Is LAME the best *encoder* for MP3?

[luv2chill]

hey can someone explain what I need to download to get Audiograbber to use this codec? I have the Fraunhoffer Professional codec installed, but I'd like to try encoding some with LAME to see if I notice an difference.

I went to the official lame site, but wasn't sure what to download.

Thanks

l2c

P.S. Besides using audiograbber, what is the best GUI front end for just encoding mp3s? Audiograbber's good for rip + encode, but if I already had the wave files, what's the best program?

 

[Possum]

NUTZ, a double post

 

[Possum]

If you already have the .wav files, then RazorLame is a good one to use with the Lame encoder/decoder. You can find links to it from the news page of www.r3mix.net.

 

[Workin']

glen I agree with you 100%. It seems many people just can't comprehend what scientifically valid testing proves. Trying to explain statistical validity is very difficult. I'm an engineer and have had much painful exposure to statistical methods and understand them well, but even many of my peers don't have a clue when it comes to statistics (by that I mean statistical methods, not "4 out of 5 dentists choose...&quot.

But from the sound of many of the arguments here, it seems many people might still believe the earth is flat!

 

[glen]

I have expensive Nike's on, and I don't care what they say, the Earth IS flat!!!!!!!!!!!!!!!!!

 

[subhuman]

glen, it's clear to me that most of the world doesn't know how to fully comprehend and listen to music, and mp3 will be fine for EASILY more than 50% of the people here on a techy-geeky BBS (where many people use $10 speakers connected to a soundblaster).

i'm done, i have nothing to prove, i'll say it again, "use your ears."

it's obvious to me that glen is here to prove that mp3's sound exactly like CDs. that's fine, it's not true, but believe what you need to believe to be happy.

to answer the original question, YES, lame is among the best MP3 encoders.

 

[glen]

dp

 

[glen]

I like you. I am not here to make you mad, or embarrass you on a message board.
Back in the old days, when we lived in caves, before science, when someone made a claim, like Zog's club is the biggest, we had no way of determining the truth. Now that we have the scientific method, we can test it out. In the case of MP3's, we have done many tests. In the case of 256 bit encoding, when people are given the original wav file and an mp3, they guess which is which 50% of the time. That is the same as if people were guessing. The proper way to dispute it is to propose a problem with the methodology, or conduct some tests and show counter results. Simply saying your ears, cave, club, or speakers are better, and add lots of !!!!!!!!!!!!!!!!, does not really help in seeking the truth.

 

[BD2003]

Now I'm curious as to why I can't get a single encoder to get decent sound on a 22khz file. Just like with 44khz files, it cuts out the top frequencies, but in this case it happens to be ~10000, something I can more than easily hear. I've tried both lame and fraunhofer on their highest settings (160kbps, wont go any higher for 22khz) and both do the same thing. Any ideas?

 

[Workin']

22khz sampling rate limits high frequency response to 11025 Hz. 44 khz limits it to 22050 Hz. No matter what encoding method you use. It's an iron-clad rule of the mathematical transforms used to encode PCM data. An mp3 encoder just makes it worse.

 

[BD2003]

I am well aware of that fact. But if a 128kbps file will encode up to like 16khz on a 44khz sound, why only to 10 on a 22khz? Doesnt make any sense...should I upsample?

 

[glen]

<< But if a 128kbps file will encode up to like 16khz on a 44khz sound ... >>


It will capture information up to 20khz.
If it stops at 16khz, it is because of the codec.



<< ...why only to 10 on a 22khz? >>


If it is sampled at 22khz, information in the wave file above 10khz does not exist.

 

[BD2003]

Ok, let me try this again...

Using a wave file and an encoded mp3 of that exact wave, there is information missing. Not only is it quite obvious to the ear, but taking a look at the spectral view in cool edit pro, a whole top slice of it has been literally cut off. Maybe its 11khz, or 10khz, or even 9, I dont know, but SOMETHINGS missing.

If a 128kbps file at 44khz can encode information well above the limit of a 22khz wave, why cant it do so at 22khz? Is it an encoder bug, or a limitation of the format? I tried encoding using fraunhofer and LAME. I tried to stay away from upsampling because of the loss in quality that can produce. Now am I stuck with upsampling (prob not a big loss, but why should I have to bother to begin with?), or am I just using the wrong encoder?

 

[glen]

Imagine you are bouncing a ball once every second.
Imagine I take a film of you doign this at a rate of one frame a second.
When you play it back, the ball will look like it is motionless.
If I film it at 2 frames a second, you will see motion.
So, basically to capture a wave, either sound or a bouncing ball, you want to sample it at at least two times the highest possible frequency.

 

[Workin']

I can't say for sure but if I had to bet on it I would say the problem is with the encoder. Don't know what you can do about it, though.