Linked by Thom Holwerda on Fri 28th Oct 2011 20:42 UTC, submitted by poundsmack
Apple Apple has open sourced ALAC. "The Apple Lossless Audio Codec is an audio codec developed by Apple and supported on iPhone, iPad, most iPods, Mac and iTunes. ALAC is a data compression method which reduces the size of audio files with no loss of information. A decoded ALAC stream is bit-for-bit identical to the original uncompressed audio file. The Apple Lossless Audio Codec project contains the sources for the ALAC encoder and decoder. Also included is an example command line utility, called alacconvert, to read and write audio data to/from Core Audio Format and WAVE files. A description of a 'magic cookie' for use with files based on the ISO base media file format (e.g. MP4 and M4A) is included as well. The Apple Lossless Audio Codec sources are available under the Apache license."
Permalink for comment 495935
To read all comments associated with this story, please click here.
RE[5]: 24bit music
by zima on Fri 4th Nov 2011 23:12 UTC in reply to "RE[4]: 24bit music"
Member since:

I don't believe that.
I do however listened to some guitar music in 24 bit flac files.
There isn't a huge difference but i do hear it, especially on the guitar.

The scale of assumed (especially when you defend some of them) differences is mostly irrelevant - if your brain "knows" what "should" be better, it WILL be perceived as better.

It just works differently in such cases, our brains are very powerful like that; those are trivial tricks for them, for example:

Your hugely complex personal neural network didn't evolve, over the last half a billion years, for accurate rating of minute differences in stimuli - it evolved for survival (which included, say, being over-sensitive to stimuli which you anticipate - even if that sound you thought you hear was NOT a predator in 99+% of the cases, you were still better at passing your genes if you assumed that)

Go through a list of cognitive biases - this is our primary mode of operation. We largely make it all up as we go (it needed to be only "good enough" within the limited energy & time available for processing the surrounding world)

Consider: split-brain patients are virtually unchanged, while we very much held the feeling of "monolithic me" - this is the (poor) level of grasp we have about our minds. There is also one localised brain trauma which results in people becoming completely blind without them realizing that

ABX is one of the methodologies we developed to deal with such biases. And if you have any doubt those methodologies work, just look at our modern, technical civilisation - brought to you by the scientific method.

(heck, if you claim you are immune to those effects, and can make accurate judgements without the need for methods like ABX, you should perhaps contact some academic neurologists - since there would be something very weird and curious about your brain ;) ...but don't even think about doing that before performing proper ABX yourself, since this is the real test of such abilities - without first doing it yourself, you'd be virtually certainly just wasting their time)

You sound like someone that knows more than me on these stuff. I'm not saying this sarcastically.

You don't seem to act very much on this impression of yours in few other posts...

I would have given you the benefit of the doubt if I had not spoken to musicians and studio people in the last few years that record their music in that format and attest the sound difference. Not so much the HZ but the jump from 16bit to 24bit. Like i said before you would have to listen to it through something like studio monitors or headphones and maybe a class A or A/B amplifier will also help.

Proper ABX test is what would help.

That's especially true with people such invested in this as musicians (most of them lose a large part of hearing over the years BTW, often a "simple" loudness damage) or studio people - heck, most of them aren't about accurate reproductions, but into what is technically noise - flavours of distortion they like.

And/or you confuse music recording and editing with music listening, they are two different things - yes, 24 bit is quite useful in the former, it eases (it's not impossible with 16 bit, just much, much harder - so why bother?) avoiding errors and undesirable distortions. But for playback, 24 bit offers little, if anything, above 16 bit (notice I am not stating this in absolute way, scientific approach can't prove a negative)

Edited 2011-11-04 23:16 UTC

Reply Parent Score: 2