OK, guess I am getting my terminology confused here. MP3 always encodes at a variable bit rate... since it breaks the frequencies up into different "queues" and encodes differently depending how much info is contained in that particular frequency range in real time.
I have not seen a VBR with the encoding software I use (Audio Grabber with Lame) but maybe I have just missed it.
I have not seen a VBR with the encoding software I use (Audio Grabber with Lame) but maybe I have just missed it.
<< MP3 always encodes at a variable bit rate >>
You are mistaken, sir! The data stream is broken into frames of fixed length (time), and each frame is encoded. With CBR, if all the available bits for the frame are not needed to encode the signal, the frame is padded to fill the required number of bits. VBR doesn't do that. Well, that's a simplistic explanation, but you get the idea.
Audio Grabber can't take advantage of all the options available for LAME. Use EAC or CDEX to rip instead.
Imported
Lifer
- Sep 2, 2000
- 14,679
- 23
- 81
After listening to it louder, I believe sample 1 is the mp3. 
Oh, and here are stats of them..
Sample 1 and 2 stats
Oh, and here are stats of them..
Sample 1 and 2 stats
I also believe sample one is the mp3, because of the same reason shark just said. The highs - the cumbals, seem a bit less precise.
I voted sample#1 as the original. There is a sound made by some sort of copper instrument at about 32 seconds into the file, and it sounds better on #1. The highs are also clearer on #1 IMO, but I could be wrong.
Man, after listening for about 5 mins, I can't even hear any difference.... I might be because I don't have exactly the best sound card in the world (Maxi Guillemot Muse).
For those who could hear the difference (you know who you are), what do you think of the following mp3, using the same sample of music? I encoded it with Fraunhofer's VBR, found within Cool Edit Pro with the MP3 ME plugin, 100% quality. The bitrate averages 202 kbit/s.
By the way, I couldn't hear a difference between sample1 and sample2, although it was trivial for me to identify the original file.
ff123's iDrive account
Look for the file called sample123.mp3
ff123
ff123's Site: Discussion of Audio Compression
By the way, I couldn't hear a difference between sample1 and sample2, although it was trivial for me to identify the original file.
ff123's iDrive account
Look for the file called sample123.mp3
ff123
ff123's Site: Discussion of Audio Compression
Wow!
I can nail the mp3 between Sample 1 and Sample 2, but no way I can hear it in sample 123.
That is astounding, unless of course there is a some flaw one can train their ears too.
But so far, I detect nothing.
I can nail the mp3 between Sample 1 and Sample 2, but no way I can hear it in sample 123.
That is astounding, unless of course there is a some flaw one can train their ears too.
But so far, I detect nothing.
That encoding is SLIGHTLY better ff123, and uses a higher bitrate, so it's obvious why. I do know that with 256kbps CBR using LAME there's no way that any human can tell a difference between an original source and that mp3. By the way, since you could hear no difference between my 2 files, how were you able to figure it out? Please PM me with your answer.
dm
dm
Spectrum Analysis of Sample One
Spectrum Analysis of Sample Two
Spectrum Analysis of Sample OneTwoThree
Notice the trend? While the 123 MP3 didn't sound as smeared and chirpy as the first file, it still isn't as "clean" as the original!
UPDATE: Here is a Spectrum Analysis of the Original File . This was encoded with the lame DLL 387 at 320 kbps CBR. It's about as close to the original as you're gonna get without taking up too much space!
Cheers!
Spectrum Analysis of Sample Two
Spectrum Analysis of Sample OneTwoThree
Notice the trend? While the 123 MP3 didn't sound as smeared and chirpy as the first file, it still isn't as "clean" as the original!
UPDATE: Here is a Spectrum Analysis of the Original File . This was encoded with the lame DLL 387 at 320 kbps CBR. It's about as close to the original as you're gonna get without taking up too much space!
Cheers!
<< I do know that with 256kbps CBR using LAME there's no way that any human can tell a difference between an original source and that mp3. >>
I think you're underestimating people. To say that NO human could possibly hear the difference is wrong, I can assure you.
Give an mp3 encoded at 256 using LAME to a mastering engineer who works with audio for a living, and ask him to tell you which is the source file, and I would bet that most of them can.
Could I? Maybe not. Could somebody? Absolutely.
Oh sharkeeper,
You spoiled the fun.
I prefer spectral view (i.e., frequency vs. time) rather than spectrum analysis which is averaged over the entire sample.
That encoding is SLIGHTLY better ff123, and uses a higher bitrate, so it's obvious why.
I couldn't tell what the average bitrate the Lame-encoded file was, so I couldn't match up the Fraunhofer VBR encode to match. If you tell me the Lame average bitrate, I could reduce the VBR quality to make the filesizes about the same.
I do know that with 256kbps CBR using LAME there's no way that any human can tell a difference between an original source and that mp3.
That's a pretty bold claim. There is a wide variation in human hearing sensitivity, and also in experience with hearing encoding artifacts. In some samples, mp3 can easily be distinguished from the original at 256 kbit/s.
ff123
You spoiled the fun.
I prefer spectral view (i.e., frequency vs. time) rather than spectrum analysis which is averaged over the entire sample.
That encoding is SLIGHTLY better ff123, and uses a higher bitrate, so it's obvious why.
I couldn't tell what the average bitrate the Lame-encoded file was, so I couldn't match up the Fraunhofer VBR encode to match. If you tell me the Lame average bitrate, I could reduce the VBR quality to make the filesizes about the same.
I do know that with 256kbps CBR using LAME there's no way that any human can tell a difference between an original source and that mp3.
That's a pretty bold claim. There is a wide variation in human hearing sensitivity, and also in experience with hearing encoding artifacts. In some samples, mp3 can easily be distinguished from the original at 256 kbit/s.
ff123
ff123,
Your web site is amazing. This may be the million dollar question, but why does the sample123 sound better?
Your web site is amazing. This may be the million dollar question, but why does the sample123 sound better?
Let's take a stroll through town at 320 kbps! This file was actually resampled so it's nowhere near the original...
HOWEVER, the imaging is incredible! Listen to the sound effects in the background. Depending on your speaker setup, if you close your eyes, it appears that the sirens are EVERYWHERE!
Cheers!
HOWEVER, the imaging is incredible! Listen to the sound effects in the background. Depending on your speaker setup, if you close your eyes, it appears that the sirens are EVERYWHERE!
Cheers!
Hey thanks for the compliment on my site, Glen.
Why does sample123 sound better? Probably mostly because it had more bits to play with, like divinemartyr said. There's no guarantee it would sound better (or even as good as) Lame at the same average bitrate.
ff123
ff123's site: Discussion of Audio Compression
Why does sample123 sound better? Probably mostly because it had more bits to play with, like divinemartyr said. There's no guarantee it would sound better (or even as good as) Lame at the same average bitrate.
ff123
ff123's site: Discussion of Audio Compression
i have no idea whats the mp3 and whats the original, but i can say this:
AudioGrabber 1.70 release 2 using LAME 3.88 latest build do VBR perfectly!
Edit:
didnt read the entire thread. interesting. what software to you use doing that spectral analisys?
AudioGrabber 1.70 release 2 using LAME 3.88 latest build do VBR perfectly!
Edit:
didnt read the entire thread
. interesting. what software to you use doing that spectral analisys?i'm glad i can't tell the difference. and even if i could, i'd rather pretend that i can't. as far as i'm concerned, turning up the volume loud so that i can think i'm hearing a difference, so that i feel bad about all my mp3s, is counterproductive.
<< When will we see mp3 or a similar format support DD and DTS? >>
Well, since both are already "lossy" compression formats, I don't think we'll ever see these available as mp3. Also, the mp3 standard includes no support for more than 2 channels, and I would suspect adding such wupport would not be trivial.
<< Give an mp3 encoded at 256 using LAME to a mastering engineer who works with audio for a living, and ask him to tell you which is the source file, and I would bet that most of them can. >>
You know maybe you're right, however, I know you're familiar with r3mix.net, why don't you go click on his 'quality' icon and see the results of the test they did with 256kbps mp3's. They tested 300 audiophiles, with both B&W Nautilus 803 speakers and Sennheiser Orpheus Electrostatic Reference headphones, and not one of those people was able to consistently pick out the mp3 from the source material. That's a pretty bold test, with some of the best equipment in the world, why would an audio engineer be able to do it any better?
I stand by my claim.
dm
TRENDING THREADS
-
Discussion Zen 5 Speculation (EPYC Turin and Strix Point/Granite Ridge - Ryzen 9000)- Started by DisEnchantment
- Replies: 25K
-
Discussion Intel Meteor, Arrow, Lunar & Panther Lakes Discussion Threads
- Started by Tigerick
- Replies: 21K
-
Discussion Intel current and future Lakes & Rapids thread- Started by TheF34RChannel
- Replies: 23K
-
-
Facebook
Twitter