Support Head-Fi.org by
starting all of your
Amazon.com shopping by
clicking here.
____________________________________________________________________
Today's Featured Head-Fi Blog: Jude's Blog
____________________________________________________________________
Please help
support Head-Fi by becoming a Contributing Member
CLICK
HERE -- Contributing Members, thank you
for your generous support! --
I have to say I consider my Ety's to be pretty decent gear, and I haven't had the urge to upgrade any further as far as IEM's go, and with those going through a cMoy amp, which is probably an average amp at best, I still am not able to discern the difference b/c 256 and 320, and only more complex tracks (tracks with more instrumental stuff and strings, which is very few in my library of music) can I really tell a very little difference b/w 192 and 256/320.
__________________ Speaker Rig: WMP11>Philips CD Player> Cerwin Vega D-3 Home Rig: WMP11>cMoy V1.48>Grado SR60 Portable Rig: Samsung YP-T9>Ety ER-6i Super Portable Rig: iPod Shuffle 2G>Super.fi 3
Other Random Canalphones/Earbuds: Sennheiser CX300, iBuds, SC Smokin' Buds Ink'd, Koss KEB24
Originally Posted by s4one
At first when I used the SF5's I noticed the bass from the SF5's made the ER6i's it's b****
I don't find it very informative. They use an outdated MP3 encoder, feed it test tones rather than music, and draw conclusions about the quality of the resulting MP3 without ever conducting a listening test.
My favorite comment about that article was posted by j7n at Hydrogen Audio: "They used JPEG on the spectrum charts. That about sums it up what they know about data compression."
I don't find it very informative. They use an outdated MP3 encoder, feed it test tones rather than music, and draw conclusions about the quality of the resulting MP3 without ever conducting a listening test.
You may not find the article informative, but I certainly did. Of course I prefer objective tests to someone else's subjective listening opinion. I'd rather decide for myself which codec sounds subjectively best to me.
The tests were specifically designed to reveal flaws in the codecs. The graphs serve to visually illustrate those flaws. Looking at the graphs shows me that the codecs do a surprisingly good job and that where they err is usually by slightly rolling off the extreme high frequencies and by raising the noise floor to around -85 db. Do you think a listening test could be so precise in identifying and quantifying these minor flaws? I don't.
IMO... the degradation is so minimal I doubt if most listeners could detect it and I find that of interest... However I understand that you may not.
IMO... the degradation is so minimal I doubt if most listeners could detect it and I find that of interest...
That's precisely my point. No-one disputes that MP3 encoding changes the source file, and has effects such as high-frequency roll-off. The significant question is whether those changes are audible. Stereophile did not conduct any listening tests, and therefore, the article does not even attempt to address that question.
Incidentally, when I speak of conducting a listening test, I am not speaking of "someone else's listening opinion." I'm speaking of ABX testing, which determines, objectively, whether the listener can tell the difference between the MP3 and the original.
That's precisely my point. No-one disputes that MP3 encoding changes the source file, and has effects such as high-frequency roll-off. The significant question is whether those changes are audible. Stereophile did not conduct any listening tests, and therefore, the article does not even attempt to address that question.
Incidentally, when I speak of conducting a listening test, I am not speaking of "someone else's listening opinion." I'm speaking of ABX testing, which determines, objectively, whether the listener can tell the difference between the MP3 and the original.
Did you even read the Stereophile article before you dismissed it as uninformative? If you did, I think you missed the points they were making. Your premise seems to hang on the fact that they didn't do listening tests between "MP3 and the original" to check for sonic differences. But right at the beginning of the article it states " MP3s and their lossy-compressed ilk do not offer sufficient audio quality for serious music listening". Implicit in this statement is their opinion that MP3 is NOT sonically equivalent to the original. Under that premise, listening test would be a foregone conclusion and pointless. Even if they had done listening tests, would that have really been relevant to you, your ears and your gear? If so, how?
Additionally, the title of the article is "MP3 vs AAC vs FLAC vs CD" which is far more comprehensive than your "MP3 versus original" concerns seem to suggest. Regardless of the fact that the article did not contain a listening test, it explored the differences between various codecs, various bit-rates and CD and quite clearly illustrates what these codecs do to the music that is so important to us.
Uninformative? Surely you jest!
Last edited by Hi-Fi Low-Life; 06-19-2008 at 04:03 AM.
:/ I'm not sure why you feel it conflicts. If the minimum allowed bitrate is 320, which is also strictly enforced by -F, and the -V is used, the MP3 essentially becomes 320 CBR. That is pointless; but conflicting?
I certainly would never do that. And I can assure that I am quite calm.
Well, with your statement, you called into question my ability to truly hear differences. How am I supposed to receive that? o_0 It sounded as if you would like to prove that it's all in my mind. Why would you want to do that?
__________________
"...stereo does not exist in the real world." -James Bongiorno
Implicit in this statement is their opinion that MP3 is NOT sonically equivalent to the original.
And that implicit in THAT statement is the assertion that they can actually hear a difference, as assertion that they have failed to support or even address.
Under that premise, listening test would be a foregone conclusion and pointless.
No. This is precisely my point. It is NOT a foregone conclusion. It is factual assertion that is verifiable, but which they failed to verify.
Regardless of the fact that the article did not contain a listening test, it explored the differences between various codecs, various bit-rates and CD and quite clearly illustrates what these codecs do to the music that is so important to us.
But it did not explore whether "what these codecs do to the music" is audible.
Originally Posted by Logistics
:/ I'm not sure why you feel it conflicts. If the minimum allowed bitrate is 320, which is also strictly enforced by -F, and the -V is used, the MP3 essentially becomes 320 CBR. That is pointless; but conflicting?
Yes, conflicting. And pointless, too, I'll agree with you there If you want a 320kbps CBR file, just use the -b 320 switch. There is no need for the -V preset.
Well, with your statement, you called into question my ability to truly hear differences. How am I supposed to receive that? o_0 It sounded as if you would like to prove that it's all in my mind. Why would you want to do that?
It has been my experience that very many people here at Head-Fi make the claim that they can differentiate between a high-bitrate MP3 and the original, but very few have actually been able to demonstrate that their claim is true. For example, HiFire started a "Public ABX Test" thread a while back in which he set up a blind ABX test of several clips encoded at -V2. Not a single person successfully demonstrated that they could tell the -V2 encoded MP3s from the original. Similarly, there was a poll a while back asking "Can you hear a difference between -V 0 and the original?" Out of approximately 100 respondents, only about 25% claimed that they could hear a difference, and out of that 25, only 2 actually substantiated their claim with ABX results as the rules of the poll required. Hydrogen Audio listening tests have show that new versions of LAME reach transparency somewhere around -V5.
I used to think that I could hear differences between 160kbps files and the original. However, when I did volume-matched blind testing, the differences that I thought I heard disappeared.
Contrary to your post, there is no way that I could prove that you cannot hear a difference, and I would never attempt to do so in any event. It is entirely possible--though unlikely--that you in fact in the very small minority of people who can distinguish a 320kbps MP3 from the original. However, given the fact that so many make that claim yet so few can actually do it, I will remain skeptical of your claim unless and until you support it with ABX test results.
That's precisely my point. No-one disputes that MP3 encoding changes the source file, and has effects such as high-frequency roll-off.
If something isn't broken to start with, you better have a good reason for "fixing" it. We agree that there is no way that an MP3 file can ever be bit equivalent to the original recording. So if one is to change the recording, with whatever risks that entails, one should have a good reason. Saving disk space? Not a good reason. Reducing download times? Not a good reason. Not in this day and age. MP3? an increasingly questionable technology, given that it can't improve on the original file, and the reasons given as justification for it are less and less relevant as time passes.
Note that I admitted in a previous post in this thread that I happily (more or less) buy VBR MP3s from eMusic. That is simply because that's what they offer, and I have other reasons for shopping there. Were they to offer FLACS, I would never download another MP3 from them.
Originally Posted by Febs
The significant question is whether those changes are audible.
I disagree, as noted above. The best an MP3 can do (theoretically) is sound as good as the orginal file. Maybe it does, maybe it doesn't. What do I get for taking that risk? Smaller file size? No thanks. Shorter download times? Again, no thanks.
If you want a 320kbps CBR file, just use the -b 320 switch. There is no need for the -V preset. 
Linear Mode
