay top dollar for high-end material: they are happy, dealers are happy.
Win-win.

Scribe, the internet is positively awash with tests, both double-blind and highly subjective, with zealots coming in at every angle trying to discredit data with incredibly convoluted high phsyics and mathematical reasoning (not to mention fringe considerations like amplification, cables, acoustics....). At times it's a debate that reminds me more of religious factions than a relatively simple matter of physics and biology.

At the end of the day, no amount of data is going to convince you that you don't hear what you think you hear (the classic "Producer's Knob" is great evidence to the subjectivity of our own convictions). If you are happy that you can hear the difference, than I'm glad you're happy.

If you are truly interested in the physics of the situation, like I said: the internet is veritably awash with pros and cons, tests and convictions, evangelism and facts.
But it's Google that is your best friend in researching that. Not I.



Well, I would be very disappointed if we'd get to a lossy-format-ONLY download situation.
Reason is taht I'd like to maintain the quality of my collection throughout the years.
Now I'm sure at some point in time MP3 will become obsolete and will be replaced by a more effective and efficient compacting format. That will render my MP3 player obsolete as well. Which means I will -at some point- probably need to re-encode my collection.

Recompacting files from an already lossy source will diminish the quality every time that excercise is repeated.
That's why I really need my source files to be lossless: so that I can re-encode them to whatever format is in vogue at whatever point in time without (incremental) quality loss.

All I know is when I purchased the digital download to The Tree of Life, I noticed a slight dropping out of sound during the final track. I wonder if the CD suffers from this...

Like I said, Scribe, I'm NOT trying to convince you.
Nor am I expecting you to convince ME.

All I said is that if you really care, it's easy to do the research you yourself ask for.

What's the point of asking for proof if you're not really interested in any?

Or in another vein: do you really think there would be any proof in the world that would convince you you don't hear what you hear? I could argue physics and biology until I'm blue in the face, an STILL be utterly unable to convince you (there's a reason I referenced the nigh-religious nature of the debate). I tend to choose my battles, and this is an unwinnable one (and a very unsatisfying one at that).

But the fact that I choose not to enter that battle doesn't mean I suddenly agree with you or anyone else claiming superpowers!

My whole point was not to start a war. I merely reported what I heard on one track that I listened to for an hour. By the way it is a damn good track.
Tom

{in response to Scribe}
I'm very sorry if you feel slighted, but there's exactly the point: the issue is very much NOT about scientific verification. Scientifically there is a MASSIVE difference between ANY bitrate MP3s (incidentally: note that my ONLY point of reference remains 320 kbps, unless specifically stated) and WAV formatted ones. And the difference is VERY easily measurable.
Math doesn't lie: you lose data when compressing.
So scientifically there is no issue at all.

But the key point is about personal bias and preference.
How much data can you lose before it becomes noticeable? How much harmonic distortion can we ignore/tolerate?
Can you HEAR it? Can ANYONE hear it? How much of what you THINK you hear is controlled by your brain and emotions? How much is polluted by fringe aspects like cabling, acoustics, encoders, amplifiers, headphone/speaker noise...?

As far as that goes, the burden of proof you lay on me is meaningless: any number of tests that do not include you leave you free to counter with "but *I* can hear it". Any conclusive evidence can only be found if you would be willing to take a double-blind test yourself.

Anyway, if you're really gung-ho on the subject, go and look up Earl Geddes who wrote a paper called "The Perception of Distortion", and later "Auditory Perception of Nonlinear Distortion—Theory" with the help of another expert on the subject. The papers showed from blind testing a group of listeners that no one can hear THD (harmonic distortion) below about 5-10%, and that compression driver speakers (known for their extreme accuracy) typically operate with 25% THD.

Some additional tests, but alas and alack, most non-blind and all subject to ...erm...well...being subjective:
http://www.pcmag.com/article2/0,2817,1159684,00.asp
http://www.erikjheels.com/?p=1236
http://www.head-fi.org/t/431522/abx-tes … ac-results
Some thoughts about double-blind tests (just to indicate how incredibly murky this water is, and to what lengths people go to work around the science).

And if you still haven't had enough: here's the true vipers' nest of geeking out on bitrates: http://www.hydrogenaudio.org/forums/ind … opic=83263 (CHOPCfull of but I warn you, it's hardly fun reading, as there is an enormous amount of zealots on that forum!)

Interestingly our own Demetris is doing a thesis on psycho-acoustics (basically how the brain interprets listening), so he'd be a very good source for you indeed. But I think unfortunately he's offline for a few weeks now.



Incidentally, I'm not entirely sure what you mean here:


I read that he says "virtually indistinguishable". So pretty much NO difference between even 256 kbps and CD (leave alone 320!)!

And now I've spent FAR too much time on this than I intended to. Again.
Again: if you think you can hear it, power to you.
(SUPERpower )

Ah, easy, Alan!
It's backwards thinking, business-wise as well as technologically!

Selling lossy formats simply isn't future-proof as I stated in an earlier post.
That very fact alone in a digital era makes it financially ridiculous to support.

There's good reason why selling cassette tapes never even came close to the sales of vinyl (even though generally they were quite a bit cheaper): having for whatever reason a need to "re-encode" (so to speak) my music to a different medium, it wouldn't matter whether I'd record my record to tape recorder, cassette tape or -later- to CD-R. The quality of the (lossy) copy would be a constant, as it was taken from the same constant-quality source.
Not so of course with cassette tapes, which are already lossy in format!
So any copy would incur a DOUBLE loss (the original one PLUS the one of the copy).

That's the technology side of things.
The problem at the business side is comparable: the very fact that psychoacoustics and ABX tests prove that 99.99999% of the world population can't distinguish between a properly encoded 218 kbps (don't ask...apparently that is some universal cut-off point found in several tests) MP3 and lossless, makes it seemingly OK to sell no more than that (even though the effort in creating/providing lossless is minimal!).

This may either be laziness, technical incomprehension or indeed cynicism.
But i am not willing to gamble on the fact that in -say- ten years time, MP3 is completely outdated, and no personal audio system will play it. Which would -in the situation above- that I'd be forced to re-encode my expensively purchased -lossy- files to ANOTHER lossy format, thereby incurring data (hence quality) loss by each such incremental step!

Unfortunately the thinking now seems to be generically like yours: "it's the best we have, so why not?"
And that means the current business practices actually pay off, so there's no incentive to change or innovate.

Well, obviously I can lament that, or be pragmatic about it. So I should be willing to accept that this means I shall either have to do without, OR in future will have to repurchase any sound file I bought before in a new format to maintain a constant quality on new platforms.