Quote:



Listen to a couple songs of a CD, then encode them at 64 kb/s. Compare the files. Any differences you hear are artifacts. It's all new characteristics (bad or good, though usually only referred to the bad) introduced during the compression process. Ever see a photo on the internet that looks blocky? Those are artifacts too (lossy JPGs). Things are "over-compressed" when the artifacts are considered to be "too much" and gets in the way of the listening/viewing experience. That's when artifacts are usually discussed.

If you encode at a high bitrate, the artifacts are less noticeable or inaudible completely, when compared to the original files, and are considered "transparent". The artifacts are still there, but much less obvious. The trick is finding the range where things become "transparent" for you. For some that's -aps. For others that's not possible with lossy (MP3, AAC, Ogg, etc.).

Some possible artifacts of too low a bitrate:

Muddiness at high frequencies (i.e. cymbals lose sharpness and clarity)
Loose bass (not controlled or tight, kick drums sound boomy and sludgy)
Rolloff of middle and upper frequencies (espically true with certain encoders such as Xing)

In general a lot of the effects of low bitrates tend to sound much like the effects of low sampling rates: muddiness, distortion, lack of clarity and crispness and poor presentation of high freqencies.

I can hear dramatic differences between 128k MP3s and 256k MP3s, using a good encoder (LAME), if I'm playing the right source material: speed metal works really well to showcase 128k MP3 flaws for me

hmmm...speed metal huh? Can you recommend any songs? I too listen too a lot of metal, will bands like Slayer or Children of Bodom work?

Anything really dynamic will show low bit rate mp3s weaknesses. It also doesnt have much of a soundstage. Low bit rate mp3 sound a lot like an old tape recording to me.

Quote:



It will be if we keep mentioning it.

Most boards I go to refer to it as SLSK or Bluebird, because more correctly-spelled references to it make it more likely that people will take notice of it. So far, it has sort of passed under the radar, and I'm hoping it stays there.

Children Of Bodom would work, I think; pick something you're really familiar with.

Impellitteri, Megadeth, etc. work well too

The artifacts that bug me most are usually flanging and pre-echo. They tend to sound like a flanger, (duh) which has sort of a wahwah type sound, or a swishy swirly watery sound that mainly affects high treble sounds like cymbals or S sounds in vocals. It's unfortunate, because once you learn to hear these you will hear them everywhere. There is a local radio station which apparently plays compressed music and the artifacts are clearly audible. I can often hear artifacts in the muzak they pipe into stores via satellite. I like to amaze my friends by pointing out when they are playing a mix CD in the car originally made from MP3s.

I just compared the Green Album CD, by Weezer, to its 128 mp3 counterpart and I found that their were only 2 songs that really showed a dramatic decrease in sound quality compared to the 128 mp3. Those two songs would be "Island in the Sun" and "Photograph". The only differences that I found were of the soundstaging and the dynamics. Also IMHO I think that the mp3 had more bass too it. Has any1 else had a comparison like this? Is there more than soundstaging that I am missing from the CD?http://www5.head-fi.org/forums/newre...hreadid=61355#
http://www5.head-fi.org/forums/newre...hreadid=61355#

opps typo at the end there

Judging from what your equipments are (from your profile), you really don't have anything that's significant enough to really show the difference between low-bitrate MP3 to higher bitrate to a significant degree.

When you move to higher level equipments, the gap become more and more obvious.

When I had lower-level equipments, I couldn't tell the difference between 128kbps mp3's and CD's...

Would getting M Audiophile USB 24/96, an Elite Sl Pimeta from Headsave, and HD600s or HD595s, show a clear difference from CD quality and 128 mp3 quality? Will getting this set up totally ruin my whole mp3 collection because they are so revealing?
What exactly am I missing from the CD version? I mean, the cans that im using right now are the sony 7506s, which have been said to be very analytical, yet the only difference I can hear in some mp3s is the soundstaging and certain dynamics.
Please help because I am racking my head over this ...

I guess the first question is, how are you testing CD versus MP3's? Are you using different equipments to test your CD versus MP3? For example, if you're testing with your iPod, are you using MP3 and WAV on it to do head to head comparison? Or are you playing CD off somewhere else? If you are using your CD, is it playing from the CD-ROM while MP3's playing from your hard drive (it does matter, a little bit)?

The thing you have to keep in mind, is while you're comparing different MP3 compressions and uncompressed, you're looking for "characteristics of sound" and not "sound". Meaning, unless you use really, really bad MP3 compression, you're unlikely to notice something like an entire chord of strings missing, or some beats simply not being there.

What you will notice, is that the string doesn't have decaying notes (only the plucking, none of the resonance and fading), drum beats might not have vibration. Bass guitars carry no visceral impact. Above are few of the examples you may be able to see.

You've already noticed some soundstage and dynamic differences... and with better equipments, those differences are just going to be more and more apparent, to the point where it drives you nuts and disdain lossy compression forever

V6/7506 is a somewhat revealing headphone, it does reveal a lot of details in the sound for its price class. However, it is not yet high-level enough to be able to reproduce a lot of nuisances in sound that you have yet to hear. There's a whole range of dynamic and detail that the 7506 doesn't even come close to approaching. It is only a detailed headphone in its price class, not comparing to higher level headphones.

For comparison, I really didn't notice how much difference there is in lossy compression versus uncompressed music 'til I got my Shure E5c (which is the first true high level headphone I got). Now in my collection, the CD3000 and A900 both reveal MP3 for the monster that it is. Once you move down to the V6 and HD280 though, it becomes much lesser of an issue.

HD600 (can't say anything about HD595, it's not out yet) does show a fair amount of details. However, I think HD580/600/650 are all headphones that make the music extremely smooth rather than revealing. I feel that HD650 reveals a lot less "flaws" in your system than the CD3000. It makes everything sound better than the source... so I wouldn't say that it's the best headphone to judge the differences between MP3 and CD (although there will still be a noticeable difference).

Wow lindrone thanks for your help. I think I am begining to understand this whole audiophile business on a whole new level.
I think that for now I will stick to the files that I have but this will proabably change as soon as I move onto higher class hardware.

Thanks LINDRONE

You're welcome, unfortunately, it's only a matter of time before you'll join us to digging through your wallet and begging on the street corners for a new headphone

Sorry to jump in on your conversation, but I have a question. Since you are talking about the aptness of various headphones for comparing audio files, do my Grado SR-80s do a good job at that? They are very revealing to me, but this is from a man who has never had another pair of decent headphones.