MP3s... Can they Blow speakers?
Originally Posted by smurf2k,Jul 27 2006, 07:07 AM
i personally have had some great high quality headphone setup and cannot discern between well-encoded mp3s and cds... but others claim so (most falsely i conclude)... but your mileage may vary. some probably do have extraordinary hearing, but most are joe blow with their klipsch or logitech z-whatever computers speakers who vehemently claim they hear an audible different.. well.. whatever.. your mileage may vary
MP3 is a lossy compression codec by nature -- it compresses the data by discarding information that the algorithm determines your ear is unlikely to detect. It's not simply a matter of "MP3s don't have lots of high frequencies" or having a lowend cutoff frequency like traditional audio quality loss issues (although many older codecs discard a lot more info under 50Hz and over 20kHz), MP3 compression introduces distortion throughout the frequency band, most of which isn't percievable -- however, certain audio data in particular can make the distortion especially apparent, if the audio is not easily predicted by the codec. Cymbals, high frequency percussion instruments, and applause tend to show MP3 artifacts a lot. Personally, even in a 320kbps LAME MP3 the swishing noises in the cymbals of any rock music drives me nuts.
(I'm an engineer, I like talking about nerdy things
)
To be precise, the MP3 algorithm doesn't remove "information your ear is unlikely to detect." The algorithm remove information your brain is unlikely able to process. I make that distinction because your ear will continue to receive and detect all of the information in a CD, but your brain is not able to process it all.
The most powerful example (and the one I use most often) is an ascending scale played in one ear and a descending scale played in the other, at the same time. Your brain will process the descending scale and ignore the ascending until the frequencies match, at which point your brain will start to ignore the descending and start paying attantion to the ascending. (Or is it the other way around? <shrug> Haven't run that trial in a loooong time.)
While the MP3 decoding algorithm is fixed in its methodology (and clearly laid out in the spec), the MP3 encoding algorithm has no such definition. It's up to the encoder creator to design the perfect algorithm. LAME is the best encoder out there, without argument, due to the sheer number of man-hours spent tuning it for various cases. They have spent countless hours working on pre-emphasis issues from sudden sharp noises, like hand claps, snares, symbols, etc. Gone (almost) are the days of encoders that totally trash the low-end of the spectrum...
However, there are always trade-offs, and one group of settings may sound great on one song and terrible on another. Unless you wish to modify settings on a packet-per-packet basis (similar to what the studio's do for DVD encoding of major films to get the cut-scene transitions perfect), we choose settings that work most of the time for "our" type of music and leave it alone. I can hear plenty of difference between CD and 128kbps, a bit at 198kbps, and (so far) none at 320kbps... I always choose variable bit rate so I have the advantage of a high quality sound file with the lowest possible size (for the quality). If I can, I prefer using AAC, though.
(Embedded DSP firmware engineer who enjoys speaking Nerd :bye: Digital audio/video pay the bills)
The most powerful example (and the one I use most often) is an ascending scale played in one ear and a descending scale played in the other, at the same time. Your brain will process the descending scale and ignore the ascending until the frequencies match, at which point your brain will start to ignore the descending and start paying attantion to the ascending. (Or is it the other way around? <shrug> Haven't run that trial in a loooong time.)
While the MP3 decoding algorithm is fixed in its methodology (and clearly laid out in the spec), the MP3 encoding algorithm has no such definition. It's up to the encoder creator to design the perfect algorithm. LAME is the best encoder out there, without argument, due to the sheer number of man-hours spent tuning it for various cases. They have spent countless hours working on pre-emphasis issues from sudden sharp noises, like hand claps, snares, symbols, etc. Gone (almost) are the days of encoders that totally trash the low-end of the spectrum...
However, there are always trade-offs, and one group of settings may sound great on one song and terrible on another. Unless you wish to modify settings on a packet-per-packet basis (similar to what the studio's do for DVD encoding of major films to get the cut-scene transitions perfect), we choose settings that work most of the time for "our" type of music and leave it alone. I can hear plenty of difference between CD and 128kbps, a bit at 198kbps, and (so far) none at 320kbps... I always choose variable bit rate so I have the advantage of a high quality sound file with the lowest possible size (for the quality). If I can, I prefer using AAC, though.
(Embedded DSP firmware engineer who enjoys speaking Nerd :bye: Digital audio/video pay the bills)
YES! this is GREAT INFO thanks guys it all makes sense (somehow)
now about this LAME encoder lol...... how the Heck do you run it?? it does not look like its a standalone sofware, please excuse the ingnorence this is the first time i hear about it thanks i've downloaded 3 versions and it doesn't run??? anybody have an idea thanks
now about this LAME encoder lol...... how the Heck do you run it?? it does not look like its a standalone sofware, please excuse the ingnorence this is the first time i hear about it thanks i've downloaded 3 versions and it doesn't run??? anybody have an idea thanks
LAME itself is an open-source encoder. Originally, the code was based upon the codebase tied to the ANSI spec (and all of its copyrighted material), so they couldn't provide a binary (i.e., compiled) version, only code. They have long since removed all vestiges of the MP3 spec test code, but it still contains copyrighted material due simply to the algorithms themselves. Other people will compile the code into a binary and include it as a library in their Windows/Linux/Mac front-end (i.e., GUI), so unless you have Microsoft's Visual C++ compiler at home, you should be looking for a pre-compiled package.
I use WinLAME... the interface is crappy as all get-out, it never saves my settings (what settings exist... which is good for beginners), and all-around could use some serious beating, but it was the first Windows GUI I came across while Googling to use the latest and greatest LAME codebase. I don't need it often, so it's no big deal.
I use WinLAME... the interface is crappy as all get-out, it never saves my settings (what settings exist... which is good for beginners), and all-around could use some serious beating, but it was the first Windows GUI I came across while Googling to use the latest and greatest LAME codebase. I don't need it often, so it's no big deal.
Originally Posted by gabster,Jul 26 2006, 11:56 PM
Here i am thinking if this could happen. i have my iPOD hooked in my car, and i notice that sound quality is not the same as if i was listening to a regular formated audio CD. therefore can this hurt speakers? I hear that distortion is what blows speakers alot more than "overpowering" them. soo, is distortion and static the same thing? is that what i hear on a lown quality MP3 tune? what would be the best kbps format for an mp3 for less "static/distortion"
i know that 128 suposed to be cd quality but is anything higher than that better?
THanks alot IA.

i know that 128 suposed to be cd quality but is anything higher than that better?
THanks alot IA.
Originally Posted by smurf2k,Jul 27 2006, 11:07 AM
redudant, extranneous sound processing will introduce distortion and artifacts.. but wont kill the speakers. clipping kills speakers...
You will find my favorite explanation of clipping here if you'd like to learn a bit more about it.
http://www.bcae1.com/2ltlpwr.htm
Sorry to bring this back from the dead... I was hoping somebody could give me their opinion on my issue?
I have Boston SL60 components powered by an Alpine MRV-F345 (rated 75w each to front channels). The amp is fed by an Alpine 9856 with an Ipod fullspeed cable. 95% of the time i'm playing music either from Ipod (using the dock connector) or via mp3 cd. Granted, most of my mp3's are low bitrate (128 kbps).
Problem is, I think my front speakers are starting to let go. The left tweeter sounds blown at certain frequencies, difficult to detect, but I believe it will worsen over time. I believe the right mid is beginning to experience the same problem.
I needed a second set of ears to listen and make sure I wasn't hearing things. Local Stereo Store Guy says they sound blown, and by the way, do I listen to alot of mp3s? I say yes, and he suggests this is the problem.
So, I have an amp pushing sufficient power to the speakers, using high quality interconnects from head unit to amp; could my music have blown my speakers?
I have Boston SL60 components powered by an Alpine MRV-F345 (rated 75w each to front channels). The amp is fed by an Alpine 9856 with an Ipod fullspeed cable. 95% of the time i'm playing music either from Ipod (using the dock connector) or via mp3 cd. Granted, most of my mp3's are low bitrate (128 kbps).
Problem is, I think my front speakers are starting to let go. The left tweeter sounds blown at certain frequencies, difficult to detect, but I believe it will worsen over time. I believe the right mid is beginning to experience the same problem.
I needed a second set of ears to listen and make sure I wasn't hearing things. Local Stereo Store Guy says they sound blown, and by the way, do I listen to alot of mp3s? I say yes, and he suggests this is the problem.
So, I have an amp pushing sufficient power to the speakers, using high quality interconnects from head unit to amp; could my music have blown my speakers?






