Can you hear the difference between a 128k and 320k mp3?

[cfranz]

Apparently yes, however, given that the results are 50/50, not really sure that it means anything.

 

[Brian]

Interesting article but, in reading it in its entirety I think it supports my position.

 

[belgianbrain]

In practice, ripping my own, 128 sounds like ass. I can live with 192, though.

I agree. 192 or above is the way to go.

 

[Netdewt]

Interesting article but, in reading it in its entirety I think it supports my position.

I understand having your position because you don't notice a difference. That is the position of the article as I understood it, and the reason lossy formats exist.

However, if you want the absolute best digital file possible, and still want to save some hard drive space versus uncompressed, lossless is still better than lossy.

I used the quote in the article to explain what lossless conversion is, not because of what the rest of the article says. The description there seemed to be more thorough than others I found.

 

[jas1203]

The voice was hard to tell, but about half way through it became obvious.

Listened with SoundBlaster X-Fi output to Pioneer VSX-405 to two Polk Monitor 50s

 

[vanquishfist]

One way I like to think of lossless files is like how a guitar fits in a formed case. Uncompressed is like a square box that the guitar will fit in, but has a lot of extra room; lossless compression is like the form fitting guitar case, the guitar still fits, but takes up much less space.

 

[Netdewt]

One way I like to think of lossless files is like how a guitar fits in a formed case. Uncompressed is like a square box that the guitar will fit in, but has a lot of extra room; lossless compression is like the form fitting guitar case, the guitar still fits, but takes up much less space.

I like that! What would lossy compression be? Smashing the guitar?

 

[Brent71]

A lossless file can not contain the same information as an uncompressed file as something has to be removed in some manner to compress the file even if it is a null.

I don't think you have a firm grip on your understanding of the lossless compression being used. It's simply shrinking the file size, not tossing data will never be seen (or heard) again. SACD uses file compression (http://en.wikipedia.org/wiki/Super_Audio_CD#DST), does that mean part of the music is being removed? No! When a lossless file is decoded properly it is bit-identical to the original.

 

[markc2]

My wife nailed the difference after listening to it only 3 times, thru a Macbook speakers. The voice is what got her, it had more dynamics to her. I of course agreed

A really cool test

Mark

 

[vanquishfist]

I like that! What would lossy compression be? Smashing the guitar?

Like trying to put an acoustic in an electric case.

 

[x_25]

Had the classical one. Picked it out after a few listens. It was the tambourine thing that gave it away. Although my Koss Pro-4A's are not terribly detailed. (would have been easier with the stax).

 

[Fletch]

In practice, ripping my own, 128 sounds like ass. I can live with 192, though.

I can easily hear the difference between 192 and 228. That is the lowest I will go. The 192K just does not sound right to my ears.

 

[ng0k]

I can't consistently tell the difference with my mid-fi ears. I've played iTunes stuff via the analog output through my stereo and it's good enough. Sure it can be better, but I'm not willing to jump through hoops for an incremental improvement.

 

[aidynphoenix]

i could hear it..

 

[RichPA]

There's a lot of variation in encodings even for the same bit rate, but 128 vs. 320 is easy for me to hear.

 

[ng0k]

Last night I ripped a CD from Vintage Flim and the BB's to iTunes at 128 and played the song Tricycle. Then I re-ripped it at 256 and played it again. I noticed the improvement, especially in the treble ranges. I also compared each version to my CDP as well. The 256 version held its own pretty well.

 

[Jonny Ramone]

I have the treble cranked on my stereo, so the lower bitrate actually sounded better- smoother and less tinny. But I've always gone for quality over size- I've never actually ripped ANYTHING to the mp3 format, as I would never consciously choose something that would give less quality than CD(other than analog).I have some files recorded on my hard drive at 192 khz for better processing and editing.

 

[georgerm]

ZIP files, so 20th century. Any digital music format losses some information compared to the original. While an algorithm may reconstruct missing information the in the file itself is not 100% complete. If it were then there would be no need for the algorithm. Even an analog recording of a musical instrument may not be lossless if the dynamic range is compressed and hence the original information is altered and the dynamic rage information is missing.

I feel the reference to zip files was only to show that compression can be performed on a file (be it an exe file or music file) and still retain 100% of the information of the original. Not 99.9% but truly 100%.

There are two types of data compression – destructive (lossy) and non-destructive (lossless). These two terms are not approximations. When a format is touted as lossless, they don’t mean kind-of lossless, it means that it retains ALL of the information of the original, and that an original can be reproduced from it.

For example, un-zipping an exe file, or any file for that matter, will un-zip to its original file size and information, or it wouldn’t be useable. Could you imagine what MS word might do to a document that was zipped, but upon unzipping had some data missing or wrong? Word would not know what to do with it. In fact a zip file will retain the exact same CRC information (if you're into that) – each and every time. File out = File in. This type of compression does nothing irreversible to the data itself (non-destructive), and actually knows how to reassemble it back to its original format. This is by definition a “loss – less” compression.

Not so with the compression of say an image file to gif or jpg (or a music file to MP3). Once compressed, it has gone through compression algorithms that make certain assumptions about the quality of the output it’s going to produce, making the file smaller, but at the same time removing data (destructive) - forever. There is no way to revert back or ‘unzip’ a jpg file to attain the original picture and/or data the original picture may have had – same with an mp3, data has been removed. This is by definition a lossy compression.

Anyway – FWIW…

 

[Netdewt]

This is by definition a “loss – less” compression.

Not so with the compression of say an image file to gif or jpg (or a music file to MP3).

Good post, however I have add this nearly useless detail. I work with digital images day in and day out.

GIF actually uses a lossless compression, it just filters out the majority of colors to save space (called indexed color). So, it does lose information in the end, but it is considered a lossless compression. PNG uses a lossless compression too (filter+compression). It just does a much better job at producing an accurate image than GIF does.

Interesting side-by-side image JPG-PNG here: http://en.wikipedia.org/wiki/Portable_Network_Graphics#Comparison_with_other_file_formats

I don't understand sound files well enough to know an analogy that works here.

 

[georgerm]

GIF actually uses a lossless compression...

Thanks Netdewt, I do stand corrected. I should not have linked GIF and JPG as the same lossy format. The link was informative, and not being real hands on with GIF files, I've always assumed they were JPG’ish. Image formats that I deal with most are NEF and JPG - NEF because they are lossless, and JPG because the format is widely used. You just can't share NEF files like you can JPG....