quote:

Originally posted by palm at the same samplerate i dont think its more to it realy.

Because mp3 is algorithm-based compression. It doesn't just subtract any old random data, or start linearly from the top or bottom of the frequency spectrum.

quote:

Originally posted by david.michael Because mp3 is algorithm-based compression. It doesn't just subtract any old random data, or start linearly from the top or bottom of the frequency spectrum.

arent u talking about VBR now?

quote:

Originally posted by palm arent u talking about VBR now?

Nah, I'm just pointing out that the mp3 encoder (LAME, etc.) starts stripping away the least important data first... stuff you don't even hear, and works its way down.

quote:

Originally posted by david.michael This is quite common and, I'd say, likely the industry standard (for digital playback). I mean, it's not like you're going to be dancing in the club, and all of a sudden one of your buddies stops you and goes "wait, wa-wa-wa-wait... stop. Do you hear that? Is that-- is that... a fucking 320kbps mp3 I hear playing?!"

hahaha....kudos, i think when i go to webster on saturday for tias im guna break that out bro. My friend is guna die when he hears it. lmao

___________________


www.myspace.com/badapple1984

quote:

Originally posted by david.michael Nah, I'm just pointing out that the mp3 encoder (LAME, etc.) starts stripping away the least important data first... stuff you don't even hear, and works its way down.

yeah in wav files with alot of air this is possible, but with fully mastered dance-tracks, with brick-wall-dynamics, i cant realy understand its possible not to loose the same amount as the "compression" does, either in dynamics or in frequency. for example if u have a track peaking at 0dB in wav it will distort when converting to wav, already then u understand it must clip and boost the freqs together somehow. i dont know tho it just dont seems logical that theres info left to "cut out" in todays music.

Well despite the fact its harder to detect the difference, the differences are still there, and you'll simply enjoy a higher bitrate song more.

I'd also like to point out (in theory anyway) that a 1440kbps mp3 would sound better then a 1440kbps wav. You would have to rip from an even better quality source though, like vinyl or whatever.

___________________


Check out my Goa and psy Trance mixes!

quote:

Originally posted by palm yeah in wav files with alot of air this is possible, but with fully mastered dance-tracks, with brick-wall-dynamics, i cant realy understand its possible not to loose the same amount as the "compression" does, either in dynamics or in frequency. for example if u have a track peaking at 0dB in wav it will distort when converting to wav, already then u understand it must clip and boost the freqs together somehow. i dont know tho it just dont seems logical that theres info left to "cut out" in todays music.

Fuckin what I'm so sick of people who have no idea about MP3 slagging it. I'd love to see all these people who complain about losing dynamics and bass take an AB test on club systems.
Let me explain real quick, it makes no difference at all where your track is peaking, MP3 won't cut out bit depth most of the time. The majority of size is shaved down by reducing the frequency range (sample rate) and then sometimes the dynamics will be reduced a little too AFAIK. You'll only notice this at 128, not even at 192 usually! Hi hats may sound a little crunchier at 192, but at 320, you can't even hear high enough (unless you can hear over 20 000 hz?) to detect the difference.
The fact is, that a brickwalled and highly frequency focussed track needs the absolute LEAST bitrate of any material, Wav is just encoding data that ISN'T there, and specifically saying, "oh you know what, there are no highs in this track, but I'm gonna spells that out so your crappy CDJ300000s can understand that."
A solo piano track with loads of dynamics and massive frequency range may get some advantage from being 320, but in a dance track, you could use 192 and no one would notice, let alone give a damn.

___________________
New Mix: March 2010 Promo
Soundcloud|Facebook

quote:

Originally posted by palm at the same samplerate i dont think its more to it realy.

Of course there is, saying '320 is 1/5 the quality of wav' is completely meaningless...

quote:

Originally posted by kitphillips Fuckin what I'm so sick of people who have no idea about MP3 slagging it. I'd love to see all these people who complain about losing dynamics and bass take an AB test on club systems. Let me explain real quick, it makes no difference at all where your track is peaking, MP3 won't cut out bit depth most of the time. The majority of size is shaved down by reducing the frequency range (sample rate) and then sometimes the dynamics will be reduced a little too AFAIK. You'll only notice this at 128, not even at 192 usually! Hi hats may sound a little crunchier at 192, but at 320, you can't even hear high enough (unless you can hear over 20 000 hz?) to detect the difference. The fact is, that a brickwalled and highly frequency focussed track needs the absolute LEAST bitrate of any material, Wav is just encoding data that ISN'T there, and specifically saying, "oh you know what, there are no highs in this track, but I'm gonna spells that out so your crappy CDJ300000s can understand that." A solo piano track with loads of dynamics and massive frequency range may get some advantage from being 320, but in a dance track, you could use 192 and no one would notice, let alone give a damn.

You clearly don't have the slightest, tiniest clue what the hell you're talking about. Wav does not contain frequencies much higher than hearing range, at 44.1khz sample-rate it contains frequencies up to the Nyquist, which is 22.05khz, this is why that sample-rate was chosen as the standard for CDs. You're right that in most cases it will be very hard to tell a 320kbps mp3 from a wav, but it is certainly possible there are differences inside the range of hearing, and even more so at 192kbps.

Last edited by Sanguis Mortuum on Sep-05-2008 at 14:59

Yes but in a club system? No chance you will hear any difference between 320 and Wav at all.

___________________
Like Progressive and Deep House? Grab my latest mix below
http://www.tranceaddict.com/forums/...s=#.UAv1rbRo3h4

quote:

Originally posted by Sanguis Mortuum Of course there is, saying '320 is 1/5 the quality of wav' is completely meaningless...You clearly don't have the slightest, tiniest clue what the hell you're talking about. Wav does not contain frequencies much higher than hearing range, at 44.1khz sample-rate it contains frequencies up to the Nyquist, which is 22.05khz, this is why that sample-rate was chosen as the standard for CDs. You're right that in most cases it will be very hard to tell a 320kbps mp3 from a wav, but it is certainly possible there are differences inside the range of hearing, and even more so at 192kbps.

Don't think so mate. Most musical material falls within 16 000 Hz, anything out of that, is still perceptible, but will come through as crunchiness more than outright unpleasantry. Realistically, in a noisy club at 3 am, I doubt most people can hear beyond 16-18 000 hertz, you know how much ear ringing you get the next day from loud clubs. Also, taking into account that most systems (and SONGS!) have a sharp rollof above 12-13 Khz so there won't be many of these frequencies anyway.
Sanguis, since you claim to have a better knowledge of it, perhaps you can explain better exactly what information is being removed in MP3 conversion? At 320, what is the typical maximum frequency you'll find etc. I'm always up for being proven wrong, but you haven't convinced me yet sorry.
OH, and by the way, in Palm's reply the main -generator was his comments about brickwall limiting and stuff, just totally confused about the relationship between song dynamics (lack thereof) and reduction of bitrate...

___________________
New Mix: March 2010 Promo
Soundcloud|Facebook

Last edited by kitphillips on Sep-05-2008 at 15:18

quote:

Originally posted by kitphillips Don't think so mate. Most musical material falls within 16 000 Hz, anything out of that, is still perceptible, but will come through as crunchiness more than outright unpleasantry. Realistically, in a noisy club at 3 am, I doubt most people can hear beyond 16-18 000 hertz, you know how much ear ringing you get the next day from loud clubs. Also, taking into account that most systems (and SONGS!) have a sharp rollof above 12-13 Khz so there won't be many of these frequencies anyway. Sanguis, since you claim to have a better knowledge of it, perhaps you can explain better exactly what information is being removed in MP3 conversion? At 320, what is the typical maximum frequency you'll find etc. I'm always up for being proven wrong, but you haven't convinced me yet sorry. OH, and by the way, in Palm's reply the main -generator was his comments about brickwall limiting and stuff, just totally confused about the relationship between song dynamics (lack thereof) and reduction of bitrate...

You're right that in a club it won't make any difference, nobody in a club will be able to notice. Im talking about listening on semi-decent monitors or headphones.

MP3 compression is not simply throwing away frequencies too high or low to hear, although that is part of it; it also uses the idea of psychoacoustic masking to throw away frequency content, across the entire spectrum, that is masked by other frequencies. The way it works can also cause artifacts such as ringing or pre-echo, particular on more dynamic sounds such as percussion.

quote:

Originally posted by Sanguis Mortuum You're right that in a club it won't make any difference, nobody in a club will be able to notice. Im talking about listening on semi-decent monitors or headphones. MP3 compression is not simply throwing away frequencies too high or low to hear, although that is part of it; it also uses the idea of psychoacoustic masking to throw away frequency content, across the entire spectrum, that is masked by other frequencies. The way it works can also cause artifacts such as ringing or pre-echo, particular on more dynamic sounds such as percussion.

I'd argue that you'd hear only negligable effects on even good monitors. From what I understand it does mainly throw away that data in the high range, where thats possible with space constraints. With 320 I imagine that you'd mostly get data loss above 12 000, which is then masked anyway... I was under the impression that masking worked by obscuring the lost data (hopefully in the high end) behind louder sounds, not by taking away slices of spectrum...

___________________
New Mix: March 2010 Promo
Soundcloud|Facebook