IDM Mailing Archive

yup, u r right, it all has to do with over-compressed mastertracks, which then being converted to mp3 formats, sound bubbly & distorted bass-less. The compression-algorithms can't handle the squashy compressed stereo-wav. I realised it by converting own, compr. trax to stereo 44.1kH mp3 format & it sucks... `SedarkA´ :(Pee-N-yoU): http://pnu.widerstand.org ______________________ ---------------------------------------- ------------------------------ Date: Tue, 31 Oct 2000 13:46:07 +0200 To: idm@hyperreal.org From: Irene McC <substar@iafrica.com> Subject: distortion Message-id: <39FECD1F.18772.11D2B79@localhost> Just a general MP3 related question: a lot of MP3's I'm listening to are distorted, either at the top end which clips, or the bass which just goes "bbsshhh" - and looking at the actual curve, it makes standing squared waves. Presumably this is not due to the extraction, but comes from the original source - is it because people use huge compression on their sound files? Is there any way around it? Even if I convert my own CD's to MP3, I cannot predict which ones will come out distorted (and many do). Selecting the "normalize" function does not seem to help - sometimes it even boots the gain. I * --------------------------------------------------------------------- To unsubscribe, e-mail: idm-unsubscribe@hyperreal.org For additional commands, e-mail: idm-help@hyperreal.org

There have been times when I have not released tracks on the internet because I cant get them to encode properly. Especially when you have a low, extended kick, and maybe a quick hihat or something going at the same time... it sounds terrible. Has anyone experimented with windows media? -----Original Message----- From: peeing knots [mailto:ultraknots@hotmail.com] Sent: Tuesday, October 31, 2000 10:02 AM To: IDM@hyperreal.org Subject: [idm] Re: something about bad MP3's yup, u r right, it all has to do with over-compressed mastertracks, which then being converted to mp3 formats, sound bubbly & distorted bass-less. The compression-algorithms can't handle the squashy compressed stereo-wav. I realised it by converting own, compr. trax to stereo 44.1kH mp3 format & it sucks... `SedarkA´ :(Pee-N-yoU): http://pnu.widerstand.org ______________________ ---------------------------------------- ------------------------------ Date: Tue, 31 Oct 2000 13:46:07 +0200 To: idm@hyperreal.org From: Irene McC <substar@iafrica.com> Subject: distortion Message-id: <39FECD1F.18772.11D2B79@localhost> Just a general MP3 related question: a lot of MP3's I'm listening to are distorted, either at the top end which clips, or the bass which just goes "bbsshhh" - and looking at the actual curve, it makes standing squared waves. Presumably this is not due to the extraction, but comes from the original source - is it because people use huge compression on their sound files? Is there any way around it? Even if I convert my own CD's to MP3, I cannot predict which ones will come out distorted (and many do). Selecting the "normalize" function does not seem to help - sometimes it even boots the gain. I * --------------------------------------------------------------------- To unsubscribe, e-mail: idm-unsubscribe@hyperreal.org For additional commands, e-mail: idm-help@hyperreal.org --------------------------------------------------------------------- To unsubscribe, e-mail: idm-unsubscribe@hyperreal.org For additional commands, e-mail: idm-help@hyperreal.org

The problem with MP3 is one of two things: crappy encoders, and mispreparation of the source material. If you want to have the best quality MP3s, you have to address those issues. Wavelab 3.0 has a 'super-high-quality' encoder that runs directly on 24bit wav files. When I mp3 my own material with this and the results are phenomenal. As far as mis-preparation goes, normalizing to 0dB (full code) is the quickest way to make the encoder screw up. It seems OK when I normalize to -0.03dB. It's also a good idea to compress the audio to raise the average level to -12dB or so. Quiet material interacts as poorly with the encoder as overly loud material. And in general, you'll get better results if you bandwidth limit the source material -- use a high pass filter set to roll off below 20hz, and a low pass to roll off about 16khz. The high end rolloff isn't that damaging to electronic music, but it can potentially make high quality acoustic recordings sound worse. Most of the MP3s you hear were either ripped directly from CD by an 'all in one' program that doesn't really accomodate differences in program material, or recorded analog by amateurs using cheap sound cards. I've heard MP3s transferred from vinyl where it was obvious that the needle needed replacing. I've also heard vinyl transfers run through 'vinyl restoration' filters with extreme noise reduction settings; that is guaranteed to sound shitty. A well prepared audio master, run through a good compressor, will not sound as good as the original, but it should be free of the most obvious artifacts. kent williams -- kent@avalon.net http://jump.to/cornwarning -- Iowa's First Techno Record Label http://www.mp3.com/chaircrusher -- tunes --------------------------------------------------------------------- To unsubscribe, e-mail: idm-unsubscribe@hyperreal.org For additional commands, e-mail: idm-help@hyperreal.org

Just find some encoder with VBR function and try to compress with lower limit of 192 bps. I personally use CDEx software with VBR 0 (maximum quality minimum compression) and minimum 256 bps. Sounds good enough. 6x time compression. 128 bps will always sound crappy. Alien np. Detroit Escalator - Black Buildings EP (recommended stuff) --------------------------------------------------------------------- To unsubscribe, e-mail: idm-unsubscribe@hyperreal.org For additional commands, e-mail: idm-help@hyperreal.org

At 09:12 AM 11/1/2000 +0200, Konstantin Minko wrote:

quoted 4 lines Just find some encoder with VBR function and try to compress with lower

>Just find some encoder with VBR function and try to compress with lower >limit of 192 bps. I personally use CDEx software with VBR 0 (maximum quality >minimum compression) and minimum 256 bps. Sounds good enough. 6x time >compression. 128 bps will always sound crappy.

Regarding the 128 bps, it really depends on the encoder. Fraunhoffer (sp) encoders tend to be the best, and if you use the codec in a good program, anything above 128 will essentially sound like 128. Of course, finding good codecs is often a problem, and a lot are commercial, so yes, using a higher bitrate is a good solution. Less compression = less problems. One interesting aspect of encoding is the VBR, or variable bit-rate, encoding capabilities of the latest version of AcidPro. In the export function, it can export to mp3 and use variable bitrate, meaning that it detects when there is more sound (uses less compression) or more silence (uses more compression). It actually seems to be highly efficient too, meaning that there aren't many mistakes, and you have well-compressed areas where there isn't much music, and the 'important' parts with music aren't compressed as bad. The result - efficient and often smaller mp3's of a surprisingly good quality. You end up having an mp3 that changes bitrates a lot, sure, but it just make's winamp/etc.'s screen upset - it doesn't affect playback. It's too bad there aren't more encoders like this. However, I'm not sure how something like that would work on decompressing... cheers, /derek --------------------------------------------------------------------- To unsubscribe, e-mail: idm-unsubscribe@hyperreal.org For additional commands, e-mail: idm-help@hyperreal.org

On 1 Nov 2000, EggyToast wrote re RE: [idm] Re: something about bad M:

quoted 1 line One interesting aspect of encoding is the VBR, or variable bit-rate,

> One interesting aspect of encoding is the VBR, or variable bit-rate,

I've done some experimenting: encoded the same track - total running time 6:46 mins, at both Variable Bit Rate using the defaults and standard 128 MP3 extraction. The VBR file weighs in at 5,513KB and the 128 file at 6,349KB. On playback, the VBR file has audible clicks and pops that are not in either the original nor the 128 version. So that's not a preferable choice from my point of view (using Xing's Audio Catalyst). I believe it's the compression that the artists / producers use on their original source material - it boosts the gain phenomenally, and I think the algorithm of the MP3 ripper cannot deal with that successfully, which results in a standing wave. As an example : take any of the noisier tracks off A Shocking Hobby, use your ripper of choice and encode to a standard of 128 Kbit's. Listen back - is there any distortion (that was *not* on the original)? NOW re-inflate the MP3 you've just made to a .wav file and using any wave editor, *look at the sine wave*. Can you spot any squared waves? To do this (compressing the MP3), I have not gone to an intermediary wave file, but used the direct ripping process. I'm getting an unusable end result - very badly distorting audio. If anybody can tell me which ripper gives you satisfactory results, PLEASE let me know ! Apparently Blade works well, but does not rip directly, only from a wave source file. Thank you for your time - I'd really appreciate your comments. I * --------------------------------------------------------------------- To unsubscribe, e-mail: idm-unsubscribe@hyperreal.org For additional commands, e-mail: idm-help@hyperreal.org

this is not the original link i'm thinking about, couldn't find it again ... http://www.airwindows.com/encoders/index.html but, despite this probably being more information than you, or i, probably need, it comes to the same conclusion that i was looking for. the original Fraunhofer encoding is inferior to the freeware/gnu LAME encoder. a free LAME based ripper for windows is cdex http://www.cdex.n3.net/ which unfortunately doesn't seem to be answering at the mo, so a mirror site is ... http://softwarecenter.net/cdex/index.html i use cdex myself, but mainly for cd -> wav so i can't say for certain whether its any good, but it may be worth a try. -martin --------------------------------------------------------------------- To unsubscribe, e-mail: idm-unsubscribe@hyperreal.org For additional commands, e-mail: idm-help@hyperreal.org

quoted 14 lines a free LAME based ripper for windows is cdex

> a free LAME based ripper for windows is cdex > > http://www.cdex.n3.net/ > > which unfortunately doesn't seem to be answering at the mo, so a > mirror site > is ... > > http://softwarecenter.net/cdex/index.html > > i use cdex myself, but mainly for cd -> wav so i can't say for certain > whether its any good, but it may be worth a try. > > -martin

I use it for mp3 encoding and enjoying it very much. Never encountered problems with VBR as well - maybe because I never use 128bps or maybe because it is easily customizable in this software? Recommended. Alien --------------------------------------------------------------------- To unsubscribe, e-mail: idm-unsubscribe@hyperreal.org For additional commands, e-mail: idm-help@hyperreal.org

quoted 3 lines Kent says:

> Kent says: > As far as mis-preparation goes, normalizing to 0dB (full code) is the > quickest way to make the encoder screw up.

I don't understand the point of normalizing at all. You're just adding another generation or processing to the sound. By normalizing you fuck with the levels that the artist made when they mastered the CD. Most hip hop (and most pop/IDM for that matter) CDs these days are mastered at very high levels, where most 'classical' and artsy music is recorded at lower, more 'proper' levels. High mastering levels to tend to deaden the subtleties of sounds, but for most pop this is not a problem. I'm guessing that the high-volume mastering is done to compensate for low-power walkmans and a going-deaf generation of walkman-addicted kids, but in any event I would never sumbit my tunes to normalization. If the levels were good enough for the artist, they're good enough for me. Unless you're suggesting that non-standard mastering levels affect the ability of the MP3 encoder to accurately encode the sound, in which case I would be full of shit. Which I am. -cf --------------------------------------------------------------------- To unsubscribe, e-mail: idm-unsubscribe@hyperreal.org For additional commands, e-mail: idm-help@hyperreal.org

On Thu, Nov 02, 2000 at 06:38:03PM -0500, Chris Fahey wrote:

quoted 4 lines I'm guessing that the high-volume mastering is done to compensate for

> I'm guessing that the high-volume mastering is done to compensate for > low-power walkmans and a going-deaf generation of walkman-addicted kids, but > in any event I would never sumbit my tunes to normalization. If the levels > were good enough for the artist, they're good enough for me.

It's also important a lot of the time that a record be cut as loud as possible if it's going to be played in a club, which is why electronic music is normalized really loud. Most records are cut about reference (0dB) in fact. -- :: atomly :: atomly@atomly.com | atomly@atdot.org | atomly@curiousnetworks.com http://www.atomly.com | http://www.mp3.com/atomly --------------------------------------------------------------------- To unsubscribe, e-mail: idm-unsubscribe@hyperreal.org For additional commands, e-mail: idm-help@hyperreal.org

Long lengthy and probably off topic explanation of normalization. Hit 'D' now if you want to get back to discussing music ;-) On Thu, 2 Nov 2000, Chris Fahey wrote:

quoted 6 lines Kent says:

> > Kent says: > > As far as mis-preparation goes, normalizing to 0dB (full code) is the > > quickest way to make the encoder screw up. > > I don't understand the point of normalizing at all. You're just adding > another generation or processing to the sound.

Not exactly. When you normalize, you find the peak level in a sound file, and compute the scaling factor X such that the peak * X == full code. Then you multiply ever sample in the file by that factor. Mathematically this is a reversable operation; according to information theory, nothing is lost. Thus normalization doesn't affect sound quality in any way except to make it louder. When you normalize, you use the maximum number of bits to represent your sound. The quieter a sound is the fewer bits of available resolution are used to represent it. If you record something at a peak level of -21db, it's only using 12 bits of resolution. You can easily prove this yourself by recording something (i.e. your favorite record) first with the gain set so that the signal peaks below -21db, and then record it so that it's as loud as possible without going over 0dB. Then normalize the two recordings to the same level and A/B compare them. The quiet recording will sound noticably coarser. The real subjective loudness of a signal, however is not a function of the peak level in the signal. The more important measurement is the RMS power of the music, which represents an overall average loudness of the track. You make a track sound louder by compressing the dynamic range -- pulling down peaks, and raising the volume of soft passages. Modern CDs in the 'pop' genres (i.e. everything but acoustic classical recordings) are usually mastered to have an RMS level in the -12 to -15 dB range. That's about as loud as you can go without making the recording sound really squashed and nasty. But to get back to MP3 encoders: They do their best work on compressed, normalized signals, because a properly compressed and normalized signal on average uses the maximum number of bits of resolution, as an average over time. The catch is that you don't want to normalize to full code -- you want to have your signal maximum peak some small amount below -- most sources recommend some fraction of a dB below full code. Why is this? I don't know precisely, but I suspect it's because full code signals present a pathological input to the encoder, which then introduces distortion and artifacts. I know through practical experience that if you go to full code normalization, your MP3 files will sound like ass. --------------------------------------------------------------------- To unsubscribe, e-mail: idm-unsubscribe@hyperreal.org For additional commands, e-mail: idm-help@hyperreal.org

What do you mean by full code? Thanks dan -----Original Message----- From: Kent williams [mailto:kent@avalon.net] Sent: Friday, November 03, 2000 7:36 AM To: Chris Fahey Cc: IDM@hyperreal.org Subject: RE: [idm] Re: something about bad MP3's Long lengthy and probably off topic explanation of normalization. Hit 'D' now if you want to get back to discussing music ;-) On Thu, 2 Nov 2000, Chris Fahey wrote:

quoted 6 lines Kent says:

> > Kent says: > > As far as mis-preparation goes, normalizing to 0dB (full code) is the > > quickest way to make the encoder screw up. > > I don't understand the point of normalizing at all. You're just adding > another generation or processing to the sound.

Not exactly. When you normalize, you find the peak level in a sound file, and compute the scaling factor X such that the peak * X == full code. Then you multiply ever sample in the file by that factor. Mathematically this is a reversable operation; according to information theory, nothing is lost. Thus normalization doesn't affect sound quality in any way except to make it louder. When you normalize, you use the maximum number of bits to represent your sound. The quieter a sound is the fewer bits of available resolution are used to represent it. If you record something at a peak level of -21db, it's only using 12 bits of resolution. You can easily prove this yourself by recording something (i.e. your favorite record) first with the gain set so that the signal peaks below -21db, and then record it so that it's as loud as possible without going over 0dB. Then normalize the two recordings to the same level and A/B compare them. The quiet recording will sound noticably coarser. The real subjective loudness of a signal, however is not a function of the peak level in the signal. The more important measurement is the RMS power of the music, which represents an overall average loudness of the track. You make a track sound louder by compressing the dynamic range -- pulling down peaks, and raising the volume of soft passages. Modern CDs in the 'pop' genres (i.e. everything but acoustic classical recordings) are usually mastered to have an RMS level in the -12 to -15 dB range. That's about as loud as you can go without making the recording sound really squashed and nasty. But to get back to MP3 encoders: They do their best work on compressed, normalized signals, because a properly compressed and normalized signal on average uses the maximum number of bits of resolution, as an average over time. The catch is that you don't want to normalize to full code -- you want to have your signal maximum peak some small amount below -- most sources recommend some fraction of a dB below full code. Why is this? I don't know precisely, but I suspect it's because full code signals present a pathological input to the encoder, which then introduces distortion and artifacts. I know through practical experience that if you go to full code normalization, your MP3 files will sound like ass. --------------------------------------------------------------------- To unsubscribe, e-mail: idm-unsubscribe@hyperreal.org For additional commands, e-mail: idm-help@hyperreal.org --------------------------------------------------------------------- To unsubscribe, e-mail: idm-unsubscribe@hyperreal.org For additional commands, e-mail: idm-help@hyperreal.org

Digital audio is represented by a signed 16 bit integer, ranging between -32768 and 32767. Full Code means a sample value of -32768 or 32767. kent williams -- kent@avalon.net http://jump.to/cornwarning -- Iowa's First Techno Record Label http://www.mp3.com/chaircrusher -- tunes --------------------------------------------------------------------- To unsubscribe, e-mail: idm-unsubscribe@hyperreal.org For additional commands, e-mail: idm-help@hyperreal.org